Ek-Hscma-Sv-002 HSC Service Manual Dec89

HSC Service Manual
Order Number EK-HSCMA-SV-002
Digital Equipment Corporation

Maynard, Massachusetts
Second Edition, December 1989
The information in this document is subject to change without notice and should not be construed as a commitment
by Digital Equipment Corporation. Digital Equipment Corporation assumes no responsibility for any errors that may
appear in this document.
The software described in this document is furnished under a license and may be used or copied only in
accordance with the terms of such license.
No responsibility is assumed for the use or reliability of software on equipment that is not supplied by Digital
Equipment Corporation or its affiliated companies.
Restricted Rights: Use, duplication, or disclosure by the U.S. Government is subject to restrictions as set forth in
subparagraph (c){1){ii) of the Rights in Technical Data and Computer Software clause at DFARS 252.227-7013.
Copyright ©1989 by Digital Equipment Corporation
All Rights Reserved.

Printed in U.S.A.
The following are trademarks of Digital Equipment Corporation:

DEC DIBOL UNIBUS
DEC/CMS EduSystem VAX
DEC/MMS lAS VAXcluster
DECnet MASSBUS VMS
DECsystem-10 PDP VT
DECSYSTEM-20 PDT
DECUS RSTS
DECwriter RSX ~DmDamDlM
HSC50 ©DEC 1983
Covered by one or more U.S. PAT. Nos.
4,475,212 4,434,487 4,413,339
4,468,035 4,543,626 4,592,072
4,241 ,399 4,338,663 4,349,871
4,450,572 and other patents pending
HSC70 ©DEC 1985
Covered by one or more U.S. PAT. Nos.
4,475,212 4,434,487 4,413,339
4,468,035 4,543,626 4,592,072
4,241 ,399 4,338,663 4,349,871
4,450,572 and other patents pending
This document was prepared using VAX DOCUMENT, Version 1.1

Contents
About This Manual xx:
1 General Information
1.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1
1.2 HSC Cabinet Layout. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-3
1.3 HSC50 Cabinet Layout. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-7
1.4 External Interfaces. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-11
1.5 HSC Hardware Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . .... .... .. . .. . . 1-13
1.5.1 Port Link Module (LINK) Functions. . . . . . . . . . . . . . . . . .... .... .. . .. . . 1-15
1.5.2 Port Buffer Module (PILA) Functions. . . . . . . . . . . . . . . . .... .... .. . .. . . 1-16
1.5.3 Port Processor Module (Kpli) Functions and Interfaces. . .... .... .. . .. . . 1-16
1.5.4 Disk Data Channel Module (K.sdi) Functions. . . . . . . . . . .... .... .. . .. . . 1-16
1.5.5 Tape Data Channel ModUle (Ksti) Functions . . . . . . . . . . .... .... .. . .. . . 1-17
1.5.6 Data Channel Module (Ksi) Functions. . . . . . . . . . . . . . . .... .... .. . .. . . 1-17
1.5.7 110 Control Processor Module (P.ioj/c) Functions. . . . . . . . .... .... .. . .. . . 1-17
1.5.8 Memory Module (M.std2) Functions . . . . . . . . . . . . . . . . . .... .... .. . .. . . 1-18
1.5.9 Memory Module (M.std) Functions . . . . . . . . . . . . . . . . . . .... .... .. . .. . . 1-19
1.6 HSC Software Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-20
1.7 HSC Maintenance Strategy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . 1-22
1.7.1 Maintenance Features .......................................... 1-22
1.8 Specifications. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-23
2 Controls and Indicators

2.1 Introduction ................................................... . 2-1
2.2 Operator Control Panel (OCP) ..................................... . 2-2
2.3 HSC Inside Front Controls and Indicators ............................ . 2-3
2.4 HSC50 Inside Front Door Controls and Indicators ...................... . 2-6
2.5 HSC50 Maintenance Access Panel Controls and Connectors ............... . 2-7
2.5.1 HSC50 dc Power Switch ........................................ . 2-8
2.5.2 HSC50 Maintenance Panel Connectors ............................. . 2-8
iii
iv Contents
2.6 Module Indicators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-8

2.7 Module Switches. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-11
2.8 881 Power Controller ............................................. 2-17
2.8.1 Operating Instructions .............. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-17
2.9 HSC50 Power Controller. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-19
2.9.1 Line Phase Indicators. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-20
2.9.2 Fuses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-21
2.9.3 Remote/Off/Local On Switch. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-21
2.9.4 Circuit Breakers (60 Hz) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 2-22
2.9.5 Circuit Breakers (50 Hz) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-22
2.9.6 Power Controller (60 Hz)--Rear View. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-22
2.9.7 Power Controller (50 Hz)--Rear View. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-23
3 Removal and Replacement Procedures

3.1 Introduction ................................................... . 3-1
3.2 Safety Precautions .............................................. . 3-2
3.3 Taking the HSC Off Line for Maintenance ............................ . 3-2
3.3.1 Single HSC in a Cluster and Clusters Running ULTRIXIUNIX .......... . 3-2
3.3.2 Multiple HSCs in a Cluster ...................................... . 3-2
3.4 Removing and Replacing Field Replaceable Units ....................... . 3--3
3.4.1 Removing HSC Power .......................................... . 3-3
3.4.2 Removing HSC50 Power ........................................ . 3-5
3.4.3 Removing Field Replaceable Units ................................ . 3--8
3.4.4 Removing the HSC Cabinet Front Door ............................ . 3-9
3.4.5 Removing the HSC50 Cabinet Front Door .......................... . 3-11
3.4.6 Removing the HSC Cabinet Back Door ............................. . 3-12
3.5 Removing and Replacing Modules. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-13
3.5.1 Removing and Replacing the Port Link Module (LINK) . . . . . . . . . . . . . . . . . 3-14
3.5.1.1 Removing the LINK Module . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-14
3.5.1.2 Setting the Replacement LINK Module Switches . . . . . . . . . . . . . . . . . . . . 3-15
3.5.1.3 Setting the Replacement LINK Module Jumpers ..... . . . . . . . . . . . . . . . 3-16
3.5.1.4 Replacing the LINK Module . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-20
3.5.1.5 Testing the LINK Module. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-20
3.5.2 Removing and Replacing Port Buffer Module (PILA) .... . . . . . . . . . . . . . . . 3-20
3.5.2.1 Removing the PILA Module ..................... . . . . . . . . . . . . . . . 3-20
3.5.2.2 Setting the Replacement PILA Module Switches .................... 3-21
3.5.2.3 Replacing the PILA Module .................................... 3-21
3.5.2.4 Testing the PILA Module ...................................... 3-21
3.5.3 Removing and Replacing the Port Processor Module (K.pli) . . . . . . . . . . . . . . 3-22
3.5.3.1 Removing the Kpli Module. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-22
3.5.3.2 Setting the Replacement K.pli Module Switches. . . . . . . . . . . . . . . . . . . . . 3-22
3.5.3.3 Replacing the Kpli Module. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-23
3.5.3.4 Testing the Kpli Module. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-23
Contents v
3.5.4 Removing and Replacing the Disk Data Channel Module (K.sdi) ......... . 3-23
3.5.4.1 Removing the Ksdi Module ................................... . 3-23
3.5.4.2 Replacing the Ksdi Module ................................... . 3-24
3.5.4.3 Testing the K.sdi Module ..................................... . 3-24
3.5.5 Tape Data Channel Module (Ksti) ................................ . 3-25
3.5.5.1 Removing the K.sti Module .................................... . 3-25
3.5.5.2 Replacing the Ksti Module .................................... . 3-25
3.5.5.3 Testing the K.sti Module ...................................... . 3-25
3.5.6 Removing and Replacing the Data Channel Module (Ksi) .............. . 3-26
3.5.6.1 Removing the K.si Module .................................... . 3-26
3.5.6.2 Setting the Replacement Ksi Module Switches .................... . 3-26
3.5.6.3 Configuration of Requestors While Replacing the Ksi Module ......... . 3-27
3.5.6.4 Replacing the K.si Module .................................... . 3-28
3.5.6.5 K.si Module External Loop Test ................................ . 3-29
3.5.6.6 Initializing the K.si Module ................................... . 3-30
3.5.6.7 Correcting K.si Module Configuration Problems .................... . 3-31
3.5.6.8 K.si Module New Boot Microcode ............................... . 3-32
3.5.6.9 Testing the K.si Module (After Initialization) ...................... . 3-33
3.5.7 Removing and Replacing the I/O Control Processor Module (P.ioj/c) ....... . 3-33
3.5.7.1 Removing the P.ioj/c Module ................................... . 3-33
3.5.7.2 Setting the Replacement P.ioj/c Module Jumpers ................... . 3-34
3.5.7.3 Replacing the P.iojlc Module ................................... . 3-34
3.5.7.4 Testing the P.ioj/c Module ..................................... . 3-35
3.5.8 Removing and~eplacing the HSC Memory Module (M.std2) ............ . 3-35
3.5.8.1 Removing ~e M.std2 Module .................................. . 3-35
3.5.8.2 Replacing the M.std2 Module ............•...................... 3-36
3.5.8.3 Testing the M.std2 Module .................................... . 3-36
3.5.9 Removing and Replacing the HSC50 Memory Module (M.std) ........... . 3-37
3.5.9.1 Removing the M.std Module ................................... . 3-37
3.5.9.2 Replacing the M.std Module ................................... . 3-37
3.5.9.3 Testing the M.std Module ..................................... . 3-38
3.6 Removing and Replacing Subunits .................................. . 3-38
3.6.1 Removing and Replacing the RX33 Disk Drive ....................... . 3-38
3.6.1.1 Removing the RX33 Disk Drive ................................ . 3-38
3.6.1.2 Setting the RX33 Disk Drive Jumpers ........................... . 3-41
3.6.1.3 Replacing the RX33 Disk Drive ................................. . 3-45
3.6.1.4 Testing the RX33 Disk Drive ................................... . 3-45
3.6.2 Removing and Replacing the TU58 Tape Drive .......... ' ............. . 3-45
3.6.2.1 Removing the TU58 Tape Drive ......... '.' ..................... . 3-45
3.6.2.2 Setting the TU58 Tape Drive Jumpers ........................... . 3-49
3.6.2.3 Replacing the TU58 Tape Drive ................................ . 3-50
3.6.2.4 Testing the TU58 Tape Drive .................................. . 3-51
3.6.3 Removing and Replacing the HSC Operator Control Panel (OCP) ........ . 3-51
3.6.3.1 Removing the HSC OCP ...................................... . 3-51
3.6.3.2 Replacing the HSC OCP ...................................... . 3-52
3.6.3.3 Testing the HSC OCP .......... : ............................. . 3-53
vi Contents
3.6.4 Removing and Replacing the HSC50 Operator Control Panel (OCP) ....... 3-53
3.6.4.1 Removing the HSC50 OCP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-53
3.6.4.2 Replacing the HSC50 OCP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-55
3.6.4.3 Testing the HSC50 OCP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-55
3.6.5 Removing and Replacing the HSC Airflow Sensor Assembly ............. 3-56
3.6.6 Removing and Replacing the HSC50 Airflow Sensor Assembly. . . . . . . . . . . . 3-58
3.6.7 Removing and Replacing the HSC Blower ................. . . . . . . . . . . 3-60
3.6.8 Removing and Replacing the HSC50 Blower. . . . . . . . . . . . . . . . . . . . . . . . . . 3-62
3.6.9 Removing and Replacing the 881 Power Controller .................... 3-64
3.6.10 Removing and Replacing the HSC50 Power Controller . . . . . . . . . . . . . . . . . . 3-67
3.6.11 Removing and Replacing the HSC Main Power Supply ................. 3-69
3.6.12 Removing and Replacing the HSC50 Main Power Supply. . . . . . . . . . . . . . . . 3-73
3.6.13 Removing and Replacing the HSC Auxiliary Power Supply . . . . . . . . . . . . . . 3-76
3.6.14 Removing and Replacing the HSC50 Auxiliary Power Supply ............ 3-79
4 Initialization Procedures
4.1 Introduction ................................................... . 4-1
4.2 Console!Auxiliary Terminal ........................................ . 4-1
4.2.1 Console Terminal Connection .................................... . 4-1
4.2.2 HSC50 Auxiliary and Maintenance Terminal Connections .............. . 4-2
4.2.3 LA12 Parameters ............................................. . 4-4
4.3 HSC Initialization ............................................... . 4-5
4.3.1 Init P.io Test (INIPIO) .......................................... . 4-7
4.3.2 INIPIO Test System Requirements ................................ . 4-7
4.3.3 INIPIO Test Prerequisites ....................................... . 4-7
4.3.4 INIPIO Test Operation ......................................... . 4-7
4.4 HSC50 Initialization ............................................. . 4-8
4.4.1 HSC50 Off-line Diagnostics Tape ................................. . 4-8
4.4.2 Init P.ioc Diagnostic ........................................... . 4-8
4.5 Fault Code Interpretation ......................................... . 4-9
5 Device Integrity Tests

5.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-1
5.1.1 Device Integrity Tests Common Areas .............................. 5-1
5.1.2 Generic Error Message Format. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-2
5.2 ILRX33 - RX33 Device Integrity Tests ............................... 5-2
5.2.1 System Requirements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-3
5.2.2 Operating Instructions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-3
5.2.3 Test Termination. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-3
5.2.4 Parameter Entry. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-3
5.2.5 Progress Reports. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-3
5.2.6 Test Summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-4
5.2.7 Error Message Example. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-4
Contents vii
5.2.8 Error Messages ............................................... . 5-4

5.3 ILTAPE-TU58 Device Integrity Test ................................. . 5-5
5.3.1 System Requirements .......................................... . 5-5
5.3.2 Operating Instructions ......................................... . 5-6
5.3.3 Test Termination .............................................. . 5-6
5.3.4 Error Messages ............................................... . 5-6
5.4 ILMEMY - Memory Integrity Tests ................................. . 5-7
5.4.4 Progress Reports .............................................. '. 5-8
5.4.5 Test Sum.maries .............................................. . 5-8
5.4.6 Error Message Example ........................................ . 5-8
5.4.7 Error Messages ............................................... . 5-9
5.5 ILDISK - DISK Drive Integrity Tests ............................... . 5-9
5.5.3 Availability .................................................. . 5-10
5.5.5 Parameter Entry .............................................. . 5-11
5.5.6 Specifying Requestor and Port ................................... . 5-12
5.5.7 Progress Reports .............................................. . 5-12
5.5.8 Test Sum.maries .............................................. . 5-12
5.5.10 Error Messages ............................................... . 5-14
5.5.11 MSCP Status Codes-ILDISK Error Reports ........................ . 5-22
5.6 ILTAPE - TAPE Device Integrity Tests .............................. . 5-23
5.6.3 User Dialog .................................................. . 5-24
5.6.4 User Sequences ............................................... . 5-27
5.6.6 Test Sum.maries .............................................. . 5-28
5.6.6.1 Interface Test Sum.mary ...................................... . 5-28
5.6.6.2 Formatter Test Summary ..................................... . 5-29
5.6.6.3 User Sequence Test Summary .................................. . 5-29
5.6.6.4 Canned Sequence Test Summary ............................... . 5-29
5.6.6.5 Streaming Sequence Test Summary ............................. . 5-29
5.6.8 Error Messages ............................................... . 5-30
5.7 ILTCOM - Tape Compatibility Test ................................. . 5-32
5.7.4 Parameter Entry .............................................. . 5-34
viii Contents
5.7.5 Test Summaries .............................................. . 5-36

5.7.7 Error Messages ............................................... . 5-36
5.8 ILEXER - Multidrive Exerciser .................................... . 5-37
5.8.4 Parameter Entry .............................................. . 5-39
5.8.5 Disk Drive Prompts ........................................... . 5-40
5.8.6 Tape Drive Prompts ........................................... . 5-42
5.8.7 Global Prompts ............................................... . 5-43
5.8.8 Data Patterns ................................................ . 5-44
5.8.9 Setting/Clearing Flags ......................................... . 5-46
5.8.11 Data Transfer Error Report ..................................... . 5-46
5.8.12 Performance Summary ......................................... . 5-46
5.8.13 Communications Error Report ................................... . 5-48
5.8.14 Test Summaries .............................................. . 5-48
5.8.15 Error Message Format ......................................... . 5-50
5.8.15.1 Prompt Error Format ........................................ . 5-50
5.8.15.2 Data Compare Error Format .................. ' ................. . 5-50
5.8.15.3 Pattern Word Error Format ................................... . 5-51
5.8.15.4 Communications Error Format ................................. . 5-51
5.8.16 Error Messages ............................................... . 5-52
5.8.16.1 Informa tional Messages ...................................... . 5-52
5.8.16.2 Generic Errors ............................................. . 5-52
5.8.16.3 Disk Errors ................................................ . 5-54
5.8.16.4 Tape Errors ............................. '................... . 5-55
6 Off-line Diagnostics
6.1 Introduction ................................................... . 6-1
6.1.1 Software Requirements ......................................... . 6-1
6.1.2 Off-line Diagnostics Load Procedure ............................... . 6-1
6.2 ROM Bootstrap ................................................. . 6-2
6.2.1 Initialization Instructions ....................................... . 6-2
6.2.2 Failures ..................................................... . 6-3
6.2.4 Error Information ............................................. . 6-4
6.2.5 Failure Troubleshooting ........................................ . 6-4
6.2.6 Bootstrap Test Summaries ...................................... . 6-4
6.2.7 Generic Error Message Format ................................... . 6-7
6.3 ODL-Off-line Diagnostics Loader .................................. . 6-7
6.3.1 Loader System Requirements .................................... . 6-8
6.3.2 Loader Prerequisites ........................................... . 6-8
Contents ix
6.3.3 Loader Operating Instructions ................................... . 6-8

6.3.4 Loader Commands ............................................ . 6-8
6.3.4.1 HELP Command ............................................ . 6-8
6.3.4.2 SIZE Command ............................................. . 6-9
6.3.4.3 TEST Command ............................................ . 6-9
6.3.4.4 LOAD Command ............................................ . 6-9
6.3.4.5 START Command ........................................... . 6-9
6.3.4.6 EXAMINE and DEPOSIT Commands ............................ . 6-10
6.3.4.7 EXAMINE and DEPOSIT Symbolic Addresses ..................... . 6-10
6.3.4.8 Repeating EXAMINE and DEPOSIT Commands ................... . 6-11
6.3.4.9 Relocation Register .......................................... . 6-11
6.3.4.10 EXAMINE and DEPOSIT Qualifiers (Switches) .................... . 6-12
6.3.4.11 Setting and Showing Defaults .................................. . 6-13
6.3.4.12 Executing INDIRECT Command Files ........................... . 6-13
6.3.5 Unexpected Traps and Interrupts ................................. . 6-13
6.3.5.1 Trap and Interrupt Vectors .................................... . 6-14
6.3.5.2 Help File .................................................. . 6-16
6.4 OFLCXT-Off-line Cache Test ...................................... . 6-16
6.4.4 Parameter Entry .............................................. . 6-17
6.4.6 Test Summaries .............................................. . 6-18
6.4.8 Error Messages ............................................... . 6-21
6.4.9 Test Troubleshooting ........................................... . 6-23
6.5 OBIT-Off-line Bus Interaction Test ................................. . 6-24
6.5.2 Off-line Bus Interaction Test Prerequisites .......................... . 6-24
6.5.5 Parameter Entry .............................................. . 6-25
6.5.7 Test Summary- ................................................ . 6-26
6.5.9 Requestor Error Summary ...................................... . 6-27
6.5.10 Memory Test Configuration ....................... ; .............. . 6-28
6.5.11 Error Messages ....................... '........................ . 6-28
6.6 OKTS -Off-line K Test Selector ..................................... . 6-30
6.6.4 Parameter Entry .............................................. . 6-31
x Contents
6.6.6 K.ci Path Status Information .................................... . 6-33

6.6.7 Test Swnmaries .............................................. . 6-33
6.6.8 Error Infonn.a tion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-35
6.6.9 Error Messages ............................................... . 6-35
6.7 OKPM-Off-line KIP Memory Test .................................. . &-41
6.7.1 System Requirements .......................................... . &-41
6.7.3 Test Termination .............................................. . &-42
6.7.4 Parameter Entry .............................................. . &-42
6.7.6 Parity Errors ................................................. . 6-43
6.7.7 Test Swnmaries .............................................. . 6-44
6.7.8 Error InforDlation ............................................. . 6-45
6.7.9 Requestor Error Summary ...................................... . 6-45
6.7.10 Error Messages ............................................... . 6-45
6.8 OMEM-Off-line Memory Test ..................................... . 6-52
6.8.4 Parameter Entry .............................................. . 6-52
6.8.6 Parity Errors ................................................. . 6-54
6.8.7 Test Summaries .............................................. . 6-54
6.8.8 Error InforDlation ............................................. . 6-55
6.8.9 Error Messages ............................................... . 6-56
6.9 OFLRXE-RX33 Off-line Exerciser .................................. . &-63
6.9.4 Parameter Entry .............................................. . 6-64
6.9.6 Test Swnmaries .............................................. . 6-65
6.9.7 Data Patterns ................................................ . 6-66
6.9.9 Error Messages ............................................... . 6-67
6.10 ORFT-Off-line Refresh Test ....................................... . 6-70
6.10.2 Opera ting Instructions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-70
6.10.4 Parameter Entry .............................................. . 6-71
6.10.5 Progress Reports .............................................. . ~71
6.10.6 Test Summaries .............................................. . ~71
6.10.7 Error Information ............................................. . ~72

6.10.8 Error Messages ............................................... . ~72
6.11 OOCP-Off-line Operator Control Panel (OCP) Test .................... . ~73

Contents xi
6.11.1 System Requirements .................... -...................... . 6-73

6.11.2 Opera ting Instructions .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-73
6.11.4 Parameter Entry .............................................. . 6-73
6.11.5 Test Summaries .............................................. . 6-75
6.11.7 Error Messages ............................................... . 6-77
6.11.8 Troubleshooting Registers and Displays through ODT ................. . 6-78
6.11.8.1 Switch Check through ODT ................................... . 6-78
6.11.8.2 Lamp Bit Check ............................................ . 6-79
6.11.8.3 SecurelEnable Switch Check ................................... . 6-80
6.11.8.4 State LED Check ........................................... . 6-81
7 Utilities
7.1 Introduction 7-1
7.2 DKUTIL - Off-line Disk Utility. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-1
7.2.1 Starting DKUTIL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-1
7.2.2 Command Syntax. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-2
7.2.3 Command Modifiers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-2
7.2.4 Command Descriptions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-3
7.2.4.1 Command Summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-3
7.2.4.2 DEFAULT Command ..................... . . . . . . . . . . . . . . . . . . . . 7-4
7.2.4.3 DISPLAY Command.. . . . . . . . . . . . . . .. .. . . . . . . . . . . . . . . . . . . . . . . . 7-5
7.2.4.4 DUMP Command ........................................... . 7-6
7.2.4.5 EXIT Command ................ ; ........................... . 7-7
7.2.4.6 GET Command ............................................. . 7-8
7.2.4.7 POP Command ............................................. . 7-8
7.2.4.8 PUSH Command ............................................ . 7-9
7.2.4.9 REVECTOR Command ....................................... . 7-9
7.2.4.10 SET Command ............................................. . 7-9
7.2.5 Sample Session ............................................... . 7-10
7.2.6 Error and Information Messages .................................. . 7-12
7.2.6.1 Error Message Variables ...................................... . 7-12
7.2.6.2 Error Message Severity Levels ................................. . 7-13
7.2.6.3 Fatal Error Messages ........................................ . 7-13
7.2.6.4 Error Messages ............................................. . 7-13
7.2.6.5 Information Messages. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 7-14
7.3 VERIFY - Off-line Disk Verifier Utility. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-15
7.3.1 Running VERIFY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-16
7.3.2 Sample Session. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-17
xii Contents
7.3.3 Error and information Messages. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-18

7.3.3.1 Variable Output Error Fields ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-18
7.3.3.2 Error Message Severity Levels. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-18
7.3.3.3 Fatal Error Messages. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-19
7.3.3.4 Warning Messages ........................................... 7-19
7.3.3.5 Information Messages. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-21
7.4 FORMAT - Off-line Disk Formatter Utility ........................... . 7-22
7.4.1 Running FORMAT ............................................ . 7-23
7.4.2 Sample Session ............................................... . 7-24
7.4.3.1 Error Message Variables ...................................... . 7-26
7.4.3.2 Message Severity Levels ...................................... . 7-26
7.4.3.4 Warning Message ........................................... . 7-27
7.4.3.5 Information Messages ........................................ . 7-27
7.4.3.6 Error Messages ............................................. . 7-28
7.4.3.7 Success Messages ........................................... . 7-28
7.5 PATCH - Off-line Load Media Modification Utility ..................... . 7-28
7.5.1 PATCH Commands ............................................ . 7-28
7.5.2 Running PATCH ................ '.............................. . 7-29
7.5.3 Sample Session ............................................... . 7-31
7.5.4.2 PATCH Error Messages ...................................... . 7-32
7.5.4.3 Warning Messages .......................................... . 7-33
7.5.4.4 Informational Messages ...................................... . 7-33
7.5.4.5 Success Messages ........................................... . 7-33
8 Troubleshooting Techniques
8.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-1
8.2 How To Use This Chapter. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-1
8.3 Initialization Error Indications ..................................... . 8-2
8.3.1 OCP Fault Code Displays ....................................... . 8-2
8.3.1.1 Fault Code Interpretation ..................................... . 8-3
8.3.2 Module LEOs ................................................ . 8-11
8.3.2.1 P.ioj/c LEDs ................................................ . 8-12
8.3.2.2 Power-up Sequence of I/O Control Processor LEOs .................. . 8-12
8.3.2.3 Memory Module LEOs ....................................... . 8-12
8.3.2.4 Data Channel LEOs ......................................... . 8-13
8.3.2.5 Host Interface LED .......................................... . 8-13
8.3.3 Communication Errors .......................................... . 8-14
8.3.4 Requestor Status for Nonfailing Requestors ......................... . 8-14
8.3.5 HSC Boot Flow and Troubleshooting Chart ......................... . 8-15
8.3.6 HSC50 Flow and Troubleshooting Chart ............................ . 8-21
Contents xiii
8.3.7 Boot Diagnostic Indications ...................................... . 8-26

8.4 Software Error Messages ......................................... . 8-26
8.4.1 Mass Storage Control Protocol Errors .............................. . 8-26
8.4.2 MSCPtrMSCP Error Format, Description, and Flags .................. . 8-26
8.4.2.1 Error FOrIllat .............................................. . 8-27
8.4.2.2 Error Message Fields ........................................ . 8-27
8.4.2.3 Format 'JYpe Codes .......................................... . 8-27
8.4.2.4 Error Flags ................................................ . 8-28
8.4.2.5 Controller Errors ............................................ . 8-29
8.4.2.6 MSCP SDI Errors ........................................... . 8-30
8.4.2.7 Disk Transfer Errors ......................................... . 8-35
8.4.3 Bad Block Replacement Errors (BBR) .............................. . 8-38
8.4.4 TMSCP Errors ............................................... . 8-39
8.4.4.1 STI Communication or Command Errors ......................... . 8-40
8.4.4.2 STI Formatter Error Log ..................................... . 8-40
8.4.4.3 STI Drive Error Log ......................................... . 8-42
8.4.4.4 Breakdown of GEDS Text Field ................................ . 8-45
8.4.4.5 Breakdown of GSS Text Field .................................. . 8-46
8.4.4.6 GSS Text Field Bit Interpretation ............................... . 8-46
8.4.5 Out-of-Band Errors ............................................ . 8-49
8.4.5.1 RX.33 Errors ............................................... . 8-50
8.4.5.2 Disk Functional Errors ....................................... . 8-52
8.4.5.3 Ta pe Functional Errors ....................................... . 8-52
8.4.5.4 Miscellaneous Errors ......................................... . 8-52
8.4.6 Traps ....................................................... . 8-53
8.4.6.1 NXM (Trap through 4) ....................................... . 8-53
8.4.6.2 Reserved Instruction (Trap through 10) .......................... . 8-53
8.4.6.3 Parity Error (Trap through 114) ................................ . 8-53
8.4.6.4 Level 7 K Interrupt (Trap through 134) .......................... . 8-54
8.4.6.5 Control Bus Error Conditions (Hardware-Detected) ................. . 8-54
8.4.6.6 Level 7 K Interrupt Example .................................. . 8-54
8.4.6.7 MMU (Trap through 250) ..................................... . 8-56
8.5 Alphabetical Listing of Software Error Messages ....................... . 8-58
A Internal Cabling Diagrams

A.1 Introduction....... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-I
A.2 HSC Internal Cabling. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-I
A.3 HSC50 Internal Cabling ................... '.' . . . . . . . . . . . . . . . . . . . . . . A-7
A.4 HSC50 (Modified) Internal Cabling. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. A-14
xiv Contents
B Exception Codes and Messages

B.1 Crash Dump Printout. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-1
B.2 SINI-E Error Printout. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-2
B.3 Submitting a Software Performance Report. . . . . . . . . . . . . . . . . . . . . . . . . . . . B-2
B.4 Exception Messages .............................................. B-4
C Generic Error Log Fields

C.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C-1
C.2 Error flags. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C-2
C.3 MSCPITMSCP Status or Event Codes ................................ C-2
D Interpretation of Status Code Bytes

D.1 Introduction.................. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D-1
D.2 K-Detected Error Example Examination. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D-2
D.3 K-Detected Failure Code Analysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D-3
E Revision Matrix Charts

E.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E-1
E.2 HSC Revision Matrix Chart ........... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E-1
E.3 HSC50 (Modified) Revision Matrix Chart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E-6
E.4 HSC50 Revision Matrix Chart E-11
Examples
6-1 Example HELP file display ........................................ . 6-16
6-2 Off-line RX33 Exerciser Data Patterns ............................... . 6-67
7-1 Example Patch of a File .......................................... . 7-31
&-1 MSCPITMSCP Error Message Format ............................... . &-27
&-2 Controller Error Message Example .................................. . &-29
&-3 MSCP SDI Error Example ........................................ . 8-30
8-4 Disk Transfer Error Example ...................................... . 8-35
&-5 Bad Block Replacement Error Example .............................. . 8-38
8-6 STI Communication or Command Error Example ....................... . 8-40
&-7 STI Formatter Error Log Example .................................. . 8-41
S-8 STI Drive Error Log Example ...................................... . 8-42
&-9 Tape Drive Related Error Message .................................. . 8-45
&-10 Additional Tape Drive-Related Error Message ......................... . 8-46
B-1 Crash Dump Example ............................................ . B-1
B-2 SINI-E Exception Code ........................................... . B-2
C-1 Error Log Fields Example ......................................... . C-1
Contents xv
D-1 K-detected Error Example. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D-1
Figures
1-1 Redundant Cluster Configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2
1-2 HSC Cabinet Front View .......................................... 1-3
1-3 HSC Cabinet Inside Front View. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1--4
1--4 HSC Module Utilization Label Example. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-5
1-5 HSC Cabinet Inside Rear View. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-6
1-6 HSC50 Cabinet Front View. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-7
1-7 HSC50 Cabinet Inside Front View. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-8
1-8 HSC50 Module Utilization Label Example. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-9
1-9 HSC50 Cabinet Rear View. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-10
1-10 HSC50 Cabinet Inside Rear View ................ : . . . . . . . . . . . . . . . . . . . 1-11
1-11 HSC External Interfaces. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-12
1-12 HSC50 External Interfaces. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-12
1-13 HSC Subsystem Block Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-14
1-14 Memory IYlap (M.std2-LOl17) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-19
1-15 Memory Map (M.std-L0106) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-20
1-16 HSC Internal Software. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-21
1-17 HSC Specifications ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-24
2-1 Operator Control Panel. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-2
2-2 ControlslIndicators Inside Front Door. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2--4
2-3 RX33 and dc Power Switch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-5
2--4 HSC50 ControlslIndicators Inside Front Door. . . . . . . . . . . . . . . . . . . . . . . . . . . 2-6
2-5 HSC50 Maintenance Access Panel ................................... 2-7
2-6 Module LED Indicators. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-9
2-7 HSC Module Utilization Label Example. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-10
2-8 LOl18 Module (DIP) Switches. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-13
2-9 L0107 Module (DIP) Switches. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-14
2-10 L0109 Module (DIP) Switches. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-15
2-11 K.si Module (LOl19-YA) Switches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-16
2-12 881 Power Controller-Front Panel Controls ........................ . . . 2-18
2-13 881 Rear Panel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-19
2-14 HSC50 Power Controller (60 Hz)-Front View. . . . . . . . . . . . . . . . . . . . . . . . . . 2-20
2-15 HSC50 Power Controller (50 Hz)-Front View. . . . . . . . . . . . . . . . . . . . . . . . . . 2-21
2-16 HSC50 Power Controller (60 Hz)-Rear View. . . . . . . . . . . . . . . . . . . . . . . . . . . 2-23
2-17 HSC50 Power Controller (50 Hz)-Rear View. . . . . . . . . . . . . . . . . . . . . . . . . . . 2-24
3-1 HSC DC Power Switch. . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3--4
3-2 HSC 881 Power Controller Circuit Breaker ............................ 3-5
3-3 HSC50 DC Power Switch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-6
3--4 HSC50 Line Power Circuit Breakers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-7
3-5 HSC FRU Removal Sequence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-8
3-6 HSC50 FRU Removal Sequence ....... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-9
3-7 HSC OCP SignallPower Line Connector. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-10
3-8 HSC50 Maintenance Access Panel Connectors. . . . . . . . . . . . . . . . . . . . . . . . . . 3-12
3-9 HSC Card Cage Cover Removal. . . . . . . . . . .. . . . . . . . . . . . .. . . . . . . . . . . . . 3-13
xvi Contents
3-10 HSC50 Card Cage Cover Removal ................................... 3-14

3-11 L0100 Node Address Switches ...................................... 3-15
3-12 L0100-E21L0118 Node Address Switches. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-16
3-13 L0100 Jumper Configuration ....................................... 3-17
3-14 L0118-B1 Jumper Configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-18
3-15 L0118-B2 Jumper Configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-19
3-16 L0109 Hardware Rev Level Switch. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-21
3-17 L0107 Hardware Rev Level Switch. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-22
3-18 K.si Switchpack ................................................. 3-27
3-19 L0105 Baud Rate Jumper. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-34
3-20 HSC DC Power Switch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-39
3-21 Removing the RX33 Cover Plate. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-40
3-22 RX33 Disk Drive Removal. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-41
3-23 Revision Al Jumper Configurations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-43
3-24 Revision A3 Jumper Configurations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-44
3-25 HSC50 DC Power Switch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-46
3-26 Removing the HSC50 TU58 Bezel Assembly. . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-47
3-27 Disconnecting the HSC50 OCP Cables .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-48
3-28 Disconnecting the HSC50 TU58 Controller Cables . . . . . . . . . . . . . . . . . . . . . . . 3-49
3-29 TU58 Baud Rate Jumpers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-50
3-30 Removing the HSC OCP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-52
3-31 Removing the HSC50 OCP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-54
3-32 881 Power Controller Circuit Breaker. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-56
3-33 Removing and Replacing the HSC Airflow Sensor Assembly ............... 3-57
3-34 HSC50 Power Controller Circuit Breaker. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-58
3-35 Removing and Replacing the HSC50 Airflow Sensor Assembly. . . . . . . . . . . . . . 3-59
3-37 Removing and Replacing the HSC Main Cooling Blower .................. 3-61
3-38 HSC50 Power Controller Circuit Breaker . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-62
3-39 Removing and Replacing the HSC50 Blower. . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-63
3-41 Removing and Replacing the 881 Power Controller .............. ;....... 3-65
3-42 881 Total Off Connector ........................................... 3-66
3-43 HSC50 Power Controller Circuit Breaker . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-67
3-44 Removing and Replacing the HSC50 Power Controller . . . . . . . . . . . . . . . . . . . . 3-68
3-46 HSC Main Power Supply Cables and Test Points. . . . . . . . . . . . . . . . . . . . . . . . 3-71
3-47 Removing and Replacing the HSC70 Main Power Supply. . . . . . . . . . . . . . . . . . 3-72
3-49 HSC50 Main Power Supply Cables and Voltage Test Points . . . . . . . . . . . . . . . . 3-75
3-50 Removing and Replacing the HSC50 Main Power Supply. . . . . . . . . . . . . . .. .. 3-76
3-52 HSC Auxiliary Power Supply Cable and Test Points. . . . . . . . . . . . . . . . . . . . .. 3-78
3-53 Removing and Replacing the HSC Auxiliary Power Supply . . . . . . . . . . . . . . . . 3-79
3-55 HSC50 Auxiliary Power Supply Cable and Voltage Test Points ............. 3-81
Contents xvii
3-56 Removing and Replacing the HSC50 Auxiliary Power Supply .............. 3-82
4-1 Console Terminal Connection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-2
4-2 Auxiliary or Maintenance Terminal Connection . . . . . . . . . . . . . . . . . . . . . . . . . 4-3
4-3 Operator Control Panel Fault Code Displays ........................... 4-6
6-1 P.ioj Switch Display Register Layout. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-79
6-2 P.ioj Control and Status Register Layout .............................. 6-81
8-1 Operator Control Panel Fault Codes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-3
8-2 OCP Fault Code 1 . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . 8-4
8-3 OCP Fault Code 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-4
8-4 OCP Fault Code 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-5
8-5 OCP Fault Code 10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-5
8-6 OCP Fault Code 11 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-5
8-7 OCP Fault Code 12 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-6
8-8 OCP Fault Code 21 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-6
8-9 OCP Fault Code 22 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-6
8-10 OCP Fault Code 23 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-7
8-11 OCP Fault Code 25 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-7
8-12 OCP Fault Code 26 . . . . . . . . . . . . . . . . . . . . . . . .. . . . . .. . . . . . . . . . . . . . . . . 8-8
8-13 OCP Fault Code 30 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-8
8-14 OCP Fault Code 31 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-9
8-15 OCP Fault Code 32 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-11
8-16 OCP Fault Code 33 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-11
8-17 HSC Boot Flow and Troubleshooting Chart ............................ 8-16
8-18 HSC50 Boot Flow and Troubleshooting Chart. . . . . . . . . . . . . . . . . . . . . . . . . . . 8-22
8-19 Request Byte Field . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-32
8-20 Mode Byte Field . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-33
8-21 Error Byte Field . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-33
8-22 Controller Byte Field ............................................. 8-34
8-23 GSS Text Field Bits Summary Breakdown. . . . . .. . . . . . . . . . . . . .. . . . . . . . . 8-46
8-24 RX33 Floppy Controller CSR Breakdown. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-51
8-25 RX33 Error Message Last Line Breakdown ............................ 8-52
8-26 MMSRO Bit Breakdown ........................................... 8-57
A-l HSC Internal Cabling. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-2
A-2 HSC50 Internal Cabling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-8
A-3 HSC50 (Modified) Internal Cabling. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. A-15
E-l HSC Revision Matrix Chart .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E-2
E-2 HSC50 (Modified) Revision Matrix Chart. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E-7
E-3 HSC50 Revision Matrix Chart. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. E-12
Tables
1-1 Differences Between HSC Models. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1
1-2 HSC Module Names . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-14
2-1 Functions of Logic Module LEDs .......... :......................... 2-10
3-1 Ksi Switchpack Options. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-27
3-2 Physical Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-28
3-3 Ksi New Microcode Load Conditions ....... . . . . . . . . . . . . . . . . . . . . . . . . . . 3-32
xviii Contents
3-4 RX33 Jumper Description. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-42

5-1 ILTAPE Test Levels .............................................. 5-30
5-2 ILTCOM Header Record . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-32
5-3 ILTCOM Data Patterns. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-33
6-1 RX33 Error Table. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . 6-5
6-2 RX33 Error Code Table . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-6
6-3 Trap and Interrupt Vectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-15
7-1 PATCH Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-28
8-1 UPAR Register Addresses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-9
8-2 Control Program Bits ...... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-9
8-3 Status of Requestors for Level 7 Interrupt. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-10
8--4P.ioj/c LEDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-12
8-5 M.std2 and M.std LEDs ......................................... . . 8-13
8-6 K.sdiJK.sti and K.si LEDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-13
8-7 K.ci (LINK, PI LA, K.pli) LEDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-13
8-8 MSCPtrMSCP Error Message Field Descriptions . . . . . . . . . . . . . . . . . . . . . . . . 8--27
8-9 MSCPtrMSCP Error Message Format Type Code Numbers. . . . . . . . . . . . . . . . 8-28
8-10MSCPtrMSCP Error Flags. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-28
8-11MSCPtrMSCP Controller Error Message Field Descriptions ............... 8--29
8-12MSCP SDI Error Field Descriptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-30
8-13Request Byte Field Descriptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-32
8-14Mode Byte Field Descriptions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-33
8-15Error Byte Field Descriptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-33
8-16Controller Byte Field Descriptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-35
8-17Disk Transfer Error Field Descriptions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-36
8-18Original Error Flags Field Descriptions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-37
8--19
Recovery Flags Field Definitions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-37
8-20Bad Block Replacement Error Field Definitions. . . . . . . . . . . . . . . . . . . . . . . . . 8-38
8-21Replace Flags Field Bit Descriptions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-39
8-22STI Communication or Command Error Printout Field Descriptions. . . . . . . . . 8--40
8-23STI Formatter Error Log Field Descriptions. . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-42
8-24STI Formatter E Log ............................................. 8-42
8-25STI Drive Error Log Field Descriptions ............................... 8-43
8-26GEDS Text . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-43
8-27STI Drive Error Log (TA78 Drive Product Specific) ...................... 8-43
8-28Status Register Summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-51
B-1 Obtaining Data Structure Information. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-3
C-1 Generic Error Log Fields. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C-1
C-2 Error Flags . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C-2
C-3 MSCPtrMSCP Status or Event Codes ................................ C-2
D-1 K.ci Status Code Bytes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D-4
D-2 K.sdi Status Code Bytes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D-7
D-3 K.sti Status Code Bytes ........................................... D-10
D-4 K.si Disk Status Code Bytes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. D-12
D-5 K.si Tape Status Code Bytes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. D-13
About This Manual
This manual contains servicing information and procedures for the HSC70, HSC50 (modified),
HSC50, and HSC40 subsystems. In this manual, HSC refers to the HSC70 and HSC40 models.
HSC50 refers to the HSC50 and HSC50 (modified) models. Individual model names are used only
when the information is model specific.
This manual describes HSC controls and indicators, error reporting, field replaceable
units, troubleshooting, and diagnostic procedures. All information in this manual is
informational/instructional and is designed to assist service personnel with HSC maintenance.
Operational theory is included wherever such background is helpful to service personnel.
Installation procedures, most HSC utilities, and detailed technical descriptions are not included in
this manual. For source material on these and other subjects not within the scope of this manual,
refer to the list of related documentation below.
Audience
This manual is intended for use by Level 1 Digital Field Service Engineers and other personnel in
maintaining the components of the HSC controller subsystem.
Scope
This manual is divided into the following chapters:
1. General Information
2. Controls and Indicators
3. Removal and Replacement Procedures
4. Initialization Procedures
5. Device Integrity Tests
6. Offline Diagnostics
7. Utilities
8. Troubleshooting Techniques
9. Appendixes:
A. Internal Cabling Diagrams
B. Exception Codes and Messages
C. Generic Error Log Fields
D. Interpretation of Status Code Bytes
E. Revision Matrix Charts
xix
About This Manual xx
Related Documentation
Documents related to the HSC are available under the following titles and part numbers:
• HSC User Guide (AA-GMEAA-TK)
• HSC Installation Manual (EK-HSCMN-IN)
• HSC70 illustrated Parts Breakdown (EK-HSC70-IP)
• HSC50 Illustrated Parts Breakdown (EK-HSC50-IP)
• HSC50 Device Integrity Tests User Documentation (EK-IHSC5-UG)
• HSC50 Offline Diagnostics User Documentation (EK-OHS-UG)
• HSC50 Utilities User Documentation (EK-UHSC5-UG)
• VT320 Owners Manual (EK-VT320-UG)
• VT320 Programmer Pocket Guide (EK-VT320-HR)
• VT320 Installation Guide (EK-VT320-IN)
• VT220 Owners Manual (EK-VT220-UG)
• VT220 Programmer Pocket Guide (EK-VT220-HR)
• VT220 Installation Guide (EK-VT220-IN)
• Installing and Using the LA50 Printer (EK-OLA50-UG)
• LA50 Printer Programmer Reference Manual (EK-OLA50-RM)
• Installing and Using the LA75 Printer (EK-OLA75-UG)
• LA75 Printer Programmer Reference Manual (EK-OLA75-RM)
• Star Coupler User Guide (EK-SCOOS-UG)
• CI7BO User Guide (EK-CI7S0-UG)
• DECwriter Correspondent Technical Manual (EK-CPL12-TM)
• TU5B DECtape II User Guide (EK-OTU5S-UG)
These documents (except for the HSC User Guide) can be ordered from Publication and Circulation
Services, 10 Forbes Road, Northboro, Massachusetts 01532 (RCS code: NR12; mail code: NR031W3).
The HSC User Guide can be ordered from the Software Distribution Center, Digital Equipment
Corporation, Northboro, Massachusetts 01532.
NOTE
Please consult the HSC Software Release Notes for the latest hardware revision levels.
General Information 1-1
1
General Information
1.1 Introduction
This chapter includes general information about the Hierarchical Storage Controllers (HSC) mass
storage server, including:
• Cabinet layout
• Software overview
• Subsystem block diagram
• Module descriptions
• Maintenance features
• Specifications
NOTE
In this manual "HSC" refers to the HSC70 and HSC40 models. "HSC50" refers to the
HSC50 and HSC50 (modified) models. Individual model names are used only when the
information is model-specific.
Table 1-1 shows the major differences between the various HSC models. Note that the HSC70
supports a combination of eight disk and tape data channels, the HSC50 supports a combination of
six disk and tape data channels, and the HSC40 supports a combination of three disk and tape data
channels.
Each disk data channel supports four drives over the standard disk interface (SDI). Each tape
data channel supports four tape formatters over the standard tape interface CSTI). Depending upon
which formatter is used, from one to four tape transports can be supported by each formatter.
Table 1-1 Differences Between HSC Models

HSC50
HSC Contents HSC70 (Modifi.ed) HSC50 HSC40
I/O control processor LOllI L0105 LOI05 L0111-YA

Memory LOl17 L0105 10105 L0117
Number of data channels (disk + 8 6 6 3
tape)
Load devices RX33 TU58 TU58 RX33
Power controller 30-24374 30-24374 70--19122 30-24374
Auxiliary power supply Yes Yes No No
Kit number HSC7X-AA1AB is available to upgrade the HSC40 to an HSC70.
1-1
1-2 General Information
The HSC controller subsystem can interface with multiple hosts using the computer interconnect
(CI) bus. One CI bus is included with the subsystem. In case of bus failure, each CI bus consists
of two paths (path A and path B). See Figure 1-1 for a sample five-node cluster configuration with
two HSCs and three host computers. In this figure, all three hosts access both HSCs over the CI
bus. Through dual-porting, both HSCs can access the tape formatter and the disks.
HOST
HOST
HOST
TERMINAL* TERMINAL*
HSC HSC
PRINTER PRINTER
I
,;")/u.
CI INTERFACE
* VIDEO OR LA12
CXO-88SB
Figure 1-1 Redundant Cluster Configuration

1.2 HSC Cabinet Layout

HSC logic and power systems are housed in a modified H9642 cross-products cabinet with both
front and rear access. Figure 1-2 shows a front view of the cabinet.
- .. -
Figure 1-2 HSC Cabinet Front View

The front of the HSC cabinet contains the operator control panel (OCP) switches and indicators.
Switch operation and indicator functions are described in Chapter 2.
To access the cabinet interior, open the front door with a key. The door key is part of the door-lock
mechanism (part number 12-25411-01). Figure 1-3 shows the HSC cabinet with the front door
open.
The upper right-hand portion of the cabinet houses two RX33 dual drives and connectors for the
OCP.
The HSC70 contains two power supplies. The HSC40 contains one power supply. The power
supplies are housed under the RX33 drives. Each power supply has a fan drawing air from the
front of the cabinet across the power unit and exhausting it through a rear duct.
AUXILIARY
POWER
CARD SUPPLY
CAGE
Figure 1-3 HSC Cabinet Inside Front View

A 14-slot card cage with a corresponding backplane provides housing for the L-series extended hex
HSC logic modules. The card cage occupies the upper left corner of the cabinet. Above the card
cage is a module utilization label indicating the slot location of each module. Figure 1--4 shows a
typical HSC module utilization label. All unassigned slots in the backplane contain baffles.
(;
(/)
(/)
(I)
()
~
Q..
ei:
o
::,:0
(1)0
a:::::
Bkhd X
Req
10 9 8 7 6 5 4 3 2 1
CXO-889A
Figure 1-4 HSC Module Utilization Label Example
NOTE
Requester slots A, B, C, D, E, F, M, and N, illustrated in Figure 1-4, are optional tape
or disk data channels. Optional slot labels are blank when no module is present.
Appropriate labels are provided with each data channel option ordered.
Open the cabinet rear door with a 5/32-inch hex key. A rear view of the cabinet with the back
door open is shown in Figure 1-5. The backplane logic modules are cooled by a blower mounted
behind the card cage. Air is drawn in through the front door louver, up through the modules, and
exhausted through the larger duct at the rear.
1--6 General Information
_~"""":IJ..I-,-- BLOWER
BLOWER
OUTLET
DUCT
W~~I1~~~~~~---INTERNAL
CICABLES
EXTERNAL
CICABLES
EXTERNAL
SICABLES
CXO-890B
Figure 1-5 HSC Cabinet Inside Rear View
NOTE
Figure 1-5 shows the blower motor outlet duct for current models. Earlier models have a
smaller blower motor outlet duct.
Two levels of cable connections are found in the HSC: backplane to bulkhead and bulkhead to
outside the cabinet. All connections to the logic modules are made through the backplane. All
cables attach to the backplane with press-on connectors.
The power controller is located in the lower left-hand rear corner of the HSC. The power control
bus, delayed output line, and noise isolation filters are housed in the power controller.
Exterior CI, SDI, and STI buses are shielded up to the HSC cabling bulkhead. These cables are
attached to bulkhead connectors located at the bottom rear of the cabinet. From the interior of the
I/O bulkhead connectors, unshielded cables are routed to the backplane.
1.3 HSC50 Cabinet Layout

HSC50 logic and power systems are housed in a modified H9642 cross-products cabinet with both
front and rear access. Figure 1-6 shows the front view of the cabinet.
---
Figure 1-6 HSC50 Cabinet Front View

On the front of the cabinet are the OCP switches and indicators. Switch operation and indicator
functions are described in Chapter 2.
To access the cabinet interior, open the front door with a key (part number 12-14664). Figure 1-7
shows the inside front view of the HSC50. The back of the front door contains two TU58 drives and
slots for tape storage.
TU58 TU58
DRIVE 0 DRIVE 1
TU58
RUN LEOs
TU58 TAPE
CARTRIDGE
STORAGE
Figure 1-7 HSC50 Cabinet Inside Front View

A 14-slot card cage with a corresponding backplane provides housing for the HSC50 L-series
extended hex logic modules. The card cage and backplane occupy the lower left corner of the
cabinet. Above the card cage is a module utilization label indicating the slot location of each
module (Figure 1-8). All unassigned slots contain baffles.
o
rJ)
Q) Qj rJ)
c c CD
c c o
C\'l
.:. C\'l o
o .&:.
o 0:
<{ C\'l
>-
cO
as
0
;!>~
o CD ._
...JD:O
Bkhd X c D E F
Req 1 4 5 6
8 7 6 5 4 3 2 1
CXO-283B
Figure 1-8 HSC50 Module Utilization Label Example
NOTE
Requester slots A through F t as shown in Figure 1-8, are optional tape or disk data
channels. Optional slot labels are blank when no module is present. Appropriate labels
are provided with each data channel option ordered.
The upper right-hand portion of the cabinet houses the maintenance access panel. A dc power
on/off switch and connectors for the TU58, the OCp, and the maintenance terminal port are located
on this panel.
Power supply units are housed under the maintenance panel. A basic HSC50-AA/AB contains one
power supply capable of providing power for three data channels. A fourth data channel requires
the addition of an auxiliary power supply. Each power supply has a fan drawing air from the front
of the cabinet across the power unit and exhausting it through a rear duct. Figure 1-9 shows the
back of the HSC50 cabinet. The rear door is opened with a 5/32-inch hex key.
1-10 General Infonnation
=---~---
",,"11"11"""""""'"
""'' ' ' ' ' "',' ,'"'111111111""""",,,,,,'' ' ' ' ' ',',
,,,,,,,,,'111' "'111""""""",,""111
"""'' ' ' ' ',',',','"' "' "'11"""""",,,,,,,,'' ' ' ' '
""'"""""
"""""""""""""",
"""""""""""""'
"II"""""""""""'""
""""""",,"""",""
!!I!!!!!!!!!!!!!!!!!!!!!
BACK DOOR LATCH

(HEX LOCK)
CXO-004A
Figure 1-9 HSC50 Cabinet Rear View

Logic modules are cooled by a blower mounted behind the card cage as shown in Figure 1-10. Air is
drawn in through the front door louver, up through the modules, and exhausted through the middle
duct at the rear.
COOLING
BLOWER
POWER
CONTROLLER
EXTERNAL
AC LINE CABLES
CORD
Figure 1-10 HSC50 Cabinet Inside Rear View

Two levels of cable connections are found in the HSC50: backplane to bulkhead and bulkhead to
outside the cabinet. All connections to the logic modules are made through the backplane. All
cables attach with press-on connectors to the backplane.
The power controller is in the lower left-hand rear corner of the HSC50. Also at the rear of the
HSC50, the power control bus and delayed output line are connected to noise isolation filters.
Exterior CI, SDI, and STI buses are shielded up to the HSC50 cabling bulkhead. These cables are
attached to bulkhead connectors located at the bottom rear of the cabinet.
From the interior of the I/O bulkhead connectors, unshielded cables are routed to the backplane
and are attached with press-on connectors.
1.4 Externallnterfaces
Figure 1-11 shows the external hardware interface lines used by the HSC, and Figure 1-12 shows
the external hardware interface lines used by the HSC50.
CI BUS----ONE OR MORE HOST COMPUTERS

(4 CABLES: 2 PATH A, 2 PATH B)
SOl B U S - - - - DISK DRIVES

(ONE CABLE PER DISK DRIVE)
STI BUS - - - - TAPE FORMATTER

(ONE CABLE PER FORMATTER)
HSC70 A S C I I - - - - - CONSOLE TERMINAL

CONTROLLER SERIAL LINE (110 BULKHEAD J60)
ASCII - - - - - (NOT USED)

SERIAL LINE
A S C I I - - - - - (NOT USED)
SERIAL LINE
RX33 DISK DRIVE }

SIGNAL INTERFACE
BACKPLANE J18
RX33 DISK DRIVE
SIGNAL INTERFACE
CXO-928C
Figure 1-11 HSC External Interfaces
CI BUS----ONE OR MORE HOST COMPUTERS

(4 CABLES: 2 PATH A, 2 PATH B)
SOl B U S - - - - DISK DRIVES

(ONE CABLE PER DISK DRIVE)
STI BUS - - - - TAPE FORMATTER

HSC50 (ONE CABLE PER FORMATTER)
CONTROLLER
A S C I I - - - - - LOCAL TERMINAL
SERIAL LINE
ASCII-----TU58 CONTROLLER
SERIAL LINE
A S C I I - - - - - HAND-HELD TERMINAL
CXO-006C
Figure 1-12 HSC50 External Interfaces

The external hardware interface lines perform the following flUlctions:
Line Function
CI bus Four coaxial cables (BNCIA-xx): two-path (path A and path B) serial
bus with a transmit and receive cable in each path. This is the
communication path between system host(s) and the H8C.
8Dlbus Four shielded wires for serial communication between the H8C and
the disk drives (one SDI cable per drive per controller) (BC26V-xx).
8TI bus Four shielded wires for serial communication between the H8C and
the tape formatter (one 8TI cable per formatter) (BC26V-xx).
ASCII serial line R8-232-C cable for local console terminal communication with the I/O
control processor module.
ASCII serial line R8-232-C cable in the HSC50 to link the TU58 controller to the
cabinet.
RX33 disk drive signal interface Cable linking RX33 drives with the RX33 controller on the M.std2
module of the H8C.
ANSII hand held terminal R8-232-C cable for hand held terminal communications with the I/O
Control Processor module of the H8C50.
1.5 HSC Hardware Overview

The HSC is a multimicroprocessor subsystem with two shared memory structures: one for control
and one for data. In addition, the HSC I/O control processor fetches its own instructions from a
private (Program) memory. Figure 1-13 shows a subsystem block diagram. Each major block is a
module unless otherwise specified.
ASCII PORT SERIAL

r------------, _ LINE INTERFACE ...
HOST INTERFACE K.CI I INPUT/OUTPUT - - TERMINAL
PORT
-
CONTROL BUS
CONTROL
PROCESSOR --
PLI BUS
-
PROCESSOR
K.PLI L0107-YA
DATA BUS
...
P.IOC/LO 105
OR
P.IOJ/L0111
--
_ PROGRAM BUS
-
L OPERATOR
CONTROL
PANEL
---. TUS8
~
PORT
BUFFER ~
MEMORY
MODULE - DRIVE
M.STD/L010S
-. OR
RX33
PILA L0109 M.STD2IL0117
-- ...
DRIVE
I"""'It"
MAGTAPE r--
..... ..
~~
III BUS PORT TAPE DATA FORMATTER

...
- LINK
L0118
r----- CHANNEL
MODULE(S)
TA78, ETC
~
f4- TAPE
TRANSPORT
OR ~}OTHER TAPE
LINK L0100
- K.SI/L0119
OR
...... DRIVES
CI
,~
CI K.STI/L01 08-YB ...... TAPE
L--PATH - - - ~- -PATH .J TRANSPORT
A 'f B
,~--,
DISK DATA
CI PATH-A
- I SC008 --.. CHANNEL
MODULE(S)
: : .} OTHER DISK
DRIVES
---. TAPE
TRANSPORT
...l STAR
COUPLER
.. ~
CI PATH B ~ \ K.SI/L0119 ~
DISK DRIVE
OR TAPE
- K.SDI/LO 108-Y A
RA81 , RASO,
ETC
~
TRANSPORT
CXO-1929B
Figure 1-13 HSC Subsystem Block Diagram

References to logic modules by their engineering terms appear throughout HSC documentation, as
well as on diagnostic printouts. Refer to Table 1-2 for a cross-reference of HSC module names.
Table 1-2 HSC Module Names

Module Eugineering Module
Name Name Designation
Port link LINK LOIOO, Rev E2

or or
InterprocessorLINK LOllS
Interface (ILl)
Port buffer PILA LOI09
Port processor K.pli LOI07
Table 1-2 (Cont.) HSC Module Names

Module Engineering Module
Name Name Designation
Disk data channel K.sdi LOI08-YA

or or or
Data channel K.si L0119-YA
Tape data channel K.sti LOI08-YB
or or or
Data channel K.si L0119-YA
Input/output control processor P.ioj LOllI (HSC70)
P.ioj LOIII-YA (HSC40)
P.ioc LOI05 (HSC50)
Memory M.std2 LOl17 (HSC)
M.std LOI06 (HSC50)
Host interface K.ci Consists of:
Port link (LINK or ILl),
Port buffer (PlLA), and
Port processor (K.pli) modules
1.5.1 Port Link Module (LINK) Functions

The port link module (LOIOO-E2 or LOl18) is a part of the host interface module set (Kci). With
all configuration switches and jumpers in default positions, the LOl18 is functionally identical with
the LOIOO-E2. The location and default positions are described in Chapter 3. The port link module
performs the following functions.
• Serializationldeserialization, encoding/decoding, dc isolation- Permits transmission of
a self-clocking stream over the CI. Information transmitted over the CI bus is serialized and
Manchester encoded. The driver circuit includes a transformer for ac coupling the encoded
signal to the coaxial cable. Information received from a CI transmission is decoded and
converted to bit-parallel form. The circuitry also provides carrier detection for determining
when the CI is in use by another node.
• Cyclic redundancy check (CRe) generation/checking-Checks the 32-bit CRe character
generated and appended to a message packet when it is received. Also generates the 32-bit
CRC character during the transmission of a packet. An incorrect CRC means either errors were
induced by noise or a packet collision occurred.
• ACKINACK generation-Generates an ACK upon receipt of a packet addressed to the LINK
if the following conditions exist:
Error-free CRC
Buffer space available for the message
Upon receipt of a packet addressed to this node, a NACK is generated if the following conditions
exist:
Error-free CRC
No buffer space available for the message
No response is made if a packet addressed to this node is received with CRC error or the node
address is incorrect.
• Packet transmission-Performs the following functions:
Executes the CI arbitration algorithm
Transmits the packet header

Moves the stored information from the transmit packet buffer to the Manchester encoder
Calculates and appends the CRC to the end of the packet
Receives the expected ACK packet
• Packet reception-Perfonns the following functions:
Detects the start of the CI transmission
Detects the sync characters
Decodes the packet header information
Checks the CRC
Moves the data from the Manchester decoder
Return.s the appropriate ACK packet
The port link modUle interfaces through line drivers/receivers directly to the CI coaxial cables.
On the HSC interior side, the port link module interfaces to the port buffer module through a set
of interconnect link signals. The port link module also interfaces to the port processor module
(indirectly through the port buffer module) using a set of port link interface (PLI) signals.
1.5.2 Port Buffer Module (PI LA) Functions

The port buffer module (LOI09) provides a limited number of high-speed memory buffers to
accommodate the difference between the burst data rate of the CI bus and HSC internal memory
buses. It also interfaces to the port link (CI link) module through the ILl signals and the port
processor module through portllink interface (PLI) signals.
1.5.3 Port Processor Module (K.pli) Functions and Interfaces

The port processor module (LOI07-YA) performs the following functions:
• Executes and validates low-level CI protocol
• Moves command/message packets to/from HSCcontrolmemory and notifies the correct server
process of incoming messages
• Moves data packets to/from HSC data memory
The port processor module interfaces to three buses:
• PLI bus interfaces the port buffer and port link modules
• Control memory bus interfaces HSCcontrolmemory
• Data memory bus interfaces HSC data memory
1.5.4 Disk Data Channel Module (K.sdi) Functions

Disk data channel module (LOI08-YA) operation is controlled by an onboard microprocessor with
a local programmed read-only memory (PROM). This data channel module performs the following
functions:
• Transmits control and status information to the disk drives
• Monitors real-time status information from the disk drives
• Monitors in real-time the rotational position of all the disk drives attached to it
• Transmits data between HSC data memory and the disk drives
• Checks the error detection code (EDC) and generates or checks the error correction code (ECC)
during read/write operations.
Commands and responses pass between the disk data channel microprocessor and other internal
HSC processes through control memory. The disk data channel module interfaces to the control
memory bus and to the data memory bus. It can also interface to four disk drives with four
individual SDI buses. Currently, combinations of up to eight disk data channel or tape data
channel modules are possible in the HSC70. The HSC50 supports combinations of up to six disk
data channel or tape data channel modules and the HSC40 supports combinations of up to three
disk data channel or tape data channel modules. Configuration guidelines are found in the HSC
Installation Manual (EK-HSCMN-IN).
1.5.5 Tape Data Channel Module (K.sti) Functions

Tape data channel module (LOl08-YB) operation is controlled by ali onboard microprocessor with
a local programmed read-only memory (PROM). The tape data channel performs the following
functions:
• Transmits control and status information to the tape formatters
• Monitors real-time status information from the tape formatters
• Transmits data between the data memory and the tape formatters
• Generates an error detection code (EDC) for each 512 bytes during a write operation. The tape
formatter generates and sends an EDC every 512 bytes during a read operation.
Commands and responses pass between the tape data channel microprocessor and other
internal HSC processes through control memory. The tape data channel module interfaces to
thecontrolmemory bus and to the data memory bus.
1.5.6 Data Channel Module (K.si) Functions

Data channel module (LOl19-YA) is an interface between the HSC and the standard disk interface
(SD!) or standard tape interface (ST!) bus and is a direct replacement for the K.sdi or K.sti data
channel modules. The K.si is configured for disk or tape interface when the HSC is initialized (see
Chapter 4). The K.si functions are the same as the functions for the K.sdi or K.sti.
1.5.7 1/0 Control Processor Module (P.ioj/c) Functions

The HSC70 P.ioj module .(LOlll) and the HSC40 P.ioj module (LOlll-YA) use a PDP-11 ISP (J-11)
processor. The HSC50 and HSC50 (modified) both use the P.ioc module (LOl05), with a PDP-11
ISP (F-ll) processor. Both contain memory management and memory interfacing logic. These
processors execute their respective HSC internal software. The input/output (110) control processor
modules also contain the following functional blocks:
• Bootstrap read-only memory (ROM)
• Arbitration and control logic for the control and data buses
• Program-addressable registers for subsystem initialization and OCP communications
• Processes for all parity checking and generation for its accesses to memory
• Program memory instruction and data cache, 8 Kbytes of direct map high-speed memory (HSC
only)
The 110 control processor modules interface to:
• Program memory on the Program memory bus
• Control memory through the signals of the backplane control bus
• Data memory through signals of the backplane Data bus

• Console tenninal RS-423 compatible signal levels (HSC only)
• TU58 tape drives (HSC50 only)
• Auxiliary terminal through an RS-232-C interface (HSC50 only)
1.5.8 Memory Module (M.std2) Functions

The HSC memory module (LO 117) contains three separate and independent system memories, each
residing on a different bus within the HSC. In addition, the memory module contains the RX33
diskette controller. The three memory systems and RX33 diskette controller are known as:
• Control memory (M.ctl)-Two banks of 256 Kbytes of dynamic RAM for subsystem control
blocks and interprocessor communication structures storage.
• Data memory (M.dat)-512 Kbytes of status RAM to hold the data from/to a data channel
module.
• Program memory (M.prog)-l megabyte of RAM for the control program loaded from the
RX33 diskette.
CAUTION
The switch pack on the M.std2 module is factory set to calibrate the RX33 diskette
controller. Do not change the setting of this switch pack; the switch settings
are unique to each module and cannot be restored outside of the manufacturing
environment.
• RX33 diskette controller (K.rx)-Resides on the Program bus and performs direct memory
access word transfers when reading or writing data to/from the RX33 diskette.
Using .physical addresses, the memory space allocations for the three memories are illustrated in
Figure 1-14.
22-BIT ADDRESS ALLOCATION
ADDRESS SPACE BUS SIZE COMMENT

17777777 110 PAGE
INTERNAL 2KW INTERNAL REGISTERS
17770000
17767777 CONTROL
WINDOWS CBUS 2KW RESERVED ADDRESSES
17760000
17757777 UNDEFINED
'r NONE 248 KW NOT ACCESSIBLE
17000000
16777777 M.CTL
CBUS 256 KB (X2) CONTROL MEMORY
16000000
15777777 M.DAT
DB US 512 KB DATA MEMORY
14000000
13777777 UNUSED
PBUS 2 MB EXPANSION ROOM
04000000
03777777 M.PROG
PBUS PROGRAM MEMORY
00000000
1 MB 0-4000 RESERVED
FOR TRAP VECTORS
CXO-931A
Figure 1-14 Memory Map (M.std2-L0117)
NOTE
Two completely redundant memory banks make up control memory. Only one bank at a
time is usable during functional operation. Bank failure detection and bank swapping
are done at boot time.
The interface to the control memory is through the backplane control bus, and to the data memory
through the backplane Data bus. The interface to the 110 control processor local Program memory
is through a set of backplane signals to the Program memory module. In addition, the memory
module houses the control circuitry for the RX33 disk drives.
1.5.9 Memory Module (M.std) Functions

The memory module (LOI06) used in the HSC50 contains the following three independent and
separate memories:
• 256 Kbytes of Program memory (M.prog}-This is space for the control program loaded from the
TU58.
• 128 Kbytes of control memory (M.ctl}-This is space for the routines initiating data transfer
action.
• 128 Kbytes of data memory (M.dat}-This is space to hold the data from/to a data channel
module.
Using physical addresses, the memory space allocations for the three memories are illustrated in
Figure 1-15.
22-BIT ADDRESS ALLOCATION
ADDRESS SPACE BUS SIZE COMMENT

17777777 1/0 PAGE
INTERNAL 2KW INTERNAL REGISTERS
17770000
17767777 CONTROL
WINDOWS CBUS 2KW RESERVED ADDRESSES
17760000
17757777 UNDEFINED
: :r NONE 248 KW NOT ACCESSIBLE
17000000
16777777 UNUSED
CBUS 64KW EXPANSION ROOM
16400000
16377777 M.CTL
CBUS 64KW CONTROL MEMORY
16000000
15777777 UNUSED
DBUS 192 KW EXPANSION ROOM
14400000
14377777 M.DAT
DBUS 64KW DATA MEMORY
14000000
13777777 UNUSED
PBUS 1.5 MW EXPANSION ROOM
01000000
00777777 M.PROG
PBUS 128 KW PROGRAM MEMORY
00000000
0-4000 RESERVED
FOR TRAP VECTORS
CXO-338B
Figure 1-15 Memory Map (M.std-L0106)

The interface to the control memory is through the backplane control bus, and to the data memory
through the backplane data bus. The interface to the I/O control processor local Program memory
is through a set of backplane signals to the control memory module.
1.6 HSC Software Overview

The HSC subsystem uses inteInal software to perform various tasks and to interface with an
operator through a dedicated terminal. TheHSC Software Release Notes for your version of
software describes the unique features of the software. These software release notes are shipped
with each HSC and with updates of the software. The major HSC software modules are shown at a
block level in Figure 1-16.
HOST CPUs TAPES DISKS
I
K.CI
I
K.STI
I
K.SDI
DIAGNOSTIC UTILITY
CI STI SOl
SUB- PROCESSES
MANAGER MANAGER MANAGER
ROUTINES
DISK TAPE DISK DIAGNOSTIC

MSCP UTILITIES
ERROR I/O I/O MANAGER
PROCESSOR MANAGER
PROCESSOR MANAGER MANAGER
CONTROL PROGRAM
TU58 OR TERMINAL
RX33 DRIVES
CXO-1928A
Figure 1-16 HSC Internal Software

The HSC control prograin is found on the system diskette for the HSC and the system tape
for the HSC50. This module is the lowest level manager of the subsystem, provides a set of
subroutines and services shared by all HSC processes. The HSC control program performs the
following functions:
• Interprets incoming utility requests
• Sets up the appropriate subsystem environment for operation of the requested utility
• Invokes the utility process
• Returns the subsystem to its normal environment upon completion of the utility execution
• Initializes and reinitializes the subsystem
• Executes all auxiliary terminal I/O
• Schedules processes (both functional and diagnostic) for execution by the P.ioclj
• Provides a set of system services and system subroutines to HSC processes
• Manages the RX33 local storage media (HSC only)
• Manages the TU58 local storage media (HSC50 only)
1-22 Generallnformation
Functional processes within the HSC communicate with each other and the HSC control program.
They communicate through shared data structures and send/receive messages.
The MSCP class server validates, interprets, and routes incoming MSCP commands and
dispatches MSCP completion acknowledgments. The following are part of the MSCP class server:
• The SDI manager handles the SDI protocol, responds to attention conditions, and manages the
on-lineloff-line status of the disk drives.
• The disk 110 manager translates logical disk addresses into drive-specific physical addresses,
organizes the data-transfer structures for disk operations, and manages the physical positioning
of the disk heads.
The CI manager handles virtual circuit and server connection activities.
The disk error processor responds to all detected error conditions. It reports errors to the
diagnostic manager and attempts to recover from errors, such as ECC, bad block replacement,
and retries. When recovery is not possible, a diagnostic is run to determine if the subsystem can
function without the failing resource. Then, appropriate action is taken to remove the failing
resource or to terminate subsystem operation.
The TMSCP class server sets up the data transfer structures for tape operations and manages
the physical positioning of the tape. The STI manager is the part of the TMSCP class server that
handles the STI protocol, responds to attention conditions, and manages the on-line/off-line status
of the tape drives.
The diagnostic manager handles all diagnostic requests, error reporting, and error logging. It
also provides decision-making and diagnostic-sequencing functions, and can access a large set of
resource-specific diagnostic subroutines.
The diagnostic subroutines run under the control of the diagnostic manager and are classified as
device integrity tests.
The utility processes perform volume-management functions such as formatting, disk-to-disk
copy, disk-to-tape copy, tape-to-disk restore. They also handle miscellaneous operations required
for modifying subsystem parameters, such as COPY, PATCH, and error dump, and are used in
analyzing subsystem problems.
1.7 HSC Maintenance Strategy

Maintenance of the HSC is accomplished with field replaceable units (FRUs). Procedures for
removal and replacement are described in Chapter 3. Do not attempt to replace or repair
component parts within FRUs.
Isolation of solid failures can be accomplished efficiently due to the logical partitioning of the
modules and extensive internal diagnostics. In addition to the device-resident diagnostics, the
HSC-resident off-line diagnostics are available to support and verify corrective maintenance
decisions.
1.7.1 Maintenance Features

The following features assist in troubleshooting the HSC:
• Self-contained and self-initiated diagnostics-On initialization, various levels of diagnostics
execute in the HSC. Read-only memory (ROM) diagnostics test each microprocessor in the disk
and tape data channels, port processor, and I/O processor modules. Pressing the HSC lnit
button starts all internal ROM diagnostics.
• Operator control panel fault code display-The OCP or the console terminal displays any
failures. If further diagnostics are needed, use the terminal to initiate diagnostics stored on the
system boot media or the off-line diagnostic media (RX33 diskettes for the HSC or TU58 tapes
for the HSC50).
• Console terminal-After initialization, the operator can use the console terminal to run on-line
device integrity tests (see Chapter 5) or off-line diagnostic tests (see Chapter 6). Also, certain
resource failure detections can initiate tests automatically.
• Module LED indicators-All logic modules have at least one LED to indicate board status. See
Chapter 2 for the location of these LEDs.
The HSC subsystem allows logical assignment of a disk drive or tape formatter to the diagnostics.
Device integrity tests allow drive diagnosis, even though other active drives are connected to the
HSC.
Background (periodic) diagnostics test HSC logic not currently in use by the subsystem. Failures
cause the HSC to reboot and execute the initialization diagnostics.
Requestor-detected data memory errors cause an initiation of the in-line memory diagnostics to test
the buffer causing the error. Failures found in any data buffer cause removal of that buffer from
service. If no failure is found, the tested buffer is returned to service. If the same buffer is sent to
test twice, it is retired from service, even though no failure is found.
1.8 Specifications
Figure 1-17 lists the HSC physical and environmental specifications.
DESCRIPTION OPTION DESIGNATION

HSCXX-AA = 60 HZ, 120/180 V
HSC MASS STORAGE SERVER
HSCXX-AB = 50 HZ, 380/415 V
MECHANICAL
MOUNTING WEIGHT HEIGHT WIDTH DEPTH

CAB TYPE
CODE LBS KG IN CM IN CM IN CM (IF USED)
MODIFIED
FS 400 181.2 42 106.7 21.3 54.1 36 91.4
H9642
POWER (AC)
AC VOLTAGE AC VOLTAGE FREQUENCY AND STEADY-STATE POWER CONSUMPTION
PHASE
NOMINAL TOLERANCE TOLERANCE CURRENT (RMS) (MAX)
120/208 104-128/180-222 60 HZ :! 1 3 17 2250 WATTS
380/415 331-443 50 HZ :! 1 3 9 2245 WATTS
POWER (AC)
AMP (MAX) BY PHASE
120 V PHASE A =1 380 V PHASE A = 1
PHASE B = 12 PHASE B = 7
PHASE C= 12 PHASE C = 7
NEUTRAL = 17 NEUTRAL =9
POWER (AC)
PLUG TYPE POWER CORD LENGTH INTERRUPT TOLERANCE APPARENT POWER (KVA)
NEMA - L21 - 30P 15 FT (4.5 M) 4 MS (MIN) 3.4 (KVA)
HUBBELL - 520 P6 15 FT (4.5 M) 4 MS (MIN) 3.4 (KVA)
POWER (AC)
HSC OPTION INRUSH CURRENT SURGE DURATION
HSCXX - AA 70 AMPS/PHASE 16 MS
HSCXX - AB 70 AMPS/PHASE 20 MS
DEVICE ENVIRONMENT
TEMPERATURE RELATIVE HUMIDITY RATE OF CHANGE HEAT DISSIPATION
OPERATING* STORAGE OPERATING STORAGE TEMP HUMIDITY 60 HZ 50 HZ
0 0 0
59 - 90 F .40 - 151 F 20 F/HR
0 0
20 - 80% 5 - 95% 0
20%/HR 7676 BTu/HR 8078 KJ/HR
15 - 32 C .40 - 66 C 11 C/HR
DEVICE ENVIRONMENT
AL TlTUDE (MAX) AIR VOLUME (AT INLET) AIR QUALITY
OPERATING STORAGE FT3/MIN M3/MIN PARTICLE COUNT (MAX)
8000 FT 30,000 FT
210 5.92 N/A
2.4 KM 9.1 KM
0
*ALTITUDE CHANGES: DERATE THE MAXIMUM TEMPERATURE 1.8 C PER THOUSAND METERS
0
(1.0 F PER THOUSAND FEET).
CXO-2023A
Figure 1-17 HSC Specifications

Controls and Indicators 2-1
2
Controls and Indicators
2.1 Introduction
This chapter describes the following controls and indicators located in five areas of the HSC and
HSC50:
• HSC
l. Operator control panel (OCP)
2. Inside front door
3. RX33 disk drives
4. Logic modules
5. Power controller
• HSC50
l. Operator control panel (OCP)
2. Inside front door (TU58 tape drives)
3. Maintenance access panel
4. Logic modules
5. Power controllers (60 Hz and 50 Hz)
2-1
2-2 Controls and Indicators
2.2 Operator Control Panel (OCP)

Figure 2-1 illustrates the controls and indicators on the OCP.
MOMENTARY MOMENTARY ALTERNATE
CONTACT CONTACT ACTION
SWITCH SWITCH SWITCH
\ I
\, /
I
@ 0
State Power
Figure 2-1 Operator Control Panel

The OCP controls and indicators are described in the following list:
• State and Init indicators-Describes the state of the HSC. Under runtime conditions, the
lnit indicator is off while the State indicator is pulsing. During initialization, these indicators
change to reflect the current initialization phase of the subsystem. Refer to the bootstrap
flowchart in Chapter 8 for details on these phases.
• Init switch-Pushing the Init switch causes the HSC to start its initialization routine. The
SecurelEnable switch must be in the enable position for this switch to be operational. Holding
the lnit switch in causes the console terminal to loop back.
• Power indicator-Goes off if the dc voltage levels drop below one-third of minimal. The power
indicator is driven from a dc comparator circuit on the I/O Control Processor module (L0111 on
the HSC70, L0111-YA on the HSC40, or L0105 on the HSC50 and HSC50 (modified)), which
constantly monitors the +5, +12, and -5.2 voltages. The power indicator also is driven by a logic
gate that monitors the Power Fail signal from the power supplies. If this signal is asserted, the
power indicator goes off.
NOTE
An on power indicator does not mean these voltages are within specification (:t5
percent).
• Fault indicator and switch-Comes on when the HSC logic detects a fault. The Fault switch
is also used for the OCP lamp test.
Fault codes-When the Fault switch is pressed and released, the lamps in Init, Online,
Fault, and the two blank switches function as an error display. If the fault code is a hard
fatal error, the fault code blinks on and off until the HSC is powered down or the Fault
switch is pressed again.
If the displayed fault code is a soft (nonfatal) failure, the fault code clears on subsequent
toggling of the Fault switch. Multiple soft fault codes can be queued in the fault code buffer.
Subsequent toggling of the Fault switch displays each soft fault code until the buffer is
emptied.
Soft fault codes are identified by the Fault indicator on (or displayed fault code) while the
State indicator is pulsing. With soft faults, the HSC continues to operate without use of
the failing resource. Hard fault codes are identified by the fault indicator on (or displayed
fault code) while the HSC State indicator is not pulsing. With hard faults, the HSC does
not continue operation until the failure is remedied.
Error codes associated with the OCP display are defined in Chapter 4 and in Chapter 8.
Lamp test-Pushing and holding the Fault switch causes all the OCP indicators to light
and function as a lamp test. Even if the Fault indicator is already on before the switch is
pushed, the lamp test can be executed.
• Online switch-Puts the HSC logic in the available state when pushed to the in position and
allows a host to establish a virtual circuit with the HSC. When this switch is released to the
out position, no new virtual circuits can be made.
• Online indicator-Shows a virtual circuit exists between the HSC and a host CPU when the
Online indicator is on. When this indicator is off, no virtual circuits are established with any
host.
• Blank indicators-Forms the lowest two bits of a five-bit fault code.
2.3 HSC Inside Front Controls and Indicators

Figure 2-2 shows the controls and indicators available when the front door is opened.
SECURE/ENABLE
SWITCH
OCP SIGNAL/POWER
LINE CONNECTOR
Figure 2-2 Controls/Indicators Inside Front Door

The following list describes the controls and indicators found on the HSC inside front door:
• SecurelEnable switch-Disables the Init switch from the OCP when in the secure position.
Also, the SET utility program cannot run and the break key from the terminal is disabled. With
the SecurelEnable switch in the enable position, the Init switch and all the utility programs can
be used.
The SHOW utility is operable with the SecurelEnable switch in either position.
• Enable indicator-Indicates the SecurelEnable switch is in the enable position when the
Enable LED is illuminated (all switches can be used). When the Enable indicator is off, the
OCP is secure.
• RX33 LEDs-When lit, indicates which particular drive is in use. There is an LED on the
front panel of each drive. When not in use, the RX33 diskettes are stored inside the front door
(Figure 2-3).
DRIVE DRIVE-
COVER IN-USE LEDs
PLATE
Figure 2-3 RX33 and de Power Switch

• dc power switch-Located on the left side of the RX33 housing (Figure 2-3). When the dc
power switch is in the 0 position, the HSC is without dc power. Moving the switch to the 1
position restores dc power.
2.4 HSC50 Inside Front Door Controls and Indicators

Figure 2-4 shows the controls and indicators on the inside of the front door.
TUSS SELF-TEST
INDICATOR
(VIEWED FROM TOP)
TUSS TAPE
CARTRIDGE
STORAGE
ENABLE TUSS RUN

LED LEOs
CXO-003B_S
Figure 2-4 HSC50 Controls/Indicators Inside Front Door

The following list describes the controls and indicators found on the inside of the front door of the
HSC50:
• SecurelEnable switch-With the SecurelEnable switch in the secure position, the lnit switch
is disabled from the OCP. Also, the SET utility program cannot run and the break key from
the terminal is disabled. With the SecurelEnable switch in the enable position, the lnit switch
and all the utility programs can be used. The SHOW utility is operable with the Secure'Enable
switch in either position.
• Enable indicator-An illuminated Enable LED indicates the SecurelEnable switch is in
the enable position (all switches can be used). When the Enable indicator is off, the OCP is
secure.
• TU58 Run indicators-When a TU58 Run indicator is on, the TU58 is currently moving tape.
Data loss can occur if the tape is removed while this indicator is on. If the indicator is off, tape
is not in motion.
• TU58 Self-Test indicator-The TU58 Self-Test indicator is found on the TU58 controller
module (Figure 2-4). The controller module is located inside the TU58 housing with the drive
mechanics. Observe the Self-Test indicator by looking down through the TU58 housing vents.
When this indicator is on, the TU58 controller has successfully completed self-diagnostics.
2.5 HSC50 Maintenance Access Panel Controls and Connectors

Removing the maintenance access panel cover reveals the dc power switch and several connectors
available for HSC50 maintenance (Figure 2-5).
DC POWER
SWITCH
OFF POSITION (0)
OCP
CONNECTOR
CONNECTORS
RESERVED FOR
FUTURE USE
MAINTENANCE
TERMINAL
SIGNAL
CONNECTOR
MAINTENANCE
ACCESS PANEL
CXO-014B
Figure 2-5 HSC50 Maintenance Access Panel

2.5.1 HSC50 dc Power Switch

When the dc power switch is in the 0 position, the HSC50 is without dc power. Moving the switch
to the 1 position restores dc power.
2.5.2 HSC50 Maintenance Panel Connectors

Two of the connectors in the maintenance access panel are used to connect the maintenance
terminal to the HSC50. One connector supplies power to the maintenance terminal and the other
is the signal connector. Additional connectors are:
• OCP connector
• TU5S connectors
• Connectors reserved for future use
2.6 Module Indicators

All logic modules have at least one LED to indicate board status. Figure 2--6 shows the locations
of these LEDs and the module utilization label. Additionally, three of these logic modules contain
specific switches.
jLABEL
MODULE UTILIZATION
NODEADDRESS----~w.f
SWITCHES
SWITCH S-3
(REV E2)
I
D1 MICRO ODT
D2 SERIAL LINE UNIT
D3 MEMORY OK
D4 SEQUENCING
LINK BOARD
STATUS
INDICATORS
e RED
o AMBER
@ GREEN
CXO-933B
Figure 2-6 Module LED Indicators

2-1 0 Controls and Indicators
NOTE
Figure 2-6 and Figure 2-7 shows a typical HSC module configuration. The disk and tape
data channel module combinations vary as follows between the HSC models:
The HSC70 supports up to 8 disk and tape data channel module combinations.
Figure 2-7 shows the HSC slot location for each of the modules.
0
U)
Q; Q; Q; Q) Qj Q) Q) Qj U)
t: t: t:
0 t: t: t: t:
t:
t:
t:
t:
t:
t:
t:
t:
t:
CD
0
U)
as as as as as as as as 0
i
fI)
(I)
0
.r:.
0
.r:.
0
.r:.
0
.r:.
0
.t:
0
.r:.
0
.r:.
0
.s:::.
0
a
~
t:
:::i
'5
m a
0
m
>:-
as m
->-
as I
as m
CiS >-
as m
Cii >-
as m
Cii >-
as <:
Cii <:
eC
eo Ceo C rl> C rl> C rl> C ~ 0 'r'" 0
o ".::t; o ".::t; o ".::t: o ".::t; o .. ~ 'r'" •• E ··0
Mod
' r ' " > f I ) 'r'" >
o (1).-
..JO::C -10::0
o
fI)
(1).-
'r'" >
o (1).-
fI)
..JQ:C ..JQ:C
o
~ .~ o (I) . - 'r'" >
'r'" > fI) CD
..JQ:C 3~~ ..Ja::::::
'r'"
o ~o
A B C D E F M N V
2 3 4 5 6 7 8 9 0
10 9 8 7 6 5 4 3 2 1
CXO-889A
Figure 2-7 HSC Module Utilization Label Example

Table 2-1 shows the functions of the various module LEDs.
Table 2-1 Functions of Logic Module LEOs

Module Color Function
LINK (LOIOO-E2) Green Board status-Indicates the node is either transmitting or receiving;
LINK (LOllS) dims or brightens relative to the amount of local CI activity.
Red Board status-Indicates the module is in the internal maintenance
mode.
PILA (LOI09) Green Board status-Indicates the operating software is running and that
all applicable diagnostics have completed successfully.
Red Board status-Indicates an inoperable module except during
initialization when it comes on during module testing.
Amber Always on when the HSC is on (used only for engineering test
purposes).
K.pli (LOI07-YA) Green Board status-Indicates the operating software is running and that
self-test module microdiagnostics have completed successfully.
Table 2-1 (Cont.) Functions of Logic Module LEDs

Module Color Function
Red Board status--Indicates an inoperable module, except during

initialization, when it comes on during module testing.
K.sdi (LOI08-YA) Green Board status-Indicates the operating software is running and that
K.sti (LOI08-YB) self-test module microdiagnostics have completed successfully.
K.si (LOI19-YA)
Red Board status--Indicates an inoperable module, except during
K.si only Amber DI-Offfor PROM load, on for RAM load.
(eight D2 through D8-Upper register #2 contents. When a
LEDs) microinstruction parity error is detected, the module clocks are
inhibited, which stops the module. The bit content of the upper error
register #2 is displayed on the LEDs. See Figure 2-6 for the location
of the LEDs.
M.std (LOI06) Green Board status--Indicates memory cycles are operating.
M.std2 (LOI17) Green Board status--Indicates the operating software is running and has
successfully tested this module.
Amber Indicates Memory Active-Lit during every memory cycle.
Red Board status--Indicates an inoperable module except during
P.ioc (LOI05) Amber State indicator (top LED)-Mirrors the OCP State indicator.
Amber Run indicator (bottom LED)-Pulses at the on-board microprocessor
run rate.
Red Board status-Indicates an inoperable module except during
Green Board status--Indicates the module has passed all applicable
diagnostics.
P.ioj (LOllI or DI amber Micro ODT - Used during J-II power-up microdiagnostics
LOIII-YA)
D2 amber Terminal port OK-Used during J-II power-up microdiagnostics.
D3 amber Memory OK-Used during J-II power-up microdiagnostics.
D4 amber Sequencing indicator-Used during J-II power-up microdiagnostics.
D5 amber State indicator-Mirrors the OCP State indicator.
D6 amber Run indicator-Pulses at the on-board microprocessor run rate.
D7red Board status--Indicates an inoperable module, except during
D8 green Board status--Indicates the module has passed all applicable
diagnostics.
2.7 Module Switches

Specific switches are found on LINK (LOIOO-E2 or LOl18), port processor (LOl07), and port buffer
(LOl09) modules as follows:
• CI port LINK module (LOIOO-E2IL0118)-Refer to Figure 2-8 for the CI node address
switches mounted on the LOIOO-E2 or LOl18 module.
NOTE
Memory module M.std2 (LOl17) contains a switch pack. These switches are factory
set to calibrate the RX33 diskette controller. Do not change the setting of this switch
pack; the switch settings are unique to each module and cannot be restored outside
of the manufacturing environment.
81
82
83
LINK L0118
MODULE
CXO-2596B
Figure 2-8 L0118 Module (DIP) Switches

Switches must be identically set to avoid CI addressing errors. See Chapter 3 for switch
positions of S1, S2, and S3.
• CI port processor and CI port buffer modules (LOI07 and LOI09)-Both the LOI07 and
LOI09 modules have dual in-line pack (DIP) switches to indicate the hardware revision leveI.
DIP switch positions should not be changed, except as directed by a Field Change Order (FCO).
Figure 2-9 shows the location of the LOI07 switches.
L0107 CI PORT BUFFER

MODULE HARDWARE REVISION
LEVEL SWITCHES
(DO NOT CHANGE EXCEPT BY FCO)
CXO-2684A

Figure 2-10 shows the location of the L0109 switches.

LEVEL SWITCHES
CXO-2683A

• K.si (LOl19-YA) data channel switchpack - Figure 2-11 shows the location of the K.si
module switches.
K.SI DATA CHANNEL

MODULE
NOTE:
ALL SWITCHES
MUST BE OFF FOR
NORMAL OPERATION.
RED STATUS LED
GREEN STATUS LED
CXO-2495A
Figure 2-11 K.si Module (L0119-YA) Switches

• P.ioj (LOllllLOlll·YA)-The P.ioj module contains two punch-out connector packs used to
assign an unique value to the P.ioj serial number register. The switch settings should never be
modified in the field.
The P.ioj module serial number is used only when a default HSC SDS-ID is generated. The
SDS-ID is a hexadecimal number uniquely identifying the HSC as a node in the cluster. This
ID is usually generated by initializing the HSC70 (toggling the Init switch on the OCP) while
holding in the OCP Fault switch until the INIPIO banner is printed on the console. For all
other reboot cases, the HSC70 P.ioj serial number is not used.
2.8 881 Power Controller

The 881 power controller is a general-purpose, three-phase controller that controls and distributes
ac power to various ac devices (power supplies, fans, blower motor, and so forth) packaged within
an HSC and HSC50. The 881:
• Controls large amounts of ac power with low level signals
• Provides ac power distribution to single-phase loads on a three-phase system
• Protects data equipment from electrical noise
• Disconnects ac power for servicing and in case of overload
In addition, the 881 features:
• Local and remote switching
• Swi tched receptacles only
• Convection cooling
• Rack mounting
• ac line filtering
• Power Control bus inputs
• Power Control bus delayed output (to allow sequencing of other controllers)
2.8.1 Operating Instructions

The two basic controls on the power controller are the circuit breaker and the BUS/OFF/ON
switch. These and all but one of the other controls are located on the front panel of the controller
(Figure 2-12).
GROMMETED
CORD
OPENING POWER CONTROL
BUS CONNECTORS
INTERNATIONAL SYMBOLS
UNDELAYED f::\ SECONDARY

DELAYED
~ ON
(0.5 SEC)
SERIAL LOGO O• OFF

SECONDARY
LABEL
~ REMOTE BUS
FUSE L:!.J CONTROL
POWER
CONNECTOR
CXO-893A
Figure 2-12 881 Power Controller-Front Panel Controls

The operator controls are described in the following list:
• Power controller circuit breaker-Controls the ac power to all outlets on the controller.
I t also provides overload protection for the ac line loads and is unaffected by switching the
BUS/OFF/ON control.
• Fuse-Protects the ac distribution system from an overload of the Power Control bus circuitry.
The fuse is located on the front panel of the power controller.
• Power Control bus connections-Used if Control bus connections to another cabinet are
required. Power Control bus connectors are JIO, JII, J12, and J13. Connectors JIO and Jll are
not delayed. Connectors J12 and Jl3 are delayed.
• BUS/OFF/ON switch-The three positions of this switch. Assuming the circuit breaker for the
power controller is on, the ac outlets are:
Energized when the BUS/OFF/ON switch is in the on position
- De-energized when the BUS/OFF/ON switch is in the off position
NOTE
The BUS position is intended for remote sensing of Digital power control bus
instructions. The switch is left in the on position when the power control bus is
not used.
• TOTAL OFF connector-A two-pin male connector on the rear panel of the power controller
(Figure 2-13). It removes power from the HSC whenever the air flow sensor detects system air-
flow loss or an over temperature condition. To reset the TOTAL OFF, cycle the circuit breaker
off and then back on again.
TOTAL OFF
CONNECTOR
CXO-934A
Figure 2-13 881 Rear Panel
2.9 HSC50 Power Controller

The 60 Hz power controller is shown in Figure 2-14 and Figure 2-16. For the 50 Hz unit, refer to
Figure 2-15 and Figure 2-17. A physical description of the power controller follows.
DELAYED
DEC POWER OUTPUT REMOTEI
CONTROL BUS CONNECTOR OFF/LOCAL
CONNECTORS ON SWITCH
LINE PHASE
LINE
INDICATOR
POWER
CIRCUIT
BREAKERS
CB1
CB2-4
(SWITCHED)
~::--t-~=::=t
FUSE
CBS
(UNSWITCHED)
CXO-013B
Figure 2-14 HSC50 Power Controller (60 Hz)-Front View
2.9.1 Line Phase Indicators

Three Line Phase indicators display the status of incoming line power. If any phase drops, the
indicator for that phase goes off.
2.9.2 . Fuses
The three line phases are fused to protect the HSC50 circuitry. These fuses are located beside the
Line Phase indicators as shown in Figure 2-14 and Figure 2-15
DELAYED
DEC POWER OUTPUT REMOTE/
CONNECTOR ON SWITCH
FUSE
LINE
POWER
CIRCUIT ACINPUT
BREAKERS LINE PHASE
INDICATOR
o CB1 I
CXO-013C
Figure 2-15 HSC50 Power Controller (50 Hz)-Front View
2.9.3 Remote/Off/Local On Switch

When this switch is in the off position, the power controller does not route ac line power to the
switched or unswitched outlets.
With the switch in the Local On position, ac power is routed to the power controller switched or
unswitched outlets.
When the switch is in the Remote position, the routing of ac power is dependent upon the Power
Control bus signals.
2.9.4 Circuit Breakers (60 Hz)

There are five power controller circuit breakers which perform the following functions:
• CB1-Protects from incoming power surges
• CB2-4-Protects the switched outlets (refer to Figure 2-13)
• CB5-Protects the unswitched outlets
2.9.5 Circuit Breakers (50 Hz)

The 50 Hz unit contains one circuit breaker (Figure 2-15). CB1 on this unit protects all circuits.
2.9.6 Power Controller {60 Hz)-Rear View

The switched outlets in Figure 2-16 are protected by CB2-4 (refer to Figure 2-14) and the bottom
(unswitched) by CB5. Both the bottom and top outlets are currently unused.
A three-pin male connector (J8) is located on the back of the power controller (Figure 2-16). It
removes power from the HSC whenever the air flow sensor detects system air-flow loss or an over
temperature condition. To reset the TOTAL OFF, cycle the circuit breaker off and then back on
again.
UNUSED
BLOWER
OUTLET
MAIN POWER
SUPPLY OUTLET
AUXILIARY POWER
SUPPLY OUTLET
UNSWITCHED
CXO-411A
Figure 2-16 HSC50 Power Controller (60 Hz)-Rear View
2.9.7 Power Controller (50 Hz)-Rear View

Outlets in Figure 2-17 are protected by CB1 (refer to Figure 2-15). Connector J3, shown at the top
of the 50 Hz power controller rear view, connects the air flow sensor.
J3
Total
Off
BLOWER
OUTLET
MAIN POWER
SUPPLY OUTLET
AUXILIARY POWER
SUPPLY OUTLET
CXO-411B
Figure 2-17 HSC50 Power Controller (50 Hz)-Rear View

Removal and Replacement Procedures 3-1
3
Removal and Replacement Procedures
3.1 Introduction
This chapter emphasizes conditions that must be met when replacing field replaceable units (FRUs),
including the following information:
• Safety precautions
• HSC failover
• FRU overviews
• Jumper configurations
• Swi tch configurations
• Test sequence to perform after FRU replacement
Observe the safety and electrostatic discharge (ESD) precautions in this section before starting
removal and replacement procedures.
This chapter covers the following replaceable HSC subunits and modules:
Modules:
Port link module (LINK)
Port buffer module (PILA)
Port processor module (K. pli)
Disk data channel module (K.sdi)
Tape data channel module (K.sti)
Data channel module (K.si)
110 control processor module (P.ioj and P.ioc)
Memory module (M.std2 and M.std)
Subunits:
RX33 disk drive
TU5S tape drive
Operator control panel
Air flow sensor
Blower
Power controller
Main power supply
Auxiliary power supply
3-1
3-2 Removal and Replacement Procedures
3.2 Safety Precautions

Because hazardous voltages exist inside the HSC, service must be performed only by qualified
people. Serious bodily injury or equipment damage can result from improper servicing. Observe the
following safety steps before servicing the HSC:
1. Turn off the dc and ac power to the HSC before removing or installing internal parts or cables.
2. To ensure absolute safety, disconnect the ac plug from its receptacle after removing power from
the HSC.
3. Remove and replace heavy subunits with care.
4. Use the Velostat (anti-static) kit (part number 29-11762) strap provided when removing and
replacing logic modules.
3.3 Taking the HSC Off Line for Maintenance

This section describes how to take an HSC off line for performing maintenance.
3.3.1 Single HSC in a Cluster and Clusters Running ULTRIXIUNIX

If there is only one HSC in the cluster, or if the cluster is running the ULTRIXIUNIX operating
system, use the following procedure:
1. Notify the system manager that the HSC is being taken off line and the drives attached to it
will not be available.
2. Dismount the drives connected to the HSC or shut down the system.
3. Place the Off-line switch in the out position.
4. Take the HSC off line with one of the following methods:
• Turn off the dc and ac power to the HSC.
• Press the Init switch to reboot the HSC.
3.3.2 Multiple HSCs in a Cluster

Most VMS system clusters have primary and secondary paths established between the host and
drives. If the HSC is the primary path, failover to the secondary path occurs and the drives remain
available to the host. Use the following procedure when taking an HSC offline from a cluster with
multiple HSCs:
1. Notify the system manager that you are taking the HSC off line.
2. Use the SETSHO command SHOW TAPE to determine which tape drives are on line to the
HSC. Dismount these tape drives using the appropriate VMS commands.
3. Determine which disk drives are on line to the HSC undergoing maintenance with the SETSHO
command SHOW DISK. The on-line drives must be either be failed over to the alternate HSC
or dismounted.
4. Dismount single-ported disk drives using the appropriate VMS commands.
5. On all dual-ported disk drives, make sure that both A and B port select switches on the drives
are pressed in. De-select the active port on the drive by pressing and releasing the illuminated
port select switch.
. 6. Verify that the illuminated switch turns off and the alternate port switch illuminates. Actual
failover time depends on the server timeout and may require 1 minute or more depending on
activity. When the alternate port light illuminates, failover to the other HSC is successful.
7. Use the SETSHO command SHOW DISK to verify that all on-line drives have failed over to the
alternate HSC.
8. If failover did not occur, reselect the ports on the drive and check the connections between the
drives and the HSC.
9. Mter all tapes and disks have been failed over, take the HSC off line using one of the following
methods:
• Turn off the dc and ac power to the HSC, or
• Place the Off-line switch in the out position, or
• Press the Init switch to reboot the HSC.
3.4 Removing and Replacing Field Replaceable Units

The following sections describe procedures for removing and replacing field replaceable units
(FRUs) in an HSC or HSC50:
3.4.1 Removing HSC Power

Before removing/replacing an FRU, turn off the ac power from the HSC. Following are the methods
for removing dc and ac power from the HSC:
DC POWER
SWITCH
OCP SIGNAU
POWER LINE
CONNECTOR
Figure 3-1 HSC DC Power Switch

1. Set the dc power switch on the side of the RX.33 housing to the off position (Figure 3-1).
WARNING
Ensure the OCP SignallPower line indicator is connected; otherwise the power
indicator on the OCP can show power off when the power is on.
2. Place the main power switch CBl on the power controller in the off position (Figure 3-2). To
ensure safety precautions, unplug the ac power plug from the ac socket.
(f)
~
o G1
:rn rn rn m:
JI3 JI2J11 J1 0
000
o CIRCUIT
BREAKER
POWER
CONNECTOR
CXO-1117A
Figure 3-2 HSC 881 Power Controller Circuit Breaker
3.4.2 Removing HSC50 Power

Following are the steps for removing dc and ac power from the HSC50:
1. Set the dc power switch on the maintenance access panel to the off position (Figure 3-3).
~ Removal and Replacement Procedures
DC POWER
SWITCH
OFF POSITION (0)
OCP
CONNECTOR
CONNECTORS
RESERVED FOR
FUTURE USE
MAINTENANCE
TERMINAL
SIGNAL
CONNECTOR
MAINTENANCE
ACCESS PANEL
CXO-014B
Figure 3-3 HSC50 DC Power Switch

2. Place the main power switch CBl on the power controller in the off position (Figure 3-2) for the
HSC50 (modified) and the line power circuit breakers (Figure 3-4) for the HSC50. To ensure
safety precautions, unplug the ac power plug from the ac socket.
DELAYED
LINE PHASE
LINE
INDICATOR
POWER
CIRCUIT
BREAKERS
CB1
CB2-4
(SWITCHED)
~:::::--t-~;r::;;::=;1
FUSE
CBS
(UN SWITCHED)
CXO-013B
Figure 3-4 HSC50 Line Power Circuit Breakers

3.4.3 Removing Field Replaceable Units

Figure 3-5 shows the sequence for removing field replaceable units (FRUs) in the HSC:
OPEN CABINET FRONT DOOR
RX33
I
OCP MODULES
OPEN CABINET BACK DOOR
POWER CONTROLLER
BLOWER
AIR FLOW
SENSOR ASSEMBLY
CABINET BACK DOOR

CABINET FRONT DOOR
MAIN POWER SUPPLY
AUXILIARY POWER SUPPLY

CXO-93SB
Figure 3-5 HSC FRU Removal Sequence

Figure 3-6 shows the FRU removal sequence for an HSC50.
OPEN CABINET FRONT DOOR
I
TU58 DRIVES MODULES
TU58 CONTROLLER MODULE OCP
OPEN CABINET BACK DOOR
POWER CONTROLLER
BLOWER
AIR FLOW
SENSOR ASSEMBLY
CABINET BACK DOOR

CABINET FRONT DOOR
MAIN POWER SUPPLY
AUXILIARY POWER SUPPLY

CXO-015C
Figure 3-6 HSC50 FRU Removal Sequence
3.4.4 Removing the HSC Cabinet Front Door

The FRUs accessed through the front door include the RX33 drives, the operator control panel
(OCP), and the logic modules. To remove the front door, use the following procedure:
1. Unlock the cabinet front door and lift the latch to open the door.
CAUTION
When performing the following steps, take care not to damage the front spring
fingers.
2. Remove HSC power by setting the dc power switch to the 0 position.
3. Disconnect the ground wire from the door.
4. Disconnect the OCP signal/power line connector at the bottom of the OCP shield (Figure 3-7).
HSC70
DC POWER
SWITCH
OCP SIGNAU
POWER LINE
CONNECTOR
Figure 3-7 HSC OCP Signal/Power Line Connector

5. Pull down on the spring-loaded rod on the top hinge inside the cabinet to disengage the door,
then lift the door off its bottom pin.
Reverse the removal procedure to replace the front door.
3.4.5 Removing the HSC50 Cabinet Front Door

The FRUs accessed through the front door include the TU5S drives, the operator control panel
(OCP), and the logic modules. To remove the front door, use the following procedure:
1. Open the cabinet front door by turning the key clockwise.
CAUTION
When performing the following steps, take care not to damage the front spring
fingers.
2. Disconnect the ground wire from the door.
3. Remove the maintenance access panel cover by loosening the four captive screws.
NOTE
Some HSC50s have a hinged maintenance access panel with only one captive screw.
4. Remove HSC50 power by setting the dc power switch to the 0 position.
5. Remove the plastic cable duct cover.
6. Disconnect the cables from the maintenance access panel connectors shown in Figure 3-8.
DC POWER
SWITCH TU58
ON POSITION (1) CONNECTORS
OFF POSITION (0)
OCP
CONNECTOR
CONNECTORS
RESERVED FOR
FUTURE USE
MAINTENANCE
TERMINAL
SIGNAL
CONNECTOR
MAINTENANCE
ACCESS PAN EL
CXO-014B
Figure 3-8 HSC50 Maintenance Access Panel Connectors

Reverse the removal procedure to replace the front door.
3.4.6 Removing the HSC Cabinet Back Door

The FRUs accessed through the back door include the power controller, blower, air flow sensor
assembly, main power supply, and auxiliary power supply. To remove the back door, use the
following procedure:
1. Open the back door with a 5/32-inch hex wrench.
Reverse the removal procedure to replace the back door.
3.5 Removing and Replacing Modules

This section contains procedures for removing and replacing modules. Refer to Figure 3-9 for the
HSC and Figure 3-10 for the HSC50 when removing the card cage cover.
DISKETTE
STORAGE
AREA
Figure 3-9 HSC Card Cage Cover Removal

NYLON
LATCHES
CARD
CAGE
COVER
CXO-019A
Figure 3-10 HSC50 Card Cage Cover Removal
WARNING:
Because hazardous voltages exist inside the HSC, service must only be performed by
qualified people. Bodily injury or equipment damage can result from improper servicing
procedures.
A Velostat (anti-static) kit (part number 29-11762) must be used during module
removal/replacement.
3.5.1 Removing and Replacing the Port Link Module (LINK)

The port link module (LINK) is a part of the host interface module set (Kci). With all configuration
switches and jumpers in default positions, the LOl18 is functionally identical with the L0100 or
L0100-E2.
3.5.1.1 Removing the LINK Module

Use the following procedure to remove the LINK module. Observe safety and ESD precautions
before starting the module removal procedure.
1. Notify users that the HSC is being taken off line and the drives attached to it will not be
available.
2. Dismount or failover any drives connected to the HSC or shut down the system.
·3. Set the dc power switch to the 0 (off) position.
On the HSC, the dc power switch is located on the side of the RX33 housing.
Removal and Replacement Procedl,Jres 3-15
On the H8C50, the dc power switch is located on the maintenance access panel.
4. Turn the two nylon latches on the card cage cover one-quarter turn.
5. Pull the H8C card cage cover up and out.
6. Locate the LINK module in slot number 14 of the card cage. This can be verified by the module
utilization label.
7. Move the door latch plate attached to the left side of the cabinet frame away from the module
removal path. In the H8C cabinet, the latch plate is swivel mounted. Lift the plate slightly and
press it flat against the cabinet frame. Remove the LINK module.
3.5.1.2 Setting the Replacement LINK Module Switches

81 and 82 are the node address switches on the LINK module. The node address switches on the
replacement module must be set identically to the switch settings on the removed LINK module.
The L0100-E2 and LOllS LINK module also have an additional switch pack (83).
The switch configurations and significance are as follows:
81/82 - Node number
83-1 - Cluster size (GT15), OFF for 16 or less nodes (default); ON for 17 or more nodes
83-2, 83-3 - Delta time/quiet slot (default) = always OFF
83-4 -10 ticks = always ON
Figure 3-11 shows the LINK (L0100) module node address switches.
II II
VALUE 8
2
4 3.
2 '
4
II
0 VALUE
2
8
3.
2 '
4
II
II 0
OF EACH
SWITCH 16
32
5
6
• P
E
N
OF EACH
SWITCH 16
32
5
6
• P
E
N
64 711 64 711
128 8
• 0
128 8 •
0
S1 52
DIP SWITCH
(EXAMPLE: OCTAL 10)
CXO-2695A
Figure 3-11 L0100 Node Address Switches

Figure 3-12 shows the LINK (L0100-E2ILOl18) module node address switches.
~~ ~~
:~~ :~~
:~ :~
'~ '~
S1 S2 S3
CXO-2696A
Figure 3-12 L0100-E2IL0118 Node Address Switches
3.5.1.3 Setting the Replacement LINK Module Jumpers

The LINK jumper configurations are as follows:
W1-Extender head, Default = Not used
W2-Active hub, Default = Not used
W3--Extender ACK timeout, Default = Not used
W4-Cluster size (GT32), Default = Less than 32
Figure 3-13 shows the LINK (L0100) jumpers.
W3
+
1++1
W1
1++1+
NOTE: BOXES INDICATE THE DEFAULT JUMPER POSITIONS.

CXO-1910B
Figure 3-13 L0100 Jumper Configuration

Figure 3-14 shows the LINK (LOl18-B1) jumpers.
W3
+
1++1
W1
1++1+
W2 1+ +I~ W4
NOTE: BOXES INDICATE THE W1 AND W3 FAULT JUMPER POSITIONS.

THE DEFAULT FOR JUMPERS W2 AND W41S "OUT".
CXO-1911B
Figure 3-14 L0118-B1 Jumper Configuration

Figure 3-15 shows the LINK (L011S-B2) jumpers.
W3
+
1++1
W1
1++1+
1++1+
W2
W4
NOTE: BOXES INDICATE THE DEFAULT JUMPER POSITIONS.
CXO-1912B
Figure 3-15 L0118·B2 Jumper Configuration

3.5.1.4 Replacing the LINK Module

This section provides the LINK module replacement procedure. Observe safety and ESD
precautions before starting the module replacement procedure.
1. Install the LINK module in slot number 14 of the card cage. This can be verified by the module
utilization label.
2. Move the door latch plate attached to the left side of the cabinet frame away from the module
installation path. In the HSC cabinet, the latch plate is swivel mounted. Lift the plate slightly
and press it flat against the cabinet frame. Install the LINK module and return the plate to its
locked position.
3. Pull the card cage cover down and in.
4. Turn the two nylon latches on the module cover plate one-quarter turn.
5. Set the dc power switch to the 1 (on) position.
On the HSC50, the dc power switch is located on the maintenance access panel.
3.5.1.5 Testing the LINK Module

Perform the following tests to verify correct LINK operation as part of the K.ci host interface
module set.
1. Boot the HSC with the off-line diagnostic media. Refer to Chapter 6 for boot procedures.
NOTE
The off-line diskette must be write protected. Place a write-protect tab over the
diskette write-enable notch.
2. Refer to Chapter 6 for test descriptions and procedures and perform the following tests:
• Off-line bus interaction test
• Off-line K test selector
• Off-line KIP memory test
3. Place the SecurelEnable switch in the secure position.
4. Boot the HSC with the system media by pressing and releasing the Init switch.
5. Bring the HSC on line by pressing and releasing the Online switch.
6. Use the SETSHO command SHOW VIRTUAL_CIRCUITS to ensure that both A and B paths
are present to all hosts.
7. Use the SETSHO command SHOW CI to verify the absence of the RTNDATIDISC datagram.
3.5.2 Removing and Replacing Port Buffer Module (PILA)

The port buffer module (PILA) is a part of the host interface module set (K.ci). The PILA interfaces
with the port link module through a set of interconnect link signals.
3.5.2.1 Removing the PILA Module

Use the following procedure to remove the PILA module. Observe safety and ESD precautions
available.
3. Set the dc power switch to the 0 (oft) position.

On the HSCSO, the dc power switch is located on the maintenance access panel.
5. Pull the card cage cover up and out.
6. Locate the PILA in slot number 13 of the card cage. This can be verified by the module
utilization label.
7. Remove the PILA module.
3.5.2.2 Setting the Replacement PILA Module Switches

The PILA module has factory-set dual in-line pack (DIP) switches to indicate the hardware revision
level. Do not change DIP switch positions, except as directed by a field change order (FCO).
Figure 3-16 shows the PILA module switch.
I, ~" , " ~r
~ LEVEL SWITCHES
CXO-2698A
Figure 3-16 L0109 Hardware Rev Level Switch
3.5.2.3 Replacing the PILA Module

This section provides the PILA module replacement procedure. Observe safety and ESD precautions
before starting the module replacement procedure.
1. Install the PILA module in slot number 13 of the card cage. This can be verified by the module
utilization label.
4. Set the de power switch to the 1 (on) position.
3.5.2.4 Testing the PILA Module

Perform the following tests to verify correct PlLA operation as part of the K.ci host interface module
set.
1. Boot the HSC with the off-line diagnostic media. Refer to-Chapter 6 for boot procedures.
NOTE
2. Refer to Chapter 6 for test descriptions and procedures and run the following tests:
• Off-line K test selector test

6. Use the SETSHO command SHOW VIRTUAL_CIRCUITS to ensure that both A and B paths
3.5.3 Removing and Replacing the Port Processor Module (K.pli)

The port processor module (Kpli) is a part of the host interface module set (K.ci). The K.pli
interfaces with the port link module indirectly through the post buffer module.
3.5.3.1 Removing the K.pli Module

Use the following procedure to remove the K.pli module. Observe safety and ESD precautions
available.
On the HSC50, the de power switch is located on the maintenance access panel.
5. Pull the HSC card cage cover up and out.
6. Locate the K pli module in slot number 12 of the card cage. This c.an be verified by the module
utilization label.
7. Remove the K pli module.
3.5.3.2 Setting the Replacement K.pli Module Switches

The Kpli module has factory-set dual in-line pack (DIP) switches to indicate the hardware revision
level. Do not change DIP switch positions, except as directed by a field change order (FCO).
Figure 3-17 shows the Kpli module switch.
I"'" ,"r
~ LEVEL SWITCHES
CXO-2697A
Figure 3-17 L0107 Hardware Rev Level Switch

3.5.3.3 Replacing the K.pll Module

This section provides the Kpli module replacement procedure. Observe safety and ESD precautions
1. Install the K pli module in slot number 12 of the card cage. This can be verified by the module
utilization label.
2. Replace the K.pli module.
3.5.3.4 Testing the K.pll Module

Run the following tests to verify correct K.pli operation as part of the Kci host interface module
set.
NOTE
6. Use the SETSHO command SHOW VIRTUAL_CIRCUITS to and ensure both A and B paths
3.5.4 Removing and Replacing the Disk Data Channel Module (K.sdi)
The Ksdi data channel interfaces between the HSC and the standard disk interface (SDI). Ksdi
operation is controlled by an on-board microprocessor with a local programmed read-only memory
(PROM). Commands and responses pass between the Ksdi microprocessor and other internal HSC
processes through Control memory.
3.5.4.1 Removing the K.sdl Module

Use the following procedure to remove the Ksdi module. Observe safety and ESD precautions
available.
2. Dismount or fallover any drives connected to the HSC.
3. Set the dc power switch to the 0 (off) position.
3-24 Removal,and Replacement Procedures
6. Check the module utilization label above the card cage for the location of the Ksdi module. The
module slots are numbered from right to left when viewed from the front.
7. Remove the K.sdi module.
3.5.4.2 Replacing the K.sdi Module

This section provides the K.sdi module replacement procedure. Observe safety and ESD precautions
1. Check the module utilization label above the card cage for the location of the module. The
2. Replace the K.sdi module.
3.5.4.3 Testing the K.sdi Module

Perform the following tests to verify correct K.sdi operation:
NOTE
The off-linfl. diskette must be write protected. Place a write-protect tab over the
3. Place the Secure!Enable switch in the secure position.
5. Bring the HSC on line by pressing and releasing the the Online switch.
6. Use the SETSHO command SHOW VIRTUAL_CIRCUITS to ensure both A and B paths are
present to all hosts.
7. Use the SETSHO command SHOW DISK to verify that all applicable drives are present.
S. Perform the disk drive integrity test ILDISK. Refer to Chapter 5 for test description and
procedure.
3.5.5 Tape Data Channel Module (K.sti)

The Ksti is an interface between the HSC and the standard tape interface (STI). Ksti operation
is controlled by an on board microprocessor with a local programmed read-only memory (PROM).
Commands and responses pass between the K.sti microprocessor and other internal HSC processes
through control memory.
3.5.5.1 Removing the K.sti Module

Use the following procedure to remove the K.sti module. Observe safety and ESD precautions
available.
2. Dismount or failover any drives connected to the HSC.
7. Remove the Ksti module.
3.5.5.2 Replacing the K.sti Module

This section provides the K.sti module replacement procedure. Observe safety and ESD precautions
2. Replace the K.sti module.
3.5.5.3 Testing the K.sti Module

Perform the following to verify correct K.sti operation:
NOTE

7. Run the tape drive integrity test ILTAPE. Refer to Chapter 5 for test description and procedure.
3.5.6 Removing and Replacing the Data Channel Module (K.si)

The K.si (LOl19-YA) data channel module is a direct replacement for the K.sdi (LOI08-YA) disk data
channel or Ksti (LOI08-YB) tape data channel modules.
NOTE
The K.si data channel initializes only with HSe Version 3.90 or higher software. Do
not use the K.si module with HSe system software with a version level lower than 3.90.
Versions lower than 3.90 will cause initialization failure.
3.5.6.1 Removing the K.sl Module

Use the following procedure to remove the K.si module. Observe safety and ESD precautions before
starting the module removal procedure.
available.
2. Dismount or failover any drives connected to the HSe.
7. Remove the Ksi, Ksdi, or Ksti module.
3.5.6.2 Setting the Replacement K.si Module Switches

The K.si has a switchpack containing four switches. Table 3-1 describes the names and functions
of the switches. SWI is on the top of the switchpack.
NOTE
The four switches must be in the OFF position to prevent errors during initialization and
normal operation.
Table 3-1 K.sl Swltchpack Options

Switch Switch Function Normal
Number Name Description Position
SWI MFG Provides loop on Off

error and single port
extemalloop.
SW2 Burn in Continuous loop. Off
Assumes external
loop and clock.
SW3 Ext loop Loops on all ports; Off
for manufacturing or
field use.
SW4 Ext Substitutes Off
clock external clock; for
manufacturing use
only.
Figure 3-18 shows the Ksi module switchpack.
CXO-2693A
Figure 3-18 K.sl Switch pack
3.5.6.3 Configuration of Requestors While Replacing the K.sl Module

To obtain maximum performance from the HSC, physically configure the tape and disk requestors
according to the following guidelines:
• Each Ksi, Ksdi, or Ksi module in the HSC backplane has a requestor priority number
according to its backplane slot. You can greatly enhance the performance of the HSC if the
attached devices are arranged according to their transfer speeds.
• Attach the slower devices to the lower priority requestors and the faster devices to the higher
priority requestors. Do this while you have the HSC shut down for K.si installation.
• With the introduction of faster devices, such as the TA90 tape drive and RA90 disk drive,
backplane configuration has become especially important. Failure to configure the devices
properly could result in data bus overrun, EDe errors and, significant performance loss.
The following table shows the device relative speeds:
Table 3-2 Physical Configuration

Device Relative Speed
TA90 tape drive Fastest

RA90 disk drive
RA82 disk drive
RA81 disk drive
RA60 disk drive
RA70 disk drive
RA80 disk drive
All STI tape drives except TA90 Slowest
Requestor priority in the HSC ascends from Requestor 2 Oowest priority) through the highest
requestor number on the HSC. Note that requestor priority levels are all relative; that is, the
individual requestor numbers have no intrinsic speed characteristic. When configuring, it is
recommended that you leave blank slots where possible to eliminate reconfiguration when adding
requestors in the future.
NOTE
The lowest priority slot is located next to the K.ci module and the highest priority slot is
located next to the P.ioj/c module.
3.5.6.4 Replacing the K.si Module

This section provides the K.si module replacement procedure. Observe safety and ESD precautions
1. Refer to Table 3-2 and verify that the requestors are arranged according to the transfer speeds
of their attached devices.
3. Replace an existing K.si, K.sdi, or K.sti module with the Ksi. If you are installing a new
requestor, install the K.si in the card cage. Attach the label accompanying the Ksi module to
the appropriate location of the module utilization label charlo
4. If you are planning to run the recommended external loop test, disconnect the SDI cables to
the bulkhead of the requestor slot for the installed Ksi. Instan the loopback connectors (part
number 70-22953-01), and refer to the section on K.si external loop tests.
On the HSC, the dc power switch is located on the side of the RX.33 housing.
3.5.6.5 K.si Module External Loop Test

Run the Ksi External Loop Test before you initialize and configure the Ksi to ensure that there
are no hidden problems with the installation. Loopback connectors (part number 70-22953-01) are
required for this test. These connectors are not included with the HSC.
1. Set SW3 of the K.si switch pack to the ON position. (SW1 is at the top of the switch pack.) All
other switches on the switch pack must be set to OFF.
2. If you have not already done so, open the rear door and remove any SDI cables from the
bulkhead of the requestor slot being tested. The K.si in slot 6, bulkhead E, is being tested in
the example in this section.
3. Install four loopback connectors to the bulkhead.
4. Set the dc power switch to 1 to restore dc power to the card cage.
5. Press and release the Init switch on the operator control panel (OCP) to run the automatic
loopback test as part of the powerup diagnostics. The following system messages appear:
INIPIO-I Booting ...
HSCxx Version V3.90 11-Feb-1988 17:00:41 System HSC009
6. If the HSC shows an error on boot, check that the loopback connectors are seated and that the
Ksi modules are fully seated in the backplane.
7. Remove the system media and insert the off-line diagnostic media into the load device.
8. Press and release the Init switch on the OCP.
The load device drive-in-use LED should light within a few seconds, indicating the bootstrap is
loading the off-line diagnostic loader to program memory.
The off-line diagnostic loader indicates it has been loaded properly by displaying the following:
HSC OFL Diagnostic Loader, Version Vnnn
Radix=Octal,Oata Length=Word,Reloc=OOOOOOOO
OOL>
The off-line loader is now ready to accept commands.

9. Issue the TEST K command and run microdiagnostics 11, 12, 13, and 14. The following example
shows the diagnostics run for requestor 6. Perform these tests for each requestor installed in
your system. The number of passes shown is for this example; you may perform more or less
passes.
OOL> TEST K I RETURN I
The test responds with:

HSC OFL K Test Selector
Requestor "* of K (1 thru 9) [] ? 6 I RETURN I
Test 41= (1 thru 17) (0) [] ? 11 ~
"* of passes to perform (D) [1] ? 10 ~~RE=T=U=RN~
End of Pass "* 0000001, 00000 Errors, 00000 Total Errors
End of Pass "* 0000010, 00000 Errors, 00000 Total Errors

Re-use parameters (YIN) [Y] ? N ~

Requestor -# of K (1 thru 9) [] ? 6 IRETURN I
Test -# (1 thru 17) (O) [] ? 12 ~
-# of passes to perform (D) [1] ? 10 ~IRE=T=U=RN=I
End of Pass -# 0000001, 00000 Errors, 00000 Total Errors

Re-use parameters (YIN) [Y] ? N I RETURN I
Requestor -# of K (1 thru 9) [ ] ? 6 IRETURN I
Test -# (1 thru 17) (0) [ ] ? 13 IRETURN I
-# of passes to perform (D) [1] ? 5 IRETURN I
Re-use parameters (YIN) [Y] ? N IRETURN I
Test -# (1 thru 17) (0) [ ] ? 14 I RETURN I
-# of passes to perform (D) [1] ? 5 I RETURN I
Re-use parameters (YIN) [Y] ? N ~
The following output is an example of a loopback test failure message:

Test -# (1 thru 17) (0) [] ? 11 ~
-# of passes to perform (D) [1] ? 5 ~IRE~T=URN~I
OKTS>00:01 T-#OOO E-#OOS U-OOO
o K Timed-Out During Init
K-Status = 370
10. If you receive a failure message, check that the loopback connectors are securely installed.
Remove the K.si module and try it in a different slot. If these steps fail, replace the K.si. Be
sure to run the tests again after fixing the fault.
11. Set SW3 of the K.si switchpack to the OFF position.
12. Install or replace all SDI cables on the HSC bulkhead.
3.5.6.6 Initializing the K.sl Module

The K.si is by default a disk data channel after initial installation or on first boot of the HSC
software. This default can result in a mismatch between the K.si configuration and the devices
attached to the K.si. The mismatch results in a series of device errors printed on the HSC console,
leading to the mismatched deviceCs) being declared inoperative by the HSC.
This mismatch can happen under the following conditions:
• Mter initial installation of the K.si and the attached devices are tape drives.
• Mter initial installation of HSC software.
• When replacing tape formatters with disk drives after the K.si has been configured for tape
formatters.
• When replacing disk drives with tape formatters after the Ksi has been configured for disk
drives.
3.5.6.7 Correcting K.si Module Configuration Problems

To correct configuration problems, connect the data channel to the proper drive or use the following
procedure to reconfigure the Ksi:
1. Press the Init switch on the OCP. The HSC prints the following message to signify that
initialization has started:
HSCxx Version 3.90 11-Feb-1988 17:00:41 System HSC009
NOTE
The term HSCxx refers to the HSC model that is receiving the K.si module.
2. Mter initialization, use the SETSHO command SHOW REQUESTORS to show the status of the
requestors. In the following example, the Ksi modules are in requestors 2, 3, 4,6, and 8; note
that these modules show up as K.sdi modules.
C'1!P:L/Y
HSCxx> SHOW RZQOBSTORS ~
Req Status Type Version Next Microcode Load
o Enabled P.ioc
1 Enabled K.ci MC- 43 DS- 2 Pila-O K.pli-32
2 Enabled K.sdi MC- 2 DS- 4
5 Enabled Empty
7 Enabled Empty
9 Enabled Empty
SETSHO-I Program Exit
3. Change the configuration of requestors 2 and 8 to tape data channels. Enter reconfiguring
commands as shown, then re-initialize the system:
NOTE
Ensure the system load media is write enabled.
C'1!P:L/Y
HSCxx> RON SETSHO I~Tmml
SETSHO> BRABLI: RBBOOT IRETURN I
SETSHO-S The HSC will reboot on exit.
SETSHO> SET RZQOBSTOR 2/TYPE=TAPE ""~=T=U=RN=
SETSHO> SET RZQOBSTOR 8/TYPE=TAPE ~
SETSHO> EXIT I~TURNI
SETSHO-Q Rebooting HSC, type Y to continue, CTRL/Y to abort: Y
HSCxx Version 3.90 11-Feb-1988 17:00:41 System HSC009
This configuration is retained on the boot media to ensure that the K.si module comes up in the
proper data channel configuration when the HSC is rebooted. If you have a K.si configured as a
tape data channel and you attach disk drives, you can reconfigure it using the above procedure and
specifying TYPE=DISK
4. Use the SHOW REQUESTORS command again to current configuration of the Ksi modules:
CTRL/Y
HSCxx> SROW REQUESTORS I RETURN I
Req Status Type Version Next Microcode Load
o Enabled P.ioc
1 Enabled K.ci MC- 43 DS- 2 Pila-O K.pli-32
2 Enabled K.sti MC- 2 DS- 3
5 Enabled Empty
7 Enabled Empty
8 Enabled K.sti MC-2 DS- 3
9 Enabled Empty
SETSHO-I Program Exit
3.5.6.8 K.si Module New Boot Microcode

When changing to a new boot media, you must power down the HSe to clear the old microcode from
the K.si. A power-down interval of about 15 seconds is adequate. Mter powering up again, reboot
and use the following procedure:
1. Use the SETSHO command SHOW REQUESTORS to check the existing configuration and
verify necessary changes.
SET REQUESTOR n/TYPE=xxxx
where:
n is the requestor number
xxxx is the requestor type (DISK or TAPE)
2. Repeat the command SET REQUESTOR for each Ksi in the HSe to set the HSe configuration
on the new boot media.
Table 3-3 describes the conditions that determine if new Ksi microcode is loaded.
Table 3-3 K.si New Microcode Load Conditions

Action New Microcode Load
Using the ENABLE REBOOT and Yes.

EXIT commands after using the SET
REQUESTOR ntrYPE=xxxx command.
Booting the system after a total power Yes.
failure or after powering down the HSC.
Changing requestor modules. (This action Yes.
requires powering down.)
Using the SET REQUESTOR No.
ntrYPE=xxxx command.
Changing to new boot media. No-The new boot media must be updated with the SETSHO
command SET REQUESTOR ntrYPE=xxxx and the HSC must
be rebooted.
Table 3-3 (Cont.) K.si New Microcode Load Conditions

Action New Microcode Load
Using the SET SCT CLEAR command. N~The new boot media must be updated with the SETSHO
command SET REQUESTOR ntrYPE=xxxx and the HSC must
be rebooted.
Holding in the Fault button while pushing N~The boot media must be updated with the SETSHO
in the Init switch. (This action clears the command SET REQUESTOR nlTYPE=xxxx and the HSC
SCT.) must be rebooted.
3.5.6.9 Testing the K.si Module (After Initialization)

The Ksi module is one of the K requestor interface modules. Perform the following to verify correct
Ksi operation:
NOTE
7. If the K.si is configured as a disk data channel, run the disk drive integrity test ILDISK. Refer
to Chapter 5 for test description and procedure.
8. If the K.si is configured as a tape data channel, run the tape drive integrity test ILTAPE. Refer
to Chapter 5 for test description and procedure.
3.5.7 Removing and Replacing the 1/0 Control Processor Module (P.ioj/c)
The P.ioj module (LOIIIILOIII-YA) uses a PDP-II ISP (J-II) processor. The P.ioc module (LOI05)
uses a PDP-II ISP (F-II) processor. Both contain memory management and memory interfacing
logic. These processors execute their respective HSC internal software.
3.5.7.1 Removing the P.ioj/c Module

Use the following procedure to remove the P.ioj/c module. Observe safety and ESD precautions
1. Press CTRLIY to get the HSC> prompt. Use the SETSHO command SHOW SYSTEM and save
the printout for reference.
available.
4. Set the de power switch to the 0 (off) position.

On the HSC, the de power switch is located on the side of the RX33 housing.
7. Locate the P.ioj/c module in slot number 1 of the card cage. This can be verified by the module
utilization label.
8. Remove the P.iojlc module.
3.5.7.2 Setting the Replacement P.ioj/c Module Jumpers

The P.ioj/c modules have factory-set jumpers. Each module has a unique serial number that
matches the pattern of the jumpers. Do not reconfigure these jumpers.
The P.ioc (L0105) also has a baud rate jumper. When the P.ioc is replaced, set the baud rate jumper
W3 to match the console terminal or duplicate the jumper configuration of the P.ioc being replaced.
Figure 3-19 shows the P.ioc module jumper location.
X149
X158
X157~ tj
CXO-2694A
Figure 3-19 L0105 Baud Rate Jumper

Set the baud rate jumper as follows:
For 9600 baud, leave jumper W3 intact.
For 300 baud, cut one elbow angle on the W3 lead and spread slightly to avoid contact with the
cut edges.
3.5.7.3 Replacing the P.loj/c Module

This section provides P.iojlc module replacement procedure. Observe safety and ESD precautions
before starting module replacement procedures.
1. Install the P.ioj/c module in slot number 1 of the card cage. This can be verified by the module
utilization label.
NOTE
Once VAXNMS recognizes an HSC, the HSC's Online indicator may show that the
HSC is alternately going on line and off line. This is because there is a discrepancy
between the nodename or ID of the HSC and the one recognized by the host for that
HSC. The HSC will be allowed to function only if the old nodename is equal to the
new nodename and the old ID is equal to the new ID.
5. Press CTRLIY to get the HSC> prompt, and issue the SETSHO command SHOW SYSTEM.
Compare this printout with the one saved during removal.
If the printout for the NodenamelID is different, use the SETSHO commands SET NAME or
SET ID to change the NodenamelID so it is the same as in the saved printout.
3.5.7.4 Testing the P.ioj/c Module

When booting the HSC with the off-line diagnostic media, the P.iojlc ROM bootstrap verifies the
basic integrity of the P.iojlc module.
Perform the following to verify correct P.ioj operation:
NOTE
• Off-line cache test (P.ioj only)
3.5.8 Removing and Replacing the HSC Memory Module (M.std2)

Memory module M.std2 (LOl17) is used in the HSC only. It contains three independent systems
memories, each residing on a different bus in the HSC. In addition, the memory module contains
the RX33 diskette controller.
3.5.8.1 Removing the M.std2 Module

Use the following procedure to remove the M.std2 module. Observe safety and ESD precautions
available.
3. Set the dc power switch to the 0 (of!) position.

6. Locate the M.std2 module in slot number 2 of the card cage. This can be verified by the module-
utilization label.
7. Remove the M.std2 module.
3.5.8.2 Replacing the M.std2 Module

This section provides the M.std2 module replacement procedure. Observe safety and ESD
CAUTION
The switch pack on the M.std2 module is factory set to calibrate the RX33 diskette
controller. Do not change the setting of this switch pack; the switch settings are unique
to each module and cannot be restored outside of the manufacturing environment.
1. Install the M.std2 module in slot number 2 of the card cage. This can be verified by the module
utilization label.
2. Replace the M.std2 module.
6. Press CTRL/C to get the HSC> prompt.
7. Issue the SET MEMORY ENABLE ALL command.
8. Mter the HSC reboots, type the command SHOW MEMORY. Check that the available memory
is equal to the maximum memory, except for 32 (decimal) words, which are disabled for lock
functionality.
3.5.8.3 Testing the M.std2 Module

Perform the following to verify correct M.std2 operation:
NOTE
• Off-line memory test
• Off-line refresh test
4. Boot the HSC with the system media by pressing and releasing the lnit switch.
6. Use the SETSHO command SHOW VIRTUAL_CIRCUITS and ensure both A and B paths are
7. Run the memory integrity test ILMEMY and refer to Chapter 5 for test description and
procedure.
3.5.9 Removing and Replacing the HSC50 Memory Module (M.std)

The HSC50 memory module (LOI06) contains three separate and independent systems memories,
each residing on a different bus within the HSC50.
3.5.9.1 Removing the M.std Module

Use the following procedure to remove the M.std module. Observe safety and ESD precautions
available.
6. Locate the M.std module in slot number 2 of the card cage. This can be verified by the module
utilization label.
7. Remove the M.std module.
3.5.9.2 Replacing the M.std Module

This section provides the M.std module replacement procedure. Observe safety and ESD
1. Install the M.std module in slot number 2 of the card cage. This can be verified by the module
utilization label.
2. Replace the M.std module.
6. Press CTRL/C to get the HSC> prompt.
7. Issue SET MEMORY ENABLE ALL command.
S. Mter the HSC50 reboots, type the command SHOW MEMORY. Check that the available
memory is equal to the maximum memory, except for 32 (decimal) words, which are disabled for
lock functionality.
3.5.9.3 Testing the M.std Module

Run the following tests to verify correct M.std operation.
NOTE
2. Refer to the Chapter 6 for test descriptions and procedures and run the following tests:
• Off-line memory test
• Off-line refresh test
7. Run the memory integrity test ILMEMY and refer to Chapter 5 for test description and
procedure.
3.6 Removing and Replacing Subunits

This section contains procedures for removing and replacing subunits.
WARNING:
Because hazardous voltages exist inside the HSC, service must be performed only by
qualified people. Bodily injury or equipment damage can result from improper servicing
procedures.
3.6.1 Removing and Replacing the RX33 Disk Drive

Two RX33 disk drives are used to load the HSC software or off-line diagnostics. The RX33 disk
drives are mounted in the HSC cabinet. A cover plate ensures proper air flow and cooling. When
removing and replacing the RX33, avoid snagging the cables attached to the rear of the drive. After
replacing an RX33, always replace the cover plate.
3.6.1.1 Removing the RX33 Disk Drive

Use the following procedure to remove the RX33 disk drive:
available.
3. Turn off the de power switch, located on the side of the RX33 housing (Figure 3-20).
HSC70
DC POWER
SWITCH
OCP SIGNAL!
POWER LINE
CONNECTOR
Figure 3-20 HSC DC Power Switch

4. Rotate the four fasteners on the RX33 cover plate one-quarter turn and remove the cover plate
(Figure 3-21).
QUARTER-TURN
FASTENER
DRIVE
COVER
PLATE
o
CXO-1118A
Figure 3-21 Removing the RX33 Cover Plate

5. Loosen the two captive screws holding the drive assembly and mounting plate to the cabinet
frame.
CAUTION
Avoid snagging the cables attached to the rear of the drives during the next step_
6. Carefully slide the drive assembly out until the housing is cleared.
7. Support the drive assembly with one hand and remove the fiat ribbon cables and power cables
from the rear of the drives.
8. Determine whether drive 0 or drive 1 should be replaced.
9. Loosen the captive screws on the drive to be replaced and remove the drive from the drive
assembly (Figure 3-22).
CAPTIVE
SCREW
RX33
MOUNTING
PLATE
Figure 3-22 RX33 Disk Drive Removal
3.6.1.2 Setting the RX33 Disk Drive Jumpers

Replacement RX33 drives are not configured for the HSC. Two identical jumpers (part number 12-
18783-00) must be added. If no extra jumpers are available, remove the jumpers from the defective
drive. Correct jumper configuration is necessary for the operation of the replacement RX33 drive.
If replacing drive 0, be sure to insert jumper DSO. If replacing drive 1, be sure to insert jumper
DSl.
The RX33 module may be revision A1 or A3. Table 3-4 shows the jumper differences and
configurations for both revisions when the drive is used in an HSC.
Table 3-4 RX33 Jumper Description

Rev A! RevA3 Status
Name Name Description In or Out
FG FG Frame ground In
HG Hi gain In
LG Lo gain Out
I SI Speed, mode 1 Out
Dual speed
II II Speed, mode 2 In
360 RPM only
DSO DO Drive select 0 In to select drive 0
DS1 D1 Drive select 1 In to select drive 1
DS2 D2 Drive select 2 Out
DS3 D3 Drive select 3 Out
U1 UO Selects mode of operation In
for loading the heads and
lighting the bezel LED (See
note)
U2 U1 See Ul/UO above In
HL HL Not applicable to HSC use Out
IU IV Not applicable to HSC use Out
ML Motor enable Out
RE Recalibration Out
DC DC1 Disk changed on pin 34 Out
DC2 Factory setting In
DC3 Not applicable to HSC use Out
DC4 Not applicable to HSC use Out
RY RY Ready on pin 34 In
NOTE
The HSe loads the heads and lights the drive-in-use LED when the DRIVE SELECT n
and READY signals are both true.
Figure 3-23 shows the jumper locations for RX33 with a revision Ai module.
POWER
CONNECTOR
LG HG
~
~
II I
DRIVE SELECT
DSO
JUMPERS
DS1
DS2
DS3
•
•
•
EDGE • RE
CONNECTOR • DC
'--_--' 20 00 RY
L..-_--'30
L..-_--'32
o
MFD CONTROL RESISTOR

BOARD TERMINATION
PACK (INSTALLED)
I. .1 INDICATES JUMPER INSTALLED

CXO-2699A
Figure 3-23 Revision A1 Jumper Configurations

Figure 3-24 shows the jumper locations for RX33 with a revision A3 module.
POWER
CONNECTOR
~II IS
KEY IL
DRIVE SELECT
U1
JUMPERS
UO
DO
D1
D2
D3
EDGE
10
I: I OLE
CONNECTOR RY ML IU HL HS
20
IWI:I:I:I:I
Ie •• 11'e e' el
DC4 DC3 DC2 DC1
30
32
0
MFD CONTROL RESISTOR

BOARD TERMINATION
PACK (INSTALLED)
,- -, INDICATES
JUMPER
INSTALLED
CXO-2700A
Figure 3-24 Revision A3 Jumper Configurations

3.6.1.3 Replacing the RX33 Disk Drive

Use the following procedure to replace the RX33 disk drive assembly:
1. Replace the drive in the drive assembly and tighten the drive captive screws.
2. Support the drive assembly with one hand and attach the flat ribbon cables and power cables to
the rear of the drives.
3. Carefully slide the drive assembly into the drive housing.
4. Replace the cover plate and tighten the captive screws holding the drive assembly.
5. Replace the ac plug in the wall socket and place the main power switch on the power controller
in the on position.
6. The drive-in-use LED lights while the drive is accessed and extinguishes after the drive access
is completed or after the drive motor stops.
3.6.1.4 Testing the RX33 Disk Drive

Mter replacing the E,X.33, use the following procedure to test the drive:
NOTE
2. Refer to Chapter 6 for a test description and procedure and run the RX33 off-line exerciser
OFLRXE.
7. Refer to Chapter 5 for test description and procedure and run the RX33 device integnty test
ILRX33.
3.6.2 Removing and Replacing the TU58 Tape Drive

Two TU5S tape drives are used to load the HSC50 software or off-line diagnostics. The TU5S tape
drives are mounted on the rear of the HSC50 front door.
CAUTION
When servicing the TU58, avoid bending the tachometer disk mounted on the drive
motor shaft. If the disk is bent but not creased, it may be straightened. If it cannot be
straightened or if it is creased, the TU58 must be replaced. The disk should not rub
against the optical sensor block or dangling wires.
3.6.2.1 Removing the TU58 Tape Drive

Use the following procedure to remove the TU5S tape drive:
available.
3. Remove the maintenance access panel cover by loosening the four captive screws.
3--46 Removal and Replacement Procedures
4. Turn off the de power switch (Figure 3-25).
DC POWER
SWITCH TU58
ON POSITION (1) CONNECTORS
OFF POSITION (0)
OCP
CONNECTOR
CONNECTORS
RESERVED FOR
FUTURE USE
MAINTENANCE
TERMINAL
SIGNAL
CONNECTOR
MAINTENANCE
ACCESS PANEL
CXO-014B
Figure 3-25 HSC50 DC Power Switch

5. Remove the two locknuts on the bottom of the TU58 bezel assembly (Figure 3-26).
• --- -- •
11/32 NUT
DRIVER
Figure 3-26 Removing the HSC50 TU58 Bezel Assembly

6. Push the bezel assembly up about 1 inch to clear the mounting hooks from their slots.
7. Pull the bezel assembly back 3 to 4 inches from the door for clearance.
8. Support the bezel assembly with one hand and disconnect J3 and J4 from the OCP
(Figure 3-27).
J3
(20 PINS)
OPERATOR
CONTROL
PANEL PCB
PHILLIPS
SCREWS (4)
.-
Figure 3-27 Disconnecting the HSC50 OCP Cables

9. Disconnect the cables from the TU58 controller module (Figure 3-28).
SECURE/
DRIVE 1 DRIVE 0 ENABLE
MECHANICS MECHANICS SWITCH
HEAD
COVER
CONTROLLER
MODULE
(PARTIALLY
PULLED OUT)
POWER BAUD RATE

CONNECTOR" JUMPERS
(FACTORY SET)
OPERATOR OPERATOR
CONTROL CONTROL
PANEL PANEL
CONNECTOR CONNECTOR
MAINTENANCE
ACCESS PANEL
CONNECTORS
.. CAUTION: CONNECTOR CAN BE REVERSED. OBSERVE PIN USAGE.
Figure 3-28 Disconnecting the HSC50 TU58 Controller Cables
NOTE
The head cover connector shown upper left in Figure 3-28 should be removed during
operation.
10. Slide the TU58 controller module out of the plastic guides.
3.6.2.2 Setting the TU58 Tape Drive Jumpers

The TU58 baud rate jumpers are factory set. Ensure the baud rate jumper setting on the new
module is the same as on the module being replaced.
Figure 3-29 shows the TU5S jumper location.
~ ••••••••••••
L:.:..:..:J •••••••••••• BAUD RATE
JUMPERS
(FACTORY SET)
SELF-TEST
INDICATOR
CXO-2692A
Figure 3-29 TU58 Baud Rate Jumpers
3.6.2.3 Replacing the TU58 Tape Drive

Following is the procedure for replacing the TU5S tape drive:
CAUTION
When servicing the TU58, avoid bending the tachometer .disk mounted on the drive
motor shaft. If the disk is bent but not creased, it may be straightened. If it cannot be
straightened or if it is creased, the TU58 must be replaced. The disk should not rub
against the optical sensor block or dangling wires.
1. Slide the controller module into the housing on the plastic guides.
2. Connect the cables to the TU5S controller module.
3. Support the bezel assembly with one hand and connect J3 and J4 to the OCP.
4. Attach the bezel assembly to the mounting hooks.
5. Replace the two locknuts on the bottom of the TU5S bezel assembly.
6. Replace the ac plug in the wall socket and place the main power switch on the power controller
in the on position.
3.6.2.4 Testing the TU58 Tape Drive

After replacing the TU58, use the following procedure to test the drive:
4. Use the SETSHO command SHOW VIRTUAL_CIRCUITS and ensure both A and B paths are
5. Run the TU58 device integrity test ILTU58. Refer to Chapter 5 for a description and procedure
for this test.
3.6.3 Removing and Replacing the HSC Operator Control Panel (OCP)
If any OCP lamp fails, replace the entire OCP.
3.6.3.1 Removing the HSC OCP

Use the following procedure to remove the HSC OCP:
1. Open the front door by turning the key clockwise and lifting the latch.
2. Turn off the dc power switch located on the side of RX33 housing.
3. Remove the four Kepnuts securing the OCP shield to the studs on the front door.
4. Remove the OCP shield.
5. Remove the four screws securing the OCP to the shield (Figure 3-30).
KEPNUTS
OCP CABLE
~
OCP "-
MOUNTING
SCREWS
INSIDE
FRONT
DOOR
CXO-938A
Figure 3-30 Removing the HSC OCP

6. Remove the two connectors from the printed circuit board on the OCP.
7. Pullout the OCP, carefully allowing for indicator and switch clearance.
3.6.3.2 Replacing the HSC OCP

Following is the procedure for replacing the OCP:
1. Replace the two connectors from the printed circuit board on the OCP.
2. Secure the OCP to the shield using the four screws that were removed.
3. Replace the OCP shield.
4. Replace the four Kepnuts securing the OCP shield to the studs on the front door.
5. Turn on the dc power switch located on the side of RX33 housing.
6. Close and secure the front door by turning the key counter-clockwise.
3.6.3.3 Testing the HSC OCP

Mter replacement, use the following procedure to test the OCP:
NOTE
2. Run the off-line OCP test. Refer to Chapter 6 for a test description and procedures.
3.6.4 Removing and Replacing the HSC50 Operator Control Panel (OCP)
OCP indicators are not field replaceable. If any lamp fails, replace the entire OCP.
3.6.4.1 Removing the HSC500CP

Use the following procedure to replace the HSC50 OCP:
1. Open the front door by turning the key clockwise.
2. Remove dc power.
3. Remove the TU58s (Section 3.6.2).
4. Remove J3 and J4 from the OCP (Figure 3-31).
5. Remove the four screws from the OCP (Figure 3--31).
6. Carefully pullout the OCp, allowing for indicator and switch clearance.
J3
(20 PINS)
OPERATOR
CONTROL
PANEL PCB
PHILLIPS
SCREWS (4)
Figure 3-31 Removing the HSC50 OCP

3.6.4.2 Replacing the HSC50 OCP

Following is the procedure for replacing the OCP:
1. Carefully replace the OCp, allowing for indicator and switch clearance.
2. Replace the four OCP screws (refer to Figure 3-31).
3. Connect J3 and J4 to the OCP (refer to Figure 3-31).
4. Replace the TU58s (Section 3.6.2).
5. Turn on the dc power.
6. Close the front door by turning the key counterclockwise.
3.6.4.3 Testing the HSC500CP

Mter replacement, use the following procedure to test the OCP:
NOTE
2. Run the off-line OCP test. Refer to Chapter 6 for a test description and procedure.
3.6.5 Removing and Replacing the HSC Airflow Sensor Assembly

Use the following procedure to remove and replace the HSe Airflow Sensor:
1. Open the back door using a 5/32-inch hex wrench.
2. Turn off the ac circuit breaker (CBl) on the 881 power controller (Figure 3--32).
3. Disconnect J70 (Figure 3-33).
4. Remove the Phillips head screw that holds the mounting clamp to the duct (Figure 3-33).
(f)
~
o (f)
:rn rn rn m:
JI3JI2J11 J10
(;) OD
o CIRCUIT
BREAKER
POWER
CONNECTOR
CXO-1117A
Figure 3-32 881 Power Controller Circuit Breaker

PHILLIPS
SCREW
SENSOR
CLAMP
Figure 3-33 Removing and Replacing the HSC Airflow Sensor Assembly
5. Slide the sensor assembly out of the duct.
6. Reverse the removal procedure to replace the airflow sensor assembly. Align the slots in the
airflow sensor tip horizontally with the floor. Mter turning on ac power to the HSC, test the
new airflow sensor for proper operation by blocking the flow of air.
3.6.6 Removing and Replacing the HSC50 Airflow Sensor Assembly

Use the following procedure to remove and replace the HSC50 airflow sensor assembly:
2. Turn off the ac circuit breaker (CBI) on the HSC50 power controller (Figure 3-34).
DELAYED
LINE PHASE
LINE
INDICATOR
POWER
CIRCUIT
BREAKERS
CB1
CB2-4
(SWITCHED)
~:::-1--~:;;;:::;:::::;t
FUSE
CBS
(UNSWITCHED)
CXO-013B
Figure 3-34 HSC50 Power Controller Circuit Breaker

3. Disconnect J70 (Figure 3-35).

4. Remove the Phillips head screw that holds the mounting clamp to the duct (Figure 3-35).
~~: ~;r :~':"': \,', :, :;,:,': ~ ,,:;.::.:~.::.:,: ;:

" ...
AIRFLOW
SENSOR . t
I •• ,
" :. :::~
. " .
" . . ..: ':.
Figure 3-35 Removing and Replacing the HSC50 Airflow Sensor Assembly
5. Slide the sensor assembly out of the duct.
6. Reverse the removal procedure to replace the airflow sensor assembly. Align the slots in the
airflow sensor tip horizontally with the :floor. Ensure sensor operability by blocking the :flow of
air. Pinching the sensor should trip CB1.
3.6.7 Removing and Replacing the HSC Blower

The blower, which provides forced air cooling for the cabinet, is removed and replaced with the
2. Turn off the ac circuit breaker (CBl) on the 881 power controller (Figure 3-36).
(f)
~
o G)
:rn mmm:
JI3JI2J11 J1 0
@OD
o CIRCUIT
BREAKER
POWER
CONNECTOR
CXO-1117A

3. Disconnect the blower power connector (Figure 3-37).

4. Disconnect the airflow sensor power connector (J70) to allow removal of the exhaust duct
(Figure 3-37).
5. Remove the exhaust duct from the bottom of the blower by lifting up the quick release latches
on each side of the duct (Figure 3-37).
PHILLIPS
SCREWS (3)
(SECURE BLOWER
MOUNTING BRACKET)
REMOVABLE
EXHAUST
DUCT
COOLING
BLOWER
POWER
CONNECTOR
AIRFLOW SENSOR AIRFLOW

POWER CONNECTOR SENSOR
(J70)
Figure 3-37 Removing and Replacing the HSC Main Cooling Blower
6. Loosen, but do not remove, the three Phillips screws holding the blower mounting bracket to
the cabinet.
7. Lift the blower and bracket up and out of the cabinet.
8. Reverse the removal procedure to replace the cooling blower.
3.6.8 Removing and Replacing the HSC50 Blower

The blower, which provides forced air cooling for the cabinet, is removed and replaced with the
2. Turn off ac power (CEl on the power controller) (Figure 3-38).
DELAYED
LINE PHASE
LINE
INDICATOR
POWER
CIRCUIT
BREAKERS
CB1
CB2-4
(SWITCHED)
~:::-i--~;;;:::;::::::;-t
FUSE
CBS
(UNSWITCHED)
CXO-013B

3. Disconnect the blower power connector (Figure 3-39).

4. Disconnect the airflow sensor power connector (J70) to allow removal of the exhaust duct
(Figure 3-39).
5. Remove the exhaust duct from the bottom of the blower by lifting up the quick release latches
on each side of the duct (Figure 3-39).
6. Loosen, but do not remove, the three Phillips screws holding the blower mounting bracket to
the cabinet (Figure 3-39).
PHILLIPS
SCREWS (3)
(SECURE BLOWER
MOUNTING BRACKET)
REMOVABLE
EXHAUST
DUCT
COOLING
BLOWER
POWER
CONNECTOR
AIRFLOW SENSOR QUICK-

AIRFLOW
POWER CONNECTOR RELEASE
SENSOR
(J70) LATCHES
Figure 3-39 Removing and Replacing the HSC50 Blower

7. Lift the blower and bracket up and out of the cabinet.
S. Reverse the removal procedure to replace the blower.
3.6.9 Removing and Replacing the 881 Power Controller

The power controller must be removed to replace a power supply.
Use the following procedure to remove and replace the 881 power controller:
2. Remove rear door latch to allow clearance for power controller removal.
3. Remove ac power by placing CBl in the off position (Figure 3-40).
(t)
~
o (t)
:rn rn rn m:
JI3JI2J11 J1 0
(!)O[J
o CIRCUIT
BREAKER
POWER
CONNECTOR
CXO-1117A

4. Unplug the power controller from the power source (Figure 3-41).
5. Remove the two top screws and then the two bottom screws securing the power controller to the
cabinet (Figure 3-41). While removing the two bottom screws, push up on the power controller
to take the weight off the screws.
MAIN POWER COOLING

SUPPLY LINE BLOWER
CORD LINE CORD
01
PHASE DIAGRAM
AUXILIARY
POWER SUPPLY
LINE CORD
POWER
~ ....-=-I CONTROLLER
SCREWS
POWER
CONTROLLER
LINE CORD
CXO-941C
Figure 3-41 Removing and Replacing the 881 Power Controller
CAUTION
Do not pull the power controller out too far because cables are connected to the back
and top_
6. Pull the power controller towards you and then out.

7. Remove the power control bus cables from connectors JI0, JII, J12, and JI3 at the front of the
power controller (Figure 3-40).
8. Disconnect the total off connector at the rear of the power controller (Figure 3-42).
TOTAL OFF
CONNECTOR
CXO-934A
Figure 3-42 881 Total Off Connector

9. Disconnect all line cords from the top of the power controller.
NOTE
Be sure to rotate the line cord elbow to the vertical position if replacing a defective
power controller with a new one. To rotate the elbow, remove the set screw, rotate
the elbow to the position shown in Figure 3-40, and replace the set screw in the other
hole.
10. Reverse the removal procedure to replace the power controller.
NOTE
To ensure proper phase distribution, reconnect the main power supply, auxiliary
power supply, and cooling blower line cords as shown in Figure 3-41.
3.6.10 Removing and Replacing the HSC50 Power Controller

The HSC50 power controller must be removed to replace a power supply.
Use the following procedure to remove and replace the HSC50 power controller:
2. Remove rear door latch to allow clearance for power controller removal.
3. Turn off ac power (CBl on the power controller) (Figure 3-43).
DELAYED
LINE PHASE
LINE
INDICATOR
POWER
CIRCUIT
BREAKERS
CB1
CB2-4
(SWITCHED)
~::::-t--~;::::;:::~
FUSE
CBS
(UNSWITCHED)
CXO-013B

4. Unplug the power controller from the power source.
5. Remove the two top screws and then the two bottom screws securing the power controller to the
cabinet (Figure 3-44). While removing the two bottom screws, push up on the power controller
to take the weight off the screws.
COLLING
BLOWER
LINE CORD
MAIN
POWER
SUPPLY
, ..
! •••
." . ...
......
- .. ,
.'.:
• - I'
. '~
'. .
"
~ .., '.'
MAIN POWER
SUPPLY
LINE CORD
::'~;{?~';::" .
AUXILIARY
POWER
SUPPLY
CONNECTORS
J1,J2, J3
AUXILIARY
POWER SUPPLY
LINE CORD
POWER
CONTROLLER
SCREWS
POWER
CONTROLLER
LINE CORD
Figure 3-44 Removing and Replacing the HSC50 Power Controller
CAUTION
Do not pull the power controller out too far because cables are connected to the back
and top.
6. Pull the power controller towards you and then out.
7. Remove the power control bus cables from connectors Jl, J2, and J3 at the front of the power
controller (Figure 3-43).
9. Reverse the removal procedure to replace the power controller.
3.6.11 Removing and Replacing the HSC Main Power Supply

Use the following procedure to remove and replace the HSC main power supply:
WARNING
The power supply is heavy. Support it with both hands to avoid dropping it.
(±)
~
o (±)
:rn rn rn rn:
JI3JI2J11 J1 0
000
o CIRCUIT
BREAKER
POWER
CONNECTOR
CXO-1117A

4. Remove the front door.
5. Remove the power controller (Section 3.6.9) to access the back of the power supply.
6. Unplug the main power supply line cord at the power controller.
NOTE
While performing 7 through 15, refer to (Figure 3-46).
7. Remove the nut from the -VI stud (ground) on the back of the power supply.
8. Remove the nut from the +VI stud (+5 volts) on the back of the power supply.
9. Remove the nut from the - V2 (ground) stud on the back of the power supply.
10. Remove the nut from the +V2 (-5.2 volts) stud.
11. Unplug J31 (+12 VDC output from the supply to backplane, power fail, and -5 volts sense line).
12. Unplug P32 (+12 VDC sense line and +5 VDC sense line).
13. Unplug J33 (to dc power switch).
14. Unplug J34 (remote on/off jumper to auxiliary power supply).
15. Unplug J35 (+12 VDC power to the airflow sensor).
Figure 3-46 shows the HSC main power supply test points.
WIRE LIST
COLOR POSITION SIGNAL COLOR POSITION SIGNAL
PURPLE TBI-3-5 12 V PURPLE TB1-3-1 12 V SENSE
PURPLE TBI-3-6 12 V BLUE TB1-2-7 ACC
BLACK BROWN TB1-2-6 AC
TBI-3-3 GND (12 V)
BLACK GRN/YEL TB1-2-5 GND
ORANGE TBI-2-2 -5 V SENSE YELLOW TB1-2-3 ON/OFF (-5, 3 V)
BLACK TBI-2-1 GND (-5 V SENSE) ORANGE TB1-2-2 -5 V SENSE (52-)
BROWN TBI-1-4 POWER FAIL BLUE TB1-1-3 ON/OFF 5 V
BLACK TBI-1-2 GND (5 V SENSE) BLACK TB1-1-2 GND (5 V SENSE)
RED TB1-1-1 5 V SENSE PURPLE TB1-3-2 12 V
BLACK TB1-3-4 GND (12 V SENSE)
MAIN POWER SUPPLY - REAR VIEW
POWER FAIL
LINE CORD
CONN ECTIONS
J33 DC
POWER
@ SWITCH
J34 AUXILIARY CONNECTOR
POWER SUPPLY
CONNECTOR
~ TO BACKPLANE
FLEXBUS
CXO-942B_S
Figure 3-46 HSC Main Power Supply Cables and Test Points
16. Turn the four captive screws on the front of the power supply counterclockwise (Figure 3-47).
MAIN POWER
SUPPLY CABLES
CAPTIVE
SCREWS
CXO-1157A
Figure 3-47 Removing and Replacing the HSC70 Main Power Supply
17. Pull the power supply out about an inch. Check the back of the cabinet to ensure the cables
and flexbus connectors are clear and will not snag when the supply is completely removed.
18. Carefully pull the power supply all the way out of the cabinet.
19. Remove the power cord from the failing unit and install it on the new power supply.
NOTE
Spare power supplies are not shipped with a power cord.
20. Reverse the removal procedure to replace the main power supply.
3.6.12 Removing and Replacing the HSC50 Main Power Supply

Use the following procedure to remove and replace the HSC50 main power supply:
WARNING
The power supply is heavy. Support it with both hands to avoid dropping it.
DELAYED
LINE PHASE
LINE
INDICATOR
POWER
CIRCUIT
BREAKERS
CB1
CB2-4
(SWITCHED)
~::::-+--,W~~
FUSE
CBS
(UN SWITCHED)
CXO-013B

5. Remove the power controller (Section 3.6.10) to access the back of the power supply.
6. Unplug the main power supply line cord at the power controller.
NOTE
While performing 6 through 14, refer to Figure 3-49, which shows the HSC50 main
power supply test points.
7. Remove the nut from the -V1 stud (ground) on the back of the power supply (Figure 3-49).
8. Remove the nut from the +V1 stud (+5 volts) on the back of the power supply (Figure 3-49).
9. Remove the nut from the -V2 (ground) stud on the back of the power supply (Figure 3-49).
10. Remove the nut from the +V2 (-5.2 volts) stud on the back of the power supply (Figure 3-49).
11. Unplug J31 (+12 VDC output from the supply to backplane, power fail, and -5 volts sense line)
(Figure 3-49).
12. Unplug P32 (+12 VDC sense line and +5 VDC sense line) (Figure 3-49). Ensure the P32 cable
is free to be removed with the power supply.
13. Unplug J33 (to dc power switch) (Figure 3-49).
14. Unplug J34 (remote on/off jumper to auxiliary power supply) (Figure 3-49).
15. Unplug J35 (+12 VDC power to the airflow sensor) (Figure 3-49).
MAIN POWER SUPPLY - REAR VIEW BACKPLANE
POWER TO MAIN POWER

AIRFLOW SUPPLY
SENSOR CONNECTORS
AUXILIARY
POWER SUPPLY
CONNECTORS
LINE CORD
CONNECTIONS
BLACK WIRES
FROM BACKPLANE (4)
Figure 3-49 HSC50 Main Power Supply Cables and Voltage Test Points
16. Turn the four captive screws·on the front of the power supply counterclockwise (Figure 3-50).
MAIN POWER
SUPPLY CABLES
CAPTIVE
SCREWS
CXO-02SB
Figure 3-50 Removing and Replacing the HSC50 Main Power Supply
17. Pull the power supply out about an inch. Check the back of the cabinet to ensure the cables are
clear and will not snag when the supply is completely removed.
18. Carefully pull the power supply all the way out of the cabinet.
19. Remove the power cord from the failing unit and install it on the new power supply.
NOTE
Spare power supplies are not shipped with a power cord.
20. Reverse the removal procedure to replace the HSC50 main power supply.
3.6.13 Removing and Replacing the HSC Auxiliary Power Supply

An HSC requires an auxiliary power supply if the total module count in the card cage is more than
eight. The auxiliary power supply is mounted directly beneath the main power supply.
Use the following procedure to remove and replace the HSC auxiliary power supply:
WARNING
This power supply is heavy. When removing the power supply, support it with both
hands to avoid dropping it.
2. Turn off ac power (CB1) on the power controller (Figure 3-51).
(t)
~
o (t)
:rn rn rn m:
JI3JI2J11 J10
000
o CIRCUIT
BREAKER
POWER
CONNECTOR
CXO-1117A

5. Remove the power controller to access the back of the power supply (Section 3.6.9).
6. Unplug the auxiliary power supply line cord at the power controller.
NOTE
While performing 7 through 10, refer to Figure 3-52.
7. Remove the nut from the +V1 stud (+5 volt) on the back of the power supply.
8. Remove the nut from the -VI stud (ground) on the back of the power supply.
9. Disconnect J50 (sense line to voltage comparator).
10. Disconnect J51 (dc on/off jumper).
WIRE LIST
COLOR POSITION SIGNAL
BLACK TBI-2 GROUND (5 V SENSE)
RED TBI-1 5 V SENSE
BROWN TBI-4 POWER FAIL
BLUE TBI-7 ACC
BROWN TBI-6 AC
GRN/YEL TBI-5 CHASSIS GROUND
BLUE TBI-3 ON/OFF
AUXILIARY POWER SUPPLY - REAR VIEW
POWER SUPPLY
TERMINAL STRI~
J51
TO BACKPLANE
J50
TO MAIN
POWER SUPPL Y _ _........1
LINE CORD
TO POWER
CONTROLLER
CXO-943B
Figure 3-52 HSC Auxiliary Power Supply Cable and Test Points
11. Figure 3-52 shows the HSC auxiliary power supply test points.
12. Turn the four captive screws on the power supply counterclockwise (Figure 3-53).
AUXILIARY POWER
SUPPLY CABLES
CAPTIVE
SCREWS
AUXILIARY POWER
SUPPLY GUIDANCE
TRACK AUXILIARY POWER
SUPPLY
CXO-1158A
Figure 3-S3 Removing and Replacing the HSC Auxiliary Power Supply
and flexbus connectors are clear.
14. Carefully slide the power supply out through the front of the HSC.
15. Remove the power cord from the failing unit and install it to the new power supply.
NOTE
Spare supplies are not shipped with a power cord.
16. Reverse the removal procedure to replace the HSC auxiliary power supply.
3.6.14 Removing and Replacing the HSC50 Auxiliary Power Supply

An HSC50 requires an auxiliary power supply if the total module count in the card cage is more
than eight. The auxiliary power supply is mounted directly beneath the main power supply.
Use the following procedure to remove and replace the HSC50 auxiliary power supply:
WARNING
This power supply is heavy. When removing the power supply, support it with both
hands to avoid dropping it.

2. Turn off the ac circuit breaker (CB1) on the HSC50 power controller (Figure 3-54).
DELAYED
LINE PHASE
LINE
INDICATOR
POWER
CIRCUIT
BREAKERS
CB1
CB2-4
(SWITCHED)
FUSE
CBS
(UNSWITCHED)
CXO-013B

5. Remove the power controller to access the back of the power supply (Section 3.6.10).
6. Unplug the auxiliary power supply line cord at the power controller.
NOTE
While performing 7 through 10, refer to Figure 3-55.
7. Remove the nut from the +VI stud (+5 volt) on the back of the power supply (Figure 3-55).
8. Remove the nut from the -VI stud (ground) on the back of the power supply (Figure 3-55).
9. Disconnect J50 (sense line to voltage comparator). (Figure 3-55).
10. Disconnect J51 (dc on/off jumper) (Figure 3-55). Refer to Figure 3-55 for the HSC50 auxiliary
power supply test points.
WIRE LIST
COLOR POSITION SIGNAL
RED TBI-1 5 V SENSE
BROWN TBI-4 POWER FAIL
BLUE TBI-7 ACC
BROWN TBI-6 AC BACKPLANE
GRN/YEL TBI-5 CHASSIS GROUND

BLUE TBI-3 ON/OFF
OUTSIDE
BLACK TBI-2 GROUND (5 V SENSE) BACKPLANE BUS
TO MAIN
POWER SUPPLY
INSIDE
BACKPLANE BUS
POWER FAIL
POWER SUPPLY
TERMINAL STRIP
J51
TO BACKPLANE
J50
TO MAIN
POWER SUPPLY ---&ao-/
+5V DC
GROUND
LINE CORD
TO POWER
CONTROLLER
CXO-027C
Figure 3-55 HSC50 Auxiliary Power Supply Cable and Voltage Test Points
11. Turn the four captive screws on the power supply counterclockwise (Figure 3-56).
MAIN
POWER
SUPPLY
CAPTIVE
SCREWS
AUXILIARY POWER
SUPPLY GUIDANCE
TRACK
AUXILIARY
POWER
SUPPLY
Figure 3-56 Removing and Replacing the HSC50 Auxiliary Power Supply
and connectors are clear.
13. Carefully slide the power supply out through the front of the HSC50.
14. Remove the power cord from the failing unit and install it to the new power supply.
NOTE
Spare supplies are not shipped with a power cord.
15. Reverse the removal procedure to replace the auxiliary power supply.
Initialization Procedures 4-1
4
Initialization Procedures
4.1 Introduction
This chapter contains procedures for connecting the console terminal on the HSC and the auxiliary
terminal on the HSC50, and initialization procedures for both HSC models.
A malfunction during initialization may be reported by a fault code displayed on the operator
control panel (OCP). These fault codes are explained in Chapter 8.
4.2 Console/Auxiliary Terminal

The console or auxiliary terminal designated for the HSC can be a VT2xx, VT3xx, VTlxx, or an
LA12 DECwriter. An LA75 or LA50 printer for hardcopy output is connected to the VT2xx, VT3xx,
and can be connected to the VT1xx if the VTlxx has the printer port option installed. Detailed
operating information is provided in the appropriate owner manuals accompanying the VTxxx and
LAxx models.
NOTE
The VT3xx series terminal can be connected to an RS-232 compatible port only.
Connection to another type of port will result in initialization failure and FCC violations.
4.2.1 Console Terminal Connection

Figure 4-1 shows the placement of the EIA terminal connectors on the HSC rear bulkhead. The
console terminal connects to the J60 connector as shown.
4-1
4-2 Initialization Procedures
CONNECT
CONSOLE
TERMINAL
TO J60
EIA TERMINAL
CONNECTORS CABLE
~ ______A_______ ~\
BULKHEAD
o o o o
J60 CONSOLE J61 J62
c=> c=> c=>
N M L K J H
00 00 00 00
O~O~O 00 D Qo D Qo D Qo 00 0
00 00 00
1~O~O~D 00 D 00 0 00 00 1
00 00 00
2~ 0 ~ 0 ~ D 00 0 00 0 00 0
3~ 0 ~ 0 ~ D 00
00 0 00
00 0 00
00
F@ E @ o @ C @ B
DATA
CHANNEL
CONNECTIONS
CABLE CONNECTORS
WITHIN A DATA CHANNEL
CXO-891B
Figure 4-1 Console Terminal Connection

Preferably, power is turned off before the console terminal is installed. However, power can be left
on while connecting the terminaL Use the following procedure for installing the console terminal
with power on or off:
1. Put the SecurelEnable switch in the secure position.
2. Change terminal state (plug in, remove power, connect EIA line, and so forth).
3. Put the SecurelEnable switch in the enable position if it is necessary to do so at this point.
NOTE
If this procedure is not followed., the HSC may enter micro-on-line debugging tool (ODT)
mode. This mode is indicated by an @ symbol on the screen. Typing a P (PROCEED)
should exit this mode.
4.2.2 HSC50 Auxiliary and Maintenance Terminal Connections

Figure 4-2 shows the placement of the two ASCII ports on the HSC50 and HSC50 (modified).
The a:uxiliary terminal can be connected. to either the rear or the front ASCII port. Two terminals
cannot be connected at the same time.
~ Ill.: I!iI
IWI t;J
P42 P41 P40
MAINTENANCE
o m TERMINAL
P45 P44 CONNECTOR
o
olli::i::i:l o / .
CABLING
BULKHEAD
MAINTENANCE
ACCESS PANEL -
YV \
o o o
TERMINAL PRINTER
AUXILIARY
TERMINAL
/(QOR,~
CONNECTOR \ I
J l
I I
FROM } . . - ____ / /
EXTERNAL /
ACPOWER /
SOURCE ~-------
Figure 4-2 Auxiliary or Maintenance Terminal Connection

Preferably, power is turned off before the console terminal is installed. However, power can be left
on while connecting the terminal. Use the following procedure for installing the console terminal
with power on or off:
1. Put the SecurelEnable switch in the secure position.
2. Change terminal state (plug in, remove power, connect EIA line, and so forth).
3. Type three space characters on the terminal keyboard.
4. If it is necessary to put the SecurelEnable switch in the enable position, do so at this point.
NOTE
If this procedure is not followed, the HSC50 may enter micro-on-line debugging tool
(ODT) mode. This mode is indicated by an @ symbol on the screen. Typing a P (proceed)
should exit this mode.
4.2.3 LA12 Parameters

Detailed information on LA12 tenninal installation and operation is found in the DECwriter
Correspondent Technical Manual (EK-CPL12-TM).
When an LA12 is used as an auxiliary tenninal, the following parameters must be established:
1. Communications:
• Auto - Ansbk = no
• Buffer = 1024
• Comm Port= EIA
• Disk - HDX = none
• Echo - Local = no
• Fault = none
• G - HDX Start Mode = Rev
• H - Hi Speed (bps) = 9600
• L - Lo Speed (bps) = 300
• M - Line Prot = FDX - Data Leads
• o - Rev Error Ovride = no
• Parity = 71M
• Q - SRTS Polarity = 10
• Restraint =XonIXoff
• S - Speed Select =hi
• Turn Char = none
• U - Power Up = line
• V - Frequency = bell 103
2. Keyboard:
• Auto - Linefeed = no
• Break = no
• C - Keyclick = no
• = normal
Keypad
• Language = USA
• Repeat = yes
3. Printer:
• A - GO Char Set =USA
• B - G1 Char Set = USA

• C - G2 Char Set = USA
• D - G3 Char Set = USA
• End-of-line = wrap
• Form Length = 264
• G - Print Cntrl Chars
• Horiz Pitch (CPI) = 10
• Newline Char = none
• Print Force = hi
• Vertical Pitch (LPI) = 6
4.3 HSC Initialization

This section describes the initialization procedures for the HSC using the system diskette. This
diskette also contains the software necessary to execute the device integrity tests and the utilities.
To boot and run the off-line diagnostics from a separate off-line diskette, refer to Chapter 7.
System initialization is started by powering on the unit or (if the unit is already on) by pressing
and releasing the Init switch with the SecurelEnable switch in the enable position. This initiates
the P.io ROM bootstrap tests and then loads the Init P.io test.
NOTE
In order to run the HSC device integrity tests, the system diskette must reside in the
RX33 drive. Customarily, this diskette resides in RX33 drive O. However, drive 1 and
drive 0 are identical, and disk placement is arbitrary.
Logic in the following areas is tested with the Init P.ioj diagnostic:
• Control processor-The rest of the instruction set not tested by the ROM bootstrap, interrupts,
memory management, and the control memory lock-cycle circuitry are included. Detected
failures result in an error code display on the OCP (Figure 4-3).
• Memory-Program memory is tested from the 110 control processor. However, the control and
data memories are tested by the highest-numbered available requestor controlled by the 110
control processor. Again, detected failures result in an OCP error code display.
• Host interface and data channels-Module status is collected and placed in a table for the HSC
operating software initialization process. As each module is enabled, it automatically executes
internal microdiagnostics. These internal diagnostics test the following:
• ROM (sequencer, checksum, parity, and so forth)
• Special logic unique to that particular module
Upon completion of diagnostics for each module, a status code is passed to the lIO control processor.
Status codes for the various modules are discussed in Chapter 5.
If the module diagnostics complete successfully, the status code represents the module type and
the green LED is turned on. If the diagnostics fail, the status code indicates the failing microtest.
In addition, detected failures cause a red LED to light on that module. Kci, Ksdi, Ksti, and Ksi
failures are also displayed on the console terminal after the boot is completed.
~ Initialization Procedures
DESCRIPTION HEX OCT BINARY
K.PLI ERROR **" 01 01 00001 OFF OFF OFF
K.SDI/K.SIINCORRECT
02 02 00010 OFF OFF OFF
VERSION OF MICROCODE **"
K.STIIK.SI INCORRECT
03 03 00011 OFF OFF
VERSION OF MICROCODE **"
P.IOJ CACHE FAILURE * 08 10 01000 OFF
K.CI FAILURE" 09 11 01001 OFF
DATA CHANNEL MODULE ERROR * OA 12 01010
P.IOJ/C MODULE FAILURE 11 21 10001
M.STD2 MODULE FAILURE ***** 12 22 10010 OFF OFF
BOOT DEVICE FAILURE ** 13 23 10011
PORT LINK NODE ADDRESS

15 25 10101
SWITCHES OUT OF RANGE
MISSING FILES REQUIRED **** 16 26 10110
NO WORKING K.CI, K.SDI,

18 30 11000
K.STI, OR K.SI IN SUBSYSTEM
INITIALIZATION FAILURE 19 31 11001
SOFTWARE INCONSISTENCY 1A 32 11010
ILLEGAL CONFIGURATION 1B 33 11011
THESE ARE THE SO-CALLED SOFT OR NONFATAL ERRORS.

POSSIBLE MEMORY MODULE/CONTROLLER ON HSC70.
INCORRECT VERSION OF MICROCODE.
**** THIS FAULT CODE WILL ALSO BE DISPLAYED IF THE L0105 MODULE IS NOT AT THE MINIMUM
REV LEVEL.
**'**"SWAP MEMORY MODULE FIRST. IF PROBLEM PERSISTS, TRY THE P.IO MODULE. CXO-90SD
Figure 4-3 Operator Control Panel Fault Code Displays
NOTE
Lighting of the red LED on the LOIOO or LOllS LINK module does not indicate a failure of
the module.
For a detailed description of the boot process, refer to the HSC Boot Flowchart in Chapter 8.
4.3.1 Init P.io Test (INIPIO)

The INIPIO test completes the P.ioj module and the HSC memory testing previously started by the
ROM bootstrap tests. All P.ioj logic not tested by the bootstrap is tested by INIPIO. In addition, the
HSC Program, Control, and Data memories are tested.
This test runs in a standalone environment (no other HSC processes are running). If a failure is
detected, the failing module is flagged by illumination of the red LED on the module. If the test
runs without finding any errors, theHSC operational software is loaded and started. The Init P.io
test is not a repair-level diagnostic. If a repair-level test is needed, run the off-line P.io test that
provides standard HSC error messages.
4.3.2 INIPIO Test System Requirements

In order to run this test, the following hardware is required:
• P.ioj (processor) module with HSC boot ROM
• K.ci
• At least one M.std2 (memory) module
• RX33 controller with at least one working drive
In. addition, an HSC system diskette (RX33 media) is required.
4.3.3 INIPIO Test Prerequisites

The INIPIO test is loaded by the HSC ROM bootstrap program. The bootstrap tests the basic J-l1
instruction set, the lower 2048 bytes of Program memory, an 8 Kword partition in Program memory,
and the RX33 subsystem used by the bootstrap. When the INIPIO test begins to execute, most J-l1
logic has been tested and is considered working. Likewise, the Program memory occupied by the
test and the RX.33 subsystem used to load the test are also considered tested and working. The
RX33 diskette is checked to ensure it contains a bootable image.
4.3.4 INIPIO Test Operation

Follow these steps to start the INIPIO test:
1. Insert the HSC system diskette in the RX.33 unit 0 drive (left-hand drive).
2. Power on the HSC or press and release the Init button on the HSC OCP with the SecureJEnable
switch enabled. The Init lamp lights and the following occurs:
• The RX.33 drive-in-use LED lights within 10 seconds, indicating the bootstrap is loading the
INIPIO test to the Program memory.
• The I/O State light is on after diskette motion stops and the INIPIO test begins testing.
• The INIPIO test displays the following message on the HSC console when it begins:
INIPIO-I BOOTING
• HSC operational software is being loaded when the State light flashes rapidly.
• HSC operational software indicates it has loaded properly when the State light blinks
slowly.
• HSC displays its name and version indicating it is ready to perform host 110.
Once initiated, the INIPIO test is terminated only by halting and rebooting the HSC. If the test
fails to load using the preceding startup procedure, perfonn the next four steps:
1. Check the OCP fault light. If the fault light is on, press the fault light once and check the fault
code (Figure 4-3).
2. Boot the diskette from the RX33 unit 1 drive (right-hand drive).
3. Boot using another diskette. If that diskette boots, the original diskette is probably damaged or
worn..
4. Boot using the HSC Off-line Diagnostic diskette. This diskette contains the off-line P.io test,
which provides extensive error reporting features. A console tenninal must be connected to run
the off-line tests.
The progress of the INIPIO test is displayed in the State LED. Before the test starts, the State
LED is off. When the test starts, the State LED is turned on, and the INIPIO-I BOOTING message
is printed on the HSC console. When the test completes with no fatal errors, the State LED begins
to blink at a steady rate. If the test detects an error, the Fault lamp on the HSC OCP is lit.
4.4 HSC50 Initialization

In order to run the HSC50 device integrity tests and utilities, the HSC50 operating software
must be initialized with both the .system and utilities cassettes loaded in the TU58 drives. Before
inserting the system tape into the TU58 drive, check the black RECORD tab. This tab must be in
the record position (as indicated by an arrow on the tab) to ensure proper system operation. The
utilities tape need not be write-enabled.
NOTE
In order to run the HSC50 device integrity tests, the system tape must reside in the TU5S
drive. Customarily, this tape resides in TU58 drive O. Drive 1 and drive 0 are identical,
and tape placement is arbitrary.
However, the utilities tape does not contain a bootable image, and if drive 0 contains the
utilities tape, the system will try to boot from drive 1.
The HSC50 can be initiated by either powering on the unit if it is powered down or, if power is
already applied, by pressing and releasing the lnit switch with the SecurelEnable switch in the
enable position. This causes the P.ioc bootstrap ROM tests to run and then load the Init P.ioc test.
4.4.1 HSC50 Off-line Diagnostics Tape

The off-line diagnostics tape can be booted in either TU58 drive and need not be write-enabled. The
off-line tape can be booted by either powering on the unit or pressing and releasing the lnit switch
with the SecurelEnable switch in the enable position. This causes the P.ioc bootstrap ROM tests to
run and then load the off-line P.ioc test.
4.4.2 Inlt P.ioc Diagnostic

The Init P.ioc test is loaded by the P.ioc ROM bootstrap test each time the HSC50 system tape is
booted. This diagnostic completes the testing of the P.ioc module and the HSC50 memories. At the
successful completion of these tests, the HSC50 operating software is loaded and started.
Logic in the following areas is tested with the lnit P.ioc diagnostic:
• Control processor-The rest of the instruction set not tested by the ROM bootstrap, interrupts,
memory management, and the control memory lock-cycle circuitry are included. Detected
failures result in an error code display on the OCP (Figure 4-3).
• Memory-Program memory is tested from the 110 Control Processor. However, the control and
data memories are tested by the highest-numbered available requestor controlled by the I/O
control Processor. Again, detected failures result in an OCP error code display.
• Host interface and data channels-Module status is collected and placed in a table for the
HSC50 operating software initialization process. As each module is enabled, it automatically
executes internal microdiagnostics. These internal diagnostics test the following:
• ROM (sequencer, checksum, parity, and so forth)
• Special logic unique to that particular module
Upon completion of diagnostics for each module, a status code is passed to the 110 control processor.
Status codes for the various modules are discussed in Chapter 5.
If the module diagnostics complete successfully, the status code represents the module type and
the green LED is turned on. If the diagnostics fail, the status code indicates the failing microtest.
In addition, detected failures cause a red LED to light on that module. K.ci, Ksdi, Ksti, and K.si
failures are also displayed on the auxiliary terminal after the boot is completed.
NOTE
Lighting of the red LED ·on the·LOIOO or LOllS LINK module does not indicate a failure of
the module.
For a detailed description of the boot process, refer to the HSC50 Boot Flowchart in Chapter 8.
4.5 Fault Code Interpretation

All failures occuni.ng during the lnit P.io test are reported on the OCP LEDs. When the Fault lamp
is lit, pressing the Fault switch results in the display of a failure code in the OCP LEDs. This code
indicates which HSC module is the most probable cause of the detected failure. The failure code
blinks on and off at 1 second intervals until the HSC is rebooted if the fault code represents a fatal
fault. A soft fault code is cleared in the OCP by pressing the Fault switch a second time. To restart
the boot procedure, press the lnit switch. To identify the probable failing module, see Figure 4-3.
For detailed descriptions of OCP fault codes, see Chapter 8.
4-1 0 Initialization Procedures
Device Integrity Tests 5-1
5
Device Integrity Tests
5.1 Introduction
Device integrity tests executing in the HSC do not interfere with normal operation other than with
the device being tested. The device integrity tests can be found on the HSC system media disk or
HSC50 utilities media tape.
The tests described in this chapter are:
• ILRX33-RX33 integrity tests
• ILTU58-TU58 integrity tests
• ILMEMY-Memory integrity tests
• ILDISK-Disk drive integrity tests
• ILTAPE-Tape device integrity tests
• ILTCOM-Tape compatibility tests
• ILEXER-Multidrive exerciser
5.1.1 Device Integrity Tests Common Areas

Device integrity tests prompts and error messages so that they conform to standard formats. All
prompts issued by these integrity tests use a generic syntax.
• Prompts requiring user action or input are followed by a question mark.
• Prompts offering a choice of responses show those choices in parentheses.
• A capital D in parentheses indicates the response should be in decimal.
• Square brackets enclose the prompt default or, if empty, indicate no default exists for that
prompt.
5-1
5-2 Device Integrity Tests
5.1.2 Generic Error Message Format

All device integrity tests follow a generic elTor message format, as follows:
XXXXXX>x>tt:tt T#aaa E*bbb U-ccc
<TEXT STRING DESCRIBING ERROR>
FRU1-dddddd FRU2-dddddd
MA-eeeeee
EXP-yyyyyy
ACT-zzzzzz
Where:
XXXXXX> is the appropriate device integrity test prompt.
x> is the letter indicating the type of integrity test
that was initiated:
0> is the demand integrity test.
A> is the automatic integrity test.
P> is the periodic integrity test
tt:tt is the current'time.
aaa is the decimal number denoting test that failed.
bbb is the decimal number denoting error detected.
ccc is the unit number of drive being tested.
FRUl is the most likely field replaceable unit (FRU).
FRU2 is the next most likely FRU.
dddddd is the name of field replaceable unit.
MA is the media address.
eeeeee is the octal number denoting offset within block.
yyyyyy is the octal number denoting data expected.
zzzzzz is the octal number denoting data actually found.
The first line of the elTor message contains general information about the elTor. The second line
describes the nature of the error. Lines 1 and 2 are mandatory and appear in all error messages.
Line 3 and any succeeding lines display additional information and are optional.
NOTE
If a P.ioj/c or M.std/2 module fails during the periodic ILMEMY tests, the FAClLITY
section of the crash code displays PRMEMY, which indicates the failure occurred during
the periodic tests.
If a Ksdi, Ksti, or Ksi modUle fails during the periodic K tests, the FACILITY section of
the crash code displays PRKSDI, PRKSTI, or PRKSI, which indicates the failure occurred
during the periodic tests.
5.2 ILRX33 - RX33 Device Integrity Tests

The ILRX33 exerciser runs a test of either of the RX33 drives attached to the HSC. ILRX33 runs
concurrently with other HSC processes and uses the services of the HSC control program and the
Diagnostic Execution Monitor (DEMON).
ILRX33 performs several writes and reads to verify the RX33 internal data paths and read/write
electronics.
A scratch diskette is not required. ILRX33 does not destroy any data on the system software.
The exerciser tests only the RX33 and the data path between the P.ioj and the RX33. All other
system hardware is assumed to be working properly.
ILRX33 verifies only a particular RX33 drive and controller combination is working or failing.
Therefore, the test should not be used as a subsystem troubleshooting aid. This test does not
support flags. If the test indicates a drive or controller is not operating cOlTectly, replace the drive
and/or controller. The controller is located on the memory module.
5.2.1 System Requirements

Hardware and software requirements include:
• P.ioj (processor) module with boot ROMs
• M.std2 memory/disk controller module
• HSC system media
• Console terminal

Press CTRLIY to get the HSC> prompt. Next, type either RUN ILRX33 or RUN DXn:ILRX33 to
initiate the tests.
NOTE
The term DXn: refers to the RX33 disk drives (DXO: or DXl:).
If ILRX33 cannot load from the specified diskette, try loading the test from the other diskette. For
example, if RUN ILRX33 fails, try RUN DXn:ILRX33.
5.2.3 Test Termination

ILRX33 can be terminated by pressing CTRLIY. The test automatically terminates after reporting
an error with one exception: if the error displayed is Retries Required, the test continues.
5.2.4 Parameter Entry

The device name of the RX33 drive to be tested is the only parameter sought by this test. When the
test is invoked, the following prompt is displayed:
Device Name of RX33 to test (DXO:, DX1:, LB:) [] ?
NOTE
The string LB: indicates the RX33 drive last used to boot the HSe control program.
One of the indicated strings must be entered. If one of these strings is not entered, the test prints
lllegal Device Name and the prompt is repeated.
5.2.5 Progress Reports

At the end of the test, the following message is displayed:
ILRX33>D>tt:tt Execution Complete
Where:
tt:tt is the current time.
5.2.6 Test Summary

The ILRX33 test summary is contained in the following paragraph.
Test 001, ReadIWrite Test-Verifies that data can be written to the diskette and read back
correctly. All reads and writes access physical block 1 of the RX33 (the RT-11 volume ID block).
This block is not used by the HSC operating software.
Initially, the contents of block 1 are read and saved. Then, three different data pattern.s are written
to block 1, read back, and verified. This checks the read/write electronics in the drive and the
internal data path between the RX33 controller and the drive. Following the read/write test, the
original contents of block 1 are written back to the diskette.
If the data read back from the diskette does not match the data written, a data compare error
is generated. The error report lists the word (MA) in error within the block together with the
EXPected (EXP) and ACTual (ACT) contents of the word.
5.2.7 Error Message Example

All error messages produced by ILRX33 conform to the HSC device integrity test error message
format (Section 5.1.2). Following is a typical ILRX33 error message:
ILRX33>D>00:00 TOOl E 003 U- 50182
ILRX33>D> No Diskette Mounted
ILRX33>D> FRU1-Drive
Other optional lines are found on different error messages.
5.2.8 Error Messages

The following paragraphs list specific information about each of the errors produced by the ILRX33.
Hints about the possible cause of the error are provided where feasible.
• Error 000, Retries Required-Indicates a Read or Write operation failed when first
attempted, but succeeded on one of the retries performed automatically by the RX33 driver
software. This error normally indicates the diskette media is degrading and the diskette should
be replaced.
• Error 001, Operation Aborted-Reported if ILRX33 is aborted by pressing CTRUY.
• Error 002, Write-Protected-Indicates the RX33 drive being tested contains a write-protected
diskette. Write enable the diskette and try again. If the diskette is not write-protected, the
RX33 drive or controller is faulty.
• Error 003, No diskette Mounted-Indicates the RX33 drive being tested does not contain a
diskette. Insert a diskette before repeating the test. If this error is displayed when the drive
does contain a diskette, the drive or controller is at fault.
• Error 004, Hard I/O Error-Indicates the program encountered a hard error while attempting
to read or write the diskette.
• Error 005, Block Number Out of Range-Indicates the RX33 driver detected a request
to read a block number outside the range of legal block numbers (0 through 2399 decimal).
Because the ILRX33 reads and writes disk block 001, it may indicate a software problem.
• Error 006, Unknown Status STATUS=xxx-Indicates ILRX33 received a status code it
did not recognize. The octal value xxx represents the status byte received. RX33 reads and
writes are performed for ILRX33 by the HSC control program's RX33 driver software. At the
completion of each Read or Write operation, the driver software returns a status code to the
RX33 test, describing the result of the operation. The test decodes the status byte to produce a
description of the error.
An unknown status error indicates the status value received from the driver did not match
any of the status values known to the test. The status value returned (xxx) is displayed to
help determine the cause of the problem. Any occurrence of this error should be reported
through a Software Performance Report (SPR). See Appendix B for detailed information on SPR
submission.
• Error 007 t Data Co:mpare Error-Indicates data subsequently read back.
MA -aaaaaa
EXP-bbbbbb
ACT-cccccc
where:
aaaaaa represents the address of the failing word
within the block (512 bytes) that was read.
bbbbbb represents the data written to the word.
cccccc represents the data read back from the word.
Because this test only reads and writes block 1 of the diskette, all failures occur while trying to
access physical block 1.
• Error 008, lllegal Device Name-Indicates the user specified an illegal device name when
the program prompted for the name of the drive to be tested. Legal device names include DXO:,
DX1:, and LB:. LB: indicates the drive from which the system was last booted. Mter displaying
this error, the program again prompts for a device name. Enter one of the legal device names
to continue the test.
5.3 ILTAPE-TU58 Device Integrity Test

The ILTU58 tests either of the TU58 drives attached to the HSC50. This test runs concurrently
with other HSC processes and uses the services of the HSC control program and the Diagnostic
Execution Monitor (DEMON). The test can be initiated with only the system tape installed.
Because the HSC50 operating system tests the TU58 every time it is used, ILTU58 performs only
minimal testing. Several read and write operations are performed to test the internal data paths
and the read/write circuitry of the TU58.
A scratch tape is not required. This test does not destroy any data on the system software.
ILTU58 tests only the TU58 and the data path between the P.ioc and the TU58. All other system
hardware is assumed to be working properly.
ILTU58 verifies that only a particular TU58 drive and controller combination is working or failing.
Therefore, the test should not be used as a troubleshooting aid. This test does not support flags.
If the test indicates that a drive or controller is not operating correctly, replace the drive and/or
controller.

Hardware and software requirements include:
P.ioc module with boot ROMs
M.std memory/disk controller module
TU58 controller with at least one working drive
HSC system media

Press CTRLIY to get the HSC> prompt. Then type RUN ILTU58 or RUN DDn::ILTU58 to initiate
the tests.

ILTU58 can be terminated by pressing CTRLIY.

The following error messages are issued:
• Retries Required-Indicates a Read or Write operation failed in the first attempt but
succeeded on one of the retries. It may also mean that the tape media is degrading and should
be replaced.
• Operation Aborted-Reported if the test is interrupted by CTRLIY.
• Write Protected- Indicates the drive being tested has a write protected tape. Try again with
Write enabled. If the tape is not write-protected, then the drive or controller is faulty.
.• No Cassette Mounted-Indicates the drive does not have a cassette. Insert a cassette tape
then repeat the test. Otherwise, the drive or the controller is faulty.
• Hard 110 Error-Indicates the program encountered a hard error while attempting to read or
write the cassette tape.
• Bad Record-Indicates the program encountered a bad record while attempting to read or
• Bad Opcode-Indicates the program encountered an illegal opcode.
• Bad Record-Indicates the program encountered a bad record while attempting to read or
• Seek Error-Indicates a seek error status was set after a SEEK command.
• Bad Unit I-Indicates the diagnostic interface could not find the unit number specified. The
drive may have been off line.
• Failed Self-Test-Indicates the HSC diagnostic interface failed.
• End of Medium-Indicates the end of the cassette tape.
• Unknown Status-Indicates the TU58 device integrity test received a status code it did not
recognize. This status code is displayed to help determine the cause of the problem.
• Data Compare Error-Indicates that data subsequently read back did not match data
written. The test writes a pattern to block 1, then reads block 1 and check that the pattern
read is the same as the pattern written. If the patterns differ, this error message is displayed.
This step is performed three times, each time with a different pattern. The patterns are:
MA -aaaaaa-represents the address of the failing word.
EXP-bbbbbb-represents the data written to the word.
ACT-cccccc-represents the data read back from the word.
Because this test only reads and writes block 1 of the cassette tape, all failures occur while
trying to access physical block 1.
• Dlegal Device Name-Indicates the user specified an illegal device name when prompted for
the name of the drive to be tested. Mter displaying this error, the program again prompts for a
device name. Enter one of the legal device names to continue the test.
5.4 ILMEMY - Memory Integrity Tests

The memory integrity test is designed to test HSC data buffers. This test can be initiated
automatically or on demand. It is initiated automatically to test data buffers that produce a
parity or nonexistent memory (NXM) error when in use by the HSC control program or any of the
Kmodules.
Buffers that fail the memory test are removed from service by sending them to the disabled buffer
queue. The disabled buffer accepts only 16 entries. When the buffer has accepted 16 entries, it acts
as a first-in-first-out (FIFO) buffer.
NOTE
The contents of the disabled buffer queue are lost during a reboot of the HSC. As a result,
all bad memory locations are lost.
Buffers sent twice to this test are also sent to the disabled buffer queue even if they did not fail the
test. Buffers that pass the memory test and have not been tested previously are sent to the free
buffer queue for further use by the HSC control program.
When the test is initiated on demand, any buffers in the disabled buffer queue are tested and the
results of the test are displayed on the terminal from which the test was initiated.
This test runs concurrently with other HSC processes and uses the services of the HSC control
program and the Diagnostic Execution Monitor (DEMON).

Hardware requirements include:
• P.ioj (processor) module with HSC boot ROMs, or P.ioc (processor) module with HSC50 boot
ROMs.
• At least one M.std2 memory module (HSC) or the M.std memory module (HSC50).
• RX33 controller with one working drive (HSC) or TU58 controller with one working drive
(HSC50).
• A console terminal for demand initiation oniy.
This program only tests Data Buffers located in the HSC Data memory. All other system hardware
is assumed to be working.
Software requirements include:
• HSC control program (system diskette or tape)
• Diagnostic Execution Monitor (DEMON)

Press CTRIJY to get the attention of the HSC keyboard monitor. The keyboard monitor responds
with the prompt:
HSCxx>
Type RUN dev:ILMEMY to initiate the memory integrity test. This program has no user-supplied
parameters or flags.
NOTE
ILMEMY tests only data memory buffers. Control/program memory errors typically
cause a reboot of the HSC.
If the memory integrity test is not contained on the specified device (dev:), an error message is
displayed.

ILMEMY can be terminated at any time by pressing CTRUY.

Error messages are displayed as needed. At the end of the test, the following message is displayed
(by DEMON):
ILMEMY>D>tt:tt Execution Complete
Where:
tt:tt = current time.
5.4.5 Test Summaries

Test 001 receives a queue of buffers for testing. If the ILMEMY is initiated automatically, the
queue consists of buffers from the suspect buffer queue.
When the HSC control program detects a parity or nonexistent memory (NXM) error in a Data
Buffer, the buffer is sent to the suspect buffer queue. While in this queue, the buffer is not used for
data transfers. The HSC periodic scheduler periodically checks the suspect buffer queue to see if it
contains any buffers. If buffers are found on the queue, they are removed and the in-line memory
test is automatically initiated to test those buffers.
If the ILMEMY test is initiated on demand, it retests only buffers already known as disabled.
If the test is initiated automatically and the buffer passes the test, the program checks to see if this
is the second time the buffer was sent to the memory integrity test. If this is the case, the buffer is
probably producing intermittent errors. The buffer is retired from service and sent to the disabled
buffer queue. If this is the first time the buffer is sent to the memory integrity test, it is returned
to the free buffer queue for further use by the HSC control program. In this last case, the address
of the buffer is saved in case the buffer again fails and is sent to the memory integrity test a second
time.
When all buffers on the test queue are tested, the memory integrity test terminates.

All error messages produced by the memory integrity test conform to the HSC integrity test error
message format (Section 5.1.2). Following is a typical ILMEMY error message:
ILMEMY>A>09:33 TOOl E 000
ILMEMY>A>Tested Twice with no Error (Buffer Retired)
ILMEMY>A>FRU1-M.std2 FRU2-
ILMEMY>A>Buffer Starting Address (physical) 15743600
ILMEMY>A>Buffer Ending Address (physical) = 15744776
When all buffers on the test queue are tested, the memory integrity test terminates.

The following list shows specific information about each of the errors displayed by the memory
integrity test.
• Error 000, Tested Twice with No Error-Indicates the buffer under test passed the memory
test. However, this is the second time the buffer was sent to the memory test and passed it.
Because the buffer has a history of two failures while in use by the control program, yet does
not fail the memory test, intermittent failures on the buffer are assumed. The buffer is retired
from service and sent to the disabled buffer queue.
• Error 001, Returned Buffer to Free Buffer Queue-Indicates a buffer failed during use
by the control program but that the memory integrity test detected no error. Because this is
the first time the buffer was sent to the memory integrity test, it is returned to the free buffer
queue for further use by the HSC control program. The address of the buffer is stored by the
memory integrity test in case the buffer again fails when in use by the control program.
• Error 002, Memory Parity Error-Indicates a parity error occurred while testing a buffer.
The buffer is retired from service and sent to the disabled buffer queue.
• Error 003, Memory Data Error-Indicates the wrong data was read while testing a buffer.
The buffer is retired from service and sent to the disabled buffer queue.
• Error 004, NXM Trap (Buffer Retired) - Indicates an unknown address or memory location
is being referenced.
• Error 005, Can't Allocate Timer, Test Aborted-Indicates the program failed to allocate a
timer for SLEEP.
5.5 ILDISK - DISK Drive Integrity Tests

ILDISK isolates disk drive-related problems to one of the following three field replaceable units
(FRUs):
1. Disk drive
2. SDI cable
3. HSC disk data channel module
ILDISK runs in parallel with disk 1/0 from a host CPU. However, the drive being diagnosed cannot
be on line to any host. ILDISK can be initiated upon demand through the console terminal or
automatically by the HSC control program when an unrecoverable disk drive failure occurs.
ILDISK is automatically invoked by default whenever a drive is declared inoperative, with one
exception: if a drive is declared inoperative while in use by an integrity test or utility.
Automatic initiation of ILDISK can be inhibited by issuing the SETSHO command SET
AUTOMATIC DIAGNOSTICS DISABLE. If the SET AUTOMATIC DIAGNOSTICS command is
issued and DISABLE is specified, ILMEMY (a test for suspect buffers) is also disabled. For this
reason, leaving ILDISK automatically enabled is preferable.
The tests performed vary, depending on whether the drive is known to the HSC disk server.
1. Drive unknown (to the HSC disk server}-It is either unable to communicate with the HSC
or was declared inoperative when it failed while commwri,cating with the HSC. In this case,
because the drive cannot be identified by unit number, the user must supply the requestor
number and port number of the drive. Then the SDI verification tests can execute. The SDI
verification tests check the-path between the KsdilK.si and the disk drive and command the
drive to run its integrity tests. If the SDI verification tests fail, the most probable FRU is
identified in the error report. If the SDI verification tests pass, presume the drive is the FRU.
2. Drive known (to the HSC disk server, that is, identifiable by unit number}-ReadlWrite/Fonnat
tests are perfonned in addition to the SDI verification tests. If an error is detected, the most
probable FRU is identified in the error report. If no errors are detected, presume the FRU is
the drive.
To find the drives known to the Disk and Tape Servers, type the SETSHO command SHOW DISKS
or SHOW TAPES.

The software requirements for this test reside on the system media and include:
• HSC executive (CRONIC)
• ILDISKDIA program
• Diagnostic process (DEMON)
• KsdilK.si microcode (installed with the KsdilK.si module)
Hardware requirements include:
• Disk drive
• Disk data channel, connected -by an SDI cable
The test assumes the I/O Control Processor module and the memory module are working.
Refer to the disk drive documentation to interpret errors that occur in the drive's integrity tests.

The following steps are used to initiate ILDISK
NOTE
10 prevent access from another HSC, deselect the alternate port switch on the drive to be
tested. The alternate port switch is the drive port switch allowing alternate HSC access
to the drive.
NOTE
The HSC system RX33 must be present at all times.
1. Press CTRLIY.
2. The following prompt appears:
HSCxx>
3. Type RUN dev:ILDISK and enter a carriage return.

4. Wait until ILDISK is read from the system software load media into the HSC Program memory.
5. Enter parameters after ILDISK is started. Refer to Section 5.5.5.
5.5.3 Availability
If the software media containing ILDISK is not loaded when the RUN ILDISK command is
entered, an error message is displayed. Insert the software media containing ILDISK and repeat
Section 5.5.2.

ILDISK is terminated by pressing CTRLIY or CTRLlC. Test termination may not take effect
immediately because certain parts of the program cannot be interrupted. An example would be
during SDI commands. If an SDI DRIVE DIAGNOSE command is in progress, interfering with the
disk drive at this time can cause the program to wait 2 minutes before aborting.

Upon demand initiation, ILDISK first prompts:
DRIVE UNIT NUMBER (U) [] ?
Enter the unit number of the disk drive for test. Unit numbers are in the form Dnnnn, where nnnn
is a decimal number between 0 and 4095 corresponding to the number printed on the drive unit
plug. Terminate the unit number response with a carriage return.
ILDISK attempts to acquire the specified unit through the HSC diagnostic interface. If the unit is
acquired successfully, ILDISK next prompts for the drive integrity test to be executed.
If the acquire fails, one of the following conditions is encountered:
1. The specified drive is unavailable. This indicates the drive is connected to the HSC, but is
currently on line to a host CPU or an HSC utility. On-line drives cannot be diagnosed. ILDISK
repeats the prompt for the unit number.
2. The specified drive is unknown to the HSC disk functional software. Drives are unknown for
one of the following reasons:
• The drive and/or disk data channel port is broken and cannot communicate with the disk
functional software.
• The drive was previously communicating with the HSC but a serious error occurred, and
the HSC has ceased communicating with the drive (marked the drive as inoperative).
In either case, ILDISK prompts for a requestor number and port number. Refer to Section 5.5.6.
Mter receiving the unit number (or requestor and port), ILDISK prompts:
RUN A SINGLE DRIVE DIAGNOSTIC (YIN) [N] ?
Answering N causes the drive to execute its entire integrity test set. Answering Y executes a single
drive integrity test. If a single drive integrity test is selected, the test prompts:
DRIVE TEST NUMBER (H) [] ?
Enter a number (in hex) specifying the drive integrity test to be executed. Consult the appropriate
disk maintenance or service manual to determine the number of the test to perform. Entering a
test number not supported by the drive results in an error 13 generated in test 5.
The test prompts for the number of passes to perform:
t OF PASSES TO PERFORM (1 to 32767) (D) [1] ?
Enter a decimal number between 1 and 32767 specifying the number of test repetitions. Pressing
RETURN without entering a number runs the test once.
5.5.6 Specifying Requestor and Port

Drives unknown to the HSC disk functional software are tested by specifying the requestor number
and port nl.lInber of the drive. The requestor number is any number 2 through 9 (HSC) or 2 through
7 (HSC50 [modified] or HSC50) specifying the disk data channel connected to the drive under test.
The port number is 0 through 3; it specifies which of four disk data channel ports is connected
to the drive under test. The requestor number and port number can be determined in one of two
ways:
1. By tracing the SDI cable from the desired disk drive to the HSC bulkhead connector, and then
tracing the bulkhead connector to a specific port on one of the disk data channels.
2. By using the SHOW DISKS command to display the requestor and port numbers of all known
drives. To use this method, exit ILDISK by pressing CTRIlY. Type SHOW DISKS in response
to the HSC prompt.
This command displays a list of all known drives including the requestor number and port
number for each dri-ve. Each disk data channel has four possible ports to which a drive can be
connected. By inference, the port number of the unknown unit must be one not listed in the
SHOW DISKS display (assuming the unknown drive is not connected to a defective disk data
channel). A defective disk data channel illuminates the red LED on the lower front edge of the
module. Refer to Chapter 2.
Mter a requestor number and a port number are supplied to ILDISK, the program checks to
ensure the specified requestor and port do not match any drive known to the HSC software. If
the requestor and port do not match a known drive, ILDISK prompts for the number of passes to
perform, as described in Section 5.5.5. If the requestor and port do match a known drive, ILDISK
reports error 08.

ILDISK produces an end-of-pass report at the completion of each pass of the integrity test. One
pass of the program can take several minutes depending upon the type of drive being diagnosed.

Test summaries for ILDISK follow.
• Test 0, Parameter Fetching-Fetches parameters is identified as test O. The user is prompted
to supply a unit number and/or a requestor and port number. This part of ILDISK also prompts
for the number of passes to perform.
• Test 01, Run KSDI Microdiagnostics--Commands the disk data channel to execute two of
its resident microdiagnostics. If the revision level of the disk data channel microcode is not up
to date, the microdiagnostics are not executed. The microdiagnostics executed are the partial
SDI test (K.sdi test 7) and the SERDESIRSGEN test (K.sdilK.si test 10).
• Test 02, Check for Clocks and Drive Available-Issues a command to interrogate the
Real-Time Drive State of the drive. This command does not require an SDI exchange, but the
real-time status of the drive is returned to ILDISK. The real-time status should indicate the
drive is supplying clocks and the drive should be in the Available state.
• Test 03, Drive Initialize Test-Issues a DRIVE INITIALIZE command to the drive under
test. This checks both the drive and the Controller Real-Time State Line of the SDI cable. The
drive should respond by momentarily stopping its clock and then restarting it.
• Test 04, SDI Echo Test-First ensures the disk data channel microcode supports the ECHO
command. If not, a warning message is issued, and the rest of test 04 is skipped. Otherwise,
the test directs the disk data channel to conduct an ECHO exchange with the drive. An ECHO
exchange consists of the disk data channel sending a frame to the drive and the drive returning
it. An ECHO exchange verifies the integrity of the write/command data and the read/response
data lines of the SDI cable.
• Test 05, Run Drive Integrity Tests-Directs the drive to run its internal integrity test. The
drive is commanded to run a single integrity test or its entire set of integrity tests depending
upon user response to the prompt:
Run a Single Drive Diagnostic ?
Before commanding the drive to run its integrity tests, the drive is brought on line to prevent
the drive from giving spurious Available indications to its other SDI port. The drive integrity
tests are started when the disk data channel sends a DIAGNOSE command to the drive. The
drive does not return a response frame for the DIAGNOSE until it is finished performing
integrity tests. This can require 2 or more minutes. While the disk data channel is waiting for
the response frame, ILDISK cannot be interrupted by a CTRLIY.
• Test 06, Disconnect From Drive-Sends a DISCONNECT command to the drive and then
issues a GET LINE STATUS internal command to the K.sdiJK.si to ensure the drive is in the
Available state. The test also expects Receiver Ready and Attention are set in drive status and
Read/Write Ready is not set.
• Test 07, Check Drive Status-Issues a GET STATUS command to the drive to check that
none of the drive's error bits are set. If any error bits are set, they are reported and the test
issues a DRIVE CLEAR command to clear the error bits. If the error bits fail to clear, an error
is reported.
• Test 08, Drive Initialize--Issues a command to interrogate the Real-Time Drive State of the
drive. The test then issues a DRIVE INITIALIZE command to ensure the previous DIAGNOSE
command did not leave the drive in an undefined state.
• Test 09, Bring Drive On Line-Issues an ONLINE command to the drive under test. Then
a GET LINE STATUS command is issued to ensure the drive's real-time state is proper for the
On-line state. Read/Write Ready is expected to be true; Available and Attention are expected to
be false.
• Test 10, Recalibrate and Seek-Issues a RECALIBRATE command to the drive. This ensures
the disk heads start from a known point on the media. Then a SEEK command is issued to the
drive, and the drive's real-time status is checked to ensure the SEEK did not result in an
Attention condition. Then another RECALIBRATE command is issued, returning the heads to
a known position.
• Test 11, Disconnect From Drive-Issues a DISCONNECT command to return the drive to
the Available state. Then the drive's real-time status is checked to ensure Available, Attention,
and Receiver Ready are true and Read/Wri te Ready is false.
• Test 12, Bring Drive On Line--Attempts to bring the disk drive to the On-line state. Test 12
is executed only for drives known to the HSC disk functional software. Test 12 consists of the
following steps:
1. GET STATUS-ILDISK issues an SDI GET STATUS command to the disk drive.
2. ONLlNE-ILDISK directs the HSC diagnostic interface to bring the drive on line.
If the GET STATUS and the ONLINE commands succeed, ILDISK proceeds to test 13. If the
GET STATUS and the ONLINE commands fail, ILDISK goes directly to test 17 (termination).
Note the on-line is performed through the HSC diagnostic interface, invoking the same software
operations a host invokes to bring a drive on line. An on-line at this level constitutes more
than just sending a SDI ONLINE command. The FCT and RCT of the drive also are read and
certain software structures are modified to indicate the new state of the drive. If the drive is
unable to read data from the disk media, the on-line operation fails. If test 12 fails, ILDISK
skips the remaining tests and goes to test 17.
• Test 13, Read Only 110 Operations Test-Tests that all read/write heads in the drive can
seek and properly locate a sector on each track in the drive read only DBN space. (DBN space
is an area on all disk media devoted to diagnostic or integrity test use.) Test 13 attempts to
read at least one sector on every track in the read only area of the drive's DBN space. The
sector is checked to ensure it contains the proper data pattern. Bad sectors are allowed, but
there must be at least one good sector on each track in the read only area. Mter each successful
DBN read, ILDISK reads one LBN to further enhance seek testing. This ensures the drive can
successfully seek to and from the DBN area from the LBN area of the disk media. ILDISK
proceeds to test 16 when test 13 completes.
• Test 14, 110 Operations Test (Read/Write 512 byte format)-Checks to see if the drive
can successfully write a pattern and read it back from at least one sector on every track in the
drive read/write DBN area. (Read/write DBN space is an area on every disk drive devoted to
diagnostic or integrity test read/write testing.) Bad sectors are allowed, but at least one sector
on every track in the read/write area must pass the test. Mter test 14 completes, ILDISK
proceeds to test 17.
• Test 17, Terminate ILDISK-Is the ILDISK termination routine. The following steps are
performed:
1. If the drive is unknown to the HSC disk functional software, or if the SDI verification test
failed, proceed to step 5 of this t~st.
2. An SDI CHANGE MODE command is issued to the drive. The CHANGE MODE command
directs the drive to disallow access to the DBN area and changes the sector size (512 or 576
bytes) back to its original state.
3. The drive is released from exclusive integrity test use. This returns the drive to the
Available state.
4. The drive is reacquired for exclusive integrity test use. This is to allow looping if more than
one pass is selected.
5. If more passes are left to perform, the test is reinitiated.
6. If no more passes are left to perform~ ILDISK releases the drive, returns all structures
acquired, and terminates.

All error messages produced by the disk drive integrity tests conform to the HSC integrity test
error message format (Section 5.1.2). Following is a typical ILDISK error message.
ILDISK>D>09:35 T 005 E 035 U-D00082
ILDISK>D>Drive Diagnostic Detected Fatal Error
ILDISK>D>FRU1-Drive FRU2-
ILDISK>D>Requestor Number 04
ILDISK>D>Port Number 03
ILDISK>D>Test 0025 Error 007F
ILDISK>D>End Of Pass 00001

Messages produced by ILDISK are described in the following list:
• Error 01, DDUSUB Initialization Failure-The HSC diagnostic interface did not initialize.
Error 01 is not recoverable and is caused by:
1. Insufficient memory to allocate buffers and control structures required by the diagnostic
interface.
2. HSC disk functional software is not loaded.
• Error 02, Unit selected Is Not a Disk-The response to the unit number prompt was not of
the form Dnnnn (refer to Section 5.5.5).
• Error 03, Drive Unavailable-The selected disk drive is not available for ILDISK.
• Error 04, Unknown Status from DDUSUB-A call to the diagnostic interface resulted in
the return of an unknown status code. This indicates a software error and should be reported
through a Software Performance Report (SPR). See Appendix B for detailed information on SPR
submission.
• Error 05, Drive Unknown to Disk Functional Code-The disk drive selected is not known
to the HSC disk functional software. The drive may not be communicating with the HSC, or
the disk functional software may have disabled the drive due to an error condition. ILDISK
prompts the user for the drive's requestor and port. Refer to Section 5.5.6 for information on
specifying requestor and port.
• Error 06, Invalid Requestor or Port Number Specified-The requestor number given was
not in the range 2 through 9 (HSC) or 2 through 7 (HSC50), or the port number given was not
in the range 0 through 3. Specify a requestor and port within the allowable ranges.
• Error 07, Requestor Selected Is Not a K.sdi-The requestor specified was not a disk data
channel (K.sdi/K..si). Specify a requestor that contains a disk data channel.
• Error 08, Specified Post contains a Known Drive-The requestor and port specified contain
a drive known to the HSC disk functional software. The unit number of the drive is supplied in
the report. ILDISK does not allow testing a known drive through requestor number and port
number.
• Error 09, Drive Can't Be Brought On Line-A failure occurred when ILDISK attempted to
bring the specified drive on line. One of the following conditions occurred:
1. Unit Is Off Line-The specified unit went to the Off-line state and now cannot communicate
with the HSC.
2. Unit Is In Use-The specified unit is now marked as in use by another process.
3. Unit Is a Duplicate-Two disk drives are connected to the HSC, both with the same unit
number.
4. Unknown Status from DDUSUB-The HSC diagnostic interface returned an unknown
status code when ILDISK attempted to bring the drive on line. Refer to error 04 for related
information on this error.
• Error 10, K.sdi Does Not Support Microdiagnostics-The KsdilKsi connected to the drive
under test does not support microdiagnostics. This indicates the K.sdilK.si microcode is not at
the latest revision level. This is not a fatal error, but the KsdilK.si should probably be updated
with the latest microcode to improve error detection capabilities.
• Error 11, Change Mode Failed-ILDISK issued an SDI CHANGE MODE command to the
drive and the command failed. The drive is presumed the failing unit because the SDI interface
was previously verified.
• Error 12, Drive Disabled Bit Set-The SDI verification test issued an SDI GET STATUS
command to the drive under test. The drive disabled bit was set in the status returned by the
drive, indicating the drive detected a serious error and is now disabled.
• Error 13, Command Failure-The SDI verification test detected a failure while attempting to
send an SDI command to the drive. One of the following occurred.
1. Did Not Complete-The drive did not respond to the command within the allowable time.
Further SDI operations to the drive are disabled.
2. Ksdi Detected Error-The K.sdilK..si detected an error condition while sending the
command or while receiving the response.
3. Unexpected Response--The SDI command resulted in an Wlexpected response from the

drive. This error can be caused by a DIAGNOSE command if a single drive integrity test
was selected, and the drive does not support the specified test number.
• Error 14, Can't Write Any Sector on Track-As part of test 04, ILDISK attempts to write a
pattern to at least one sector of each track in the read/write area of the drive DBN space. (DBN
space is an area on every disk drive reserved for diagnostic use only.) During the write process,
ILDISK detected a track with no sector that passed the read/write test. (ILDISK could not
write a pattern and read it back successfully on any sector on the track.) The error information
for the last sector accessed is identified in the error report. The most probable cause of this
error is a disk media error.
If test 03 also failed, the problem could be in the disk read/write electronics, or the DBN
area of the disk may not be formatted correctly. To interpret the MSCP status code, refer to
Section 5.5.1!'
• Error 15, ReadlWrite Ready Not Set in On-line Drive-The SDI verification test executed
a command to interrogate the Real-Time Drive State line of the drive. The line status reported
the drive was in the On-line state, but the Read/Write Ready bit was not set in the status.
This could be caused by a failing disk drive, bad RJW logic, or bad software media.
• Error 16, Error Releasing Drive-ILDISK attempted to release the drive under test. The
release operation failed. One of the following occurred.
1. Could Not Disconnect-An SDI DISCONNECT command to the drive failed.
2. Unknown Status from DDUSUB-Refer to error 04.
• Error 17, Insufficient Memory, Test Not Executed-The SDI verification test could not
acquire sufficient memory for control structures. The SDI verification test could not be
executed. Use the SETSHO command SHOW MEMORY to display available HSC memory.
If any disabled memory appears in the display, consider further testing of the memory module.
If no disabled memory is displayed, and no other integrity test or utility is active on this HSC,
submit an SPR. See Appendix B for detailed information on SPR submission.
• Error 18, K Microdiagnostic Did Not Complete-The SDI verification test directed the
disk data channel to execute one of its microdiagnostics. The microdiagnostic did not complete
within the allowable time. All drives connected to the disk data channel may now be unusable
(if the microdiagnostic never completes) and the HSC probably must be rebooted. The disk data
channel module is the probable failing FRU.
• Error 19, K Microdiagnostic Reported Error-The SDI verification test directed the
disk data channel to execute one of its microdiagnostics. The microdiagnostic completed and
reported an error. The disk data channel is the probable FRU.
• Error 20, DCB Not Returned, K Failed for Unknown Reason-The SDI verification test
directed the disk data channel to execute one of its microdiagnostics. The microdiagnostic
completed without reporting any error, but the disk data channel did not return the dialog
control block (DCB). All drives connected to the disk data channel may now be unusable. The
disk data channel is the probable FRU and the HSC probably will have to be rebooted.
• Error 21, Error in DCB on Completion-The SDI verification test directed the disk data
channel to execute one of its microdiagnostics. The microdiagnostic completed without reporting
any error, but the disk data channel returned the dialog control block (DCB) with an error
indicated. The disk data channel is the probable FRU.
• Error 22, Unexpected Item on Drive Service Queue-The SDI verification test directed
the disk data channel to execute one of its microdiagnostics. The microdiagnostic completed
without error, and the disk data channel returned the dialog control block (DCB) with no
errors indicated. However, the disk data channel sent the drive state area to its service queue,
indicating an Wlexpected condition in the disk data channel or drive.
• Error 23, Failed To Reacquire Unit-In order for ILDISK to allow looping, the drive under
test must be released and then reacquired. (This method is required to release the drive from
the On-line state.) The release operation succeeded, but the attempt to reacquire the drive
failed. One of the following conditions occurred:
1. Drive Unknown to Disk Functional Code-A fatal error caused the HSC disk functional
software to declare the drive inoperative, so the drive unit number is not recognized. The
drive must now be tested by specifying requestor and port number.
2. Drive Unavailable-The specified drive is now not available for integrity test use.
3. Unknown Status from DDUSUB-Refer to Error 04.
The drive may be allocated to an alternate HSC. Check the drive port lamp to see if this
caused the error.
• Error 24, State Line Clock Not Running-The SDI verification test executed a command
to interrogate the Real-Time Drive State of the drive. The returned status indicates the drive
is not sending state line clock to the disk data channel. Either the port, SDI cable, or drive is
defective or the port is not connected to a drive.
• Error 25, Error Starting I/O Operation-ILDISK detected an error when initiating a disk
Read or Write operation. One of the following conditions occurred:
1. Invalid Header Code-ILDISK did not supply a valid header code to the HSC diagnostic
interface. This indicates a software error and should be reported through a Software
Performance Report (SPR). See Appendix B for detailed information on SPR submission.
2. Could Not Acquire Control Structures-The HSC diagnostic interface could not acquire
sufficient control structures to perform the operation.
3. Could Not Acquire Buffer-The HSC diagnostic interface could not acquire a buffer needed
for the operation.
4. Unknown Status from DDUSUB-The HSC diagnostic interface returned an unknown
status code. Refer to Error 04.
NOTE
Retry ILDISK during lower HSC activity for the second and third problems if these
errors persist.
• Error 26, Init Did Not Stop State Line Clock-The SDI verification test sent an SDI
INITIALIZE command to the drive. When the drive receives this command, it should
momentarily stop sending state line clock to the disk data channel. The disk data channel
did not see the state line clock stop after sending the initialize. The drive is the most probable
FRU.
• Error 27, State Line Clock Did Not Start Up After Init-The SDI verification test sent an
SDI INITIALIZE to the drive. When the drive receives this command, it should momentarily
stop sending state clock to the disk data channel. The disk data channel saw the state clock
stop, but the clock never restarted. The drive is the most probable FRU.
• Error 28, I/O Operation Lost-While ILDISK was waiting for a disk Read or Write operation
to complete, the HSC diagnostic interface notified ILDISK that no I/O operation was in
progress. This error may have been induced by a hardware failure, but it actually indicates
a software problem, and the error should be reported by a software performance report (SPR).
See Appendix B for detailed information on SPR submission.
• Error 29, Echo Data Error-The SDI verification test issued an SDI ECHO command to the
drive. The command completed but the wrong response was returned by the drive. The SDI set
and the disk drive are the probable FRUs.
• Error 30, Drive Went Off Line-The drive, previously acquired by the integrity test, is now
unknown to the disk functional code. This indicates the drive spontaneously went off line or
stopped sending clocks and is now unknown. The test should be restarted using the requestor
and port numbers instead of drive unit number.
• Error 31, Drive Acquired But Can't find Control Area-The disk drive was acquired, and
ILDISK obtained the requestor number and port number of the drive from the HSC diagnostic
interface. However, the specified requestor does not have a control area. This indicates a
software problem and should be reported through a Software Performance Report (SPR). See
Appendix B for detailed infonnation on SPR submission.
• Error 32, Requestor Does Not Have Control Area-ILDISK cannot find a control area for
the requestor supplied by the user. One of the following conditions exists:
1. The HSC does not contain a disk data channel (or other type of requestor) in the specified
requestor position.
2. The disk data channel (or other type of requestor) in the specified requestor position failed
its initialization integrity tests and is not in use by the HSC.
Open the· HSC front door and remove the cover from the card cage. Locate the module slot in
the card cage that corresponds to the requestor. Refer to the module utilization label above the
card cage to help locate the proper requestor. If a blank module (air baffle) is in the module slot,
the HSC does not contain a requestor in the specified position. If a requestor is in the module
slot, check that the red LED on the lower front edge of the module is lit. If so, the requestor
failed and was disabled by the HSC. If the red LED is not lit, a software problem exists and
should be reported through a Software Perfonnance Report (SPR). See Appendix B for detailed
information on SPR submission.
• Error 33, Can't Read Any Sector on Track-As part of test 03, ILDISK attempts to read
a pattern from at least one sector of each track in the read-only area of the drive DEN space
(DBN space is an area on every disk drive reserved for diagnostic or integrity test use). All
drives have the same pattern written to each sector in the read only DBN space.
During the read process, ILDISK detected a track that does not contain any sector with the
expected pattern. Either ILDISK detected errors while reading or the read succeeded, but the
sectors did not contain the correct pattern. The error information for the last sector accessed
is supplied in the error report. The most likely cause of this error is a disk media error. If test
04 also fails, the problem may be in the disk read/write electronics, or the DBN area of the disk
may not be formatted correctly. To interpret the MSCP status code, refer to Section 5.5.11.
• Error 34, Drive Diagnostic Detected Error-The SDI verification test directed the disk
drive to run an internal integrity test. The drive indicated the integrity test failed, but the
error is not serious enough to warrant removing the drive from service. The test number and
error number for the drive are displayed (in hex) in the error report. For the exact meaning of
each error, refer to the service documentation for that drive.
• Error 35, Drive Diagnostic Detected Fatal Error-The SDI verification test directed the
disk drive to run an internal integrity test. The drive indicated the integrity test failed and the
error is serious enough to warrant removing the drive from service. The test and error number
are displayed (in hex) in the error report. For the exact meaning of each error, refer to the
service manual for that drive.
• Error 36, Error Bit Set in Drive Status Error Byte-The SDI verification test executed an
SDI GET STATUS command to the drive under test. The error byte in the returned status was
nonzero indicating one of the following conditions:
1. Drive error
2. Transmission error
3. Protocol error
4. Initialization integrity test failure
5. Write lock error

For the exact meaning of each error, refer to the service manual for that drive.
• Error 37, Attention Set Mter SEEK-The SDI verification routine issued a SEEK command
to the drive which resulted in an unexpected ATTENTION condition. The drive status is
displayed with the error report. Refer to the service manual for that drive.
• Error 38, Available Not Set In Available Drive-The SDI verification routine executed a
command to interrogate the Real-Time Drive State Line of the drive. ILDISK found Available
is not set in a drive that should be Available.
• Error 39, Attention Not Set in Available Drive-The SDI verification routine executed a
command to interrogate the Real-Time Drive State Line of the drive and found Attention is not
asserted even though the drive is Available.
• Error 40, Receiver Ready Not Set-The SDI verification routine executed a command to
interrogate the Real-Time Drive State Line of the drive. The routine expected to find Receiver
Ready asserted, but it was not.
• Error 41, ReadlWrite Ready Set in Available Drive-The SDI verification routine executed
a command to interrogate the Real-Time Drive State Line of the drive and found Available
asserted. However, ReadlWrite Ready also was asserted. ReadlWrite Ready should never be
asserted when a drive is in the Available state.
• Error 42, Available Set in On-line Drive-The SDI verification routine issued an ONLINE
command to the disk drive. Then a command was issued to interrogate the Real Time Drive
State Line of the drive. The line status indicates the drive is still asserting Available.
• Error 43, Attention Set in On-line Drive-The SDI verification routine issued an ONLINE
command to the drive. The drive entered the On-line state, but an unexpected Attention
condition was encountered.
• Error 44, Drive Clear Did Not Clear Errors-When ILDISK issued a GET STATUS
command, error bits were set in the drive response. Issuing a DRIVE CLEAR failed to clear the
error bits. The drive is the probable FRU.
• Error 45, Error Reading LBN-As part of test 14, ILDISK alternates between reading DBNs
and LBNs. This tests the drive's ability to seek properly. The error indicates an LBN read
failed. The drive is the probable FRU.
• Error 46, Echo Framing Error-The framing code (upper byte) of an SDI ECHO command
response is incorrect. The EXPected and ACTual ECHO frames are displayed with the error
message. The K.sdilK.si cable and the drive are the probable FRUs.
• Error 47, K.sdi Does Not Support ECHO-The disk data channel connected to the drive
under test does not support the SDI ECHO command because the disk data channel microcode
is not the latest revision level. This is not a fatal error, but the disk data channel microcode
should be updated to allow for improved isolation of drive-related errors.
• Error 48, ReqIPort Number Information Unavailable-ILDISK was unable to obtain
the requestor number and port number from HSC disk software tables. The drive may have
changed state and disappeared while ILDISK was running. This error also can be caused by
inconsistencies in HSC software structures.
• Error 49, Drive Spindle Not Up to Speed-ILDISK cannot continue testing the drive
because the disk spindle is not up to speed. If the drive is spun down, it must be spun up before
ILDISK can completely test the unit. If the drive appears to be spinning, it may be spinning
too slowly or the drive may be returning incorrect status information to the HSC.
• Error 50, Can't Acquire Drive State Area-ILDISK cannot perform the low-level SDI tests
because it cannot acquire the Drive State Area for the drive. The Drive State .Area is a section
of the K Control Area used to communicate with the drive through the SDI interface. To
perform the SDI tests, ILDISK must take exclusive control of the Drive State .Area; otherwise,
the HSC operational software may interfere with the tests. The Drive State.Area must be in
an inactive state (no interrupts in progress) before it can be acquired by ILDISK If the drive
is rapidly changing its SDI state and generating interrupts, ILDISK may be unable to find the
drive in an inactive state.
• Error 51, Failure While Updating Drive STATUS-When in the process of returning the
drive to the same mode as ILDISK originally found it, an error occurred while performing an
SDI GET STATUS command. When a drive is acquired by ILDISK, the program remembers
whether the drive was in 576-byte mode or 512-byte mode (reflected by the S7 bit of the mode
byte in the drive status). When ILDISK releases the drive (once per pass of the program),
the drive mode is returned to the state the drive was in when ILDISK first acquired it. In
order to ensure the HSC disk functional software is aware of this mode change, ILDISK calls
the diagnostic interface routines to perform a GET STATUS to the drive. These routines also
update the disk functional software information on the drive to reflect the new mode.
Error 51 indicates the drive status update failed. The diagnostic interface returns one of three
different status codes with this error:
1. DRIVE ERROR-The GET STATUS command could not be completed due to an error
during the command. If informational error messages are enabled (through a SET ERROR
INFO command), an error message describing the failure should be printed on the console
terminal.
2. BAD UNIT NUMBER-The diagnostic interface could not find the unit number specified.
The drive may have spontaneously transitioned to the Off-line state (no clocks) since the
last ILDISK operation. For this reason, the unit number is unknown when the diagnostic
interface tries to do a GET STATUS command.
3. UNKNOWN STATUS FROM DDUSUB-Refer to Error 04.
• Error 52, 576-Byte Format Failed-The program attempted to perform a 576-byte format to
the first two sectors of the first track in the read/write DBN area. No errors were detected
during the actual formatting operation, but subsequent attempts to read either of the
reformatted blocks failed. The specific error detected is identified in the error report.
• Error 53, 512-Byte Format Failed-The program attempted to perform a 512-byte format to
the first two sectors of the first track in the read/write DBN area. No errors were detected
during the actual formatting operation, but subsequent attempts to read either of the
reformatted blocks failed. The specific error detected is identified in the error report.
• Error 54, Insufficient Resources to Perform Test-This error indicates further testing
cannot complete due to lack of required memory structures. To perform certain drive tests
ILDISK needs to acquire timers, a dialog control block (DCB), free control blocks (FCBs), data
buffers, and enough Control memory to construct two disk rotational access tables (DRATs). If
any of these resources are unavailable, testing cannot be completed. Under normal conditions
these resources should always be available.
• Error 55, Drive Transfer Queue Not Empty Before Format-ILDISK found a transfer
already queued to the K.sdilK.si when the format test began. ILDISK should have exclusive
access to the drive at this time, and all previous transfers should have been completed before
the drive was acquired. To avoid potentially damaging interaction with some other disk process,
ILDISK aborts testing when this condition is detected.
• Error 56, K.sdi Detected Error During Format-K.sdilK.si detected an error during a
Format operation. Each error bit set in the fragment request block (FRB) is translated. into a
text message that accompanies the error report.
Device ·Integrity Tests 5-21
• Error 57, Wrong Structure on Completion Queue-While formatting, ILDISK checks each
structure returned by the K.sdilK.si to ensure the structure was sent to the proper completion
queue. An Error 57 indicates one of these structures was sent to the wrong completion queue.
This type of error indicates a problem with the K.sdilK.si micro sequencer or a Control memory
failure.
• Error 58, Read Operation Timed Out-To guarantee the disk is on the correct cylinder and
track while formatting, ILDISK queues a Read operation immediately preceding the FORMAT
command. The Read operation did not complete within 16 seconds indicating the KsdiJK.si
is unable to sense sector/index pulses from the disk, or the disk is not in the proper state to
perform a transfer. ILDISK aborts the format test following this error report.
• Error 59, K.sdi Detected Error in Read Preceding Format-To guarantee the disk is on
the correct cylinder and track while formatting, ILDISK queues a Read operation immediately
preceding the FORMAT command. The Read operation failed, so ILDISK aborts the format
test. Each error bit set in the fragment request block (FRB) is translated into a text message
which accompanies the error report.
• Error 60, Read DRAT Not Returned to Completion Queue-To guarantee the disk is on
the correct cylinder and track while formatting, ILDISK queues a Read operation immediately
preceding the format command. The Read operation apparently completed successfully because
the fragment request block (FRB) for the read was returned with no error bits set. However,
the Disk Rotational Access Table (DRAT) for the Read operation was not returned indicating a
problem with the K.sdilK.si.
• Error 61, Format Operation Timed Out-The K.sdilKsi failed to complete a Format
operation. A Format operation consists of a Read followed by a format. The Read completed
successfully, but after waiting a 16-second interval the Format was not complete. A change in
drive state may prevent formatting, the drive may no longer be .sending sector/index information
to the K.sdilK.si, or the K.sdilK.si may be unable to sample the drive state. The format test
aborts on this error to prevent damage to the existing disk format.
• Error 62, Format DRAT Was Not Returned to Completion Queue-The K.sdilK.si failed
to complete a Format operation. A Format operation consists of a read followed by a format.
The Read completed successfully, and the fragment request block (FRB) for the format was
returned by the K.sdilK.si with no error indicated. However, the disk rotational access table
(DRAT) for the Format operation was never returned, indicating a probable K.sdi.lK.si failure.
Mter reporting this error, the format test aborts.
• Error 63, Can't Acquire Specified Unit-ILDISK was initiated automatically to test a disk
drive declared inoperative. When initiated by the disk functional software, ILDISK was given
the requestor number, port number, and unit number of the drive to test. ILDISK successfully
acquired the drive by unit number, but the requestor and port number of the acquired drive did
not match the requestor and port given when ILDISK was initiated. This indicates the HSC
is connected to two separate drives with the same unit number plugs. To prevent inadvertent
interaction with the other disk drive, ILDISK performs only the low-level SDI tests on the unit
specified by the disk functional software. Read/write tests are skipped because the drive must
be acquired by unit number to perform read/write transfers.
• Error 64, Duplicate Unit Detected-At times during the testing sequence, ILDISK must
release, then reacquire, the drive under test. ..After releasing the drive and reacquiring it,
ILDISK noted the requestor and port number of the drive it was originally testing do not
match the requestor and port number of the drive just acquired. This indicates the HSC is
connected to two separate drives with the same unit number. If this error is detected, ILDISK
discontinues testing to prevent inadvertent interaction with the other disk drive.
• Error 65, Format Tests Skipped Due to Previous Error-To prevent possible damage to
the existing disk format, ILDISK does not attempt to format if any errors were detected in the
tests preceding the format tests. This error message informs the user that formatting tests will
not be performed.
• Error 66, Testing Aborted-ILDISK was automatically initiated to test a disk drive declared
inoperative by the disk functional code of the HSC. The disk drive had previously been
automatically tested at least twice and somehow was returned to service. Because the tests
performed by ILDISK may be causing the inoperative drive to be retu.rn.ed to service, ILDISK
does not attempt to test an inoperative drive more than twice. On all succeeding invocations
of ILDISK, an Error 66 message prints and ILDISK exits without performing any tests on the
drive. This prevents ILDISK from automatically initiating and dropping the drive from the test
over and over again.
• Error 67, Not Enough Good DBNs for Format-In order to guarantee the disk is on the
proper cylinder and track, all formatting operations are immediately preceded by a Read
operation on the same track where the format is planned. This requires the first track in the
drive's read/write DBN area to contain at least one good block that can be read without error.
An Error 67 indicates a good block was not found on the first track of the read/write DBN area,
so the fonnatting tests are skipped.
5.5.11 MSCP Status Codes-iLDISK Error Reports

This section lists some of the MSCP status codes that may appear in ILDISK error reports. All
status codes are listed in the octal radix. Further information on MSCP status codes is provided in
Appendix C.
007-Compare Error
OlO-Forced Error
052-SERDES Overrun
053-SDI Command Timeout
lO3-Drive Inoperative
llO-Header Compare or Header Sync Timeout
ll2-EDC Error
ll3-Controller Detected Transmission Error
l50-Data Sync Not Found
l52-Internal Consistency Error
l53-Position or Unintelligible Header Error
2l3-Lost ReadlWrite Ready
253-Drive Clock Dropout
3l3-Lost Receiver Ready
350-Uncorrectable ECC Error
353-Drive Detected Error
4l0-0ne Symbol ECC Error
412-Data Bus Overrun
4l3-State or Response Line Pulse or Parity Error
450-Two Symbol ECC Error
452-Data Memory NXM or Parity Error
453-Drive Requested Error Log
5lO-Three Symbol ECC Error
5l3-Response Length or Opcode Error
550-Four Symbol ECC Error
553-Clock Did Not Restart Mter lnit
6lO-Five Symbol ECC Error
6l3-Clock Did Not Stop Mter Init
650-Six Symbol ECC Error
653-Receiver Ready Collision
710-Seven Symbol ECC Error
7l3-Response Overflow
750-Eight Symbol ECC Error
5.6 ILTAPE - TAPE Device Integrity Tests

The following tests can be initiated through ILTAPE:
• Tape formatter-resident integrity tests
• Functional test of the tape transport
• Full test of the K.stiIK.si interface
When a full test is selected, the KstilK.si microdiagnostics are executed, line state is verified,
an ECHO test is performed, and a default set of formatter tests is executed. See the DRIVE
UNIT NUMBER prompt in the Section 5.6.3 for information on initiating a full test. Detected
failures result in fault isolation to the FRU level.
Three types of tape transport tests are listed below. See Section 5.6.6 for a summary of each.
Fixed canned sequence
User sequence supplied at the terminal
Fixed streamer sequence
Hardware requirements necessary to run ILTAPE include:
• HSC subsystem with K.stiIK.si
• STI-compatible tape formatter
• TA7B, TAB!, or other DSA tape drive (for transfer commands only)
• Console terminal
• RX33 disk drive or equivalent (HSC)
• TU5B tape device or equivalent (HSC50)
In addition, the 110 control processor, program memory, and control memory must be working.
Software requirements necessary to run ILTAPE include:
• CRONIC
• DEMON
• K.stilK.si microcode (installed with the K.stilK.si module)

The following steps outline the procedure for running ILTAPE. The test assumes an HSC is
configured with a terminal and STI interface. If the HSC is not booted, start with step 1. If
the HSC is already booted, proceed to step 2.
1. Boot the HSC.
Press the lnit button on the HSC OCP. The following message appears:
INIPIO-I Booting •..
The boot process takes about 1 minute, and then the following message appears:
ase Version xxxx Date Time System n
2. Press CTRI1Y.
This causes the KMON prompt:

HSC>
3. Type R DXn:ILTAPE.
This invokes the tape device integrity test program ILTAPE. The DXn is the HSC device name. The
n refers to the unit number of the specific HSC drive. For example, DX1: refers to RX33 Drive 1
(HSC) and DD1: refers to TU58 Drive 1 (HSC50). The following message appears:
ILTAPE>D>hh:mm Execution Starting

The test can be terminated by pressing CTRUC. Certain errors that occur during execution will
cause ILTAPE to terminate automatically.
5.6.3 User Dialog

The following paragraphs describe ILTAPE/user dialog during execution of ILTAPE. Note that the
default values for input parameters appear within the brackets of the prompt. The absence of a
value within the brackets indicates the input parameter is not defaultable.
DRIVE UNIT NUMBER (U) []?
To run formatter tests or transport tests, enter Tnnn, where nnn is the MSCP unit number (such
as T316).
For a full interface test, enter Xm, where m is any number. Typing X instead of T requires a
requestor number and slot number. The following two prompts solicit requestor/slot numbers:
ENTER REQUESTOR NUMBER (2-9) []?
Enter the requestor number. The range includes numbers 2 through 9, with no default value.
ENTER PORT NUMBER (0-3) []?
Enter the port number. The port number must be 0, 1, 2, or 3 with no default value. After this
prompt is answered, ILTAPE executes the K.stilK.si interface test.
EXECUTE FORMATTER DIAGNOSTICS (YN) [Y]?
Enter RETURN to execute formatter tests. The default is Y. Entering N will not run formatter
tests.
MEMORY REGION NUMBER (H) [O].?
This prompt appears only if the response to the previous prompt was RETURN. A formatter test is
named according to the formatter memory region where it executes. Enter the memory region (in
hex) in which the formatter test is to execute. ILTAPE continues at the prompt for iterations. Refer
to the appropriate tape drive service manual for more information on formatter tests.
EXECUTE TEST OF TAPE TRANSPORT (YN) [N]?
To test the tape transport, enter Y (the default is N). If no transport testing is desired, the dialog
continues with the ITERATIONS prompt. Otherwise, the following prompts appear:
IS MEDIA MOUNTED (YN) [N]?
This test writes to the tape transport, requiring a mounted scratch tape. Enter Y if a scratch tape
is already mounted.
FUNCTIONAL TEST SEQUENCE NUMBER (D) [1]?
Select one of five transport tests. The default is 1 (the canned sequence). Enter 0 if a new user
sequence will be input from the terminal. Enter 2, 3, or 4 to select a user sequence previously input
and stored on the HSC device. User sequences are described in Section 5.6.4. Enter 5 to select the
streaming sequence.
INPUT STEP 00:
This prompt appears only if the response to the previous prompt was o. See Section 5.6.4 for a
description of user sequences.
ENTER CANNED SEQUENCE RON TIME IN MINUTES (D) [1]?
Answering this prompt determines the time limit for the canned sequence. It appears only if the
canned sequence is selected. Enter the total run time limit in minutes. The default is 1 minute.
SELECT DENSITY (O=ALL, 1=1600, 2=6250) [O]?
This prompt permits selection of the densities used during the canned sequence. It appears only if
the canned sequence is selected. One or all densities may be selected; the default is all.
SELECT DENSITY (1=800, 2=1600, 3=6250) [3]?
This prompt appears only if a user-defined test sequence was selected. The prompt permits
selection of anyone of the possible tape densities. The default density is 6250 bits per inch (bpi).
Enter 1, 2, or 3 to select the desired tape density.
1 = 800 bpi
2 = 1600 bpi
3 = 6250 bpi
The next series of prompts concerns speed selection. The particular prompts depend upon the type
of speeds supported (fixed or variable). ILTAPE determines the speed types supported and prompts
accordingly.
If fixed speeds are supported, ILTAPE displays a menu of supported speeds, as follows:
Fixed Speeds Available:
(1) ssss ips
(2) ssss ips
(3) ssss ips
(4) ssss ips
The supported speed in inches per second is shown as ssss. The maximum number of supported
speeds is 4. Thus, n cannot be greater than 4. The prompt for a fixed speed is:
SELECT FIXED SPEED (D) [1]?
To select a fixed speed, enter a digit (n) corresponding to one of the above displayed speeds. The
default is the lowest supported speed. ILTAPE continues at the data pattern prompt.
If variable speeds are supported, ILTAPE displays the lower and upper bounds of the supported
speeds as follows:
VARIABLE SPEEDS AVAILABLE:
LOWER BOUND = III ips
UPPER BOUND = uuu ips
NOTE
If only a single speed is supported, ILTAPE does not prompt for speed. It runs at the
single speed supported
To select a variable speed, enter a number within the bounds, inclusively, of the displayed supported
variable speeds. The default is the lower bound. The prompt for a variable speed is:
SELECT VARIABLE SPEED (D) [0 = LOWEST]?
The next prompt asks for the data pattern.

DATA PATTERN NUMBER (D) [3]?
Choose one of five data patterns.

O-User supplied
I-All zeros
z..:-All ones
3-Ripple zeros
4-Ripple ones
The default is 3. If the response is 0, the following prompts appear:
HOW MANY DATA ENTRIES (D) []?
Enter the number of unique words in the data pattern. Up to 16 words are permitted.
DATA ENTRY (H) []?
Enter the data pattern word (in hex), for example, ABCD. This prompt repeats until the all data
words specified in the previous prompt are exhausted.
SELECT RECORD SIZE (GREATER THAN OR EQUAL TO 1) (D) [8192]?
Enter the desired record size in decimal bytes. The default is 8192 bytes. The maximum record
size that can be specified is 12288.
NOTE
This prompt does not appear if streaming is selected.
ITERATIONS (D) [1]?
Enter the number of times the selected tests are to run. After the number of iterations is entered,
the selected tests begin execution. Errors encountered during execution cause display of appropriate
messages at the terminal.
5.6.4 User Sequences

To test/exercise a tape transport, write a sequence of commands at the terminal. This sequence
may be saved on the HSe device and be recalled for execution at a later time. Up to three user
sequences can be saved on the HSC device.
Following is a list of supported user sequence commands:
WRT-Write one record
RDF-Read one record forward
RDFC-Read one record forward with compare
RDB-Read one record backward
RDBC-Read one record backward with compare
FSR-Forward space one record
FSF-Forward space one file
BSR-Backspace one record
BSF-Backspace one file
REW-Rewind
RWE-Rewind with erase
UNL--Unload (after rewind)
WTM-Write tape mark
ERG-Erase gap
Cnnn -Counter set to nnn (0 = 1000.)
Dnnn -Delay nnn ticks (0 = 1000.)
BRnn-Branch unconditionally to step nn
DBnn-Decrement counter and branch if nonzero to step nn
TMnn-Branch on Tape Mark to step nn
NTnn-Branch on no Tape Mark to step nn
ETnn-Branch on EOT to step nn
NEnn-Branch on not EOT to step nn
QillT-Terminate input of sequence steps
To initiate the user sequence dialog, type 0 in response to the prompt:
FUNCTIONAL TEST SEQUENCE NUMBER (D) [1]?
The following paragraphs describe the ILTAPE user dialog during a new user sequence.
INPUT STEP nn
Enter one of the user sequence commands listed previously. ILTAPE keeps track of the step
numbers and automatically increments them. Up to 50 steps may be entered. Typing QUIT in
response to the INPUT STEP prompt terminates the user sequence. At that time, the following
prompt appears:
STORE SEQUENCE AS SEQUENCE NUMBER (0,2,3,4) [OJ?
The sequence entered. at the terminal may be stored on the HSe load device in one of three files. To
select one of these files, type 2, 3, or 4. Once stored, the sequence may be recalled for execution at a
later time by referring to the appropriate file (typing 2, 3, or 4 in response to the sequence number
prompt).
Typing <BOLD>(the default) indicates the user sequence just entered should not be stored. In this
case, the sequence cannot be run at a later time.
An example of entering a user sequence follows:
INPUT STEP 00 REW ;Rewind the tape

INPUT STEP 01 C950 ;Set counter to 950
INPUT STEP 02 WRT ;Write one record
INPUT STEP 03 ET07 ;If EOT branch to step 7
INPUT STEP 04 RDB ;Read backward one record
INPUT STEP 05 FSR ;Forward space one record
INPUT STEP 06 DB02 ;Decrement counter, branch
;to step 2 if nonzero
INPUT STEP 07 REW ;Rewind the tape
INPUT STEP 08 QUIT ;Terminate sequence input
STORE SEQUENCE AS SEQUENCE NUMBER (0,2,3,4) [O]? 3
This sequence writes a record, reads it backwards, and skips forward over it. If an EOT is
encountered prior to writing 950 records, the tape is rewound and the sequence terminates. Note,
the sequence is saved on the HSC device as sequence number 3 and can be recalled at a later
execution of ILTAPE.

When transport testing is finished, a summary of soft errors appears on the terminal upon
completion of the test. The format of this summary is:
SOFT ERROR SUMMARY: READ WRITE COMPARE
xxxxxx. xxxxxx xxxxxx
Successful completion of a formatter test is indicated by the following message on the terminal:
TEST nnnn DONE
The formatter test number is represented by nnnn.

When an error is encountered, an appropriate error message is printed on the terminal.

The following sections summarize the tests contained in ILTAPE.
5.6.6.1 Interface Test Summary

This portion of ILTAPE tests the standard tape interface (STI) of a specific tape data channel
and port. It also performs low-level testing of the formatter by interfacing to the K.stilK.si drive
service area (port) and executing various Level 2 STI commands. The testing is limited to dialog
operations; no data transfer is done. The operations performed are DIAGNOSE, READ MEMORY,
GET DRIVE STATUS, and READ LINE STATUS.
K.stilK.si microdiagnostics are executed to verify the tape data channel. A default set of formatter
tests (out of memory region 0) is executed to test the formatter, and an echo test is performed to
test the connection between the port and the formatter.
Failures detected are isolated to the extent possible and limited to tape data channels, the STI set,
or the formatter. The STI set includes a small portion of the K..stilK..si module and the entire STI
(all connectors, cables, and a small portion of the drive). The failure probabilities of the STI set are:
1. STI cables or connectors (most probable)
2. Formatter
3. K.sti!K.si (least probable)
When the STI set is identified as the FRU, replacement should be in the order indicated in the
preceding list.
5.6.6.2 Formatter Test Summary

Fonnatter tests are executed out of a formatter memory region selected by the user. Refer to the
tape drive service manual for a description of the formatter tests. Failures detected identify the
fonnatter as the FRU.
5.6.6.3 User Sequence Test Summary

User sequences are used to exercise the tape transport. The particular sequence is totally user-
defined. Refer to Section 5.6.4.
5.6.6.4 Canned Sequence Test Summary

The canned sequence is a fixed routine for exercising the tape transport. The canned sequence first
performs a quick verify of the ability to read and write the tape at all supported densities. Using
a user-selected record size, the canned sequence then writes, reads, and compares the data written
over a 200-foot length of tape. Positioning over this length of tape is also performed. Finally,
random record sizes are used to write, read, compare, and position over a 50-foot length of tape.
Errors encountered during the canned sequence are reported at the terminal.
5.6.6.5 Streaming Sequence Test Summary

The streaming sequence is a fixed sequence that attempts to write and read the tape at speed
(without hesitation). The entire tape is written, the tape is rewound, and the entire tape is read
back. Execution may be terminated at any time by pressing CTRUY.
NOTE
In reading the tape, ll..TAPE uses the ACCESS command. This allows the tape to move at
speed.. This is necessary because of the buffer size restrictions existing for test programs.

ILTAPE conforms to the test generic error message format (Section 5.1.2). An example of an
ILTAPE error message follows:
ILTAPE>D>09:31 T # 011 E # 011 U-T00101
ILTAPE>D>COMMAND FAILURE
ILTAPE>D>MSCP WRITE MULTIPLE COMMAND
ILTAPE>D>MSCP STATUS: 000000
ILTAPE>D>POSITION 001792
The test number reflects the state level where ILTAPE is executing when an error occurs. This
number does not indicate a separate test that can be called. Table 5-1 defines the ILTAPE test
levels.
Table &-1 ILTAPE Test Levels

Test Number ILTAPE State
o Initialization of tape software interface

1 Device (port, formatter, unit) acquisition
2 STI interface test in execution
3 Formatter tests executing in response to Diagnostic Request (DR) bit
4 Tape transport functional test
5 User-selected formatter test executing
6 Termination and clean-up
The optional text is dependent upon the type of error.

The following list describes ILTAPE error messages.
• Error 01, Initialization Failure--Tape path software interface cannot be established due to
insufficient resources (buffers, queues, timers, and so forth).
• Error 02, Selected Unit Not a Tape--Selected drive is not known to the HSC as a tape.
• Error 08, Invalid RequestorlPort Number--Selected requestor number or port number is
out of range or requestor selected is not known to the system.
• Error 04, Requestor Not A K.sti-Selected requestor is not known to the system as a tape
data channel.
• Error 05, Timeout Acquiring Drive Service Area-While attempting to acquire the drive
service area (port) in order to run the STI interface test, a timeout occurred. If this happens,
the tape functional code is corrupted. ILTAPE invokes a system crash.
• Error 06, Requested Device Unknown-Device requested is not known to the tape
subsystem.
• Error 07, Requested Device Is Busy-Selected device is on line to another controller or host.
• Error 08, Unknown Status from Tape Diagnostic Interface--An unknown status was
returned from the test software interface TDUSUB.
• Error 09, Unable to Release Device--Upon termination of ILTAPE or upon an error
condition, the device(s) could not be returned to the system.
• Error 10, Load Device Write Error-CHECK IF WRITE LOCKED-An error occurred
while attempting to write a user sequence to the HSC device. Check to see if the HSC load
device is write-protected. The prompt calls for a user sequence number. To break the loop of
reprompts, press CTRUY.
• Error 11, Command Failure--A command failed during execution of ILTAPE. The command
in error may be one of several types, such as an MSCP or Level 2 STI command. The failing
command is identified in the optional text of the error message. For example:
ILTAPE>D>MSCP READ COMMAND
ILTAPE>D>MSCP STATUS: nnnnnn
• Error 12, Read Memory Byte Count Error-The requested byte count used in the read
(formatter) memory command is different from the actual byte count received.
EXPECTED COUNT: xxxx ACTUAL COUNT: yyyy --
• Error 13, Formatter Diagnostic Detected. Error-A test running in the formatter detects
an error. Any error text from the formatter is displayed.
• Error 14, Formatter Diagnostic Detected Fatal Error-A test running in the formatter
detects a fatal error. Any error text from the formatter is displayed.
• Error 15, Load Device Read Error-While attempting to read a user sequence from the load
device, a read error was encountered. Ensure a sequence has been stored on the load device as
identified by the user sequence number. The program reprompts for a user sequence number.
To break the loop of reprompts, press CTRLIY.
• Error 16, Insufficient Resources to Acquire Specified Device-During execution, ILTAPE
was unable to acquire the specified device due to a lack of necessary resources. This condition
is identified to ILTAPE by the tape functional code through the diagnostic interface, TDUSUB.
ILTAPE has no knowledge of the specific unavailable resource.
• Error 17, K Microdiagnostic Did Not Complete-During the STI interface test, the
requestor microdiagnostic timed out.
• Error 18, K Microdiagnostic Reported Error-During the STI interface test, an error
condition was reported by the K microdiagnostics.
• Error 19, DCB Not Returned, K Failed for Unknown Reason-During the STI interface
test, the requestor failed for an undetermined reason and the Diagnostic Control Block (DCB)
was not returned to the completion queue.
• Error 20, Error in DCB upon Completion-During the STI interface test, an error condition
was returned in the DCB.
• Error 21, Unexpected Item on Drive Service Queue-During the STI interface test, an
unexpected entry was found on the drive service queue.
• Error 22, State Line Clock Not Running-During the STI interlace test, execution of an
internal command to interrogate the Real-Time Formatter State line of the drive indicated the
state line clock is not running.
• Error 23, Init Did Not Stop State Line CLock-During the STI interlace test, after
execution of a formatter INITIALIZE command, the state line clock did not drop for the time
specified in the STI specification.
• Error 24, State Line Clock Did Not Start Up After Init-During the STI interface test,
after execution of a formatter INITIALIZE command, the state line clock did not start up
within the time specified in the STI specification.
• Error 25, Formatter State Not Preserved Across lnit-The state of the formatter prior to
a formatter initialize was not preserved across the initialization sequence.
• Error 26, Echo Data Error-Data echoed across the STI interface was incorrectly returned.
• Error 27, Receiver Ready Not Set-Mter issuing an ONLINE command to the formatter, the
Receiver Ready signal was not asserted.
• Error 28, Available Set in On-line Formatter-Mter successful completion of a formatter
ONLINE command to the formatter, the Available signal is set.
• Error 29, Load Device Errol"-File Not Found-During the user sequence dialog, ILTAPE
was unable to locate the sequence file associated with the specified user sequence number.
Ensure load device media is properly installed. The program reprompts for a user sequence
number. To break the loop of reprompts, press CTRIJY.
• Error 30, Data Compare Error-During execution of the user or canned sequence, ILTAPE
encountered a software compare mismatch on the data written and read back from the tape.
The software compare is actually carried out by a subroutine in the diagnostic interlace,
TDUSUB. The results of the compare are passed to ILTAPE. Informatipn in the text of the error
message identifies the data in error.
• Error 31, EDe Error-During execution of the user or canned sequence, ILTAPE encountered
an EDC error on the data written and read back from the tape. This error is actually detected
by the diagnostic interface, TDUSUB, and reported to ILTAPE. Information in the text of the
error message identifies the data in error.
• Error 32, Invalid Multiunit Code from GUS Command-Mter a unit number is input to
ILTAPE and prior to acquiring the unit, ILTAPE attempts to obtain the unit's multiunit code
through the GET UNIT STATUS command. This error indicates a multiunit code of zero was
returned to ILTAPE from the tape functional code. Because a multiunit code of zero is invalid,
this error is equivalent to a device unknown to the tape subsystem.
• Error 33, Insufficient Resources To Acquire Timer-ILTAPE was unable to acquire a timer
from the system; insufficient buffers are available in the system to allocate timer queues.
• Error 34, Unit Unknown or On Line to Another Controller-The device identified by the
selected unit number is either unknown to the system or it is on line to another controller.
Verify the selected unit number is correct and run ILTAPE again.
5.7 ILTCOM - Tape Compatibility Test

ILTCOM tests the compatibility of tapes that may have been written on different systems and
different drives with STI compatible drives connected to an HSC through the STI bus. ILTCOM
may generate, modify, read, or list a compatibility tape. Data read from the compatibility tape is
compared to the expected pattern. A compatibility tape consists of file groups (called bunches) of
specific data pattern records.
Each bunch contains a header record and several data records of different sizes and is terminated
by a tape mark. The last bunch on a tape is followed by an additional tape mark (thus forming
logical EOT).
Each bunch contains a total of 199 records: one header record followed by 198 data records. The
header record contains 48 (decimal) bytes of 6 bit-encoded descriptive information, as follows:
Table 5-2 ILTCOM Header Record

Field Description Length Example
1 Drive type 6 bytes TA78

2 Drive serial number 6 bytes 123456
3 Processor type 6 bytes HSC
4 Processor serial number 6 bytes 123456
5 Date 6 bytes 093083
6 Comment 1 18 bytes Comment
lILTCOM can read but cannot generate a comment field.
The data records are arranged as follows:

• Sixty-six records 24 (decimal) bytes in length. These records sequence through 33 different data
patterns. The 1st and 34th records contain pattern 1, the 2nd and 35th records contain pattern
2, and so forth, through the 33rd and 66th records containing pattern 33.
• Sixty-six records 528 (decimal) bytes in length. These records sequence through the 33 data
patterns as described above.
• Sixty· six records 12,024 (decimal) bytes in length. These records sequence through the 33 data
patterns in the same manner as the preceding data patterns.
The data patterns used are shown in Table 5-3.
Table 5-3 ILTCOM Data Patterns

Pattern
Number Pattern Description
1 377 Ones
2 000 Zeros
3 274,377,103,000 Peak shift
4 000,377,377,000 Peak shift
5 210,104,042,021 Floating one
6 273,167,356,333 Floating zero
7 126,251 Alternate bits
8 065,312 Square pattern
9 000,377 Alternate frames
10 001 °
Track on
11 002 Track 1 on
12 004 Track 2 on
13 010 Track 3 on
14 020 Track 4 on
15 040 Track 5 on
16 100 Track 6 on
17 200 Track 7 on
18
19
376
375
°
Track off
Track 1 off
20 373 Track 2 off
21 367 Track 3 off
22 357 Track 4 off
23 337 Track 5 off
24 277 Track 6 off
25 177 Track 7 off
26 207,377,370,377 Bit peak shift.
27 170,377,217,377
28 113, 377, 264, 377
29 035,377,342,377
30 370,377,207,377
31 217,377,170,377
32 264,377,113,377
33 342,377,035,377

The hardware requirements necessary to run ILTCOM include:
• An HSC subsystem with KstilK.si
• STI-compatible tape formatter
• Tape drive
Because ILTCOM is not diagnostic in nature, all of the necessary hardware is assumed to be
working. EITors are detected and reported, but fault isolation is not a goal of ILTCOM.
ILTCOM software requirements include:

• CRONIC
• DEMON
• KstilK.si microcode
• TFUNCT
• TDUSUB

The following steps outline the procedure for running ILTCOM. ILTCOM assumes the HSC is
configured with a terminal, STI interface, and a TA78 tape drive (or STI-compatible equivalent). If
the HSC is already booted, proceed to step 2. If the HSC needs to be booted, start with step 1.
1. Boot the HSC.
Press the Init button on the OCP of the HSC. The following message appears:
INIPIO-I Booting •..
The boot process can take several minutes, and then the following message appears:
HSC Version xxxx Date Time System n
2. Press CTRLIY.
This causes the KMON prompt to appear:
HSC>
3. 1)rpe R DXn.:ILTCOM. The variable n equals the number of the RX33 drive containing the HSC
system diskette. When running ILTCOM on an HSC50, use DDn: to access the TU58 tape
drive.
This invokes the compatibility test program ILTCOM. The following message appears:
ILTCOM>D>hh:mm Execution Starting
The subsequent program dialog is described in the next section.

ILTCOM is terminated normally by selecting the exit function (EXIT) or by pressing CTRUY or
CTRUC. Certain errors that occur during execution cause ILTCOM to terminate automatically.

ILTCOM allows the writing, reading, listing, or modifying of compatibility tapes. The following
describes the user dialog during the execution of ILTCOM.
DRIVE UNIT NUMBER (U) []?
Enter the tape drive MSCP unit number (such as T21).

SELECT DENSITY FOR WRITES (1600, 6250) []?
Enter the write density by typing (up to) four characters of the density desired (1600 for 1600 bpi).
SELECT FUNCTION (WR=WRITE,REA=READ,ER=ERASE,
LI=LIST,REW=REWIND,EX=EXIT) [J?
Enter the function by typing the characters that uniquely identify the desired function (for instance,
REA for read).
The subsequent dialog is dependent upon the function selected.
• WRITE-The write function writes new bunches on the compatibility tape. Bunches are either
written one at a time or over the entire tape. Bunches are written from the current tape
position. If the write function is selected, the following prompts occur:
PROCEED WITH INITIAL WRITE (YN) [NJ?
Type Y to proceed with the initial write. The default is no, in which case program control is
continued at the function selection prompt. If the response is yes, the following prompt occurs.
WRITE ENTIRE TAPE (YN) [NJ?
Type Y if the entire tape is to be written. Writing of bunches begins at the current tape position
and continues to physical EOT. Type the default N if the entire tape is not to be written. rn this
case, only one bunch is written from the current tape position. This prompt only appears on
the initial write selection. Mter the bunch has been written, control continues at the function
selection prompt.
• READ-The read function reads and compares the data in the bunches with an expected
(predefined) data pattern. As the reads occur, the bunch header information is displayed at the
terminal. The format of the display is shown in the following example:
BUNCH 01 WRITTEN BY TA78 SERIAL NUMBER 002965
ON A HSC SERIAL NUMBER 005993 ON 09-18-84
The number of bunches to be read is user selectable. All reads are from beginning of tape
(BOT). If the read function is selected, the following prompt appears:
READ HOW MANY BUNCHES (D) [O=ALLJ?
Type the number of bunches to be read. The default (0) causes all bunches to be read. Mter the
requested number of bunches have been read and compared, control continues at the function
selection prompt.
• LIST-The list function reads and displays the header of each bunch on the compatibility tape
from BOT. The display is the same as the one described under the read function. The data
contents of the bunches are not read and compared. Mter listing the tape bunch headers,
control continues at the function selection prompt.
• ERASE-The erase function erases a user-specified number of bunches from the current tape
position toward BOT. ILTCOM backs up the specified number of tape marks and writes a second
tape mark (logical EOT). This effectively erases the specified number of bunches from the tape.
Thus, for example, if the current tape position is at bunch 5 and the user wishes to erase two
bunches, three bunches are left on the tape after the ERASE command completes.
ILTCOM does not allow the user to erase all bunches. At least one bunch must remain. For
example, with five bunches on the tape, only four bunches can be erased.
If the erase function is selected, the following prompt appears at the terminal:
ERASE HOW MANY BUNCHES FROM CURRENT POSITION (D) [OJ?
Type the number of bunches to be erased. The default of 0 results in no change in tape contents
or position. Control continues at the function selection prompt.
• REWIND-The rewind function rewinds the tape to BOT.
• EXIT-The exit function rewinds the tape and exits the tape compatibility program ILTCOM.

ILTCOM writes, reads, and compares compatibility tapes upon user selection. The testing that
takes place looks for compatibility of tapes written on different drives (and systems).
As incompatibilities due to data compare errors or unexpected formats are found, they are reported.
ILTCOM makes no attempt to isolate faults during execution; it merely reports incompatibilities
and other errors as they occur.

ILTCOM conforms to the test generic error message format (Section 5.1.2). An example of an
ILTCOM error message follows:
ILTCOM>D>09:29 T 000 E 003 U-T00100
ILTCOM>D>COMMAND FAILURE
ILTCOM>D>OPTIONAL TEXT
Where:
E nnn is an error number.
U-Txxxxx indicates the Tape MSCP unit number.
The optional text is dependent upon the type of error. Some error messages contain the term object
count in the optional text. Object count refers to tape position (in objects) from BOT.

The following are the ILTCOM error messages.
• Error 01, Initialization Failure-Tape path cannot be established due to insufficient
resources.
• Error 02, Selected Unit Not a Tape-User selected a drive not known to the system as a
tape.
• Error 03, Command Failure-A command failed during execution of ILTCOM. The command
in error may be one of several types (MSCP level, STI Level 2, and so forth). The failing
command is identified in the optional text of the error message. For example:
ILTCOM>D>tt:tt T 000 E 003 U-T00030
ILTCOM>D>COMMAND FAILURE
ILTCOM>D>MSCP READ COMMAND
ILTCOM>D>MSCP STATUS: nnnnnn
• Error 05, Specified Unit Not Available-The selected unit is on line to another controller.
• Error 06, Specified Unit Cannot Be Brought On Line-The selected unit is offline or not
available.
• Error 07, Specified Unit Unknown-The selected unit is unknown to the HSC configuration.
• Error 08, Unknown Status from TDUSUB-An unknown error condition returned from the
software interface TDUSUB.
• Error 09, Error Releasing Drive-After completion of execution or after an error condition,
the tape drive could not successfully be returned to the system.
• Error 10, Can't Find End of Bunch-The compatibility tape being read or listed has a bad
format.
• Error 11, Data Compare Error-A data compare error has been detected. The ACTual and
EXPected data are displayed in the optional text of the error message. For example:
ILTCOM>D>tt:tt T 000 E 011 U-T00030
ILTCOM>D>DATA COMPARE ERROR
ILTCOM>D>EXPECTED DATA: XXXXXX ACTUAL DATA: YYYYYY
ILTCOM>D>NUMBER OF FIRST WORD IN ERROR: nnnnn
ILTCOM>D>NUMBER OF WORDS IN ERROR: mmmmm
ILTCOM>D>OBJECT COUNT = cccccc
• Error 12, Data EDC Error-An EDC error was detected. ACTual and EXPected values are
displayed in the optional text of the error message.
5.8 ILEXER - Multidrive Exerciser

ILEXER exercises the various disk drives and tape drives attached to the HSC subsystem. The
exerciser is initiated upon demand. Drives to be tested are selected by the operator. The exerciser
issues random READ, WRITE, and COMPARE commands to exercise the drives. The results of the
exerciser are displayed on the terminal from which it was initiated.
The reports given by ILEXER do not provide any analysis of the errors reported, nor explicitly call
out a specific FRU. This is strictly an exerciser.
ILEXER runs with other processes on the HSC subsystem. It is loaded from the RX33 or TU58 and
uses the services of the Diagnostic Execution Monitor (DEMON) and the HSC control software.

In order for the ILEXER program to run, the following hardware and software items must be
available.
1. HSC subsystem:
a. Console terminal
b. P.io
c. K.sdi, K.sti, or K.si
d. Program, Control, and Data memories
e. RX33 (HSC) system diskette or TU58 (HSC50) system tape load device
2. SDI compatible disk drive
and/or
3. STI compatible tape drive
4. HSC system software, including:
a. HSC internal operating system
h. DEMON
c. K.sdilK.si microcode
and/or
d. KstilK.si microcode
e. SDI manager
and/or
f. STI manager or equivalent
g. Disk functional code
and/or
h. Tape functional code
1. Error Handler
j. Diagnostic Interface to Disk functional code
and/or
k. Diagnostic Interface to Tape functional code
Tests cannot be performed on drives if their respective interfaces are not available (K.sdi, Ksti, or
Ksi).

Perform the following steps to initiate ILEXER:
1. Press CTRLIY.
2. The HSC responds with:
HSCxx>
3. Type RUN DXO:ILEXER.

DXO: refers to RX33 Drive 1. On an HSC50, DD1: refers to the TU58 Drive 1.
The system loads the program from the specified local HSC load media (any appropriate media with
the image ILEXER in an RT 11 format). When the program is successfully loaded, the following
message is displayed:
ILEXER>D>hh:mm Execution Starting
Where:
hh:mm is the current time.
ILEXER then prompts for parameters. After all prompts are answered, the execution of the test
proceeds. Error reports and performance summaries are returned from ILEXER.
When ILEXER has run for the specified time interval, reported any errors found, and generated a
final performance summary, the exerciser concludes with the following message:
ILEXER>D>hh:mm Execution Complete

Upon completion of the exercise on each selected drive, reporting of any errors found, and displaying
of final performance summary, ILEXER terminates normally. All resources, including the drive
being tested, are released. The operator may terminate ILEXER before normal completion by
pressing CTRLIY. The following output is displayed, plus a final performance summary:
ILEXER>D>hh:mm DIAGNOSTIC ABORTED
ILEXER>D>PLEASE WAIT -- CLEARING OUTSTANDING I/O
Certain parts of ILEXER cannot be interrupted, so the CTRLIY may have no effect for a brief
moment and may need repetition. Whenever ILEXER is terminated, whether normally or
by operator abort, ILEXER always completes any outstanding I/O requests and prints a final
performance summary.

The parameters in ILEXER follow the format:
PROMPT DESCRIPTION (DATATYPE) [DEFAULT]?
"","
Where:
• PROMPT DESCRIPTION explains the type of information ILEXER needs from the operator.
• The DATATYPE is the form ILEXER expects and can be one of the following:
YIN-Yes/no response
D -Decimal number
U -Unit number (see form below)
H -Number (in hex)
• DEFAULT is the value used if a carriage return is entered for that particular value. If a default
value is not allowed, it appears as [J.
The next prompt is:
DRIVE UNIT NUMBER (U) [] ?
Enter the unit number of the drive to be tested. This prompt has no default. Unit numbers are
either in the form Dnnnn or Tnnnn, where nnnn is a decimal number between 0 and 4095 that
corresponds to the number printed on the drive's unit plug. The D or T indicates either a disk
drive or tape drive, respectively. Terminate the unit number with a carriage return. ILEXER
attempts to acquire the specified unit through the HSC Diagnostic Interface. If the unit is acquired
successfully, ILEXER continues with the next prompt. If the acquire fails with an error, one of the
following conditions was encountered:
1. The specified drive is unavailable. This indicates the drive is connected to the HSC but is
currently on line to a host CPU or HSC utility. On-line drives cannot be diagnosed. ILEXER
repeats the prompt for the unit number.
2. The specified drive is unknown to the HSC disk functional software. Drives are unknown for
one of the following reasons:
• The drive and/or KsdilK.si port is broken and cannot communicate with the disk functional
software.
• The drive was communicating with the HSC when a serious error occurred and the HSC
ceased communicating with the drive.
In either case, ILEXER asks the operator if another drive will be selected. If so, it askS for the
writ number. If not, ILEXER begins to exercise the drives selected. If no drives are selected,
ILEXER terminates.
When a disk drive is specified, one set of prompts is presented. When a tape drive is selected, an
entirely different set of prompts is presented. Pressing CTRUZ at any time during parameter input
selects the default values for the remaining parameters.
Mter a drive is selected and ILEXER has both acquired the drive and brought it on line, or if a
nondefaultable parameter is encountered, the following prompt appears:
ILEXER>D>hh:mm Nondefaultable Parameter
Select up to 12 drives to be exercised: either all disk drives, all tape drives, or a combination of the
two.
5.8.5 Disk Drive Prompts

The following prompts are presented if the drive selected is a disk drive.
ACCESS USER DATA AREA (YIN) [N]?
Answering Y to this and the next prompt directs ILEXER to perform testing in the user data area.
It is the operator's responsibility to see that the data contained there is either backed up or of no
value. If this prompt is answered with an N or carriage return, testing is confined to the disk area
reserved for diagnostics or integrity tests (DBN area). When testing is confined to the DBN area,
the following five prompts are not displayed.
ARE YOU SURE (YIN) [N]?
Answering N causes the DBN area to be exercised. Answering Y allows the exercise to take place
in the user data area of the disk.
DO YOU WANT BBR (YIN) [Y]?
Answer N if the drive is suspected as bad. If you are positive the drive is good, answer Y to enable
BBR.
START BLOCK NUMBER (D) [OJ?
This value specifies the starting block of the area ILEXER exercises when the user data area is
selected. If block 0 is specified, ILEXER begins with the first LBN on the disk.
END Block NUMBER (D) [O=MAX]?
This parameter specifies the ending block of the area ILEXER exercises when the user data area
is selected. If block 0 is specified as the ending block, ILEXER exercises up to the last LBN on the
disk.
INITIAL WRITE TEST AREA (YIN) [N]?
Answering Y to this prompt causes ILEXER to write the entire test area before beginning random
testing. If the prompt is answered with an N or a carriage return, the prompt immediately
following is omitted.
TERMINATE TEST ON THIS DRIVE FOLLOWING INITIAL WRITE (YIN) [N]?
This question allows an initial write on the drive and terminates the test at that point. The default
answer (N) permits this initial write. After completing the initial write, the test continues to
exercise the drive.
NOTE
The following prompts specify the test sequence for that part of the test following the
initial write portion. That is, even if the operator requests read-only mode, the drive will
not be write-protected until after any initial write has been completed.
SEQUENTIAL ACCESS (YIN) [N]?
The operator has the option of requesting all disk data access be performed in a sequential manner.
READ ONLY (YIN) [N]?
If answered N, the operator is asked for both a pattern number and the possibility of write-only
mode. If the answer is Y, ILEXER does not prompt for write-only mode, but only asks for a data
pattern number if an initial write was requested.
DATA PATTERN NUMBER (0-15) (D) [15]?
The operator has the option of selecting one of 16 disk data patterns. Selecting data pattern 0
allows selection of a pattern with a maximum of 16 words. The default data pattern (15) is the
factory format data pattern.
WRITE ONLY (YIN) [N]?
This option permits only Write operations on a disk. This prompt is not displayed if read-only mode
is selected.
DATA COMPARE (YIN) [N]?
If this prompt is answered with an N or a carriage return, data read from the disk is not checked;
for example, disk data is not compared to the expected pattern. If the prompt is answered with a Y,
the following prompt is issued. The media must have been previously written with a data pattern
in order to do a data compare.
DATA COMPARE ALWAYS (YIN) [N]?
Answering a Y causes ILEXER to check the data returned by every disk Read operation. Answering
with an N or carriage return causes data compares on 15 percent of the disk reads.
NOTE
Selection of data compares significantly reduces the number of disk sectors transferred
in a given time interval.
ANOTHER DRIVE (YIN) []?
Answering with a Y permits selection of another drive for exercising. This prompt has no default.
Answering with an N causes ILEXER to prompt:
MINIMUM DISK TRANSFER LENGTH IN SECTORS (1 TO ,400) [10]?
MAXIMUM DISK TRANSFER LENGTH IN SECTORS (1 TO 400) [10J?
These prompts request the range of size in sectors of each data transfer issued to the disk drives.
The default disk transfer length is 10 sectors.
Once the preceding parameters are entered, ILEXER continues with the prompts listed as global
user prompts (Section 5.8.7).
5.8.6 Tape Drive Prompts

ILEXER displays the following prompts if the drive selected is a tape drive.
IS A SCRATCH TAPE MOUNTED (YIN) [N]?
Answering N results in a reprompt for the drive unit number. Answering Y displays the next
prompt.
ARE YOU SURE (YIN) [N]?
If the answer is N, the operator is reprompted for the drive unit number. If answered with a Y, the
following prompts are displayed.
DATA PATTERN NUMBER (16-22) (D) [21]?
Seven data patterns are available for tape. The default pattern (pattern 21) is defined in
Section 5.8.8.
DENSITY (1=800, 2=1600, 3=6250) (D) [2] ?
The response to this prompt is 1, 2, or 3. Any other response is illegal, and the prompt is displayed
again. The default is 2 or a density of 1600 bpi.
SELECT AUTOMATIC SPEED MANAGEMENT (YIN) [N]?
Either Automatic Speed Management (if the feature is supported) or a tape drive speed is selected
at this point. If the choice is Automatic Speed Management, the available speeds are not displayed.
ILEXER>D>FIXED [VARIABLE] SPEEDS AVAILABLE:
This is an informational message identifying the speeds available for the tape drive. If the speeds
are fixed, the value is presented. If the speed is variable within a range, the range is listed, and
the next prompt asks the operator to select a speed. See the tape drive user documentation for
available speeds.
SELECT FIXED [VARIABLE] SPEED (D) [1]?
This prompt allows selection of the variable speed for the tape drive selected. See the tape drive
user documentation for available speeds.
RECORD LENGTH IN BYTES (1 to 12288) (D) [8192)?
Response to this prompt specifies the size in bytes of a tape record. Maximum size is 12K bytes.
The default value is 8192, the standard record-length size for 32-bit systems. Constraints on the
HSC diagnostic interface prohibit selection of the maximum allowable record length of 64K bytes.
DATA COMPARE (YIN) [N]?
Answering N results in no data compares performed during a read from tape. Answering Y causes
the following prompt:
DATA COMPARE ALWAYS (YIN) [N]?
Answering Y selects data compares to be performed on every tape Read operation. Answering N
causes data compares to be performed on 15 percent of the tape reads.
ANOTHER DRIVE (YIN) []?
If answered with Y, the prompts beginning with the prompt for DRIVE UNIT NUMBER are
repeated. If answered with N, the global prompts in Section 5.8.7 are presented. This prompt has
no default, allowing the operator to default all other prompts and to be able to set up another drive
for this pass of ILEXER.
5.8.7 Global Prompts

The following prompts are presented to the operator when no more drives or drive-specific
parameters are to be entered into the testing sequence. These prompts are global in the sense
they pertain to all the drives.
RUN TIME IN MINUTES (1 TO 32767) [10]?
The minimum time is 1 minute, and the default is 10 minutes. After the exerciser has executed for
that period of time, all testing terminates and a final performance summary is displayed.
HARD ERROR LIMIT (D) [20]?
The number of hard errors allowed for the drives being exercised can be specified. The limit can be
set from 0 to 20. When a drive reaches this limit, it is removed from any further exercising on this
pass of ILEXER. Hard errors include the following types of errors:
• Tape drive BOT encountered unexpectedly
• Invalid MSCP response received from functional code
• UNKNOWN MSCP status code returned from functional code
• Write attempted on write-protected drive
• Tape formatter returned error
• Read compare error
• Read data EDC error
• Unrecoverable read or write error
• Drive reported error
• Tape mark error (ILEXER does not write tape marks)
• Tape drive truncated data read error
• Tape drive position lost
• Tape drive short transfer occurred on Read operation
• Retry limit exceeded for a tape Read, Write, or Read Reverse operation
• Drive went OFFLINE or AVAILABLE unexpectedly
The prompt next calls for:
NARROW REPORT (YIN) [N]?
Answering Y presents a narrow report which displays the performance summaries in 32 columns.
The default display, selected by answering N or carriage return, is 80 columns. The format of this
display is described in further detail in Section 5.8.12. This report format is intended for use by
small hand-held terminals.
ENABLE SOFT ERROR REPORTS (YIN) [N]?
Answering Y enables soft error reports. By default, the operator does not see any soft error reports
specific to the number of retires required on a tape I/O operation.
Answering N results in no soft error report. Soft errors are classified as those errors that eventually
complete successfully after explicit controller-managed retry operations. They include Read, Write,
and Read-Reverse requested retries.
DEFINE PATTERN 0 -- HOW MANY WORDS (16 MAX) (D) [16]?
If data pattern 0 was selected for any preceding drive, the size of the data pattern must be defined
at this time. The pattern can contain as many as 16 words (also the default). If a number larger
than 16 is supplied, an error message is displayed and this prompt is presented again. When a
valid response is presented, the following prompt is displayed the specified number of times.
DATA IN HEX (H) [OJ?
ILEXER expects a 4 character hex value as the answer to this prompt.
5.8.8 Data Patterns

The data patterns available for use with ILEXER are listed in the following sections. Note that
pattern 0 is a user-defined data pattern. Space is available for a repeating pattern of up to 16
words.
The following are data patterns for disks:
Pattern 0 Pattern 1 Pattern 2 Pattern 3

User Defined 105613 031463 030221

Shifting 1s Shifting Os Alter 1s,Os B1011011011011001
000001 177776 000000 133331
000003 177774 000000
000007 177770 000000
000017 177760 177777
000037 177740 177777
000077 177700 177777
000177 177600 000000
000377 177400 000000
000777 177000 177777
001777 176000 177777
003777 174000 000000
007777 170000 177777
017777 160000 000000
037777 140000 177777
077777 100000 000000
177777 000000 177777

B0101 .. /B1010 .. B110 ... 26455/151322
052525 155554 026455 066666
052525 026455
052525 026455
125252 151322
125252 151322
125252 151322
052525 026455
052525 026455
125252 151322
125252 151322
052525 026455
125252 151322
052525 026455
125252 151322
052525 026455
125252 151322

Ripple 1 Ripple 0 Manufacture Patterns
000001 177776 155555 155555
000002 177775 133333 133333
000004 177773 155555 066666
000010 177767 155555 155555
000020 177757 133333 133333
000040 177737 155555 066666
000100 177677 155555 155555
000200 177577 133333 133333
000400 177377 155555 066666
001000 176777 155555 155555
002000 175777 133333 133~33
004000 173777 155555 066666
010000 167777 155555 155555
020000 157777 133333 133333
040000 137777 155555 066666
100000 077777 155555 155555
The following are data patterns for tapes:

Alternating All ones Alternating Alternating
one and zero bytes of all bytes of all
bits ones ones and all
125252 zeros
125252

Alternating Alternating Alternating Alternating
two bytes four bytes three bytes of bytes of ones
ones and two ones and four ones and one and zeros with
bytes zeros bytes zeros byte zeros high byte in
pattern number
also zero*
*Pattern 23 is used with odd record sizes.
5.8.9 Setting/Clearing Flags

The Enable Soft Error Report display prompt in Section 5.8.7 allows the operator to inhibit the
display of soft error reports. No other error reports can be inhibited.

ILEXER has three basic forms of progress reports: the Data Transfer error report, the performance
summary, and the Communication error report.
1. The Data Transfer error report is printed each time an error is encountered in one of the drives
being tested.
2. The Performance summary is printed when ILEXER completes a pass on each drive being
exercised or when the operator terminates the pass through a CTRLIY. This Performance
summary also is printed on a periodic basis during the execution of ILEXER.
3. The Communication error report is sent to the console terminal any time ILEXER is unable to
establish and maintain communications with the drive selected for exercising.
5.8.11 Data Transfer Error Report

The ILEXER Data Transfer error report is printed on the terminal each time a data transfer error
is found during execution of ILEXER. The report describes the nature of the error and all data
pertinent to the error found. The Data Transfer error report is a standard HSC error log message.
It contains all data necessary to identify the error. The only exception to this is when the error
encountered by ILEXER is a data compare error. In this case, ILEXER has performed a check and
found an error during the compare, resulting in an ILEXER error report.
5.8.12 Performance Summary

The Performance summary is printed on the terminal at the end of a manually terminated testing
session, or after the specified number of minutes for the periodic Performance summary. This report
provides statistical data which is tabulated by ILEXER during the execution of this test.
The Performance summary presents the statistics which are maintained on each drive. This
summary contains the drive unit number, the drive serial number, the number of position
commands performed, the number of 0.5 Kbytes read and written, the number of hard errors,
the number of soft errors, and the number of software correctible transfers. For tape drives being
exercised by ILEXER, an additional report breaks down the software correctible errors into eight
different categories.
The frequency of report display is altered in the following fashion:

1. Press CTRL/G during the execution of ILEXER.
2. The following prompt is displayed:
Interval time for performance summary in seconds (D) [30]?
The format of the Performance summary follows:

PERFORMANCE SUMMARY (DEFAULT)
UNIT R SERIAL POSI KBYTE KBYTE HARD SOFT SOFTWARE
NO NUMBER TION READ WRITTEN ERROR ERROR CORRECTED
Dddd HHHHHHHHHHH ddddd dddddddddd dddddddddd ddddd ddddd ddddd
Tddd HHHHHHHHHHH ddddd dddddddddd dddddddddd ddddd ddddd ddddd
A Performance summary is displayed for each disk drive and tape drive active on the HSC. The
following list explains the performance summary:
• Unit Number-The unit number of the drive. D is for disk, T is for tape. The nUmber is
reported in decimal.
• R--The status of the drive. If an asterisk (*) appears in this field, the drive was removed
from the test and the operator was previously informed. If the field is blank, the drive is being
exercised.
• Serial Number-The serial number (in hex) for each drive.
• Position-The number of seeks.
• Kbyte Read-The number of Kbytes read by ILEXER on each drive.
• Kbyte Written-The number of Kbytes written by ILEXER.
• Hard Error-The number of hard errors reported by ILEXER for a particular drive.
• Soft Error-The number of soft tape errors reported by the exerciser if enabled by the operator.
• Software Corrected-The number of correctible ECC errors encountered by ILEXER. Only
ECC errors above the specific drive ECC error threshold are reported through normal functional
code error reporting mechanisms. ECC errors below this threshold are not reported through an
error log report, but are included in this count maintained by ILEXER.
If any tape drives are exercised, the following summary is displayed within each performance
summary:
UNIT MEDIA DOUBLE DOUBLE SINGLE SINGLE OTHER OTHER OTHER
NO ERROR TRKERR TRKREV TRKERR TRKREV ERR A ERR B ERR C
Tddd ddddd ddddd ddddd ddddd ddddd ddddd ddddd ddddd
An explanation of the summary columns follows:

• Media Error-The number of bad spots detected on the recording media.
• Double TRKERR-The number of double-track errors encountered during a read or write
forward.
• Double TRKREV-The number of double-track errors encountered during a read reverse or
write reverse.
• Single TRKERR-The number of single-track errors detected during a read or write in the
forward direction.
• Single TRKREV-The number of single-track errors encountered during a read reverse or

write reverse.
• Other Err A-C-Reserved for future use.
PERFORMANCE SUMMARY (NARROW)
ILEXER>O>PER SUM
O[T]ddd
SN HHHHHHHHHHHH
P ddddd
R dddddddddd
W dddddddddd
HE ddddd
SE ddddd
SC ddddd
This report is repeated for each drive tested.

If tape drives are being tested, the following report is issued for each tape drive following the disk
drive performance summaries:
ILEXER>O>ERR SUM
ILEXER>O>Tddd
ILEXER>O>ME ddddd
ILEXER>O>OF ddddd
ILEXER>O>OR ddddd
ILEXER>O>SF ddddd
ILEXER>O>SR ddddd
ILEXER>O>OA ddddd
ILEXER>O>OB ddddd
ILEXER>O>OC ddddd
5.8.13 Communications Error Report

Whenever ILEXER encounters an error that prevents it from communicating with one of the drives
to be exercised, ILEXER issues a standard error report. This report gives details enabling the
operator to identify the problem. For further isolation of the problem, the operator should run
another test specifically designed to isolate the failure (ILDISK or ILTAPE).

The test numbers in ILEXER correspond to the module being executed within ILEXER itself. The
main module is called MDE, and it calls all other modules.
• Test number It Main Program: MDE-Multidrive Exerciser is the main program within
ILEXER. It is responsible for calling allnther portions of ILEXER. It obtains the buffers and
control structures for the exerciser. It verifies that disk or tape functionalities are available
before allowing ILEXER to continue.
• Test number 2t INITI-INITT is called to initialize drive statistic tables. It obtains the
parameters and verifies the values of each one entered. This routine calls INICOD to obtain
drive-specific parameters.
• Test number 3 t INICOD-INICOD is the initialization code for ILEXER. It gets the various
parameters for the drives from the operator and fills in the drive statistic tables with initial
data for each drive. It also verifies the validity of the input for the parameters. INICOD, in
turn, calls ACQUIRE to acquire the disk and/or tape drive.
• Test number 4 t ACQUIRE-ACQUIRE is responsible for acquiring the drives as specified by
the operator. It brings all selected drives on line to the controller and spins up the disk drives.
Errors reported in this routine cause the removal of the drive from the exercise.
• Test number 5, INITD-INITD initializes the disk drives for the exercise. This routine clears
all disk access control blocks and invokes the initial write.
• Test number 6, TPINIT-TPINIT initializes the tape drives for the exercise. It rewinds all
acquired tape drives and verifies the drives are at the BOT. If an error occurs, the drive is
removed from the exerciser. TPINIT is also responsible for obtaining buffers for each acquired
tape drive.
• Test number 7, Exerciser-EXER is the main code of the exerciser. It dispatches to the disk
exerciser (QDISK and CDISK) and the tape exerciser (TEXER). It continuously queues up I/O
commands to disk and tape, and checks for I/O completion. The subroutines EXER calls are
responsible for sending commands and checking. for I/O completion.
• Test number 8, QDISK-QDISK is part of the disk exerciser that selects commands to send
to the disk drives. If the initial write is still in progress, it returns to EXER. QDISK calls a
routine to select the command to exercise the disk drive. The following scenario is the algorithm
used to select the command:
a. If the drive is read only and data compare is not requested, a Read operation is queued to
the drive.
b. If read only and data compare (occasional) are requested, a Read operation is queued along
with a random choice of compare/not-compare.
c. If read only and data compare (always) are requested by the operator, a READ-COMPARE
command is queued to the drive.
d. If write only is requested, and data compare is not, then a write request is queued up to the
disk drive.
e. If write only and data compare (occasional) are requested, a Write operation is queued along
with a random choice of compare/not-compare.
f. If write only and data compare (always) are requested, a WRITE-COMPARE command is
queued to the drive.
g. If only data compare (occasional) is requested, then a random selection of read/write and
compare/not-compare is done.
h. If only data compare (always) is requested, a COMPARE command is paired with a random
selection of read/write.
QDISK randomly selects the number of blocks for the selected operation.
• Test number 9, RANSEL-RANSEL is the part of the tape exerciser that is responsible for
sending commands to the tape drives. This routine is called by TEXER, the tape exerciser
routine. RANSEL selects a command for a tape drive using a random number generator.
Following are some constraints for the selection process:
a. No reads when no records exist before or after the current position.
h. No writes when records exist after the current position.
c. No position of record when no records exist before or after the current position.
d. Reverse commands are permitted on the drive when 16 reverse commands previously
have been selected. That is, lout of every 16 reverse commands are sent to the drive.
Immediately following a reverse command, a position to the end-of-written-tape is
performed. The reason for forward biasing the tape is to prevent thrashing.
The following commands are executed in exercising the tape drives:
1. READ FORWARD
2. WRITE FORWARD
3. POSITION FORWARD
4. READ REVERSE
5. REWIND
6. POSITION REVERSE
RANSEL randomly selects the number of records to read, write, or skip.
• Test number 10, CDISK-CDISK checks for the completion of disk 110 specified by QDISK.
CDISK checks the return status of a completed 110 operation and if any errors occur, they are
reported.
• Test number 11, TEXER--TEXER is the main tape exerciser which selects random writes,
reads, and position commands. TEXER processes the I/O once it is completed and reports any
errors encountered.
• Test number 12, EXCEPT-EXCEPT is the ILEXER exception routine. This is the last
routine called by MDE. EXCEPT is called when a fatal error occurs, when ILEXER is stopped
with a CTRUY, or when the program expires its allotted time. It cleans up any outstanding
110, as necessary, returns resources, and returns control to DEMON.
5.8.15 Error Message Format

ILEXER outputs four types of error formats: prompt errors, data compare errors, pattern word
errors, and communication errors. These formats agree with the generic test error message format
(Section 5.1.2).
5.8.15.1 Prompt Error Format

Prompt errors occur when the operator enters the wrong type of data or the data is not within the
specified range for a parameter. The general format of the error message is:
ILEXER>D>error message
The error message is an ASCII string describing the type of error discovered.
5.8.15.2 Data Compare Error Format

A data compare error occurs when an error is detected during the exercise of a particular drive.
The two formats for the data compare error are data word compare error and pattern word error.
A data word compare error occurs when the data read does not match the expected pattern. The
format of the data compare error is:
ILEXER>D>hh:mm T ddd E ddd U-uddd

ILEXER>D>Error Description
ILEXER>D>MA -- HHHHHHHHHH
ILEXER>D>EXP -- HHHH
ILEXER>D>ACT -- HHHH
ILEXER>D>MSCP STATUS CODE = HHHH
ILEXER>D>FIRST WORD IN ERROR = ddddd
ILEXER>D>NUMBER OF WORDS IN ERROR = ddddd
Where:
hh:mm is the current system time.
T is the test number in the exerciser.
E corresponds to the error number.
U is the unit number for which the error is being reported.
MA is the media address (block number) where the error occurred.
EXP is the EXPected data.
ACT is the data (or code) actually received.
MSCP STATUS CODE is the code received from the operation.
FIRST WORD IN ERROR describes the number of the first word
found in error.
NUMBER OF WORDS IN ERROR -- Once an error is found, the routine
continues to check the remainder of the data returned and counts
the number of words found in error.
5.8.15.3 Pattern Word Error Format

The format for the pattern word error is slightly different from the data word compare error. A
pattern word error occurs when the first data word in a block is not a valid pattern number. The
format is:
ILEXER>D>MA -- HHHHHHHHHH
ILEXER>D>EXP HHHH
ILEXER>D>ACT -- HHHH
The MSCP status code, first word in error, and number of words in error are not relevant for this
type of error. The other fields are as described for the data compare error.
5.8.15.4 Communications Error Format

Communications errors occur when ILEXER cannot establish/maintain communications with a
selected drive. The error message appears in the following format:
ILEXER>D>Optional Data lines follow here
Where:
hh:mm is the time stamp for the start of ILEXER.
T is the test number in the exerciser.
E corresponds to the error number.
U is the unit number for which the error is being reported.
Error Description is an ASCII string describirig the
error encountered.
Optional Data lines -- A maximum of eight optional lines per report.

The following section lists the informational and error messages and explains the cause of the error.
A typical error message is:
ILEXER>D>09.32 T*006 E*204 U-T00100
ILEXER>D>Comm Error: TBUSUB call failed
5.8.16.1 Informational Messages

The informational messages are not fatal to the exerciser and are intended only to:
• Alert the user to incorrect input to parameters
• Indicate missing interfaces
• Provide user information
The following list describes informational messages.
• Number Must Be Between 0 and IS-Reported when the user entered an erroneous value
for the data pattern on a disk.
• Pattern Number Must Be Within Specified Bounds-Reported when the operator tries to
specify a disk pattern number for a tape.
• You May Enter at Most 16 Words in a Data Pattern-Reported if the operator specifies
more than 16 words for a user-defined pattern, and the operator is reprompted for the value.
• Starting LBN Is Either Larger than Ending LBN or Larger than 'IOtal LBN on Disk-
Reprompts for the correct values. The operator selected a starting block number for the test
which is greater than the ending block number selected, or it is greater than the largest block
number for the disk.
• Please Mount a Scratch Tape-Appears after an N response to the prompt asking if the
scratch tape is mounted on the tape drive to be tested.
• Disk Interface Not Available-Indicates the disk functionality is not available to exercise
disk drives. This means the K.sdilK.si is not available or not operable.
• Tape Interface Not Available-Indicates the tape functionality is not available to exercise
tape drives. This means the K.stilK.si is not available or not operable.
• Please Wait-Clearing Outstanding I/O-Printed when the operator presses CTRUY to stop
ILEXER. All outstanding 110 commands are aborted at this time.
5.8.16.2 Generic Errors

The following list indicates the error number, text, and cause of errors displayed by ILEXER.
• Error 01, No Disk or Tape Functionality••.Exerciser Terminated-Neither the K.sdi,
Ksti, nor K.si interfaces are available to run the exercise. This terminates ILEXER.
• Error 02, Could Not Get Control Block for Timer-Stopping n,EXERr-ILEXER could
not obtain a transmission queue for a timer. This should occur only on a heavily loaded system
and is fatal to ILEXER.
• Error 03, Could Not Get Timer-Stopping ILEXERr-The exerciser could not obtain a timer.
Two timers are required for ILEXER. This should only occur on a loaded system and is fatal to
ILEXER.
• Error 04, Disk Functionality Unavailable-Choose Another Drive-The disk interface is
not available. A previous message is printed at the start of ILEXER if any of the interfaces are
missing. This error prints when the operator still chooses a disk drive for the exercise.
• Error 05, Tape Functionality Unavailable-Choose Another Drive-The tape interface is

not available. A previous message is printed at the start of ILEXER if any of the interfaces are
missing. This error prints when the operator still chooses a tape drive for the exercise.
• Error 06, Couldn't Get Drive Status-Choose Another Drive-ILEXER was unable to
obtain the status of a drive for one of the following conditions:
1. The drive is not communicating with the HSC. Either the formatter or the disk is not on
line.
2. The cables to the Ksdi, K.sti, or Ksi are loose.
• Error 07, Drive Is Unknown-Choose Another Drive-The drive chosen for the exerciser
is not known to the HSC functional software for that particular drive type. Either the drive is
not communicating with the HSC, or· the functional software has been disabled due to an error
condition on the drive.
• Error 08, Drive Is Unavailable--Choose Another Drive-This may be the result of:
1. The drive port button is disabled for that port.
2. The drive is on line to another controller.
3. The drive is not able to communicate with the controller on the port selected.
• Error 09, Drive Cannot Be Brought On Line-ILEXER was unable to bring the selected
drive on line. One of the following conditions occurred:
1. The unit went into an Off-line state and cannot communicate with the HSC.
2. The unit specified is now being used by another process.
3. There are two drives of same type with duplicate unit numbers on the HSC.
4. An unknown status was returned from the HSC diagnostic interface when ILEXER
attempted to bring the drive on line.
• Error 12, Could Not Return Drive to Available State-The release of the drive from
lLEXER was unsuccessful. This is the result of a drive being taken from the test due to
reaching an error threshold or going off line during the exercise.
• Error 13, User Requested Write on Write-Protected Unit-The operator should check the
entry of parameters and also check the write protection on the drive to make sure they are
consistent.
• Error 14, No Tape Mounted on Unit...Mount and Continue-The operator specified a
scratch tape was mounted on the tape drive selected when it was not mounted. Mount a tape
and continue.
• Error 15, Record Length Larger than 12K or 0 - The record length requested for the
transfer to tape was either greater than 12K or O.
• Error 16, This Unit Already Acquired-A duplicate unit number was specified for a drive
and the drive has already been acquired.
• Error 18, Invalid Time EnterecL ••Must Be from 1 to 3599-The user entered an erroneous
value to the performance summary time interval prompt.
• Error 20, Could Not Get Buffers for Transfers-The buffers required for a tape transfer
could not be acquired.
• Error 21, Tape Rewind Commands Were Lost - Cannot Continue-The drive was
unloaded during ILEXER execution.
5.8.16.3 Disk Errors

The following list includes the error number, text, and cause of ILEXER disk errors.
• Error 102, Drive Spindle Not Up to Speed-Spin Up Drive and Restart-The disk drive
is not spun up.
• Error 103, Drive No Longer Exercised-A disk drive reached the hard error limit or the
drive went off line to the HSC during the exercise.
• Error 104, Couldn't Put Drive in DBN Space - Removed from Test-An error or
communication problem occurred during the delivery of an SDI command to put the drive
in DBN space.
• Error 105, No DACB Available-Notify Customer Service, Submit SPR-This is reported
if no DACBs can be acquired. If this happens, contact Customer Service as soon as possible and
submit an SPR.
• Error 106, Some Disk I/O Failed to Complete-An 1/0 transfer did not complete during an
allotted time period.
• Error 107, Command Failed-Invalid Header Code-ILEXER did not pass a valid header
code to the diagnostic interface for the HSC.
• Error 108, Command Failed-No Control Structures Available-The diagnostic interface
could not obtain disk access control blocks to run the exercise. The HSC could be overloaded.
Try ILEXER on a quiet system. If the error still occurs, test the HSC memory.
• Error 109, Command Failed-No Buffer Available-The diagnostic interface could not
obtain buffers to run the exercise. The HSC could be overloaded. Try ILEXER on a quiet
system. If the error still occurs, test the HSC memory.
• Error 111, Write Requested on Write-Protected Drive-The operator requested an initial
Write operation on a drive that was already write-protected. The operator should pop out the
write-protect button on the drive reporting the error or have ILEXER do a read-only operation
on the drive.
• Error 112, Data Compare Error-Bad data was detected during a Read operation.
• Error 113, Pattern Number Error-The first two bytes of each sector, which contain the
pattern number, did not match.
• Error 114, EDC Error-Error Detection Code error: invalid data was detected during a Read
operation.
• Error 116, Unknown Unit Number Not Allowed in U,EXER-The operator attempted to
enter in a unit number of the form Xnnnn, which is not accepted by ILEXER.
• Error 117, Disk Unit Numbers Must Be Between 0 and 4094 decimal-The operator
specified a disk unit number out of the allowed range of values.
• Error 118, Hard Failure on Disk-A hard error occurred on the disk drive being exercised.
NOTE
The following disk errors identify the function attempted by IT,EXER that caused an
error to occur. Error logs do not indicate the operation attempted.
• Error 119, Hard Failure on Compare Operation-A hard failure occurred during a compare
of data on the disk drive.
• Error 120, Hard Failure on Write Operation-A hard fault occurred during a Write
operation on the specified disk drive.
.• Error 121, Hard failure on Read Operation-A hard failure occurred during a Read
operation on the disk drive being exercised.
• Error 123, Hard Failure on INITIAL WRITE Operation-A hard failure occurred during
the first write to the disk drive.
• Error 124, Drive No Longer On Line-A drive that was being exercised went into an
Available state. This could be caused by the operator releasing the port button on the drive. A
fatal drive error could also cause the drive to go into this state.
5.8.16.4 Tape Errors

The following list includes the error number, text, and cause of ILEXER tape errors.
• Error 201, Couldn't Get Formatter Characteristics-A communication problem with the
drive is indicated. It could be caused by the unit not being on line.
• Error 202, Couldn't Get Unit Characteristics-The drive is not communicating with
ILEXER. The unit could be off line.
• Error 203, Some.Tape 110 Failed To Complete-The drive or formatter stopped functioning
properly during a data transfer.
• Error 204, Comm Error: TDUSUB Call Failed-ILEXER cannot communicate with the
drive through interface structures. They have been removed. Either the drive went available
from on line, the drive is offline, or there is a fault.
• Error 205, Read Data Error-A Read operation failed during a data transfer, and none was
transferred.
• Error 206, Tape Mark Error...rewinding to restart-ILEXER does not write tape marks. If
this error occurs, it indicates a drive failure.
• Error 207, Tape Position Lost...rewinding to restart-An error occurred during a data
transfer or a retry of one.
• Error 209, Data Pattern Word Error Defect-The first two bytes of a record containing the
data pattern did not match.
• Error 210, Data Read EDC Error-Error Detection Code error; incorrect data was detected.
• Error 211, Could Not Set Unit Char..•removing from test-The drive is offline and not
communicating.
• Error 213, Truncated Record Data Error•••rewinding to restart-More data was received
than expected, indicating a drive problem.
• Error 214, Drive Error...Hard Error-A hard failure occurred with the drive being exercised.
• Error 215, Unexpected Error Condition.••removing drive from test-This is caused by

MSCP error conditions, which are not allowed (invalid commands, unused codes, write-protected
drive write, and so forth).
• Error 216, Unexpected BOT Encountered.•.will try to restart-The drive is experiencing
a positioning problem.
• Error 217, Unrecoverable Write Error.••rewinding to restart-A hard error occurred
during a Write operation. The write did not take place due to this error.
• Error 218, Unrecoverable Read Error.••rewinding to restart-A hard error occurred
during a Read operation and a data transfer did not take place.
• Error 219, Controller Error...Hard Error•.rewinding to restart-A communications
problem exists between the controller and the formatter.
• Error 220, Formatter Error••.Hard Error-A communications problem exists between the
formatter and the controller and/or drive.
• Error 221, Retry Required on Tape Drive-A failed ReadlWrite operation required a retry
before succeeding.
• Error 222, Hard Error Limit Exceeded. ..removing drive from test-The drive exceeded
the threshold of hard errors determined by a global user parameter (Section 5.8.7). The drive is
then removed from the exercise.
• Error 224, Drive Went Off Line...removing from test-The drive went off'line during the
exercise. This is caused by the operator taking the drive off line or a hard failure forcing the
drive off line.
• Error 225, Drive Went Available...removing from test-The drive became available to
ILEXER and was not at the beginning of the exercise.
• Error 226, Short Transfer Error...rewinding to restart-Less data was received than
transferred.
• Error 227, Tape Position Discrepancy-The tape position was lost, indicating a hard failure.
Off-line Diagnostics 6-1
6
Off-line Diagnostics
6.1 Introduction
This chapter describes the off~line diagnostics, how to run them, errors that can occur, and
summaries of th~ tests in each diagnostic. Included in the off-line diagnostics are:
• ODL-Off-line diagnostics loader
• OFLCXT-Off-line cache test (HSC only)
• OBIT-Off-line bus interaction test
• OKTS-Off-line K test selector
• OKPM-Off-line KIP memory test
• OMEM-Off-line memory test
• OFLRXE-Off-line RX33 exerciser (HSC only)
• ORFT-Off-line refresh test
• OOCP-Off-line operator control panel (OCP) test
The off-line diagnostics contain specific common characteristics, which are discussed in the following
three sections. These characteristics are listed below:
• Identical software requirements
• Common load procedure
• Identical bootstrap initialization procedures
• Generic error message format
6.1.1 Software Requirements

All off-line diagnostics require boot media containing a bootable image of the diagnostics software
programs. For an HSC, an RX33 off-line diagnostic diskette is required. For an HSC50, a TU5S
off-line diagnostic tape cassette is required.
6.1.2 Off-line Diagnostics Load Procedure

For the HSC, the off-line diagnostics diskette boots from either RX33 drive and should not be write-
enabled. This diskette contains the necessary software to run all the HSC off-line diagnostics.
Booting is done either by powering on or by pressing and releasing the Init switch with the
SecurelEnable switch in the enable position. This causes the P.ioj ROM bootstrap tests to run
followed by the off-line P.ioj test.
6-1
6-2 Off-line Diagnostics
The off-line diagnostics TU58 cassette boots from either TU58 drive and should not be write-
enabled. This TU58 contains the necessary software to run all the HSC50 off-line diagnostics.
Booting is accomplished either by powering on or by pressing and releasing the Init switch with
the SecurelEnable switch in the enable position. This causes the P.ioc ROM bootstrap tests to run
followed by the off-line P.ioc test.
NOTE
For off-line diagnostics, the HSe must be booted with the SecurefEnable switch in the
enable position. If a hardware error occurs during boot, the software executes a halt
instruction on certain errors. A halt instruction, even in Kernel mode, is valid only if the
SecurefEnable switch is in the enable position. Otherwise, the result can be an illegal
instruction trap in addition to the error causing the halt.
In order for the bootstrap to complete successfully, the following must be operational:
• Basic instruction set of the PDP-11
• First 2048 bytes of program memory plus 8 Kwords of contiguous Program memory below
address 160000
• RX.33 controller and at least one RX33 drive containing a diskette with a bootable image for the
HSC
• TU58 controller and at least one TU58 drive containing a cassette with a bootable image for the
HSC50
Before control is turned over to the HSC bootstrap ROMs, internal microcode tests execute in
the J11 (HSC) or F11 (HSC50) chip sets. Refer to Table 2-1 for definitions of the J111F11 module
(P.ioj/c) LEDs.
6.2 ROM Bootstrap

The P.ioj/c ROM bootstrap verifies the basic integrity of the P.ioj/c module, part of the Program
memory, and the boot device. The goal of the bootstrap tests is to test enough of the HSC to allow
further test loading from the boot device.
The bootstrap test is the first step in the HSC initialization process. It is run for every bootstrap or
reload of the HSC operating system (CRONIC). The bootstrap is initiated automatically each time
the HSC is powered on and also is initiated by CRONIC when a software reboot is required.
The bootstrap is a PDP-11 program written to execute in a DCJ11IDCF11 CPU in a standalone
environment. This means no other software processes coexist with the bootstrap.
Bootstrap failures are reported through the fault lamp mechanism, which specifies the module
most likely causing the problem. See Figure 4--3 for the fault code definitions. An error table is
maintained in Program memory addresses 00000400 through 00000412. These addresses contain
the reasons for each load device boot failure.
6.2.1 Initialization Instructions

The following procedure lists the operating instructions for the P.ioj/c ROM bootstrap. Refer to
Section 6.2.4 if this procedure fails.
1. HSC-Insert the off-line diagnostics diskette with a bootable image into the RX33 unit 0 drive
Oeft-hand drive).
2. HSC50-Insert the off-line diagnostics tape with a bootable image into either of the TU58
drives.
3. Turn power ON.
4. Set the SecurelEnable switch to the enable position, then press the Init switch. The bootstrap
initiates automatically.
At this point, the P.ioj/c module executes internal microdiagnostics and then begins to execute from
the boot ROM. The Init lamp lights on the HSC operator control panel (OCP) when the bootstrap
PDP-II tests are done. The load device drive-in-use LED lights within 8 to 10 seconds, indicating
the bootstrap is attempting to load software into Program memory. If the load is successful, the
bootstrap transfers control to the first instruction of the image just loaded from the diskette.
6.2.2 Failures
Most bootstrap failures result in lighting the fault lamp on the HSC OCP. When this happens, press
the Fault switch momentarily, and read the failure code displayed in the OCP lamps. Section 6.2.5
indicates the HSC modules most likely causing the bootstrap failure. Momentarily pressing the Init
switch on the OCP reinitiates the bootstrap.
The microdiagnostic LEDs on the JI11FII module indicate if a hard fault exists causing the JIIIFI1
to hang before control is passed the boot ROM. Section 6.2.5 contains an explanation of these LEDs.
If a failure occurs in the tests of the PDP-II basic instruction set, the fault lamp mechanism does
not report the failure. Instead, the PDP-II executes a Branch dot (BR .) and does not continue the
bootstrap program. A failure of this type is easily detected because the Init lamp does not light.
(The Init lamp does light immediately after the basic PDP-1I tests successfully complete.)
When a console terminal is connected to the P.ioj/c, the exact instruction that failed is determined
by pressing the terminal break key and noting the address displayed on the terminal. With a
bootstrap listing, this address indicates the instruction that failed.
NOTE
The bootstrap does not accept user-modifiable flags.

The bootstrap does not issue progress reports in the usual sense; however, certain indications of
bootstrap. progress are evident. These indications are given in the following list:
• Lamps clear-Clears all of the HSC OCP lamps. If the lamps fail to clear immediately after
the bootstrap is initiated, a failure of the P.ioj/c is probable. (Circuitry on the P.ioj/c module is
responsible for initiating the bootstrap program.)
• Init lamp-Lights as soon as the basic tests of the PDP-II instruction set are finished. These
tests normally complete within milliseconds after the bootstrap is initiated. Failure of the lnit
lamp to light indicates a failure in the P.ioj/c PDP-II processor.
• RX33 drive-in-use-Lights as the bootstrap tries to load the lnit P.io test (or off-line P.ioj test)
from the RX33 following the test of the PDP-II and Program memory.
• State lamp-Lights when the bootstrap completes and initiates the lnit P.ioj/c test (or off-line
P.ioj/c test). When the State lamp is ON, the lnit lamp is OFF.
• Fault lamp-Lights during the boot process if the ROM bootstrap tests have detected a fatal
error.
6.2.4 Error Information

Specific error codes for the P.ioj/c bootstrap (Codes 21, 22, and 23) are described in detail in
Chapter 4.
Because the bootstrap operates in a standalone environment, it does not use the terminal as an
error-reporting mechanism. Instead, the HSC OCP lamps are used to report errors and to indicate
the module most likely causing the error.
When the bootstrap detects an error, it lights the fault lamp on the OCp. When the Fault switch is
pressed, the bootstrap displays a failure code in the OCP lamps. The failure code blinks on and off
at one-half second intervals.
6.2.5 Failure Troubleshooting

The ODT program (built into the PDP-II microcode) contains further information about bootstrap
failures. This information iE' shown in the following list:
• Init is off, Fault is lit-A failure was detected after control was passed to the bootable image
loaded from the diskette.
• Init and Fault both lit-The fault code displays when the fault lamp is momentarily pressed.
The program is halted by pressing the break key on the console terminal. If 17772340 is typed,
ODT responds by displaying the contents of address 17772340, the test number. Use the test
number to refer to the appropriate test in Section 6.2.6.
• Init and Fault lamps are both off-Either the bootstrap program was not automatically
initiated or the bootstrap PDP-II instruction test failed.
Before proceeding, ensure the SecurelEnable switch is set to the enable position. If the
switch was not in the enable position when the Init switch was pressed, the HSC did not
initiate its boot sequence. If the SecurelEnable switch is in the correct position, the JI11Fli
microdiagnostics may have failed.
To check the microdiagnostics, remove the card cage cover and examine the four LEDs on the
central edge of the JIIlFll module. At powerup, set all the LEDs and then turn them off as the
JI11Fli proceeds through its microdiagnostic sequence.
6.2.6 Bootstrap Test Summaries

This section summarizes the bootstrap tests.
• Test 0, Basic PDP-II Instruction Set-This test verifies the correct operation of a PDP-
11 instruction subset. This instruction subset includes only those instructions required for
completion of the bootstrap. The following instructions are tested:
a. Single operand instructions (both word and byte mode):
ADC, CLR, COM, INC, DEC, NEG, TST, ROR, ROL, ASR, ASL, SWAB, Nap
h. Double operand instructions (both word and byte modes):
MOV, CMp, BIT, BIC, BIS, ADD, SUB
c. Branch instructions:
BR, BNE, BEQ, BPL, BMI, BCC (BHIS), BCS (BLO), BGE, BLT, BGT, BLE, BHI, BLOS,
BVC, BVS
d. Jump and miscellaneous instructions:
JMp' JSR, RTS, SOB, MTPS, MFPS, CCC, CLN, CLY, CLZ, SEN, SEY, SEZ
e. Addressing modes:
All eight addressing modes

The PDP-II instruction set test uses two methods of reporting errors. During the initial
part of the test, errors result in an infinite program loop at the location of the detected error.
During the latter part of the test (when enough instructions have been tested), the fault lamp
mechanism is used to report failures. Refer to Section 6.2.2.
• Test It Program Memory (Swap Bank)-The memory (Swap Bank) HSC memory module
includes special logic that permits changing the address range of Program memory. This
address range is controlled by the Swap Banks bit in the P.ioj/c control and status register
(CSR). This test verifies the Swap Banks bit can be set and cleared. (The actual memory
switching is not tested, only the setting and clearing of the bit is tested.) A failure in this test
indicates the P.ioj/c module must be replaced.
• Test 2t Program Memory (Vector Area)-In order for the HSC control program to function,
the first 2048 bytes (addresses 00000000 through 00003777) of Program memory must be
working. This test verifies the first part of Program memory is operating properly. If the
test fails, the Swap Banks feature is used, attempting to swap a portion of memory into the
00000000 through. 00093777 address range. If the test still fails after Swap Banks has been
invoked, a Program memory error is reported through the fault lamp mechanism (Section 6.2.2).
A failure in this test indicates the M.std2fM.std module must be replaced.
• Test 3 t Program Memory (8-Kword Partition)-Mter verifying the first part of Program
memory is working, the bootstrap tries to find an 8-Kword piece of Program memory between
address 00004000 and address 00160000. This partition is used to load the Init P.ioj/c test
from the load device. If insufficient memory is available, a Program memory error is reported
through the fault lamp mechanism.
A failure in this test indicates the M.std2fM.std module must be replaced.
• Test ~ RX33 Controller Test-This test verifies basic functionality of the control logic on the
M.std2 module. The four controller registers are tested for stuck bits. The DMA hardware is
checked for correct cycling and addressing. The interrupt logic is checked to ensure interrupts
are properly acknowledged. With the control hardware verified, proceed to the next step, and
try to read data from one of the drives.
• Test 5 t RX33 DrivelInterface Test-The goal of this test is to find a working RX33 drive
containing a diskette with a bootable image. Such an image is identified by a PDP-II NOP
instruction in the first word of the image. The intended drive is checked for DRIVE READY
from the interface. Then RECALIVERIFY commands the drive to seek to track O. This
command then reads the diskette header to verify the RECAL did move the head to track
O.
Mter a suitable drive is found, the first eight blocks of the diskette are loaded into the 8-Kword
partition found in test 3. The eight blocks loaded consist of the first five blocks of the lnit P.ioj/c
test (or Off-line P.ioj/c test), the RT-ll volume ID block, and the first RT-ll directory segment
on the diskette. (The directory blocks are loaded at this time to save directory look-up time in
the Init P.ioj/c test or the Off-line P.ioj/c test.)
RX33 drive 0 is tested first. A failure with drive 0 causes the bootstrap to proceed to drive 1 and
begin the tests again. If neither RX33 drive is working correctly, an RX33 error is displayed by
the fault lamps. An error table is maintained in Program memory addresses 00000400 through
00000412, which remembers why each rejected RX33 drive failed the boot. Table 6-1 shows the
RX33 error table addresses and meanings.
Table 6-1 RX33 Error Table

Address Meaning
00000400 Contains controller error code (code 1 or code 2)

Table 6-1 (Cont.) RX33 Error Table

Address Meaning
00000402 RX33 address being accessed, if applicable

00000404 Expected result
00000406 Actual result
00000410 Drive error code, byte-encoded: drive l/drive 0
(high-byte/low-byte)
NOTE
It is not possible to simultaneously have information in addresses 00000400 and 00000410.
If the boot fails with a RX33 error, the ODT feature of the PDP-II is used to examine the RX33
error table to determine why each RX33 drive failed the test. (Remember, the bootstrap tries both
drives before declaring an error.) Use the following sequence to examine the RX33 error code table.
1. Press the break key on the console terminal.
The terminal displays the address of the current instruction of the bootstrap, then prompts
for input with an @ character.
2. Type the appropriate address, nnn.
The terminal displays the (octal) contents of that address.
3. Press LINEFEED to examine the Table 6-2 controller error and related failure
information.
Table 6-2 RX33 Error Code Table

Controller Error Failure Information
1 NXM occurred while accessing RX33 registers.

2 A bit was stuck in the registers. See EXPected/ACTual for more information.
3 Force mode interrupt did not occur.
4 DMA test mode hardware error occurred.
5 DMA address counters were wrong after transfer.
6 Incorrect data found after DMA test operation.
7 Data parity was bad after DMA test operation.
10 Drive was not ready (no diskette inserted or door was open).
11 Hard error (CRC or Record Not Found) occurred on RECAUverify.
12 Track 0 bit was not set after RECAL.
13 SEEK command timeout occurred.
14 Seek error (CRC or Record Not Found) occurred.
15 READ SECTOR command timeout.
16 Hard error (CRC or Record Not Found) occurred on read.
17 N onbootable image (non-NOP instruction is the first word).
Failure information for both drives in address 00000410 is possible. In this case, nonzero data
is in both bytes. Only when failures are detected on both drives does the boot ROM generate a
LOADFAL failure code and branch to the fault light routine.
• Test 6, Transfer Control to Loaded Image-This part of the bootstrap is not actually a test.
However, it is given a test number in case an error occurs in this section of code. The PDP-11
general registers are loaded with certain parameters (CSR and unit of load device, base address,
and size of partition, and so forth). The image loaded from the RX33 is initiated by jumping to
the first instruction. Any errors occurring in this part of the bootstrap are probably unexpected
traps or interrupts caused by intermittent P.ioj/c or M.std2/M.std failures. When the loaded
image is started, the State lamp is lit, and the lnit lamp is turned off.
6.2.7 Generic Error Message Format

All of the diagnostics use a comman method of reporting errors and a common error message
format. All errors are reported on the console terminal as they occur. In all off-line diagnostics,
error messages conform to the HSC diagnostic error message format.
The first line of an error message contains general information concerning the error and is
mandatory. The second line of an error message consists of text describing the error and also is
mandatory. The third and succeeding lines of the message are used for additional information
where required, and are optional.
The generic error message format follows:
XXXXXX>hh:mm Tn En 0000
SEEK error detected during positioning operation
optional line 1
optional line 2
optional line 3
Where:
XXXXXX> is the prompt for the particular diagnostic in question
such as OFLCXT> or OBIT>.
hh:mm is the number of hours and minutes since system boot.
Tn is a test number.
En is an error number with a range of 1 through 77 (octal).
0000 is the unit number.
The final field in the first line appears only in diagnostics where such information is appropriate.
Each error number has a unique text string associated with it. For errors that consist of results
that did not compare with the expected value, the diagnostic uses the optional lines to show
EXPected/ACTual (EXP/ACT) data. Errors on data transfers and SEEK commands use the optional
lines to print out the LBN, track, sector, and side to help isolate problems to the media or the drive.
6.3 ODL-Off-line Diagnostics Loader

The off-line diagnostics loader provides a software environment for the HSC off-line diagnostics.
The loader supports a command language that loads and executes an off-line diagnostic from the
load device into Program memory. The loader command language also permits the display and
modification of any address contents in the HSC Program, Data, or Control memories.
The software environment provided for off-line diagnostics includes a load device driver and a
terminal driver. A standard software interface between the diagnostics and the load device and
terminal devices takes the place of individual interface routines within the diagnostics. The loader
also maintains a timer that keeps track of the relative time since the loader was last booted. This
allows diagnostic error messages to be time-stamped.
6.3.1 Loader System Requirements

Hardware required to run the off-line diagnostics loader includes:
• 110 Control Processor module with HSC Boot ROM
• At least one memory module
or
• TU58 with at least one working drive
• Terminal connected to 110 Control Processor console interface
6.3.2 Loader Prerequ isites

In the process of loading the off-line diagnostics loader, several diagnostics are run. The ROM
bootstrap tests the basic PDP-II instruction set, tests a partition in Program memory, and tests the
load device used for the boot. Then the bootstrap loads the off-line P.ioj/c test, which completes the
PDP-1I tests and the remainder of the I/O Control Processor module tests.
Mter these tests, the off-line diagnostics loader is loaded from the load device to memory, and
control is passed to the loader. Due to the sequence of tests that precede the loader, the loader
assumes the 110 Control Processor module and the RX.33 are tested and working.
6.3.3 Loader Operating Instructions

Follow these steps to start the off-line loader:
1. Insert the HSC off-line diagnostics media into the load device.
2. Power on the HSC, or press and release the Init button on the HSC OCP.
3. The load device drive-in-use LED lights within a few seconds, indicating the bootstrap is loading
the off-line diagnostic loader to Program memory.
4. In less than 30 seconds, the off-line diagnostics loader indicates it has loaded properly by
displaying the following:
HSC OFL Diagnostic Loader, Version Vnnn
Radix=Octal,Data Length=Word,Reloc=OOOOOOOO
ODL>
5. The off-line loader is now ready to accept commands. Section 6.3.4 contains information on the
loader command language.
6.3.4 Loader Commands

The following sections describe the commands recognized by the off-line loader.
6.3.4.1 HELP Command

The HELP command supplies an abbreviated list of all commands the loader recognizes. In
response to the HELP command, the loader reads the file OFLLDR.HLP from the load device
and displays the contents of this file on the HSC console terminal. Section 6.3.5.2 contains a listing
of the loader help file.
6.3.4.2 SIZE Command

The off-line system sizer is invoked by the SIZE command. The sizer determines the sizes of the
HSC Program, Control, and Data memories, and the type of requestor in each HSC requestor
position. The term requestor position refers to the priority of a particular requestor on the Data
and Control memory buses. It does not match the numbering of module slots.
6.3.4.3 TEST Command

The off-line diagnostics loader TEST command is used to invoke the various off-line diagnostics
available on the HSC. The following list shows the particular form of the TEST command used to
invoke each diagnostic. In general, the TEST command format allows specification of the system
component to be tested; for instance, the TEST MEMORY command invokes the off-line memory
test.
• Off-line Cache Test-Verifies the full functionality of the onboard cache. The off-line cache
test is invoked by the TEST CACHE command.
• Bus Interaction Test-Invoked by the TEST BUS command. The bus interaction test
generates contention on the HSC Data and Control memory buses by two or more Ks
simultaneously testing different sections of the Control and Data memories. '1Wo or more
working requestOrs are required to run this test (including the K.ci).
• K Test Selector-Invoked by the TEST K command. The K test selector allows specific
requestor microdiagnostics to run.
• KIP Memory Test-Invoked by the TEST MEMORY BY K command. The KIP memory test
uses one of the HSC requestors to test either Data or Control memory. This test runs faster
than the off-line memory test because a requestor is- roughly seven times faster than the 110
Control Processor. Program memory cannot be tested using the KIP memory test as the Ks do
not have an interface to the Program memory bus.
• Off-line Memory Test-Invoked by the TEST MEMORY command. This test uses the 110
Control Processor to test Program, Control, or Data memories.
• Off-line RX33 Exerciser-A combined hardware diagnostic and exerciser for the M.std.2lRX33
subsystem of the HSC. Invoke the off-line RX33 exerciser by the TEST RX command.
• Memory Refresh Test-Invoked by the TEST REFRESH command. The memory refresh test
allows the refresh feature of the memories to be tested.
• OCP Test-Invoked by the TEST OCP command. The OCP test checks the HSC lights and
switches. The test requires manual intervention by an operator.
6.3.4.4 LOAD Command

The LOAD command loads a program into HSC Program memory without starting it. The
command format is LOAD <filename>, where <filename> is the name of any file on the HSC
OFFLINE diskette. The loader finds the specified file and loads it into Program memory. This
command is for patching a program image in before starting execution. After the patch is made,
the program can be initiated through the START command.
6.3.4.5 START Command

The START command initiates the loader program currently loaded in Program memory. The
START command can be used in conjunction with the LOAD.command, or it may be used to
reinitiate the last loaded off-line diagnostic. This saves the time required to reload the program
from the load device. For example, you have previously typed SIZE to initiate the off-line system
sizer program and after the sizer completes, you wish to run it again. Typing START restarts the
sizer without reloading the program from the load device, saving many seconds of load time.
6.3.4.6 EXAMINE and DEPOSIT Commands

The EXAMINE and DEPOSIT commands are used to display or modify the contents of any location
in the asc Program, Control, and Data memories. Qualifiers (switches) can be used with these
commands to display bytes, words, long words, or quad words. The radix (octal, decimal, hex) of
the displayed data also can be controlled by qualifiers. Alternately, the SET DEFAULT command
can be used to set the default data length and radix for all EXAMINE and DEPOSIT commands
(Section 6.3.4.11).
EXAMINE Command:
The EXAMINE command is used to display the contents of any location in the asc Program,
Data, or Control memories. The format of the command is: EXAMINE <address>. The
<address> can be a string of digits in the current (default) radix. Certain symbolic addresses
also are permitted (Section 6.3.4.7).
In the following example, the user entered a command to examine the contents of location
14017776. (Notice the EXAMINE command can be abbreviated to a single E.) When the loader
displays the contents of location 14017776, the address is preceded by a (D) indicating the
location is within Data memory. The display shows the location contains the value 125252.
ODL> E 14017776
(D) 14017776 125252
DEPOSIT Command:
The DEPOSIT command is used to modify the contents of any location in the asc Program,
Control, or Data memories. The format of the command is: DEPOSIT <address> <data>. The
<address> can be a string of digits in the current (default) radix. Certain symbolic addresses
also are permitted (Section 6.3.4.7).
In the next example, the user entered a command to store the value 123456 in the contents of
address 14017776. The previous contents of this Data memory location are replaced with the
value specified in the DEPOSIT command (123456).
ODL> D 14017776 123456
6.3.4.7 EXAMINE and DEPOSIT Symbolic Addresses

The four symbols used as symbolic addresses in a DEPOSIT or EXAMINE command are described
in the following list.
• Asterisk (*)-Indicates the loader is to use the same address as used in the last EXAMINE or
DEPOSIT command. For example, if the contents of address 16012344 were just examined and
the value 1234 is to be deposited into the same address, type DEPOSIT * 1234 instead of typing
DEPOSIT 16012344 1234.
• Plus sign (+)-This sign also is used as a symbolic address. This symbol means the loader is
to use the address following the last address used by an EXAMINE or DEPOSIT command.
When the loader sees a plus sign (+) as an address, it takes the last address used by
EXAMINE or DEPOSIT and adds an offset, which depends on the current default data length
(Section 6.3.4.11).
If the current default data length is a byte, the loader adds one to the last address. If the
default is a word, the loader adds two to the last address. The offset is four for longword data
length and eight for quadword. This feature is useful when examining a number of items stored
in successive locations.
For example, if you want to examine a table of words beginning at address 14125234, examine
the first location by typing EXAMINE 14125234. The next location can now be examined by
typing EXAMINE + instead of typing EXAMINE 14125236.
• Minus sign (-)-This sign also is used as a symbolic address. It indicates the loader is to use
the address preceding the last address used by either command. When the loader sees a minus
sign (-) as an address, the loader takes the last address used by an EXAMINE or DEPOSIT
and subtracts an offset, which depends on the current default data length (Section 6.3.4.11).
If the current default data length is a byte, the loader subtracts one from the last address. If
the default is a word, the loader subtracts two from the last address. The loader subtracts four
for longword data length and eight for quadword. This feature is useful in the same way as the
+ symbol, but examines a table starting at the highest address and proceeding down to lower
addresses.
For example, if a table of words that ends at address 14012346 is to be examined, the operator
would examine the last location of the table by typing EXAMINE 14012346. The preceding
location in the table could now be accessed by typing EXAMINE - instead of typing EXAMINE
14012344.
• At (@)-The @ symbol also is used as a symbolic address. This symbol means the loader uses
the data from the last EXAMINE or DEPOSIT command as an address. This feature is useful
when following linked lists. For example, first examine location 123434 which contains a
pointer to a linked list. Now type EXAMINE @ to examine the location pointed to by the first
location.
6.3.4.8 Repeating EXAMINE and DEPOSIT Commands

When troubleshooting memory problems, continuously executing an EXAMINE or DEPOSIT
command is sometimes useful. The REPEAT command is used for this continuous execution. Type
REPEAT EXAMINE or DEPOSIT. To stop a repeated command, type CTRUC.
In the following example of repeating a DEPOSIT command, the value 125252 is continuously
deposited into address 14017776. The format of the DEPOSIT command does not change. The
DEPOSIT command is just preceded by the word REPEAT. Also, the REPEAT command can be
abbreviated to RE:
REPEAT DEPOSIT 14017776 125252
or
BE D 14017776 125252
In the repeating an EXAMINE command example, the contents of address 14017776 can be
continuously examined. The format of the EXAMINE command does not change. The EXAMINE
command is just preceded by the word REPEAT.
REPEAT BXAMDm 14017776
or
BE E 14017776
In the examples shown, the contents of location 14017776 are displayed continuously on the
terminal. This slows down the repetition of the command and wastes paper on hardcopy devices.
Stop output to the terminal by typing a CTRUO. However, the loader also provides a special
EXAMINE command qualifier (!INHIBIT) for suppressing output to the terminal. This qualifier is
discussed in Section 6.3.4.10.
6.3.4.9 Relocation Register

The loader provides a relocation register. It can be used to reduce the number of address digits
typed for an EXAMINE or DEPOSIT command when all addresses are in either the Control
or Data memories. The contents of the relocation register are added to the address given with an
EXAMINE or DEPOSIT command. The relocation register contains a 0 when the loader is initiated,
so it normally has no effect on the addresses typed in an EXAMINE or DEPOSIT command.
Use the following example to examine many locations in Data memory.

ODL> SET RELOCATION:14000000
ODL> EXAMINE 0
(D) 14000000 123432
ODL> EXAMINE 1234
(D) 14001234 154323
Load the relocation register with the address of the first location in Data memory (14000000).
When an EXAMINE command with an address of 0 is issued, the loader adds the relocation
register to the address given, resulting in the examination of address 14000000. Likewise, when an
EXAMINE command with an address of 1234 is issued, the loader displays the contents of location
14001234.
The following example shows how to examine many locations in Control memory.
ODL> SET RELOCATION:16000000
ODL> EXAMINE 0
(C) 16000000 125252
ODL> EXAMINE 4320
(C) 16004320 125432
The relocation register is loaded with the address of the :first location in Control memory
(16000000). When an EXAMINE command is issued with an address of 0, the loader adds the
relocation register to the address given, displaying the contents of address 16000000. Likewise,
when the user issues an EXAMINE command with an address of 4320, the loader displays the
contents of location 16004320. You can stop the display with a CTRUC.
6.3.4.10 EXAMINE and DEPOSIT Qualifiers (Switches)

The following list describes the EXAMINE and DEPOSIT qualifiers.
• /NEXT-Allows an EXAMINE or DEPOSIT command to work on successive addresses. When
used with a valid EXAMINE command, it specifies that after the command location has beeD:
displayed, the loader also displays the next number of locations following the first. For example,
the command E 10001NEXT:5 results in the display of locations 1000, 1002, 1004, 1006, 1010,
and 1012 (assuming the default data length is a word). The number of the argument can be any
value in the current default radix that can be contained in 15 binary bits or less. For instance,
if the default radix is octal, the number of the argument can be any value between 1 and 77777.
The /NEXT qualifier works the same way for the DEPOSIT command, except that the data
given with the DEPOSIT command is stored in the location specified and the next number of
locations following.
• IBYTElWORDILONG/QUAD--These qualifier switches are used to control the data length of
examined or deposited data. Normally, the loader uses the default data length (Section 6.3.4.11)
when data is examined or deposited. However, the data length qualifiers can be used to
override the default for a single examine or deposit. For instance, assume the default data
length is currently a word, and a byte quantity at address 16001234 is to be examined. Typing
EXAMINE 160012341BYTE would display the proper byte without affecting the default data
length.
• IOCTAUDECIMAIIIIEX-These qualifier switches can be used with an EXAMINE command
to control the radix of the address and data displayed. They are not used to control the radix of
the address supplied in the EXAMINE command. The radix of the address and data displayed
by an EXAMINE command is usually controlled by the current default radix (Section 6.3.4.11),
but the IOCTAlJDECIMAUHEX qualifiers are used to override the default radix for a single
EXAMINE command. For example, assume the default radix is octal. Typing EXAMINE
140012341HEX displays the contents of address 14001234(8) in the hexadecimal radix. The
EXAMINE display would be as follows: (D) 30029C HHHH. HHHH represents the contents
(hex) of the location displayed. The address is also displayed in hex.
• /INHIBIT (abbreviated to IINH)-This qualifier switch inhibits the display of examined data
when repeating an EXAMINE command. This is useful both for saving paper on hardcopy
devices and for speeding up the EXAMINE operation for scope-loop purposes. For example, the
command REPEAT EXAMINE 16012346IINH results in the loader continuously reading the
contents of location 16012346 without displaying anything at the console.
6.3.4.11 Setting and Showing Defaults

The SET DEFAULT command is used to change the default radix and/or data length. The default
radix controls the radix of parameters supplied with EXAMINE or DEPOSIT commands and the
radix of data displayed by the EXAMINE command. The default data length controls the length
(byte, word, long, quad) of data displayed by the EXAMINE command or data stored by a DEPOSIT
command.
The default radix may be set to octal, decimal, or hexadecimal. When the off-line loader first starts,
it sets the default radix to octal. Type SET DEFAULT HEX to set the default radix to hexadecimal.
Mter the default radix is set, it remains so until another SET DEFAULT command is issued or the
loader is rebooted.
The default data length may be set to byte, word, longword, or quadword. When the loader is
first started, it sets the default data length to word (16 bits). Type SET DEFAULT LONG to set
the default data length to longword (32 bits). Setting the default data length to longword causes
an EXAMINE command to display longword quantities and causes the DEPOSIT command to
store longword quantities. (Because the loader is executing in a PDP-II, longwords are stored and
retrieved as two successive 16-bit words.) After the default data length is set, it remains so until
changed by another SET DEFAULT command or until the loader is rebooted.
6.3.4.12 Executing INDIRECT Command Files

The loader is capable of executing indirect command files stored on the load device. These command
files consist of valid off-line loader commands terminated by a carriage return «CR» and a line
feed «LF». Comments may also be placed in indirect command files by preceding a comment line
with an exclamation mark (t). Comment lines must also be tenninated with a <CR> and <LF>. As
an example, the Off-line Loader Help file (Section 6.3.5.2) is an indirect command file that contains
only comments.
Indirect command files cannot be created by the loader or by CRONIC. The command files must be
created in RT-ll format and stored on the off-line diagnostics diskette. Any editor that does not
insert line numbers in the output files can be used to create command files.
6.3.5 Unexpected Traps and Interrupts

When the loader detects an unexpected trap or interrupt, the following message is displayed:
Unexpected trap through www, VPC=xxx, PSW=yyy
Error Address = zzz
Where:
www is the address of the trap or interrupt vector.
xxx is the virtual PC of the loader at the time of trap.
yyy are the contents of PSW at the time of trap.
zzz is the address of the location causing NXM or parity trap.
The first line of the unexpected trap report is issued for all unexpected traps or interrupts.
The second line is issued only if the trap was through vector addresses 000004 (NXM trap) or
000114 (parity trap). The address of the vector is a direct clue to the cause of the trap. Refer to
Section 6.3.5.1 for a list of the devices and error conditions associated with each vector.
The virtual PC (VPC) of the instruction executing when the trap occurs is sometimes useful in
determining the cause of the trap. The VPC can be referenced in the listing to find the instruction
causing the trap. The contents of the VPC includes the address of the instruction following the
instruction executing when the trap occurred. Notify Customer Service to analyze such failures.
NXl\f traps can be caused by EXAMINE or DEPOSIT commands if an address not contained in a
particular HSC is specified. For example, if an HSC contains only Data memory from addresses
14000000 through 14177776, and an EXAMINE or DEPOSIT is tried for address 14200000, the
loader reports an NXM trap. In this example, the NXM trap would not represent an error condition.
Parity traps can be caused by an EXAMINE command if a user examines an address not initialized
with good parity. For example, when the HSC memories are powered on, the parity bits are in
random states. Thus, if a user examines a location not written since power-on, the location may
generate a parity error. This does not constitute an error condition.
However, if a location produces a parity error and that location has been written since power-on, a
memory error is indicated. .
NOTE
The 110 Control Processor and Ks have hits allowing them to write had parity for testing
the parity circuit. These bits are for diagnostics engineering purposes only.
6.3.5.1 Trap and Interrupt Vectors

Table 6-,.3 is a list of trap and interrupt vectors for various devices and error conditions recognized
by the 110 Control Processor PDP-II processor.
Table 6-3 Trap and Interrupt Vectors

Vector Device or Error Condition
000004 Nonexistent memory, stack overflow, halt in user mode, and odd address trap
000010 illegal instruction
000014 BPI' instruction
000020 lOT instruction
000024 Power fail interrupt
000030 EMT instruction
000034 TRAP instruction
000060 Console terminal-receiver interrupt
000064 Console terminal-transmitter interrupt
000100 Line clock interrupt
000114 Parity trap
000120 Control bus interrupt---level 4
000130 Control businterrupt---level6
000230 RX33 interrupt
000250 MMU abort (trap)
000300 SLU (Serial Line Unit) #1, receiver interrupt
000304 SLU (Serial Line Unit) #1, transmitter interrupt
000314 SLU (Serial Line Unit) #2, receiver interrupt
000310 SLU (Serial Line Unit) #2, transmitter interrupt
6.3.5.2 Help File

The help file display is started by entering HELP at the ODL prompt. This file is unique to each
version of software, as shown by Vnnn in Example 6-1. This display is the complete help facility
for the off-line diagnostics. Exit the help file display by typing a CTRUC.
HSC OFL Diagnostic Loader Help File - Vnnn

Capital letters = required input, lower case optional
Commands (terminated by CR):

'Examine <address>' ;display data at <address> specified
'Deposit <address> <data>' ;deposit <data> to <address>
<address> = digit string in current default radix, or:
,*, use same address as last ex or de
'+' use address following last address
'-' use address preceding last address
'@' use <data> from last ex or de as address
'HElp' ;print this file
'@<filenarne>' ;execute indirect command file
'Load <filename>' iload file to diagnostic partition
'REpeat <command>' ";repeat specified command until AC
'RUN <filename>' i Do implicit LOAD and START of <filename>
'SEt Default <option>,<option>;set default radix or data length
<options> = Byte,Word, Long, Quad, Hex, Octal,Decimal
'SEt Relocation:.' ;set relocation register to •
relocation register is 22-bit positive •
; added to address of all 'Ex' and 'De'
commands.
'SHow' ;display defaults and Loader version
'SIze' ;Size memories and display K status
'Start' ;start program in diagnostic partition
'Test Bus' ;load and start the OFL Bus Interaction test
'Test K' iload and start the OFL K Test Selector
'Test MEmory' iload and start the OFL Memory Test
'Test MEmory By K' ;load and start the OFL K/P Memory Test
'Test OCP' ;load and start the OFL OCP Test
'Test Refresh' iload and start the OFL Refresh Test
'Test Cache' ;load and start the OFL cache test (LOlll-xx)
'Test Rx' iload and start the OFL Rx33 test (LOlll-xx)
'WCS' iload and start the OFL control store loader
!Qualifiers (switches) for 'Ex' and 'De':

'/Next:.' irepeat Ex or De on next 'i' addresses
'/Byte,/Word,/Long,/Quad' iuse specified data length instead of default
'/Octal,/Decirnal,/Hex' iuse specified radix for Ex display
'/INH' iinhibit display of examined data
<end of help file>
Example 6-1 Example HELP file display
6.4 OFLCXT-Off-line Cache Test

The off-line cache test (OFLCXT) is a diagnostic that rWlS Wlder the off-line loader in a standalone
environment. It provides in-depth testing of the cache logic on the Jll P.ioj and verifies the full
fWlctionality of the onboard cache. Execution time for a single pass is between 16 seconds and 4
minutes, depending on the options selected.

OFLCXT is loaded into memory through the off-line loader. This test requires 8-Kwords of memory
to run. One-half of this memory space contains the program; the other half is used as a cached
buffer. All terminal 110 and handling of the line dock is done by the off-line loader.

This section contains operating instructions specific to OFLCXT. If the HSC is not booted and
running the off-line loader, necessary instructions are found in Section 6.1.2 and in Section 6.2. To
run the off-line loader, enter the TEST CACHE command at the ODL> prompt.
This command loads OFLCXT from the media and transfers control to the diagnostic. When it
starts, OFLCXT displays the following:
HSC OFFLINE Cache Test Vxxx
Where: Vxxx is a three-digit version/edit number.

OFLCXT can be terminated by typing CTRUY.

Following are the three user-modifiable parameters for OFLCXT. In each case, the default, invoked
by a "," (comma), is shown in brackets. If no default is possible, the brackets are empty.
• Select Data Reliability test-This is the first user-modifiable parameter, an optional selection
of the data reliability tests. It is a moving-inversions style test for exercising the RAM array.
The off-line cache test prints:
Run extended cache ram test (Y/N) [N]?
Selection of this optional test increases test time per pass to about 4 minutes. It is useful for
the manufacturing burn-in and test areas. It is not necessary to run this optional test in order
to fully verify the health of the cache.
• Leave Cache Enabled-Determines the cache state at the termination of the diagnostic.
OFLCXT prints:
Leave cache enabled after successful completion (Y/N) [N] ?
This feature allows enabling the cache for further use after running the diagnostic to verify the
cache is working. If the diagnostic detects any hard failures in the cache, it is not enabled at
the end of the diagnostic. This prevents complications if the cache contains hard failures and is
inadvertently turned on.
• Number of Passes-Accepts a total number of passes from 1 to 32767 (decimal). The test
prompts for this number as follows:
* of passes to perform (D) [D) ?
Any decimal number up to 32767 can be used. Fatal errors can cause the diagnostic to
terminate before the specified number of passes executes.
At the completion of the total passes requested by the user, the diagnostic prompts:
Reuse parameters (Y/N) [Y] ?
To repeat the last test specified using the parameters, answer this prompt with Y or RETURN.
To cause the test to prompt for new parameters, answer the prompt with N.
Use the DELete key to delete mistyped parameters before terminating the entry with RETURN.
If an error in a parameter was terminated with RETURN, type CTRUC to return to the initial
prompt and re-enter all parameters.

OFLCXT provides summary information at the end of each pass. The end of pass message is
similar to the following:
End of Pass 00001, 00000 Errors, 00000 Total Errors
The errors field contains the number of errors for the pass. The total errors field contains a running
total of errors accumulated since the start of the diagnostic.
For any OFLCXT prompts, use the DELete key to delete mistyped parameters before terminating
the entry with RETURN. If an error in a parameter was terminated with RETURN, type CTRUC
to return to the initial prompt and re-enter all parameters.

Following are summaries of the OFLCXT tests 1 through 16:
• Test 1, Cache register access test-Checks for the presence of the necessary cache
controVstatus registers, the cache control register (17777746), the hit/miss register (17777752),
and the memory system error register (17777744). To perform further diagnosis, these registers
must respond.
• Test 2, Cache control register hits-Tests the read/write bits of the cache control register
(17777746) for stuck-at faults. In addition, bits (8,11:15), which are write-only, are checked for
read data of O. Bits 6 and 10 which cause data and tag parity to be written incorrectly on new
data allocated to cache are treated as special cases. Mter writing/reading each of these bits, the
cache is :flushed to remove any bad parity locations.
• Test 3, Force miss action-Verifies all references made with either bit 3 or bit 2 of the cache
control register set that cause a cache miss and leave the cache entry unchanged. To perform
this test, first write a test address with bits 3:2 cleared to allocate cache and place a known
data pattern into the cache. Then bit 2 is set, and the same test location is written again. With
bit 2 set, the cache will not update, and the data in cache is still considered valid. When bit 2 is
cleared and the test location is accessed again, the old data from cache is the result. If not, the
force miss action of bit 2 did not work. The same sequence is repeated for bit 3, and the same
results are expected.
• Test 4, Hit/miss register, part I-Checks the basic operation of the hit/miss register in
logging hit/miss information on instruction fetches and data reads/writes. The hit/miss register
is critical to further cache diagnosis because it is the window into what is actually going on
inside the cache.
First, a test location is allocated with cache enabled. Then cache is bypassed, and the test
location is accessed again by a write. This write goes directly to main memory and bypasses
the cache. The cache is enabled, and a read access to the test location results in a hit condition
in the hit/miss register. Then the test location offset by 8-Kwords is accessed. This results in a
miss, since the upper bits of the address (tag) will not match.
• Test 5, Hit/miss register, part IT-Checks all the combinations of the six bits in the hit/miss
register for a single miss at different bit positions. This is done by caching a certain sequence
of instructions and executing them, with miss conditions forced at each bit position. At the
completion of this test the hit/miss register has been checked for both 1's and O's at each bit
position.
• Test 6, Byte accesses-Ensures byte references to the cache are handled correctly by the
control logic. The first operation is a byte-write to the test location not allocated followed by
a byte-read of the test location. The read results in a miss. Then the entire word at the test
location is allocated. The upper byte of the test location is modified, and a cache hit is expected.
The entire word is also read and compared against the expected result to see if the byte-write
occurred. A similar chain of events follows, this time modifying the low byte.
• Test 7, PDR Cache bypass test-Tests all of the Kernel PDRs <0:7>, as well as the User
PDRs. It is very important for the bypass cache bit (bit 15 of any PDR) to work correctly in the
multiprocessing environment of the HSC.
PDR bypass is tested by remapping all PDRs to point to control memory. Control memory is
written by the MMU writing a data pattern and allocating cache. Control memory windows
are used to write Control memory to a second pattern without involving the cache control logic.
When Control memory is read through the lMMU with the bypass bit set, the actual Control
memory content (second pattern) is the result if the bypass bit is actually set. If the old content
(first pattern) is read back, the bypass bit is not working. PARs 1,2, 3, 5, and 6 are tested in
this way.
PARs 0, 4, and 7 are treated as special cases due to programming environment restrictions.
They are tested by allocating cache with some location mapped by the PARlPDR under test and
then setting the bypass bit. When the test location is read, the hit/miss register records a hit
and then invalidates the location. If the location is written or read again, it results in a miss as
long as the bypass bit is set.
After all the Kernel PARlPDR registers are tested, the program maps user space that is
identical to Kernel space and switches into user mode to re-execute all the tests. After all
User PARlPDR pairs have been tested, the program swaps back. into Kernel mode and proceeds
to the next test.
• Test 8, Cache :Bush action-Allocates all 4 Kwords of cache, and then executes a flush
command by setting bit 8 in the cache control register. The cache control logic then writes every
location in cache with the data value 17777746 and resets the valid bit for each location. All 4
Kwords of cache allocated before the flush are read again, and if any location responds with a
hit when read, an error is declared.
• Test 9, Unconditional bypass to main memory-Checks the correct operation of bit 9 of the
cache control register. Bit 9 is used to bypass cache in a fashion similar to the bypass bits in
the PARJPDRs. Any location allocated in cache before the bypass bit is set results in a hit on
the first access, and further accesses all show as misses.
This function is used when it is desirable to temporarily disable the cache in a fashion that does
not leave the cache with stale data when re-enabled. A test location is allocated, and then the
bypass bit is set. The first access of the test location is a hit, and the second is a miss.
• Test 10, Force tag/data parity errors-Forces parity errors in the tag and data fields of the
cache array to test the parity detection logic. A special diagnostic mode is used, with bit 0 of
the cache control register and one of the force parity error bits set. When bit 0 is set, any trap
through 114 is disabled on a parity error detected in cache. If a parity trap does occur, an error
is declared.
First, tag errors are forced using bit 10 in the cache control register. When this bit is set,
locations allocated to cache do so with bad tag parity. When accessed again (resulting in a
cache hit), the tag parity error bit is set (bit 5 in the memory system error register). The force
data parity error bit (bit 6 of the cache control register) is checked next. After a location is
allocated to cache with bad data parity, further reads of that location result in setting the data
parity error hits (bits 6:7) of the memory system error register. After using the force bad parity
bits, the program flushes the cache to remove these parity errors.
• Test 11, Abort/interrupt on parity errors-Uses the force parity error bits in the cache
control register to force parity errors in the cache array. Because testing of the detection of such
errors has been done, testing of the other logic related to cache data or tag parity errors can be
done.
Different combinations of tag and parity errors are forced, with the cache control register set to
interrupt through trap 114, or abort through trap 114 on parity errors. An interrupt through
trap 114 sets the correct error bites) in the memory system error register. Also, the instruction
detecting the parity error completes.
On an abort through trap 114, the correct error bites) is set, but the instruction does not
complete. If the parity error is detected on the fetch of the source data, the data in the
destination of the instruction is not modified. The PC on the stack after each interrupt or
abort instruction is checked against the PC that is expected.
• Test 12, DMA invalidate-Modifies a location resulting in· the cache acquiring stale data
unless cache logic detects the DMA change. The RX33/.M:.std2 subsystem is used to generate
DMA operations to Program memory. A DMA write to a Program memory location allocated to
cache results in a cache miss when it accesses after the DMA write.
• Test 13, Check blockage of parity error on NXM abort-Generates simultaneous NXM
and parity errors. The NXM trap occurs, overriding the parityerror.
• Test 14, Cache data RAM test---Tests the cache data RAMs by mapping one PAR and using
the cache solely for data storage. A data pattern to detect dual-addressing is written to the
cache. Failures of the cache data to match the EXPected data on read-back are considered
miscompare errors. The test is first done using word addresses and test values, and then
repeated with byte addresses and byte data patterns. Each location allocated is expected to be
a hit from cache, and the content is checked as well.
• Test 15, Tag store RAM test-Checks the tag bits of the cache array for dual address errors
and stuck-at faults. With the cache flushed and completely deallocated, the first 256 locations
of the cache are written with a unique data value in each address. Then the entire cache is
read. Only the 256 locations written are cache hits, and only these locations have the EXPected
data pattern. Then the upper address bits are changed so a new combination of tag bits results.
This test is repeated 15 times until all of the tag bits have been tested.
• Test 16, Data RAM reliability test---Performs a modified moving inversions test on the cache
data RAM array. Due to the geometry of the data RAMs, every fourth bit is done concurrently
to save time. This results in using the same pattern in both nibbles of the data word. This
test must be selected by the user as it does not normally run by default. About 4 minutes are
required to complete one pass of this test.

OFLCXT displays the errors detected during execution on the console terminal. Error messages
follow the diagnostics generic error message format preceded by an OFLCXT> prompt.
A typical OFLCXT error message format follows:
OFLCXT>hh:mm T aaa E bbb U-OOO

< Text describing error >
MA-xxxxxxxx
EXP-yyyyyy
ACT-zzzzzz
where:
hh is the elapsed hours since last bootstrap.
rom is the elapsed minutes.
aaa is the decimal number denoting test.
bbb is the decimal number denoting the error detected.
MA-xxxxxxxx is the address of location causing the error.
EXP-yyyyyy is the data pattern that was expected.
ACT-zzzzzz is the data pattern that was actually found.
Each error number has a unique text string associated with it. For errors with results that did not
compare with the expected value, the diagnostic uses the optional lines to show EXPected/ACTual
data.
Soft errors (such as cache parity errors) can accumulate to a point where the diagnostic classes
them as fatal. The test then terminates on a fatal error.

The following list describes in detail each possible error message. The errors are listed in numerical
order.
• Error 00, Memory Parity Error, VPC = x x 8 I 8 I Applicable to all tests. Can occur at any
time during execution of the diagnostic. The virtual PC on the stack is printed to help identify
the program area where the error occurred. The content of the error address register also is .
displayed.
Both the virtual PC and the error address register content are optional lines. Detection of this
error causes the testing to cease. Then the diagnostic returns to the Reuse Parameters prompt.
• Error 01, NXM Trap, VPC = XXXI81 Applicable to all tests. Causes the diagnostic to return
to the Reuse Parameters prompt. Additional data (such as the virtual PC of the instruction that
caused the trap and the physical address contained in the error address register) are printed as
optional lines.
• Error 02, Cache Parity Error, VPC = 8 18888 Applicable to tests 2 through 16. Results
when a trap through the parity error vector is detected and the cache is enabled. The virtual
PC where the error was detected is printed, as well as the content of the error address register.
If the 22-bit value in the error address register is 177770024, no main memory error was
present. Assume the parity error is from the cache.
• Error 03, Bit Stuck in Cache Control Register-Applicable to test 2. Indicates a bit is
stuck-at-fault in the cache control register. The EXPected and ACTual data values are printed
as optional lines.
• Error 04, Forced Miss Operation Failed-Applicable to test 3. Bit 2 of the cache control
register does not prevent the cache from allocating a test location. This could be a problem in
the cache control gate array or in the hit/miss compare logic.
• Error 05, Forced Miss with Abort Failed-Applicable to test 3. Bit 3 did not prevent the
cache from allocating when set. Failures of this nature mean the cache cannot be disabled,
and all memory references may be allocating cache regardless of the intent of the code being
executed. The cache control gate array or the tag compare logic may be at fault.
• Error 06, Expected Cache Hit Did Not Occur-Applicable to tests 4, 6, 9, 12, and 14. Did
not allocate a given test location to the cache as expected, causing a miss condition in the
hit/miss register.
• Error 07, Expected Cache Miss Did Not Occur-Applicable to tests 7, 9, and 10. Shows a
test location not expected to be allocated, or valid, as a hit on access.
• Error 10, Value in HitJMiss Register Incorrect-Applicable to test 5. Indicates the 6-
bit value in the hit/miss register was incorrect after a certain sequence of instructions. The
expected values, as well as the actual contents of the hit/miss register, are printed as optional
lines.
• Error 11, Write Byte Operation Caused Cache Update-Applicable to test 6. A byte
operation (on a miss) did not cause cache to deallocate the test location. Thus, when the test
location was read back, a cache hit resulted.
• Error 12, Write Byte Did Not Cause Cache Update-Applicable to test 6. A byte-value did
not get written into cache or main memory.
• Error 13, Cache Failed To Flush Successfully-Applicable to test 8. When checking cache
after a flush command was executed, one or more locations still contained valid data (were
detected as cache hits).
• Error 14, Access with Force Bypass Did Not Cause Invalidate-Applicable to test 9. The
second access to an allocated location, with the force bypass bit (bit 9) set in the control register,
did not result in a miss as expected.
• Error 15, Tag Parity Error Did Not Set-Applicable to test 10. The diagnostic could not
set the tag parity error bit in the memory system error register when faced with an actual tag
parity error.
• Error 16, Abort on Cache Parity Error Did Not Occur-Applicable to test 11. The cache
logic did not abort the instruction under execution when a cache parity error was forced, and
the abort bit (bit 7) was set in the control register.
• Error 17, Unexpected Parity Trap During Abort Test-Applicable to test 10. Although
expected to, cache control bit 0 did not prevent the cache logic from taking a trap on bad parity.
The address where the trap occurred is printed as optional information.
• Error 20, Content of Memory System Error Register Incorrect-Applicable to test 11.
The error bits in the memory system error register (1777744) do not reflect the correct status
for the operation. under test. The EXPected and ACTual content are printed as optional lines.
• Error 21, Return PC Wrong During Abort/Interrupt Test-Applicable to test 11. The
return PC on the stack is not equal to the value expected during an abort or interrupt operation
caused by a cache parity error. The state sequencer gate array is most likely defective.
• Error 22, Cache Data Parity Bit(s) Did Not Set-Applicable to test 10. The diagnostic was
unable to set the data parity error bitCs) in the memory system error register on a forced parity
error. The parity logic may not be detecting parity errors or one of the bits in the memory
system error register may be stuck low.
• Error 23, Interrupt on Parity Error Did Not Occur-Applicable to test 11. The cache did
not interrupt through vector 114 on a forced parity error. The state sequencer or the parity
detection logic may be faulty.
• Error 24, Expected NXM Trap Did Not Occur-Applicable to test 13. A NXM trap was not
detected during an access to location 1777757776. The timeout logic that detects a NXM may
be defective, or some problem may exist in the cache data path gate array that prevents it from
acting on timeout.
• Error 25, Parity Error Was Not Blocked By NXM-Applicable to test 13. When accessing
a location expected to result in a NXM, the parity error flag set instead, and a trap occurred
through vector 114. The NXM signal may not have been detected by the cache data path gate
array.
• Error 26, Cache Data Miscompare on Word Operation-Applicable to test 14. A word
address in the cache array did not have the correct data when read. This may indicate address
line faults or data path faults allowing the location to be rewritten after the test value was
placed there. The EXPected/ACTual data values are printed as optional lines.
• Error 27, Cache Data Miscompare on Byte Operation-Applicable to tests 14 and 15.
A location in the cache, when addressed in a byte fashion, did not have the EXPected data
pattern. This may indicate address line faults or data path control faults which allowed
overwriting the EXPected value.
• Error 30, DMA Write to Memory Did Not Cause Cache To Invalidate-Applicable to test
12. A DMA write by the RX33 controller to a test location, allocated to cache, still resulted in a
hit status after the transfer. The cache has stale data.
• Error 31, Instruction Still Completed During Abort Condition-Applicable to test 11.
With the abort bit set in the cache control register, an instruction set up to detect a parity error
on an operand fetch still finished execution modifying the destination of the instruction.
• Error 32, Load Device Error During DMA Test-Applicable to test 12. The load device
subsystem did not respond correctly to the DMA test operation. There may be faults in the load
device controller or the interrupt service logic. This message is information only.
• Error 33, PDR Cache Bypass Failed-Applicable to test 7. Setting the PDR bypass bit in
the PARJPDR pair under test did not bypass the cache. This points to a Ml\{U or cache data
path gate array problem. The PDR number and the CPU execution mode (Kernel or User) are
printed as optional lines in the error message.
• Error 34, Tag Store Address mt Failure-Applicable to test 16. Changing the value of the
tag bits (bits 16:22 of the physical address) still resulted in a hit condition (even though the
address should not have compared) forcing a fetch to main memory. There may be a problem in
the tag RAMs or the tag compare logic in the cache data path is not working.
• Error 35, Tag Store Address Miss Failure-Applicable to test 16. When going through the
possible values for the tag bits (16:22 of the physical address), the cache failed to allocate for
some combination of the bits. Possible problems are stuck bits in the address lines going to the
cache array, bad RAMs in the cache array, or a fault in the tag compare logic.
• Error 41, Processor Type Is Not JI1-Applicable to test 1. The processor type register does
not show the correct value for a J11 chip set. Attempting to run this diagnostic on anything
other than a J11 produces this error.
6.4.9 Test Troubleshooting

All of the logic under test is contained on the J11 (P.ioj) module with the exception of the memory
used by the diagnostic. Main memory parity errors usually point to the memory module. Because
much of the logic tested is buried within the two gate arrays on the module, troubleshooting is often
limited to a best-guess replacement of one or both of these gate arrays.
Cache parity errors and data miscompare errors can usually be traced to specific RAMs if proper
attention is paid to the data content and address.
For scope loops, the cache test is run with a large number of passes, and a CTRUO typed on the
console to inhibit error message printout.
Constant hit/miss errors, or tag address hit problems, also may be caused by the tag compare logic,
which is separate from the gate arrays and the data path.
6.5 OBIT-Off-line Bus Interaction Test

The off-line bus interaction test (OBIT) creates Control and Data bus contention among the
requestors in the HSC subsystem. The contention is generated by simultaneously testing different
portions of the same memory (Control and/or Data) from different requestors. In the process of
testing the memories, the various requestors in the subsystem contend with each other for the use
of the Control and Data buses.
In addition to the bus contention generated by the requestors, 110 Control Processor interaction
can be selected with the Program, Control, and Data memories, with the OCP, and/or with the
load device. If 110 Control Processor interaction is selected, it occurs simultaneously with the bus
contention generated by the requestors.
This test requires a minimum of two working requestors in order to operate and uses a maximum of
nine requestors if they are available. The more requestors available for use by this test, the greater
the amount of bus contention. A larger number of requestors makes it easier to isolate failures to a
particular source. Also, the run time of this test increases linearly as the number of requestors is
increased.
If the OBIT fails, it must first be determined if the failure was caused by an interaction problem.
This is determined by running the off-line KIP memory test (test memory by K). When the test
prompts for parameters, specify the requestor number of the requestor that detected the failure in
OBIT. Also specify the same starting and ending addresses displayed with the error report from the
bus interaction test. If the requestor also fails the off-line KIP memory test, the original problem
was not an interaction problem. The problem is localized in the same manner as any ordinary
memory failure.

Hardware required to run OBIT is shown in the following list:
• 110 Control Processor module with HSC boot ROMs
• Memory module
• Working Control and Data memories
• Load device with at least one working drive
• At least two working requestors (K.sdi, K.sti, or K.ci)
6.5.2 Off-line Bus Interaction Test Prerequisites

Booting procedures and testing through successful loading of the off-line diagnostics loader program
are described in Section 6.1.2 and in Section 6.2.
Due to the sequence of tests that precede the memory test, OBIT assumes the 110 Control Processor
module and the load device are tested and working. OBIT also assumes the Control and Data
memories were previously tested with the off-line memory test or the off-line KIP memory test, and
are working.

At the loader prompt ODL>, the operator types the TEST BUS command and OBIT is loaded and
started. The test indicates it has been loaded properly by displaying the following:
Hse OFL Bus Interaction Test
The test then sizes the Program, Control, and Data memories and determines the number of
requestors available for testing.

OBIT can be terminated by typing CTRL/C. OBIT may continue running for a few seconds after it
is terminated.

Mter displaying the program name and version, the Program, Control, and Data memories are
sized. The bounds of each memory are displayed on the terminal.
NOTE
For any of the OBIT prompts, use the DELete key to delete mistyped parameters. If an
error in a parameter already terminated with is noted, type a CTRUC to return to the
off-line loader. Then typ~ START to restart the test from the beginning.
The test prompts for selection of the requestors used for the test, as follows:
Use requestor #001, K.ci (YIN) [Y] ?
Answer with a Y if the K.ci should be used. Answer with N if the K.ci should not be used.
At least two working requestors must be used to run the bus contention test because one requestor
cannot generate bus contention by itself. The program displays the following error message if less
than two requestors remain after the requestors that should be used have been indicated:
Not Enough Ks Available for Test
Next, the program prompts for the type of I/O Control Processor interaction desired.
P.ioj Memory Interaction desired (YIN) [Y] ?
Answer the prompt with a RETURN (or Y) if I/O Control Processor interaction with memory is
wanted. Answer with N if I/O Control Processor interaction with memory is not wanted. If the
prompt is answered with an N, the following three prompts are skipped. If the prompt is answered
with a RETURN, the following prompts are displayed:
Interact with Program memory (YIN) [Y] ?
Interact with Control memory (YIN) [Y] ?
Interact with Data memory (YIN) [Y] ?
For each prompt, answer with a RETURN if I/O Control Processor is to interact with the specified
memory while the requestors are generating contention on the Control and Data buses. Answer
with N if the I/O Control Processor is not to interact with the specified memory. (If I/O Control
Processor interaction is selected, the I/O Control Processor interacts with the memory at the same
time the requestors are generating Control and Data bus contention.) The program next prompts
for OCP interaction.
OCP Interaction Desired (YIN) [Y] ?
If I/O Control Processor interaction with the OCP is wanted, answer with RETURN. If OCP
interaction is not wanted, answer with N. The test then prompts for load device interaction.
Interact with load device (YIN) [Y] ?
If 110 Control Processor interaction with the load device is wanted, answer with RETURN. If such
interaction is not wanted, answer with N. The program then prompts:
Number of passes to perform (D) [1] ?
Enter a decimal number between 1 and 2,147,483,647 (omitting commas) to specify the number of
times the bus interaction test is repeated. Entering 0 causes one pass of the test. Mter the nwnber
of passes is entered, the bus contention test begins. The test can be aborted at any time by typing
CTRUC. The test may continue running for a few seconds after CTRUC is typed.
After the specified number of passes is completed, the following prompt is issued:
Reuse parameters (YIN) [Y] ?
To repeat the last test specified using the parameters, answer this prompt with Y or RETURN. To
cause the test to prompt for new parameters, answer the prompt with N.
If an error in a parameter was terminated with RETURN, type CTRLlC to return to the initial

Each time the program completes one full set of bus contention tests, an end-of-pass report is
displayed. A pass consists of completing a full set of contention tests, including: Control bus tests,
Data bus tests, and combined Control and Data bus tests. The end-of-pass message is displayed as
follows:
End of Pass nnnnnn, xxxxxx errors, yyyyyy total errors.
Where:
nnnnnn is the decimal count of the number of passes completed.
xxxxxx is the decimal count of the number of errors detected
on the current pass.
yyyyyy is the decimal count of the total number of errors detected
since the test was initiated.
6.5.7 Test Summary

The moving inversions memory test (MOVI) is used to generate bus contention among the
requestors. Each requestor in an HSC contains the moving inversions test as part of its
microdiagnostic software set. The moving inversions RAM test is used to detect data and
addressing problems in dynamic semiconductor memories.
The following are the steps in the moving inversions algorithm:
1. Write 000000 in each location being tested.
2. Read all locations in order from lowest to highest. Mter reading a location and checking for a
0, rewrite the same location with a 1 in the least significant bit. Then reread the location and
verify the write worked correctly.
3. Again, read all locations in order from lowest to highest, checking to see each location contains
the data previously written. Then rewrite the data fOWld with a single additional 1 bit and
reread to check that the write worked properly.
4. Repeat step 3 Wltil the test pattern consists of a word containing all 1's (pattern 177777).
5. Repeat steps 1 through 4, but this time start at the highest memory address and work down
to the lowest each time. However, instead of adding an additional 1, add an additional O. This
changes each memory location from all 1's back to all O's.
6. End of test. All memory is cleared to 000000.

Error messages produced by this test conform to the HSC generic diagnostic error message format.
Off-line bus interaction test error messages are preceded by an OBIT> prompt.
Following is a typical OBIT error message.
OBIT>hh:mm Ttaaa Etbbb U-OOO

Memory Test Error
Detected By K.sdi, requestor 006
MA-xxxxxxxx
EXP-yyyyyy
ACT-zzzzzz
< K-error-Summary-Info >
Memory Test Configuration:
K.ci , requestor 001, M.ctl 16000700 16100274
K.sdi ,requestor 006, M.ctl 16100300 16177674
where:
hr is the hours since Off-line Loader was last booted.
mm is the minutes since Off-line Loader was last booted.
< K-error-Summary-Info >
Refer to Section 6.5.9 for information on Requestor Error Summary, and to Section 6.5.10 for
information on Memory Test Configuration.
6.5.9 Requestor Error Summary

When the requestor reports a memory test failure to the I/O Control Processor, the following
information is supplied:
a. Address of the failing memory location
b. Data EXPected and data ACTually found
c. Error summary information
The error summary information is supplied as a 3-bit field, including:
a. A bit indicating a parity error occurred while reading the location
b. A bit indicating an NXM: error occurred while accessing the location
c. A hit indicating a Control bus (CBUS) error occurred while accessing the location
When a memory error report is issued for an error detected by the requestor, the last line of the
error report includes a list of the error summary bits that were set, if any.
A Control bus (CBUS) error indicates the requestor asserted an illegal combination of the three
CCYCLE lines when accessing Control memory. Because these lines were previously tested from
the I/O Control Processor (in the OFL P.ioj/c test), a Control bus error is probably caused by a
problem with the requestor's drivers that assert the CCYCLE lines.
6.5.10 Memory Test Configuration

The memory test configuration lists each requestor being used for OBIT along with the section
of memory each requestor was testing when the failure occurred. The configuration information
consists of:
1. Type of requestor (K.ci, K.sdi, K.sti, or K.si) and the requestor number
2. Memory being tested by the requestor (M.ctl = Control memory, M.data = Data memory)
3. First address of the chunk of memory being tested
4. Last address of the chunk of memory being tested

The following list describes the nature of the failure indicated by each error number:
• Error 000, Memory Test Error-Indicates one of the requestors detected a memory error in
the Control or Data memories. The following is a sample error report.
Memory Test Error
Detected by K.ci, requestor 001
MA -16010234
EXP-000177
ACT-000377
Parity error
K.ci ,requestor 001, M.ctl 16000700 -- 16100274
K.sdi ,requestor 007, M.ctl 16100300 -- 16177674
Where:
MA is the 22-bit address of the failing location.
EXP is the data pattern EXPected by the requestor.
ACT is the data pattern found by the requestor.
Memory Test Configuration are the other requestors that
enabled when failure occurred.
This sample error report indicates the Ksdi detected a memory parity error while reading
address 16010234 of Control memory (M.ctl). The requestor expected to find the value 000177
in the location but instead found the value 000377. At the time the error occurred, the Kci
in requestor 1 was testing addresses 16000700 through 16100274 of Control memory, and the
Ksdi in requestor 7 was testing addresses 16100300 through 16177674 of the Control memory.
• Error 001, K Timed Out During Init--Displayed when a requestor fails to complete its lnit
sequence in time. This error usually indicates the specified requestor failed one of its internal
microdiagnostics. A sample error report follows:
K Timed-out During Init
K.ci , requestor 001, Status = 104
Other Ks Enabled:
K.sdi, requestor 6
K.sdi, requestor 7
This sample error report indicates the K.ci in requestor 1 did not finish its initialization
diagnostics in the required time. The requestor status displayed with the error report indicates
the requestor failed test 4 of its microdiagnostics (lxx in status = failed test xx). Two other
requestors were enabled at the time the requestor K.ci timed out, and one of these requestors
may be responsible for the time-out.
When the 110 Control Processor enables the requestor to perform the memory test, the requestor
begins its initialization sequence, including execution of certain microdiagnostics. At the
end of the requestor's Init sequence, the requestor indicates it found the K Control Area by
complementing a pointer word in Control memory. If the requestor fails to complement this
pointer word within 50 milliseconds (4.2 seconds for the K.ci) after being enabled, error 001 is
reported. The contents of the K status register are displayed with the error report.
• Error 002, K Timed Out During 'lest-Indicates the specified requestor failed- to complete its
memory test within the expected time. A sample error report follows:
K Timed-out During Test
K.sdi, requestor 007, Status = 002
K.ci , requestor 1, M.ctl 16000700 16100274
K.sdi, requestor 7, M.ctl 16100300 16177674
The sample error report indicates the K.sdi in requestor 7 never completed the memory test
it was assigned. (Ks are allowed up to 1 minute to complete a memory test.) The memory
configuration displayed with the error report shows all Ks testing at the same time the K.ci
timed out. In this example, the K.ci in requestor 1 was also testing at the time the K.sdi timed
out.
Test time-out failures may be caused by a failure in the requestor that timed out. They may
also be caused by a failure in one of the other requestors that was testing at the same time.
• Error 003, Parity Trap-Indicates the 110 Control Processor detected a parity error. The
22-bit address of the location causing the error is displayed as the MA data in the error report,
where:
• MA is the address causing the parity trap.
• VPC is the Virtual PC of the memory test at the time the trap occurred. Reference this
address in the listing to locate the area of the test where the error occurred.
The data is lost when a parity trap occurs so no EXPected or ACTual data can be displayed.
• Error 004, NXM Trap-Indicates the 110 Control Processor detected a Nonexistent Memory
(NXM) error. An NXM error is caused when no memory responds to a particular address. The
MA data in the error report indicates the address which produced the NXM trap. Mter the trap
is reported, the program attempts to restart the test from the beginning. The MA and VPC
fields have the same meanings as error 003.
If this error occurs at a memory address that should be in the memory configuration, the
memory in question is not supplying an ACK to the I/O Control Processor when the specified
address is presented on the memory bus. The most probable point of failure is the logic on
the memory module that compares addresses on the memory bus with the range of addresses
to which the module is to respond. Also, the comparator itself could be faulty or the [C IN, C
OUT], [D IN, D OUT], or [P IN, P OUT] lines on the backplane could be installed in error.
• Error 005, Memory Test Error (P.ioj/c detected)-Indicates the 110 Control Processor
detected an error while testing Program memory. This error can only occur if I/O Control
Processor interaction with Program memory is selected. This interaction consists of:
1. A series of PDP-II instructions that perform read/modify/write (RMW) cycles to selected
Program memory locations
2. Quick-verify tests of the entire Program memory (done 6 Kwords at a time)
Error 005 can be caused by cross-talk between the Program memory bus and either the Control
or Data bus. It can also be caused by a failure in the Program memory logic which inhibits
refresh cycles in the middle of a RMW cycle.
• Error 006, TU5S Synchronization Failure - This is an HSC50-only error. The TU58 drive
was unable to properly establish synchronization.
• Error 007, General TU58 Error- This is an HSC50-only error. Text is provided with the
error message to indicate details about the TU5S failure.
• Error 008, TU58 Checksum Error- This is an HSC50-only error. The data checksum read
from the TU5S did not match the one generated during the read operation.
• Error 009, TU58 End Packet Error- This is an HSC50-only error.
• Error 010 (12 octal), Cache Parity Trap, VPC = iii*XX Can happen during any test. The
Jll trapped through the parity vector. The error was caused by the cache.
NOTE
Errors 011 through 017 can occur on an HSC when load device interaction is enabled.
• Error 011, RX33 Drive Not Ready-The drive selected for the operation was not ready. The
door may be open or the diskette absent during a READ or POSITION command.
• Error 012, RX33. CRC Error During Seek- The RX33 detected a CRC error during a seek.
The RX33 could not verify position when reading header information from the diskette.
• Error 013, RX33 Track 0 Not Set on Recalibrate---A Recalibrate (seek to track 0) operation
is performed before each block of Read operations. The RX33 did not show correct status after
the RECAL command.
• Error 014, RX33 Seek Timeout- The RX33 did not respond by interrupting during a seek.
• Error 015, RX33 Seek Error-Sets the seek error bit (bit 4 of the CSR). At the end of a Seek
operation, the RX33 was not where it thought it should be.
• Error 016, RX33 Read Timeout-Indicates the RX33 did not interrupt at the end of a READ
command.
• Error 17, RX33 CRCIRNF Error on Read Command-Can be caused by a soft error or a
bad spot(s) on the disk. For informational purposes, the following additional message prints
out:
First LBN In Transfer = xxxx
Where xxx is the LBN of the first block in the transfer. The off-line interaction bus test
performs reads in blocks of four.
6.6 OKTS -Off-line K Test Selector

The off-line K test selector (OKTS) allows a K to perform an internal microdiagnostic self-test on
command. OKTS executes from the P.ioj/c and uses the HSC K Control Area for instruction. Select
the K for testing and the test number of the microdiagnostic test for execution.

The following hardware is required to run OKTS:
• P.ioj/c module with HSC Boot ROMs
• M.std2lM.std memory module
• A working section of Control memory for use as a K Control Area
• One working load device drive
• Terminal connected to the P.ioj/c console interface
Due to the sequence of tests that precede this test, assume the P.ioj/c, Program memory, and load
device are working.

If the HSC is not booted and loaded, refer to Section 6.1.2 and Section~6.2. If the loader prompt
ODL> is displayed, follow these steps to start the K test selector:
1. Type TEST K. The load device drive-in-use LED lights as the test is loaded.
2. Test K indicates it has been loaded properly by displaying the following:
HSC OFL K Test Selector
3. The test next prompts for parameters.

OKTS can be terminated by typing CTRUC.

This section gives detailed information on how to enter the test parameters for the OKTS. Items
in square brackets are the default value for each particular prompt. If no default is possible, the
brackets are empty.
OKTS first prompts:
* of K (1
I
Requestor through 9) [] ?
Answer this question with a single digit (1 through 9) that specifies the requestor number of the K
to be used. Terminate the response by typing RETURN. After the requestor number is supplied, a
K Control Area is located in Control memory and tested. This area is required for communicating
with the K that will run its microdiagnostics. The test then prompts:
Test * (1 through 20) (0) [] ?
Legal test numbers are octal numbers between 1 and 20, except for test 5. Test 5 is the K's Control
and Data memory test, which is supported by the OFL KIP memory test. Terminate the test
number entry with RETURN.
Refer to the following lists for the names of each microdiagnostic. Included in each list is the type
of K being used and the failing test number.
1. Kci microdiagnostics-The following list shows the test number and name of each of the Kci
microdiagnostics:
Test 0 - Sequencer test
Test 1 - ALU test
Test 2 - Data bus test
Test 3 - Control bus test
Test 4 - PROM parity test
Test 5 - Memory test (unavailable through K test selector)
Test 6 - RAM test
Test 7 - PLI interface test
Test 10 - Packet buffer test
Test 11 - Link test
2. Ksdi microdiagnostics-The following list shows the test number and name of each of the Ksdi
microdiagnostics:
Test 1 - ALU test

Test 4 - PROM parity test
Test 5 - Memory test (not available through K test selector)
Test 6 - RAM test
Test 7 - SERDESIRSGEN test
Test 10 - Partial SDI interface test
Test 11 - Control memory access test
Test 12 -Lock test
3. Ksti microdiagnostics-The following list shoVls the test number and name of each of the K.sti
microdiagnostics:
Test 1 - ALU test
Test 4 - PROM parity "test
Test 6 - RAM test
Test 7 - SERDES test
Test 10 - Partial STI interface test
Test 11 - Control memory access test
Test 12 - Lock test
4. Ksi microdiagnostics-The following list shows the test number and name of each of the Ksi
I
microdiagnostics:
Test 0 - 2911 test
Test 1 - ALU t.est
Test 2 - ROM parity/traps test
Test 3 - Scratchpad test
Test 5 - MOVI for system test
Test 7 - RTS gate array test
Test 10 - SIECL test
Test 11- Frame test (U => L)
Test 12 - Frame test (L => U)
Test 13 - Sector test (U => L)
Test 14 - Sector test (L => U)
Test 15 - WCS load/verify test
Test 16 - WCS MOVI test
Test 17 - WCS EDC check
Test 20 - INIT packet search test
The test then prompts:
# of passes to perform (D) [1] ?
Enter a decimal number between 1 and 2147483647 to specify the number of times the memory test
should be repeated. (Entering 0, or just RETURN, results in the performance of one pass.)
The P.ioj/c next instructs the K to perform the selected test, and allows up to 4.2 seconds for the K
to complete its test. If the K completes the test within this time, the P.ioj/c displays an end-of-pass
message. If the K fails to complete within 4.2 seconds, the P.ioj/c displays a K time-out elTor (elTor
009).
The K microdiagnostics are designed to hang when an error is detected, so all failures in the
microdiagnostics are reported as time-out elTors. The current test may be aborted at any time by
typing CTRUC.
Mter the first test has been specified and completed, the following prompt is issued:
Reuse parameters (YIN) [YJ ?

Each time the K completes one full pass through the test specified, an end-of-pass report is
displayed. A full pass is defined as:
1. The K completes the test with no errors detected.
2. The K fails its test, and the P.ioj/c times out.
The end-of-pass message is displayed as follows:
End of Pass nnnnnn, xxxxxx Errors, yyyyyy Total Errors
where:
nnnnnn is the number of passes.
xxxxxx is the number of errors detected during the current pass.
yyyyyy is the number of errors detected for all passes.
6.6.6 K.ci Path Status Information

Whenever a Kci is enabled, it runs the CI link test as part of its microdiagnostics. The link test
performs loop-back tests on CI paths A and B of the Kci. To pass the link test, one of the paths
must work (one failing path is not a fatal error). The micro diagnostics then return information in
the K Control Area, which specifies which paths worked and how many retries were required. (The
test retries 64 times before declaring a failure.)
The off-line K test selector reports the CI path status each time the Kci is initialized. If the link
test is selected (K..ci test 11), the path status is reported only after the link test completes. (When
the Kci is enabled, it runs all of its microdiagnostics, including the link. test. If the link. test is
selected, the K.ci runs that test once more.)
The CI path status display indicates which path failed the link test, if any. If both paths fail, the
microdiagnostics fail in test 11, and no path status information is displayed. The status display
also includes the number of retries required for paths that passed the link test.

The following is a list of OKTS test summaries.
• Test 000, Moving Inversions Test-The moving inversions (MOVI) memory test is used by
the P.ioj/c to test a K Control Area. The K Control Area is used to pass memory test parameters
to the K and to return the results of memory tests to the P.ioj/c. The moving inversions RAM
test is used to detect data and addressing problems in dynamic semiconductor memories.
2. Read all locations in order from lowest to highest .. Mter reading a location and checking for
a 0, rewrite the same location with a single 1 in the least significant bit. Then reread the
location and verify the write worked correctly.
3. Again read all locations in order from lowest to highest. Check that each location contains
the data previously written. Rewrite the data found with a single additional 1 bit. Reread
it to verify the Write operation worked properly.
4. Repeat step 3 until the test pattern consists of a word containing all 1's (pattern 177777).
5. Repeat step 3, but this time substitute a single extra 0 each time, instead of a 1.
6. Continue step 5 until the test pattern consists of a word of all O's (pattern 000000).
7. Repeat steps 1 through 6, but this time start at the highest memory address and work
down to the lowest each time. This will work each memory location from all O's to all l's,
and back to all O's.
• Thst 001 through test 020, K Microdiagnostics-Refer to the following four lists for the
names of each microdiagnostic. Included in each list is the type of K being used and the failing
test number.
1. K.ci microdiagnostics-The following list shows the test number and name of each of the
K.ci microdiagnostics:
Test 1 - ALU test
Test 2 - Data Bus test
Test 3 - Control Bus test
Test 4 - PROM Parity test
Test 5 - Memory test (unavailable through K test selector)
Test 6 - RAM test
Test 7 - PLI Interface test
Test 10 - Packet Buffer test
Test 11 - Link test
2. K.sdi microdiagnostics-The following list shows the test number and name of each of the
K.sdi microdiagnostics:
Test 1 - ALU test
Test 6 - RAM test
Test 7 - SERDESIRSGEN test
Test 10 - Partial SDI Interface test
Test 11- Control Memory Access test
Test 12 - Lock test
3. K.sti microdiagnostics-The following list shows the test number and name of each of the
K.sti microdiagnostics:
Test 1 - ALU test
Test 6 - RAM test
Test 7 - SERDES test
Test 10 - Partial STI Interface test
Test 11- Control Memory Access test
Test 12 - Lock test
4. K.si microdiagnostics-The following list shows the test number and name of each of the
'~ K.si microdiagnostics:
Test 0 - 2911 test

Test 1 - ALU test
Test 2 - ROM ParitylTraps test
Test 3 - Scratchpad test
Test 5 - MOVI for System test
Test 7 - RTS Gate Array test
Test 10 - SIECL test
Test 11 - Frame test (U ~ L)
Test 12 - Frame test (L ~ U)
Test 13 - Sector test (U ~ L)
Test 14 - Sector test (L ~ U)
Test 15 - WCS LoadNerify test
Test 16 - WCS MOVI test
Test 17 - WCS EDC check
Test 20 - INIT Packet Search test

Error messages produced by this test conform to the HSC generic diagnostic error message format.
Off-line K selector test error messages are preceded by an OKTS> prompt.
A typical OKTS error message format follows:
OKTS>hh:mm T aaa E bbb 0-000

MA -xxxxxxxx
EXP-yyyyyy
ACT-zzzzzz
< K-Error-Surnmary-Info >
where:
mm is the elapsed minutes.
bbb is the decimal .number denoting the error detected.

Errors detected by OKTS fall into one of three classes:
1. Control memory errors occurring when the P.ioj/c is testing the portion of Control memory
used to communicate with the K. (The P.ioj/c does not test Data memory.) Error numbers 000
through 007 are all Control memory errors detected by the P.ioj/c. The difference between these
errors is the exact step in the memory test where they are detected. The step where an error
was detected can be a helpful clue to the cause of the error.
2. Failures in a K microdiagnostic detected by a time-out. Error 008 indicates the K failed to
initialize properly. Error 009 indicates the K failed the selected microdiagnostic.
3. Unexpected traps detected by the P.ioj/c (NXM: and Parity). Errors 010 and 011 are unexpected
trap errors detected by the P.ioj/c. Error 010 signifies a parity trap occUlTed, and error 011
indicates a nonexistent memory trap. The reports for unexpected trap errors differ slightly from
a data error report since they do not display EXPected and ACTual data. Error 012 indicates no
working Control memory could be found for a K Control Area. Error 13 is a cache parity trap.
The following list describes the nature of the failure indicated by each error number:
• Error OOO-Occurs in the moving inversions test when the P.ioj/c is testing the K Control Area
at a memory location that did not contain the expected pattern, where:
• MA is the address of the failing location.
• EXP is the data pattern EXPected.
• ACT is the data pattern ACTually found.
This error can be caused by a data error in the address specified, or it may indicate a dual-
addressing problem (the location was incorrectly addressed and written when some other
location was written). At this step in the test, a dual-addressing problem is characterized by:
1. The ACTual data contains a single additional 1.
2. The additional 1 bit occurs immediately to the left of the left-most 1 in the EXPected data.
For example:
EXP=000377, ACT=000777
EXP=077777, ACT=177777
EXP=OOOOOO, ACT=OOOOOl
For the first example, the location in error was probably written with the pattern 000777
when a lower numbered address was being written with the same pattern. When the
location in error was subsequently checked to ensure it still contained the previous pattern
(000377), it contained the next pattern (000777).
Data errors at this step of the test fall into one of the following classes:
a. The ACTual and EXPected data differ by more than one bit:
EXP=017777, ACT=017477
b. The ACTual data contains fewer 1's than the EXPected data:
EXP=003777, ACT=001777
c. The bit in error is not in the bit position immediately to the left of the left-most 1 in the
EXPected data:
EXP=000777, ACT=002777
• Error OOl-Occurs in the moving inversions test when the P.ioj/c is testing the K Control Area
at a location written with a pattern. Immediately after the write, the location was read and
found to contain an incorrect pattern, where:
This error indicates a memory data problem. One of the following hardware failures is
indicated:
1. A bit was picked up or dropped when the location was written.
2. A bit was picked up or dropped when the location was read.

If the error occurs repeatedly but only in a single location, the memory chip containing the
failing bit for that address is probably defective.
If the error occurs in many locations, but only occurs in a particular nibble (4-bit field), one of
the bus data transceivers for that nibble probably is defective.
If the error occurs in many locations and the bits in error are randomly spaced throughout the
word, the memory or bus timing is probably faulty.
If the error occurs in more than one location but the addresses of the failing locations are
similar, there could be crosstalk between the memory data and addressing lines. For instance,
all failing addresses end with either 2 or 6.
• Error 002-0ccurs in the moving inversions test when the P.ioj/c is testing the K Control Area.
A memory location did not contain the expected pattern, where:
addressing problem. (The location was incorrectly addressed and written when some other
location was being written.) At this step in the test, a dual-addressing problem is characterized
by:
1. The ACTual data contains one more 0 than the EXPected data.
2. The additional 0 occurs in the same bit position as the left-most bit in the EXPected data.
For example:
EXP=003777, ACT=001777
EXP=000017, ACT=000007
EXP=177777, ACT=077777
In the first example, the location in error was probably written with the pattern 001777 when
a lower numbered address was being written with the same pattern. When the location in
error was subsequently checked to ensure it still contained the previous pattern (003777), it
contained the next pattern (001777).
Data errors in this step of the moving inversions test fall into one of the following categories:
1. The ACTual and EXPected data differ by more than one bit:
EXP=177777, ACT=174777
2. The ACTual data contains more 1's than the EXPected data:
EXP=037777, ACT=077777
3. The bit in error is not in the same bit position as the left-most bit in the EXPected data:
EXP=001777, ACT=001377
A location was written with a pattern. Immediately after the write, the location was read and

indicated:
word, the memory or bus timing probably is faulty.
2. The additional 1 bit occurs immediately to the left of the left-most bit in the EXPected data.
For example:
EXP=000377, ACT=000777
EXP=077777, ACT=177777
a higher numbered address was being written with the same pattern. When the location in
contained the next pattern (000777). Data errors at this step of the test fall into one of the
following classes:
EXP=017777, ACT=017477
2. The ACTual data contains fewer 1's than the EXPected data:
EXP=003777, ACT=001777
3. The bit in error is not in the bit position immediately to the left of the left-most bit in the
EXPected data:
EXP=000777, ACT=002777
• Error 005--0ccurs in the moving inversions test when the P.ioj/c is testing the K Control Area.
indicated:
failing bit for that address is probably defective. If the error occurs in many locations, but
only occurs in a particular nibble (4-bit field), one of the bus data transceivers for that nibble
probably is defective.
location was being written). At this step in the test, a dual-addressing problem is characterized
by:
1. The ACTual data containing one more 0 than the EXPected data.
2. The additional 0 occurring in the same bit position as the left-most bit in the EXPected
data. For example:
EXP=003777, ACT=001777
EXP=000017, ACT=000007
EXP-177777, ACT-077777
contained the next pattern (001777). Data errors in this step of the moving inversions test fall
into one of the following categories:
1. The AC'fual and EXPected data differ by more than one bit:
EXP=177777, ACT-174777
EXP=037777, ACT=077777
EXP=001777, ACT=001377
indicated:
If the error occurs in many locations, and the bits in error are randomly spaced throughout the
similar, there could be crosstalk between the memory data and addressing lines. For example,
• Error OOS-Indicates the selected K did not complete its Init sequence properly. When the
P.ioj/c enables the K to perform a test, the K begins its Init sequence (which includes executing
certain microdiagnostics). At the end of the K's Init sequence, the K indicates it found the
K Control Area by complementing a pointer word in the Control memory. If the K fails to
complement this pointer word within 4.2 seconds of being enabled, error 008 is reported.
The contents of the K status register are displayed with the error report.
If this error occurs, make sure the requestor number parameter given matches the actual
requestor number of the K.
• Error 009-Indicates the K failed the selected microdiagnostic test. This usually indicates a
serious hardware problem in the K. The contents of the K status register are displayed with the
error report.
• Error OIO-Indicates the P.ioj/c detected a parity trap. The 22-bit address of the location that
caused the trap is displayed as the MA data in the error report, where:
• VPC is the virtual PC of the memory test at the time the trap occurred. Reference this
Because the data is lost when a parity trap occurs, no EXPected or ACTual data is displayed.
Mter the trap is reported, the program attempts to restart the test from the beginning.
• Error OIl-Indicates the P.ioj/c detected a nonexistent memory trap. A NXM error is caused
when no memory responds to a particular address. The MA data in the error report indicates
the address which produced the NXM trap. After reporting the trap, the program attempts to
restart the test from the beginning, where:
• MA is the address causing the NXM trap.
If this error occurs at a memory address that should be in your memory configuration, the
memory in question is not supplying an ACK to the P.ioj/c when the specified address is
presented on the memory bus. The most probable point of failure is the logic on the memory
module that compares addresses on the memory bus to the range of addresses the module is to
respond to. Also, the comparator itself could be faulty, or the [C IN, C OUT], [D IN, D OUT], or
[P IN, P OUT] lines on the backplane could be in error.
• Error OI2-Indicates no working Control memory could be found for a K Control Area. A K
Control Area is required to communicate with a K The Control memory must be repaired before
the K test selector can be used to test a K Use the off-line loader command TEST MEMORY to
test Control memory.
• Error OI3--Cache parity trap, vpe = xxxxxs This can happen during any test. The
JIl1Fll trapped through the parity vector. The error was caused by the cache.
During the run of the diagnostic, the JIl1Fll took a trap through the parity error vector. This
is a cache error and the virtual PC at the time of the trap is printed.
6.7 OKPM-Off-line KIP Memory Test

The off-line KIP memory test (OKPM) tests the HSC control and data memories from a K.sdi, K.sti,
K.si, or Kci. OKPM executes from the I/O control processor and uses the HSC K control area to
instruct one of the subsystem requestors to test either the control or data memories.
Select the K to be used, as well as the starting and ending addresses of the section of memory to be
tested. The test algorithm used by the K stresses the memories detecting transient errors caused
by bus and memory timing problems. Errors are reported at the console terminal as they occur.

Hardware required by OKPM includes:
• Memory module
• Load device and controller with at least one working drive
• Terminal connected to I/O Control Processor console interface
• At least one working Ksdi, K.sti, K.si, or K.ci
• Working Control memory for a K Control Area

If the HSC is not booted and loaded, refer to Section 6.1.2 and Section 6.2. If these preceding steps
are complete, the ODL> prompt is present. Follow these next steps to start OKPM.
1. Type TEST MEMORY BY K in response to the loader prompt ODL>. The load device LED
lights as the memory test is loaded.
2. OKPM indicates it has been loaded properly by displaying the following:

Hse OFL KIp Memory Test

OKPM can be terminated by typing CTRUC. The test may continue running for a few seconds after
it is terminated.

This section describes the various parameters for OKPM.
NOTE
For any of the OKPM prompts, use the DELete key to delete mistyped parameters before
terminating the entry with RETURN. If an error in a parameter entry was terminated
with RETURN, type CTRUC to return to the initial prompt and re-enter all parameters.
OKPM first prompts:
Requestor # of K (1 through 9) [] ?
Answer this question with the single digit (1 through 9) that specifies the requestor number to be
used. Terminate the response by typing RETURN. After the requestor number is supplied, a K
Control Area is located in Control memory and tested. This area is required for communicating
with the requestor that performs tests of Data and Control memory. The test then prompts:
Control (O) or data (1) memory [O]?
Type 0 to test Control memory or type 1 to test Data memory. Type RETURN to terminate the
response. (Typing just RETURN selects the Control memory test.) The memory test next prompts
for the first address to test.
First (min=XXXXXXXX) [min] ?
Enter the first address to be tested. Addresses are 8 octal digits in length. The [min] address
displayed is the lowest address that may be entered for the memory chosen. After typing the
address, terminate the response with RETURN. (Typing just RETURN causes the first address to
default to the [min] address.)
NOTE
Because requestors test Control memory in 4-byte units, the lowest 2 bits of the starting
address are ignored (treated as binary O's). For example, if address 16000223 is entered
as the first address, the requestor starts testing at address 16000200.
Because requestors test Data memory in 64-byte units, the lower 6 bits of the starting
address are ignored (treated as binary O's). For example, if address 14012376 is entered
as the first address, the K starts testing at address 14012300.
The test next prompts for the last address to test:
Last (max=XXXXXXXX) [] ?
Enter the last address to be tested. The max address displayed is the highest address still within
the memory chosen. If the system being worked on does not have a fully populated memory, the
last address that may be tested is less than the max address displayed. If a last address that
exceeds the amount of memory in this system is chosen, the memory test displays a Nonexistent
Memory (NXM) error when the test reaches the first address beyond the end of the memory. (Use
the off-line loader command SIZE to determine the actual last address in a given HSC.)
NOTE
Because requestors test Control memory in 4-byte units, the lower 2 bits of the ending
address are ignored (treated as binary l's). For instance, if address 16023400 is specified
as the last address, the K will test up to and including address 16023403.
Because requestors test Data memory in 64-byte units, the lower 6 bits of the ending
address are ignored (treated as binary l's). If address 14005400 is specified as the last
address, the requestor will test up to and including, address 14005477.
Finally, the memory test prompts:
t of passes to perform (D) [1] ?
Enter a decimal number between 1 and 2147483647 to specify the number of times the memory
test is to be repeated. (If 0 or just RETURN is entered, the test performs one pass.) The test can
be aborted at any time by typing CTRUC.
Mter the first memory test completes, the following prompt is issued:

Each time the requestor completes one full pass through the memory specified, an end-of-pass
report is displayed. A full pass is defined as:
1. A complete test of the memory specified with no errors detected.
2. Testing the memory speci...lied until an error occu..Y"S.
The end-of-pass message is displayed as follows:
where:
nnnnnn is the number of passes.
xxxxxx is the number of errors detected during the current pass.
yyyyyy is the number of errors detected for all passes.
6.7.6 Parity Errors

When a parity error occurs, it is desirable to know whether the error was produced by the loss or
gain of a data bit or by the loss or gain of a parity bit. When a parity trap occurs in the 110 Control
Processor, the data that was read is discarded by the PDP-II. However, a feature of the 110 Control
Processor allows parity traps to be disabled. Using this feature, a user can determine if a parity
error is being caused by a data or parity bit as follows:
1. Mter a parity trap (P.ioj/c detected) is reported, type CTRUC to terminate the memory test.
2. Type another CTRUC to return to the OFL diagnostic loader. The loader prompts ODL>.
3. Type EX 17770042 RETURN. The contents of the 110 Control Processor switch control and
status register (SWCSR) are displayed as (I) 17770042 nnnnnn.
4. Type De * nnnn4n. The nnnn4n represents the previous contents of the register, including a 1
in bit 5. I/O Control Processor parity traps are now disabled.
5. Return to the memory test by typing START.
6. Rerun the memory test with the original parameters.
If the location that previously produced a parity trap then produces a data error, the original
parity trap was caused by a data bit problem. The error report indicates the failing bit through
the EXPected and ACTual data displayed.
If the location that previously produced a parity trap does not fail again when the memory test
is rerun, the original parity trap was caused by an error in one of the parity bits (high or low
byte) for that word.
7. Type a CTRUC to return to the loader, and re-enable parity errors by typing De 17770042
nnnnOn. The nnnnOn represents original contents of the I/O Control Processor SWCSR, before
parity traps were disabled (refer to step 5).

The following is a summary of individual KIP memory tests.
• Test 000, moving inversions test from P.ioj/c-This is the moving inversions (MOV!)
memory test used by the I/O Control Processor to test a requestor control area. The K Control
Area is used to pass memory test parameters to the requestor and to return the results of
memory tests to the I/O Control Processor. The moving inversions RAM test is used to detect
data and addressing problems in dynamic semiconductor memories.
2. Read all locations in order from lowest to highest. Mter reading a location and checking for
a 0, rewrite the same location with a single 1 in the least significant bit. Then reread the
location and verify the Write worked correctly.
3. Again read all locations in order from lowest to highest. Check each location for the data
previously written. Rewrite the data found with a single additional 1 bit. Reread it to verify
the Write operation worked properly.
4. Repeat step 3 until the test pattern consists of a word containing all 1's (pattern 177777).
5. Repeat step 3, but this time substitute a single extra 0 each time instead of a 1.
6. Continue step 5 until the test pattern consists of a word of all O's (pattern 000000).
7. Repeat steps 1 through 6, but this time start at the highest memory address each time and
work down to the lowest. This changes each memory location from all O's to all 1's and back
to all O's.
• Test 001, moving inversions test from K-This is the moving inversions test implemented in
the K microcode. The algorithm is identical to that described in the previous test, except steps
5 and 6 are omitted to save time.
When the requestor detects an error, the remainder of the test is aborted, and the information
concerning the error is retum.ed to the 110 Control Processor through the K Control Area. The
I/O Control Processor is responsible for displaying the error report.

Errors produced by OKPM can be caused by a memory error detected either by the 110 control
processor or by the requestor being used to test memory. Errors detected by the 110 control
processor occur when the 110 control processor is testing the portion of control memory used to
communicate with the K. (The 110 control processor does not test data memory.)
A typical OKPM error message follows:
OKPM>hh:mm T aaa E bbb 0-000

MA -xxxxxxxx
EXP-yyyyyy
ACT-zzzzzz
< K-Error-Summary-Info >
where:
6.7.9 Requestor Error Summary

When the requestOr reports a memory test failure to the 110 Control Processor, the following
information is supplied: .
• Address of the failing memory location
• Data EXPected and data AC'fually found
• Error summary information
The error summary information is supplied as a 3-bit field, including the following:
• A bit indicating a parity error occurred while reading the location.
• A bit indicating an NXM error occurred while accessing the location.
• A bit indicating a Control bus (CBUS) error occurred while accessing the location.
When a memory error report is issued for an error detected by the K, the last line of the error
report includes a list of the error summary bits that were set, if any.
A control bus (CBUS) error indicates the requestor asserted an illegal combination of the three
CCYCLE lines when accessing Control memory. As these lines were previously tested from the 110
Control Processor (in the OFL P.ioj/c test), a Control bus error is most likely caused by a problem
with the requestor's drivers that assert the CCYCLE lines.

Error messages produced by OKPM can be caused by a memory error detected either by the 110
Control Processor or by the requestor being used to test memory. Errors detected by the 110 Control
Processor occur when the 110 Control Processor is testing the portion of Control memory used to
communicate with the K. (The 110 Control Processor does not test Data memory.)
To determine whether the 110 Control Processor or the requestor detected an error, examine the
second line of the error message. The text begins either with a (P) or a (K). If the text begins With
a (P), the 110 Control Processor detected the error. If the text begins with a (K), the requestor
detected the error.
Error numbers 000 through 007 are all Control memory errors detected by the 110 Control
Processor. The difference between these errors is the exact step in the memory test where they
are detected. The step where an error is detected can be a helpful clue to the cause of the error.
Error 008 indicates the requestor failed to initialize properly.
Error 009 indicates a Control or Data memory error detected by the K In addition to the normal
error information, the last line of the error report contains a K error summary.
Errors 010 and 011 are unexpected trap errors detected by the 110 Control Processor. Error 010
signifies a parity trap occurred. Error 011 indicates a Nonexistent Memory (NXM) trap. The
reports for unexpected trap errors differ slightly from a data error report because they do not
display EXPected and ACThaI data.
Error 012 indicates no working Control memory could be found for a K Control area. Error 013
indicates a parity trap caused by cache.
The following list describes the nature of the failure indicated by each error number.
• Error OOO-Occurs in the moving inversions test when the 110 Control Processor is testing the
K Control Area. A memory location did not contain the expected pattern, where:
2. The additional 1 bit occurs immediately to the left of the left-most bit in the EXPected data,
such as:
EXP=000377, ACT=000777
EXP=077777, ACT=177777
EXP=017777, ACT=017477
2. The ACTual data contains fewer l's than the EXPected data:
EXP=003777, ACT=001777
EXPected data:
EXP=000777, ACT=002777
• Error OOl-Occurs in the moving inversions test when the 110 Control Processor is testing
the K Control Area. A location was written with a pattern. Immediately after the write, the
location was read. It contained an incorrect pattern, where:
indicated:
failing bit for that address probably is defective.
word, the memory or bus timing is probably the problem.
If the error occurs in. more than one location, but the addresses of the failing locations are
similar, crosstalk between the memory data and addressing lines may be present. For example,
• Error 002-0ccurs in the moving inversions test when the 110 Control Processor is testing the
location was being written.)
At this step in the test, a dual-addressing problem is characterized by:
2. The additional 0 occurs in the same bit position as the left-most bit in the EXPected data,
such as:
EXP=003777, ACT=001777
EXP=000017, ACT=000007
EXP=177777, ACT=077777
error was subsequently checked to 'ensure it still contained the previous pattern (003777), it
EXP=177777, ACT=174777
EXP=037777, ACT=077777
EXP=0017777, ACT=00377
• Error 003-0ccurs in the moving inversions test when the 110 Control Processor is testing
indicated:
similar, crosstalk between the memory data and addressing lines could be present. For example,
• Error 004--0ccurs in the moving inversions test when the 110 Control Processor is testing the
2. The additional 1 bit occurs immediately to the left of the left-most hit in the EXPected data,
such as:
EXP=000377, ACT=000777
EXP=077777, ACT=177777
EXP=017777, ACT=017477
EXP=003777, ACT=001777
3. The bit in error is not in the hit position immediately to the left of the left-most bit in the
EXPected data:
EXP=000777, ACT=002777
• Error 005-0ccurs in the moving inversions test when the I/O Control Processor is testing
indicated:
• Error 006-0ccurs in the moving inversions test when the I/O Control Processor is testing the
This error can be caused by a data error in the address specified or it may indicate a dual-
2. The additional 0 occurs in the same bit position as the left-most bit in the EXPected data,
such as:
EXP=003777, ACT=001777
EXP=000017, ACT=000007
EXP=177777, ACT=077777
EXP=177777, ACT=174777
EXP=037777, ACT=077777
EXP=001777, ACT=001377
• Error 007-0ccurs in the moving inversions test when the 110 Control Processor is testing
indicated:
similar, crosstalk between the memory data and addressing lines may be present. For example,
• Error OOS-Indicates the selected requestor did not complete its lnit sequence properly. When
the I/O Control Processor enables the requestor to perform the memory test, the requestor
begins its Init sequence (which includes executing certain microdiagnostics). At the end of the
requestor's Init sequence, the requestor indicates it found the K Control Area by complementing
a pointer word in Control memory. If the requestor fails to complement this pointer word within
50 milliseconds (4.2 seconds for K.ci) of being enabled, error 008 is reported.
The contents of the K status register are displayed with the error report. If this error occurs,
make sure the requestor number parameter given matches the actual requestor number.
• Error 009-Indicates a Control or Data memory error detected by the K, where:
• MA is the 22-bit address of the failing location.
• EXP is the data pattern EXPected by the K.
• ACT is the data pattern found by the K
In addition to the address and the EXPected/ACTual data, the K returns an error summary,
displayed as the last line of the error report. The error summary information indicates whether
the error was caused by a parity error, a Nonexistent Memory (NXM) error, or a Control bus
(CBUS) elTor. If the error was not caused by any of the these, the error summary line does not
appear in the elTor report. Refer to Section 6.7.9 for further information on the elTor summary.
• Error OlO-Indicates the I/O Control Processor detected a parity trap. The 22-bit address of
the location that caused the trap is displayed as the MA data in the error report, where:
Because the data is lost when a parity trap occurs, no EXPected or ACTual data can be
displayed. To further localize the problem, disable parity errors and rerun the test as described
in Section 6.7.6. If the original failure was in a data bit position, the memory test detects and
reports the error, displaying the EXPected and ACTual data. This helps to trace the error to a
particular address and/or bit position. If no further errors are detected after disabling parity
errors, the original failure was in one of the parity bits for the address displayed in the parity
trap report.
• Error OIl-Indicates the I/O Control Processor detected a Nonexistent Memory (NXlM) trap. A
NXM error is caused when no memory responds to a particular address. The MA data in the
error report indicates the address that produced the NXM trap. Mter the trap is reported, the
program attempts to restart the test from the beginning, where:
• MA is the address causing the NXM trap.
address in the listing to locate the area of the test where the elTor occurred.
If this error occun at a memory address that should be in the memory configuration, the
memory in question is not supplying an ACK message to the I/O Control Processor when the
specified address is presented on the Memory bus. The most probable point of failure is the
compare logic on the memory module. This logic compares addresses on the Memory bus with
the range of addresses to which the module is to respond. The comparator itself could be faulty
or the [C IN, C OUT], [D IN, D OUT], or [P IN, P OUT] lines on the backplane could be in error.
• Error Ol2-Indicates no working Control memory could be found for a K Control Area. A K
Control Area is required to communicate with a requestor. Control memory must be repaired
before the KIP memory test can be used. Use the off-line loader command TEST MEMORY to
test the Control memory.
• Error 013, Cache Parity Trap, vpe = *1**** Indicates the Jl1 took a trap through the
parity elTor vector during the run of the diagnostic. This is a cache error; the virtual PC at the
time of the trap is printed.
6.8 OMEM-Off-line Memory Test

The off-line memory test (OMEM) exercises the HSC memories. Control, Data, or Program memory
may be selected for testing. Three memory testing algorithms are used: the quick verify algorithm,
the moving inversions algorithm, and the walking 1's algorithm.
The quick verify algorithm quickly uncovers stuck data and address bits. The other two algorithms
stress the memories, attempting to detect transient errors caused by bus and memory timing
problems.
Errors are reported at the console terminal as they occur. After reporting a data error, or a parity
error from a location being tested, testing continues where it left off. If an NXM error occurs during
the memory test, testing is restarted from the beginning.

Following are the OMEM hardware requirements:
• Memory module
• Load device with at least one working drive

If the HSC is not booted and loaded, refer to Section 6.1.2 and Section 6.2. If the HSe is booted and
loaded, the terminal displays an ODL> prompt. At this point, follow these steps to start OMEM:
1. 1)rpe SIZE in response to the loader prompt ODL>.
The load device drive-in-use LED lights as the off-line system sizer is loaded. The sizer displays
the bounds of the various memories in the HSC. The memory size information includes the last
address of each memory.
2. 1)rpe TEST MEMORY in response to the loader prompt ODL>.
The load device drive-in-use LED lights as OMEM is loaded. OMEM test indicates it has been
loaded properly by displaying the following:
HSC OFL Memory Test

OMEM can be terminated by typing CTRUC.

This section describes the OMEM parameter entry.
NOTE
For any of the OMEM prompts, use the DELete key to delete mistyped parameters before
the typing RETURN. If an error in a parameter already terminated with RETURN is
noted, type CTRI1C to return to the initial prompt and re-enter all parameters.
The following are the parameters that can be modified:
Control(O), Data(l) , or Program(2) Memory [0] ?
Type 0 to test control memory.

Type 1 to test data memory.
Type 2 to test program memory.
The memory test next prompts for the first address to test.
First (min=XXXXXXXX) [min] ?
Enter the first address to be tested. Addresses are 8 octal digits long. The default is the lowest
address that may be entered for the memory chosen
The test next prompts for the last address to test.
Last (max=XXXXXXXX) [] ?
Type the last address to be tested. The max address displayed is the highest address in the
memory chosen. Use the memory size information displayed by the ODL SIZE command to answer
this prompt with the correct address for the HSC under test.
If an address exceeds the memory in the system, the memory test displays a nonexistent memory
(NXM) error when the test reaches the first address beyond the end of the memory.
The test then prompts:
* of passes to perform (D) [1] ?
times the memory test should be repeated. (Entering 0 results in one pass.)
Mter the first memory test is complete, the following prompt is issued:

A complete pass through OMEM consists of one pass through the quick verify test, one pass through
the moving inversions test, and one pass through the walking 1's test. 'Mter each complete pass, an
end-of-pass message is displayed as follows:
where:
nnnnnn is a decimal total of the complete passes made.
xxxxxx is the number of errors detected on the current pass.
yyyyyy is number of errors detected during the passes completed so far.
NOTE
A complete pass through the memory test for program memory may take about 8 hours.
Unless exhaustive memory testing is required, allow this test to run only until the quick
verify pass complete message is displayed. This takes no more than 10 minutes.
6.8.6 Parity Errors

When a parity error occurs, it is desirable to know whether the error was produced by the loss or
gain of a data bit or by the loss or gain of a parity bit. When a parity trap occurs in the 110 Control
Processor, the data that was read is discarded by the PDP-II. However, a feature of the 110 Control
Processor allows parity traps to be disabled. Using this feature, a user can determine if a parity
error is being caused by a data or parity bit as follows:
1. Mter a parity trap (P.ioj/c detected) is reported, type CTRLlC to terminate the memory test.
2. Type another CTRLlC to return to the OFL diagnostic loader. The loader prompts ODL>.
3. Type Ex 17770042. The contents of the 110 Control Processor switch control and status register
(SWeSR) are displayed as (I) 17770042 nnnnnn.
4. Type De * nnnn4n. The nnnn4n represents the previous contents of the register, including a 1
in bit 5. I/O Control Processor parity traps are now disabled.
5. Return to the memory test by typing START.
6. Rerun the memory test with the original parameters.
If the location that previously produced a parity trap then produces a data error, the original
parity trap was caused by a data bit problem. The error report indicates the failing bit through
the EXPected and ACTual data displayed.
If the location that previously produced a parity trap does not fail again when the memory test
is rerun, the original parity trap was caused by an error in one of the parity bits (high or low
byte) for that word.
7. Type a CTRUC to return to the loader, and re-enable parity errors by typing De 17770042
nnnnOn. The nnnnOn represents original contents of the I/O Control Processor SWCSR, before
parity traps were disabled (refer to step 5).

The following list describes the three algorithms used by OMEM.
• 'lest 000, quick verify test-Quickly detects stuck bits and dual-addressing problems. The
algorithm used by the quick verify test is as follows:
Write 000000 to each location of the memory
FOR i = First to Last address
IF < location i does not contain a >
THEN < display error >
Write test pattern to location i (146314(8»
IF < location i does not contain pattern >
Write complement of pattern to location i (031463(8»
IF < location i does not contain complement >
NEXT i
• 'lest 001, moving inversions test-Detects data and addressing problems in dynamic
semiconductor memories.
The moving inversions algorithm performs the following:
1. Writes 000000 in each location of the memory.
2. Reads all locations in order from lowest to highest. Mter reading a location and checking
for a 0, rewrites the same location with a single 1 in the least significant bit. Then rereads
the location and verifies the Write worked correctly.
3. Again reads all locations in order from lowest to highest. Checks that each location contains
the data previously written. Rewrites the data found with a single additional 1 bit. Rereads
it to verify the Write operation worked properly.
4. Repeats step 3 until the test pattern consists of a word containing all l's (pattern 177777).
5. Repeats step 3 but this time substitutes a single extra 0 each time instead of a 1.
6. Continues step 5 until the test pattern consists of a word of all O's (pattern 000000).
7. Repeats steps 1 through 6 but this time starts at the highest memory address each time
and works down to the lowest. This writes each memory location from all O's to alII's and
back to all O's.
8. Clears all memory to 000000.
• 'lest 002, walking-l's test-An algorithm that stresses semiconductor memories and is
effective in locating timing problems on the memory module or on the bus.
The walking I's algorithm performs the following:
1. Writes all memory to O's (pattern = 000000).
2. Checks all memory for O's. Declares error 008 if not o.
3. Sets TESTADDRESS equal to the first address to test.
4. Writes 177777 to contents of TESTADDRESS.
5. Checks that all other locations are equal to 000000. Declares an error 009 if not equal to
000000.
6. Checks that TESTADDRESS contains 177777. Declares an error 010 if not equal to 177777.
7. Writes 000000 to contents of TESTADDRESS.
8. IF TESTADDRESS is the last address to be tested, testing is complete. If TESTADDRESS
is not the last address to be tested, 2 will be added to TESTADDRESS and the process will
go back to step 4. This will continue until TESTADDRESS is the last address to be tested.

OMEM displays the errors detected during execution on the console terminal. All error messages
follow the diagnostics generic error message format preceded by an OMEM> prompt.
A typical OMEM error message format follows:
OMEM>hh:mrn T aaa E bbb U-OOO

MA-xxxxxxxx
EXP-yyyyyy
ACT-zzzzzz
where:
Parity trap and NXM trap errors do not include expected and actual data.

Error messages produced by OMEM can be classed as either data errors or unexpected traps. Error
numbers 000 through 010 are all memory data errors. The only difference between these errors
is the exact step in the testing algorithm where they are detected. The step at which a data error
occurs can be an important clue to the cause of the error. Errors 000 through 007 are declared in
the moving inversions algorithm; errors 008 through 010 are declared in the walking 1's algorithm..
Errors 011 and 012 are unexpected trap errors. Error 011 signifies a parity trap occurred and error
012 indicates a nonexistent memory trap. The reports for unexpected trap errors differ slightly
from a data error report because they do not display EXPected and ACTual data.
• Error OOO-Occurs in the moving inversions test (Section 6.8.7). A memory location did not
contain the expected pattern, where:
addressing problem. In the second case, the location was incorrectly addressed and written
when some other location was written. At this step in the test, a dual-addressing problem is
characterized by:
2. The additional 1 bit occurs immediately to the left of the left-most bit in the EXPected data,
such as:
EXP=000377, ACT=000777
EXP=077777, ACT=177777
EXP=017777, ACT=017477
EXP=003777, ACT=001777
EXPected data:
EXP=000777, ACT=002777
• Error OOl-Occurs in the moving inversions test (Section 6.8.7) when the 110 Control Processor
was testing the K Control Area. A location was written with a pattern. Immediately after the
write, the location was read. It contained an incorrect pattern, where:
• EXP is the data pr:ttern EXPected.
indicated:
If the error occurs repeatedly, but only in a single location, the memory chip containing the
word, the memory or bus timing probably is the problem.
similar and crosstalk could exist between the memory data and addressing lines. For example,
• Error 002-0ccurs in the moving inversions test (Section 6.8.7). A memory location did not
contain the expected pattern, where:
For example:
6-58 Off-iine Diagnostics
EXP=003777, ACT=001777
EXP=000017, ACT=000007
EXP=177777, ACT=077777
1. The AC'fua1 and EXPected data differ by more than one bit:
EXP=177777, ACT=174777
EXP=037777, ACT=077777
EXP=001777, ACT=001377
• Error OOS-Occurs in the moving inversions test (Section 6.8.7). A location was written with a
pattern. Immediately after the write, the location was read and found to contain an incorrect
pattern, where:
This error indicates a memory data problem and one of the following hardware failures is
indica.ted:
similar, crosstalk could be present between the memory data and addressing lines. For example,
• Error 004-0ccurs in the moving inversions test (Section 6.8.7) when the 110 control processor
is testing the K Control Area. A memory location did not contain the expected pattern, where:
addressing problem. In the latter case, the location was incorrectly addressed and written when
some other location was written.

1. The ACTual data containing a single additional 1.
2. The additional 1 bit occurring immediately to the left of the left-most bit in the EXPected
data. For instance:
EXP=000377, ACT=000777
EXP=077777, ACT=177777
1. The ACThaI and EXPected data differ by more than one bit:
EXP=017777, ACT=017477
2. The ACThaI data contains fewer 1's than the EXPected data:
EXP=003777, ACT=001777
EXPected data:
EXP=000777, ACT=002777
• Error 005--0ccurs in the moving inversions test (Section 6.8.7) when the 110 control processor
is testing the K Control Area. A location was written with a pattern. Immediately after the
write, the location was read and it contained an incorrect pattern, where:
indicated:
If the error occurs in more than one location but the addresses of the failing locations are.
• Error 006-0ccurs in the moving inversions test (Section 6.8.7) when the 110 control processor
is testing the K Control Area. A memory location did not contain the expected pattern, where:

For example:
EXP=003777, ACT=001777
EXP=000017, ACT=000007
EXP=177777, ACT=077777
EXP=177777, ACT=174777
EXP=037777, ACT=077777
EXP=001777, ACT=001377
• Error 007-0ccurs in the moving inversions test (Section 6.8.7) when the I/O control processor
is testing the K Control Area. A location was written with a pattern. Immediately after the
write, the location was read and found to contain an incorrect pattern, where:
indicated:
similar, crosstalk may be present between the memory data and addressing lines. For example,
• Error 008-0ccurs in the walking 1's -test (Section 6.8.7). All locations in the memory under
test were written with the pattern 000000. Then all locations were read to check that they
contained 000000. When the location specified in the error report was read, it did not contain
000000, where:
• EXP is the data pattern EXPected (000000).
Because all locations were cleared to 000000 before this error was detected, a dual-addressing
problem is unlikely. More likely, a bit was picked up when the word was written or read.
If the error occurs repeatedly but only in one location, the memory chip containing the bit in
error for that address is probably marginal.
If the error occurs in many locations, but always occurs in a particular nibble (4-bit field), one
of the bus data transceivers for that nibble probably is marginal.
If errors occur in many locations and the bits in error are randomly spaced throughout the
words, the memory or bus timing is probably marginal.
• Error 009-0ccurs in the walking 1's test (Section 6.8.7). One location in the memory under
test was written with the pattern 177777 and all the other locations should contain the
pattern 000000. While reading to check that all other locations are clear, a location was found
containing something·other than 000000, where:
This error is either a data error or a dual-addressing error. (The location was incorrectly
addressed and written when some other location was being written.)
At this step of the test a dual-addressing failure is possible if the ACTual data is 177777.
During this part of the test, one location in the memory was written to 177777. When this
write was performed, the failing location may also have been addressed and written with
the same data. When the test was checking that all other locations were clear, it found the
second location with the pattern 177777. If this is a true dual-addressing problem, the error is
repeated on each pass of the test.
At this step of the test, a data error is probable if the ACTual data is not 177777. Some clues to.
the possible causes of a data error follow.
If the error occurs repeatedly but only in a particular bit in a single location, the memory chip
that contains the failing bit for that location is defective.
If errors occur in many locations, but only occur in a particular nibble (4-bit field), one of the
bus data transceivers for that nibble probably is marginal.
• Error 010-0ccurs in the walking 1's test (Section 6.8.7). At this step of the test, one location
in the memory under test was set to the pattern 177777 and all other locations were cleared to
000000. Mter checking that all other locations contain 000000, the location that should contain
177777 was read. It contained some other pattern, where:

Because only Read operations were performed after writing the 177777, a dual-addressing
problem is highly improbable.
If the error occurs repeatedly but only in a particular bit of a single location, the memory chip
that holds that bit for the failing location is defective.
If errors occur in many locations, but only occur in a particular nibble (4-bit field), one of the
bus data transceivers for that nibble probably is marginal.
If errors occur in more than one location but the addresses of the failing locations are similar,
crosstalk may be present between the memory data and addressing lines. For example, all
failing addresses end in 2 or 4.
• Error Oil-Indicates a parity trap occurred. The parity trap probably occurred in a location
under test but may have been caused by Program memory where the memory test itself resides.
The MA data in the error report indicates the address of the location causing the parity trap.
Mter reporting the parity trap, the memory test continues if the parity error occurred in a
memory location under test, where:
• MA is the address of the location causing the parity trap.
Because the data is lost when a parity trap occurs, no EXPected or ACTual data is displayed.
To further localize the problem, disable parity errors and rerun the test. (Refer to Section 6.7.6.)
If the original failure was in a data hit position, the memory test detects and reports the error,
displaying the EXPected and ACTual data. This helps trace the error to a particular address
and/or bit position. If no further errors are detected after disabling parity errors, the original
failure was in one of the parity bits for the address displayed in the parity trap report.
• Error OI2-Indicates a nonexistent memory (NXM) trap occurred. An NXM error is caused
when no memory responds to a particular address. The MA data in the error report identifies
the address that produced the NXM trap. Mter reporting the error, the program attempts to
restart testing from the beginning, where:
• MA is the address being tested at the time the NXM: trap occurred.
• VPC is the PC of the memory test at the time the trap occurred. Reference this address in
the listing to locate the area of the test where the error occurred.
This error frequently occurs when trying to test beyond system memory addresses.
If this error occurs at a memory address that should be within your memory configuration, the
memory in question is not supplying an ACK to the 110 Control Processor when the specified
address is presented on the memory bus. The most probable point of failure is the logic on the
memory module that compares addresses on the Memory bus with the range of addresses to
which the module is to respond. The comparator itself could be faulty or the [C IN, C OUT], [D
IN, D OUT], or [P IN, P OUT] lines on the backplane could be in error.
• Error OI3-0ccurs in the quick verify test. This error may indicate a dual-addressing problem.
The quick verify test consists of clearing the entire memory, then writing two patterns to each
location and checking that the writes worked properly. Before writing the first pattern to each
location, the contents of the location is O. Error 013 indicates a location contain something
besides a 0 before the first pattern was written.
If the ACTual data in the error report is 031463(8) or 146314(8), a dual-addressing problem
probably is the cause of the error. (When an address lower in memory was written with a test
pattern, the failing location also was written with the same pattern.) Dual-addressing problems
are normally caused by shorts between memory address bits.
If the ACTual data is other than 031463(8) or 146314(8), the problem probably is caused by
a memory bit or bits stuck in the 1 state. The first pattern written is 146314(8). The second
pattern written is the 1's complement of the first pattern, 031463(8).
• Error 014-0ccurs in the quick verify test. The MA in the error report shows the failing
address. The ACTual data shows the bit or bits that failed.
• Error 015-0ccurs when an NXM trap occurs as the memory under test is initially being
cleared. The last address to test (operator-supplied) exceeds the amount of memory actually
installed in the HSC or part of the memory under test is not responding. If the NXM: occurs at
an address that should respond, use CTRUC or CTRUY to return to the off-line loader. Use the
loader's REPEAT EXAMINE (address that caused trap) to set up a scope loop for isolating the
problem.
• Error 016-Cache Parity Trap, VPC = xxxxxx Indicates the J11 took a trap through the
parity error vector during the run of the diagnostic, and the error was determined tQ be from
the cache. The virtual PC at the time of the trap is printed.
6.9 OFLRXE-RX33 Off-line Exerciser

OFLRXE is a combined hardware diagnostic and exerciser for the HSC M.std2lRX33 subsystem.
Diagnosis of the DMA. hardware and diskette controller are provided, as well as a read/write
exerciser to provide exercise for the actual drive portion of the subsystem.
OFLRXE is a standalone diagnostic running under the off-line diagnostic loader. This loader
provides terminal 110 service, time keeping, string conversions, and interrupt handling.
OFLRXE is an 8-Kword program of which approximately half is control code and half is mapped for
data buffer transfers.

To run OFLRXE, the HSC must be booted from the off-line diagnostic diskette. When the system is
booted from this media, the ODL> prompt is displayed. Hardware and software requirements are:
• P.ioj module
• M.std2 memory/controller module
• At least one RX33 drive
• One scratch diskette for each drive to be tested (maximum of two)
• Testing of the J11 chip set and cache is assumed if it is turned on
• Two tested 4-Kword partitions of memory

If the HSC is not booted and loaded, refer to Section 6.1.2 and Section 6.2. If the HSC is already
booted and displaying the off-line loader prompt ODL>, proceed as follows:
At the ODL> prompt, invoke OFLRXE by typing TEST RX. This loads the OFLRXE from the media,
and transfers control to the diagnostic. At the start, the diagnostic prints out the following string:
Hse Off-line RX33 Exerciser Vxxx

Where: Vxxx is a 3-digit version/edit number.
NOTE
H unable to boot from drive 0, move the diskette to drive 1, try again, or use a backup
copy of the off-line diagnostics diskette.

OFLRXE can be terminated by typing CTRUC or CTRLIY. OFLRXE also terminates on expiration
of the allotted time or on fatal errors.

Following are the OFLRXE modifiable parameters. Drive selection is prompted for by the program
in the following manner:
Test drive n (Y/N) [YJ ?
Where: n is the drive number (0 or 1)
The default is Y. The prompt repeats for each available diskette on the HSC. The test prompts if
the initial Write operation is to be performed:
Perform initial write on this drive (YIN) [Y] ?
The default is Y. This lays down a background pattern on the entire disk in preparation for the
random read/write exerciser. Selecting this option adds 10 minutes of test time per drive.
As soon as the previous prompts have been answered, the program directs placement of a scratch
diskette in the selected drive:
Insert a scratch diskette in the drive, type a carriage return to continue.
At this point, insert the scratch diskette. The random read/write exercise takes place over the
entire surface of the diskette, so be sure the diskette is a scratch one only to be used for the
exercise. Run time of the exerciser is user-selectable and is prompted for by the program as follows:
* of minutes to exercise (D) [30] ?
Enter a number between 1 and 32767. The default is 30 minutes. This 30 minutes starts after the
initial patterning of the disk (if selected) so the total test time with two drives and initial patterning
is amount of time selected plus 20 minutes. A value of 1440 minutes gives a 24-hour run time for
burn-in purposes. The 30-minute default is sufficient for installation use and repair verification.
At the end of the amount of time allotted for the exerciser, the program prompts you by printing:

OFLRXE does not run in a conventional sense. There are no pass-completed messages. Instead,
informational messages are printed indicating what the exerciser is doing. The program has a
user-requested status available. If you type CTRLfr, the program responds:
Number of sectors transferred = xxxxxx, yyyyyy errors.
where:
xxxxxx is a 16-digit number of sectors successfully transferred.
yyyyyy is a 6-digit cumulative number of errors detected.
At the end of the initial write test (if selected), the exerciser prints:
Initial write completed on drive OOOn
Where: n is the drive number (0 or 1).
When the exerciser begins the random read/write phase of the testing, the following message is
printed:
Beginning random exerciser
The random exerciser is now in progress. It runs for the amount of time requested by you. When
the requested time has expired, the program prints the following string:
Exerciser completed.
The program then returns to the parameter entry routine.

The following is a summary of OFLRXE tests.
• Test 1, RX33 controller registers-Performs stuck-at testing on the RX33 controller registers
at 17777400, 17777402, 17777404, and 17777406. A simple walking 1's test is performed on
each register, except for the CSR register at 177400, which only has the high byte tested.
• Test 2, interrupt hardware--Exercises the interrupt hardware on the M.std2. The interrupts
generated are also tested for the correct priority when they occur.
• Test 3, DMA logic and counters-Checks out all of the DMA handshake signals, the data
path, and the address path. A special DMA test mode in the controller is used to perform
one read or write to/from each memory location loaded in the DMA address registers. Correct
incrementing action from the counters is checked. The ACTual data loaded to memory on a
DMA write is checked as well.
• Test 4, parity logic-Also uses DMA test mode in addition to the force bad parity function (bit
11 of the CSR) to prove parity errors can be detected, and correct parity is written to memory
by the DMA control logic. NXM action also is lumped into this test. Correct handling of NXM
errors and correct reporting by the error bit in the CSR is checked.
• Test 5, verify track counters and registers-Uses the step function of the diskette controller
chip to verify that all cases of the track counter bits internal to the diskette controller chip work
as advertised. Step functions are performed for each power of two in the diskette track register
(step four times, step eight times more, and so forth). The verify option is set on each step
command so the diskette controller reads headers on each track to verify position.
• Test 6, oscillating seek test-Performs an oscillating seek test using the algorithm:
oscillating seek test
begin
incnt = 0
outcnt = 124
while incnt<> outcnt do
begin
seek outcnt;
CHECK STATUS;
If outcnt <> rxtrk then error 11
outcnt =outcnt-1;
seek incnt;
CHECK STATUS;
if Incnt <> rxtrk then error 11
incnt =incnt + 1;
end;
end { oscillating seek test. }
In this manner, all seeks are performed in both directions with all seek counts between <0:77>.
Verification is performed on each track to check the step logic.
• '.lest 7, sequential read/write test-Performs the basic patterning of the diskette with a
background pattern. This test is user-selected. If selected, this test writes each LBN on the
RX33 diskette in ascending order with a unique pattern consisting of the track, sector, and
side of that LBN, and then an incrementing-byte pattern for the remainder of the 512-byte
sector. Each LBN so written is then read back, and each word is compared to the data that was
written. This test takes about 10 minutes per drive.
• '.lest 8, random reads/writes-Does random Reads and Writes to the selected drives. If both
drives are selected for test, operations on each drive are performed in groups of five.
This test runs until the allotted time for the exercise expires, or the user terminates the test
with CTRUC. The mechanism of this test is as follows:
A random number is generated. The value of this number determines if the operation is a Read
or a Write, and which LBN is used.
If the command is a READ, the appropriate LBN is read from the disk. The header bytes (0:5)
of the data read are then compared against the values expected. The pattern number bytes
(6:7) are then compared against a list to see which pattern is to be used to compare the rest of
the buffer (10:512).
If the command is a WRITE, other bits of the random number are used to select one of four
different patterns to write to the disk. A buffer is then set up with the correct header bytes for
the LBN to be written and the correct background data pattern. This buffer is then written out
on the diskette.
Descriptions of the data patterns used are found in the following section.
6.9.7 Data Patterns

Four unique data patterns were selected to give maximum delta of frequency with the modified
frequency modulation (MFM) encoding used on the RX33. These patterns are shown in
Example 6-2.
PATTERN NUMBER I PATTERN VALUE

----------------+---------------------------------------
177400 I Incrementing by bytes starting at 2404
11111 I 1000101110001011 binary, 105613 octal
22222 I 0011001100110011 binary, 031463 octal
33333 I 0011000010010001 binary, 030221 octal
44444 I 0000101110001011 binary, 005613 octal
----------------+---------------------------------------
Example 6-2 Off-line RX33 Exerciser Data Patterns

A generic message format for all off-line diagnostic errors is found in Section 6.2.7. The following
section contains information on specific errors associated with OFLRXE.
A typical OFLRXE error message follows:
OFLRXE>S2:22 T 008 E 010 D 001
SEEK error detected during positioning operation
LBN = 004356
Track = 000114
Sector =000007
Surface = 00000
Soft errors, such as seek errors, can build up to a. point where a diagnostic defines them as fatal
and terminates on a fatal error. The internal bias for soft errors is currently set to 20. When this
number is exceeded, the exerciser determines the errors are fatal and terminates.

The following is a list of errors associated with test failures.
• Error 00, Parity Trap, VPC = * * * * * * Applicable to all tests. Occurs at any time during
execution of the diagnostic. The virtual PC on the stack is printed to help identify the program
area where the error occurred. Both the content of the error address register and the virtual
PC are displayed as optional lines. This error terminates the test. The diagnostic retmns to
the reuse parameters prompt.
• =
Error 01, NXM Trap, vpe * *i * ** Applicable to all tests. Causes the diagnostic to return
to the reuse parameters prompt. Additional data, such as the virtual PC of the instruction
which caused the trap, and the physical address contained in the error address register are
printed as optional lines.
• Error 02, Bit Stuck in Register-Applicable to test 1. Indicates a stuck-at fault is present in
one of the RX33 control registers. The register address and the EXPected and ACTual data are
printed as optional lines in the error message. If the error is in the low byte, the problem is the
diskette controller chip. If the error is in the high byte, the problem is with the MAR register at
that address. If more than one register shows the same bit(s) in error, the problem is probably
in the bus transceivers.
• Error 03, Interrupt Occurred Without Enable Set-Applicable to test 2. Indicates there
is a stuck-at fault in the register, .or the etch going into the DC003 interrupt control chip. The
interrupt enable bit, <13> of the CSR, does not disable interrupts.
• Error 04, RX33 Interrupt Occurred at Wrong Priority-Applicable to test 2. Indicates the
RX33 interrupt OCCUlTed with the priority at five or greater. The virtual PC where the interrupt
occurred is printed out as an optional line. Using the listing of the program, the priority at the
time of the interrupt can be determined.
• Error 05, Unexpected Interrupt from RX33-Applicable to all tests. Indicates an

unexpected interrupt. An interrupt that occurs at any time when a command to the RX33
is not in progress is defined as unexpected. The virtual PC where the interrupt occurred is
printed as an optional line.
• Error 06, Track 0 Did Not Set after RECALIBRATE Command-Applicable to test
5. Indicates the track 0 status bit (bit 2 of the CSR) did not set upon completion of a
RECALIBRATE command. The drive may not be sending the signal or the cable to the drive
may be faulty.
• Error 07, RX33 Did Not Interrupt as Expected-Applicable to test 2. Indicates an expected
interrupt never occurred. The interrupt control chip (De003) may be at fault, or the diskette
controller chip interrupt signal is stuck at 1. The Jl1 may be unable to recognize interrupts
from the diskette controller or the backplane etches carrying interrupt control signals are open.
• Error 10, Seek Error Detected during Positioning Operation-Applicable to tests 5, 6, 7,

and 8. Indicates a seek error status (bit 4 of the CSR) was set after a SEEK or RECALIBRATE
command. The problem may be in the diskette controller chip or the diskette. If the errors
are occurring mostly in test 5 starting with track 0, the problem probably is fundamental; the
controller cannot read the diskette at all. If the errors occur in a random fashion, the problem
probably is the diskette.
• Error 11, Current Track Register Incorrect-Applicable to tests 5 and 6. Indicates the
values in the track register of the diskette controller chip are not as expected after a given
operation. This problem probably is in the diskette controller chip.
• Error 12, CRC Error in Header Detected during Position verify-Applicable to tests 5,
6, 7, and 8. Detects a CRe error when reading a header during a position verify. This error
occurs when a valid header has been found and read, but the CRC at the end is incorrect.
This probably is the diskette. If the controller is able to detect the address and data marks
that precede a header (so that it knows that a header is being read), the data separation logic
probably is working.
• Error 13, Processor Type Is Not JII-Applicable to test O. Does not contain the value which
defines a Jll. This error causes the diagnostic to terminate.
• Error 14, Drive Under Test Is Not Ready-Applicable to tests 5,6,7, and 8. Indicates the
diskette drive is sending NOT READY status to the controller. The door may open on the drive,
or no diskette is inserted. If these conditions are not the cause of the fault, the ready signal
from the drive may be stuck.
• Error 15, Last Command Did Not Complete-Applicable to tests 5, 6, 7, and 8. Indicates
the last command issued to the diskette controller never interrupted to show completion. This
error points to the diskette chip since it occurs after the interrupt logic has already been tested.
• Error 16, RX33 Header Does Not Compare-Applicable to tests 7 and 8. The header
information written in the data area of a sector is not what it should be for that sector and side,
written as part of the data in that sector. This error happens when an undetected positioning
error has occurred, either during the read or the write of the sector involved. The LBN, track,
sector, and side are displayed as optional lines.
• Error 17, Record Not Found during Read (Could Also Say Write)-Applicable to tests
7 and 8. Indicates the controller was unable to find that sector on the current track when
attempting to read or write a given sector. Either a misposition occurred, or that sector is
unreadable. Because this error occurs after basic read capability has been tested, the most
probable culprit is the diskette, with the diskette chip being the next most probable problem
point. The LBN, track, sector, and side are displayed as optional lines.
• Error 20, CRC Error in Data During Read (Could Also Say Write}-Applicable to tests
7 and 8. Indicates the controller detected a CRC error when reading the desired sector. If the
error occurs multiple times in a row for a given sector, the problem is most likely the diskette
(or the drive it is installed in). Single errors when an LBN has this error only once are soft
errors. The LBN, track, sector, and side information is printed as optional lines.
• Error 21, Lost Data Detected During Read (Could Also Say Write}-Applicable to tests
7 and 8. Indicates the DMA logic did not service an I/O request of the diskette controller chip
in time. There are probably problems in the DMA logic, or stuck-at faults exist in the etch
between the controller chip and the DMA logic.
• Error 23, Invalid Pattern Code in Buffer-Applicable to test 8. Indicates the data word,
defined as the pattern code, read from the diskette does not match any of the possible patterns
used. It is unlikely the data was read incorrectly from the diskette and not detected as a CRC
error. Usually this error occurs when a diskette is not written with the initial data pattern.
The LBN, track, sector, and side are displayed as optional lines.
• Error 24, Drive Is Write-Protected-Applicable to tests 7 and 8. Indicates the drive is
sending write protect status. Either the interface is bad, or the drive is in error (assuming
there is not a write-protected diskette in the drive). This error terminates the diagnostic, as a
write-protected diskette cannot be written on.
• Error 25, CRC Error in Header during Read (Could Also Say Write)-Applicable to tests
7 and 8. Indicates the controller detected bad eRC in the header it was reading as part of a
data transfer command. This probably is a diskette error. The LBN, track, sector, and side are
displayed as optional lines.
• Error 26, Data Incorrect after DMA TEST MODE Command-Applicable to tests 3 and
4. Indicates the memory content after a DMA test mode command was not correct. There
are either stuck-at faults in the DMA registers, or the transfer did not happen at all (that is,
the memory is unchanged). This is a fundamental error in the diskette logic; the diagnostic
terminates after detecting it.
• Error 27, Data Compare Error-Applicable to tests 7 and 8. Indicates a manual check
of data read by the diskette turned up an error. Either the transfer did not complete, an
intermittent error occurred in the data or address path, or what was written on the disk was
written incorrectly. The LBN, track, sector, and side are displayed as optional lines.
• Error 30, RX33 Detected Parity Error during Read (Could Also Say Write)-Applicable
to tests 7 and 8. Indicates the RX.33 detected a parity error when doing a DMA read from
memory. Either Program memory is bad or the parity logic on the controller is in error.
• Error 31, RX33 Detected NXM during Read (Could Also Say Write)-Applicable to tests 7
and 8. Indicates the RX33 detected a NXM during a DMA operation. Either the DMA address
was loaded wrong and pointed to a nonexistent location, or the handshake logic on the M.std2
board is in error.
• Error 32, RX33 MAR Value Incorrect after DMA 'rransfer-Applicable to test 3. Indicates
the value of the MAR address counters was in error after a DMA test operation. The problem is
probably in the counters or the etch associated with them. The EXPected and AC'fua1 data are
printed out as optional lines.
• Error 33, Parity Error Was Not Forced in Main Memory-Applicable to test 4. Indicates
a write to Program memory with bad parity set (bit 11 of the CSR) did not result in bad parity
in memory. There is either a stuck-at fault in the parity logic or the operation never wrote
memory in the first place.
• Error 34, Parity Error Did Not Set in CSR-Applicable to test 4. Indicates a DMA read of
a location with known bad parity did not set the parity error bit (bit 15 of the CSR). Either the
data was never read or there is a stuck-at fault in the parity logic.
• Error 35, NXM Did Not Set in CSR-Applicable to test 4. Indicates a DMA read of a location
expected to give a NXM did not set NX]M in the CSR. Look for stuck-at faults in the NX]M
detection logic.
• Error 36, Parity Error Set Along with NXM in CSR-Applicable to test 4. Indicates both
the parity error and the NXM error set simultaneously in the CSR. On a NXM error, the parity
error should not set. Check for stuck-at faults in the NXM/parity error logic.
• Error 37, Cache Parity Error, VPC = * * ***" Applicable to all tests. Indicates the J11
took a trap through the parity error vector, a cache error during the run of the diagnostic. The
virtual PC at the time of the trap is printed.
6.10 ORFT-Off-line Refresh Test

The off-line memory refresh test (ORFT) finds memory problems related to refresh. Patterns are
written to memory and then checked after waiting 1 minute. Three separate patterns are used to
test each memory bit (including parity bits) in both the 1 and 0 states. All three HSC memories
are tested (Program, Control, and Data), although only the Program and Control memories require
refreshing. Tests of Data memory are included because some static RAM failures resemble refresh
problems.
ORFT can find problems in the memories not detected by the normal memory tests. ORFT is not
intended to be run on memories that fail the normal memory tests.

The following hardware is required to run ORFT:
• I/O Control Processor module with HSC boot ROMs
• Memory module that passes the off-line memory test and/or the off-line KIP memory test
• HSC load device with at least one working drive
ORFT assumes the HSC memories pass both the off-line memory test and the off-line KIP memory
test. In addition, ORFT assumes the memories are working except for the refresh circuitry.

If the HSC is not booted and loaded, refer to Section 6.1.2, and Section 6.2. If the HSC is already
1. 1YPe TEST REFRESH in response to the prompt ODL>.
2. ORFr indicates it is loaded properly by displaying the following:
HSC OFL Memory Refresh Test
3. The refresh test now prompts for parameters.

ORFT can be terminated by typing CTRUC.

This section describes the prompts for the ORFr parameters.
NOTE
For any of the ORFl' promptst use the DELete key to delete mistyped parameters before
typing RETURN. If an error in a parameter already terminated with RETURN is noted,
type CTRUC to return to the initial prompt and re-enter all parameters.
ORFT first prompts with:
# of passes to perform (D) [1] ?
times the refresh test is to be repeated. (Entering a 0 or just a carriage return results in one pass.)
After selection of the number of passes the test begins. The test can be aborted at any time by
typing CTRUC. Each pass of the test requires three minutes to complete.
After the refresh test completes, the following prompt is issued:

Each time the refresh test completes one full pass, an end-of-pass report is displayed. Each pass of
the test requires three minutes to complete. The end-of-pass message is displayed as follows:
where:
nnnnnn is a decimal total of the complete passes made.
xxxxxx is the number of errors detected on the current pass.
yyyyyy is number of errors detected during the passes completed so far.

The following are the ORFT test summaries.
• Test 01t pattern 177777-Fills the memories with the pattern 177777. This sets all data bits
and also sets the upper and lower byte parity bits. The entire Control and Data memories are
filled with the pattern. All of Program memory not occupied by the refresh test and the off-line
loader is also filled with the pattern. After filling the memories, the program delays for one
minute, then each memory location is read and checked for the pattern. Any errors detected are
reported on the terminal.
• Test 02 t pattern OOOOOO-Fills the memories with the pattern 000000. This clears all data
bits and sets the upper and lower byte parity bits. The entire Control and Data memories are
filled with the pattern. All of Program memory not occupied by the refresh test and the off-line
loader is also filled with the pattern. After filling the memories, the program delays for one
minute, then each memory location is read and checked for the pattern. Any errors detected are
reported on the terminal.
• Test 03, pattern 100001-Fills the memories with the pattern 100001. This sets data hits 0
and 15 and clears data bits 1 through 14. Both parity bits are also cleared. The entire Control
and Data memories are filled with the pattern. All of Program memory not occupied by the
refresh test and the off-line loader is also filled with the pattern. After filling the memories, the
program delays for 1 minute, then each memory location is read and checked for the pattern.
Any errors detected are reported on the terminal.

All error messages produced by ORIT conform to the HSC diagnostic error message format (refer
to Section 6.2.7). Following is a typical ORFT error message.
ORFT>hh:mm T aaa E bbb 0-000

MA-xxxxxxxx
EXP-yyyyyy
ACT-zzzzzz
where:

• Error Ol-Indicates the test detected a parity error when reading the pattern from the
indicated location. The EXPected and ACTual data are included in the error report. This
error indicates a data bit or parity bit was not refreshed (assuming the memory in question
passed the off-line memory test). If the EXPected and ACTual data are the same, one of the
parity bits was not refreshed..
• Error 02--Indicates the test detected a data compare error when reading the pattern from the
indicated location. The EXPected. and ACTual data are displayed in the error report. Note: a
parity error did not occur so more than 1 bit must have failed to refresh.
• Error 03-Indicates the I/O Control Processor detected a parity error. The 22-bit address of
the location that caused the trap is displayed as the MA data in the error report, where:
• VPC is the virtual PC of the memory test at the time the trap occurred.. Reference this
Because the data is lost when a parity trap occurs, no EXPected. or ACTual data can be
displayed.. The parity error occurred within the program itself, not within the memory being
tested. After the trap is reported., the program attempts to restart the test from the beginning.
• Error 04-Indicates the I/O Control Processor detected a NXM: trap. An NXM: error is caused
when no memory responds to a particular address. The MA data in the error report indicates
the address that produced the NXM trap. After the trap is reported, the program attempts to
restart the test from the beginning. (The MA and VPC fields have the same meanings as those
in error 03.)
If this error is at a memory address that should be in the memory configuration, the memory
in question is not supplying an ACK to the I/O Control Processor when the specified address is
presented on the Memory bus. The most probable point of failure is the logic on the memory
module that compares addresses on the Memory bus with the range of addresses to which the
module is to respond. The comparator itself could be faulty or the [C IN, C OUT], [D IN, D
OUT], or [P IN, P OUT] lines on the backplane could be in error.
• Error 05, Cache Parity Trap, vep = x x x x x x: Indicates the J11 took a trap through the
parity error vector during the run of the diagnostic. This is a cache error. The virtual PC at the
time of the trap is printed.
6.11 OOCP-Off-line Operator Control Panel (OCP) Test

OOCP checks the operation of the HSC lamps and switches. Testing includes the five OCP lamps
and switches, the State LED, SecurelEnable switch, and the enable LED.
This section includes troubleshooting procedures for localizing faults detected by this test.

The following hardware is required to run OOCP:
• 1/0 Control Processor module with HSC boot ROMs
• Memory module
• HSC load device with at least 1 working drive
• OCP
Due to the sequence of tests that precede this test, it is safe to assume the I/O Control Processor
module, Program memory, and HSC load device are tested and working.

If the HSC is not booted and loaded, refer to Section 6.1.2, and Section 6.2. If the HSC is already
Type TEST OCP in response to the ODL> prompt. The HSC load device in motion LED is ON.
The test indicates it is loaded properly by displaying the following message:
HSC OFL OCP Test

The test may be aborted at any time by typing CTRUC.

OOCP first checks the position of the SecurelEnable switch through a bit in the I/O Control
Processor control and status register (address 17770040). If the switch is in the secure position, the
following prompt is issued. Otherwise, OOCP skips to the next prompt.
Put Secure/Enable switch into enable position
If the SecurelEnable switch is in the enable position and the above prompt is issued anyway, a
problem is indicated with the bit in the 110 Control Processor CSR that monitors the SecurelEnable
switch. Refer to the troubleshooting procedures in Section 6.11.8. The program waits until the
SecurelEnable switch is changed to the enable position and issues the following message:
(Enable LED is lit, State LED is blinking)
Check to verify the enable LED is lit and the OCP State LED is blinking. There are two State
LEDs: one is to the left of the lnit switch on the HSC OCP, and the other is located on the I/O
Control Processor module (the fourth LED from the bottom of the rightmost module in the HSC
card cage). If either LED is not blinking, refer to the troubleshooting procedures in Section 6.11.8.
The test next prompts for a lamp test.
Press Fault (all OCP lamps should light) (Y/N) [YJ ?
Press the fault lamp and observe that all OCP lamps light. If none of the lamps light, a problem
may be present in the lamp test logic on the OCP assembly. If all lamps light properly, type a
carriage return to continue the test. If the lamp test fails, replace the OCP.
Next, the program checks that all OCP switches are OFF (out position). If any switch bits in the
110 Control Processor switch/display register read as l's (ON), the program lights the lamps for
those switches and prompts:
Put all lit switches in OFF (out) position (Y/N) [YJ ?
If the fault or lnit lamps are lit (nonlocking switches), a problem exists with the wiring in those
switches or with their respective bits in the switch/display register. Replace the OCP.
Otherwise, press all lit switches to release their locks and type a carriage return. If the message
repeats and one or more lamps remain lit even though the switches are OFF (out position), refer to
the troubleshooting procedures in Section 6.11.8.
The program then tests each of the OCP switches, one at a time. A switch lights and the following
prompt is displayed:
Press and release the lit switch
Press the switch that is lit. The program allows about 1 second for the switch to be released after
it is pressed and then continues to the next prompt. If the program fails to respond when a switch
is pressed, refer to the troubleshooting procedures in Section 6.11.8. For those switches that lock in
the ON position (online switch and the two unmarked switches), the program prompts:
Press and release the lit switch again
Press the switch again to return it to the OFF (out) position. If the online switch or either of the
unmarked switches fails to lock in the ON position, the switch is defective, and the OCP should be
replaced.
After the OCP switch tests are complete, several features of the SecurelEnable switch are tested.
The program begins these tests by prompting:
Put Secure/Enable switch into secure position
The program waits until the SecurelEnable switch is in the proper position before continuing.
If the program fails to respond when the switch is moved to the secure position, refer to the
troubleshooting procedures in Section 6.11.8. When the program detects the switch is in the secure
position, it prompts with:
(Enable LED should turn off)
Ensure the enable LED is off. If this LED fails to turn off when the switch is in the secure position,
a short or wiring problem is probable.
Next, the program prompts:
Press Init (HSC should not re-boot) (Y/N) [YJ ?
Press the lnit switch. When the SecurelEnable switch is in the secure position, pressing the lnit
switch has no effect. (Do not press any other switch or an error message results.) If the HSC starts
to perform a bootstrap (Init lamp turns on and green LED on I/O Control Processor turns off), the
SecurelEnable switch is not disabling the action of the lnit switch. After pressing the lnit switch,
type RETURN to continue. The test responds with the following prompt:
Press terminal break key (HSC should not halt) (Y/N) [Y] ?
Press the break key as directed. When in secure mode, the break key does not cause the JI11Fll
processor to halt (enter ODT). If the terminal displays the @ character when break is pressed, the
SecurelEnable switch is not disabling the action of the break key. Refer to the troubleshooting
procedures in Section 6.11.8. After pressing the break key, type RETURN to continue the test. The
final prompt of the test is:
Put Secure/Enable switch into enable position.
The test waits until the SecurelEnable switch is returned to the enable position. At that point the
test terminates and returns to the off-line loader.
If an error in a parameter was terminated with RETURN, type CTRIlC to return to the initial

The following sections summarize test 000 through test 009.
• Test 000, observe enable and State LEDs-Performed by the operator, because the program
cannot tell whether the enable or State LEDs are lit. If the enable LED is off, a wiring problem
may be the cause (LED not connected to power/ground source) or the LED itself may be faulty.
If the State LED on the OCP fails to blink, check the State LED on the 110 Control Processor
module (fourth LED from the bottom of the nghtmost module in the HSC card cage). If neither
State LED is blinking, the problem probably is caused by the bit in the I/O Control Processor
CSR register that controls the State LED (refer to Section 6.11.8.4). If one of the State LEDs is
blinking but the other is not, the nonblinking LED probably is wired wrong or is faulty.
• Test 001, lamp test through Fault switch-Performs an automatic lamp test. When the
Fault switch is pressed, all lamps light and remain lit until the switch is released.
If none of the lamps light when the Fault switch is pressed, the problem is probably in the
lamp test circuitry on the OCP assembly. It is possible all lamps are defective or they are not
installed. Replace the OCP.
If some lamps light when fault is pressed but others do not, replace the OCP.
• Test 002, check all switches OFF-Reads the 110 Control Processor switch/display register
to see if any of the switch bits read as ON (switch bit is a 1). If the bit for any switch reads as
ON, the corresponding lamp is lit and the program prompts to turn off any switch that is lit.
The program will not proceed until all switch bits read as OFF.
If a lamp remains ON, even though the corresponding switch is OFF (out position), the switch
is either wired incorrectly or the bit in the I/O Control Processor switch/display register for that
switch is faulty. Refer to Section 6.11.8.1 to localize the problem.
• Test 003 t Fault switch-Directs pressing the lit Fault switch. The program then monitors the
switch bits in the I/O Control Processor switch/display register and waits for the Fault switch
bit to set. If any other switch bit sets, an error is reported and the program terminates.
If pressing the Fault switch has no effect, one of the following could be the cause:
Fault switch is broken.
Fault switch is not properly wired.
Fault switch bit in the 110 Control Processor CSR cannot be set.
Refer to the troubleshooting procedures in Section 6.11.8.
If pressing the Fault switch results in an error message, refer to Section 6.11.7.
• Test 004t Online switch-Directs pressing the lit Online switch. The program then monitors
the switch bits in the I/O Control Processor switch/display register and waits for the Online
switch bit to set. If any other switch bit sets, an error is reported and the program is
terminated.
If pressing the Online switch has no effect, one of the following could be the cause:
Online switch is broken.
Online switch is not properly wired.
Online switch bit in the I/O Control Processor CSR cannot be set.
If pressing the Online switch results in an error message, refer to Section 6.11.7.
• Test 005t first unmarked switch-Directs pressing the lit first unmarked switch. The
program then monitors the switch bits in the I/O Control Processor switch/display register and
waits for the first unmarked switch bit to set. If any other switch bit sets, an error is reported
and the program is terminated.
If pressing the first unmarked switch has no effect, one of the following could be the cause:
First unmarked switch is broken.
First unmarked switch is not properly wired.
First unmarked switch bit in the I/O Control Processor CSR cannot be set.
If pressing the first unmarked switch results in an error message, refer to Section 6.11.7.
• Test 006t second unmarked switch-Directs pressing the lit second unmarked switch. The
program then monitors the switch bits in the I/O Control Processor switch/display register
and waits for the second unmarked switch bit to set. If any other switch bit sets, an error is
reported and the program terminates.
If pressing the second unmarked switch has no effect, one of the following could be the cause:
Second unmarked switch is broken.
Second unmarked switch is not properly wired.
Second unmarked switch bit in the I/O Control Processor CSR cannot be set.
If pressing the second unmarked switch results in an error message, refer to Section 6.11.7.
• Test 007, enable LED off-Begins with a prompt to put the SecurelEnable switch into the
secure position. The program waits until bit 15 of the 110 Control Processor control and status
register reads as a 0, indicating the switch is in the secure position. Then the program tells the
operator to observe that the enable LED 'is OFF.
If the enable LED fails to turn off when the switch is in the secure position, replace the OCP.
• Test 008,!nit switch in secure mode-Checks that the Init switch has no effect when the
SecurelEnable switch is in the secure position. The test prompts for the Init switch to be
pressed while the program monitors the switch bits in the 110 Control Processor switch/display
register. Monitoring ensures that pressing the Init switch does not cause any switch bits to set.
If pressing the Init switch causes the HSC to reboot, the secure position of the SecurelEnable
switch is not disabling the Init switch. Replace the OCP.
If pressing the Init switch causes one of the switch bits in the switch/display register to set, an
error message is displayed. Refer to Section 6.11. 7 for further information.
• Test 009, break key in secure mode-Checks if the terminal break key has no effect when
the SecurelEnable switch is in the secure position. (Normally the break key causes the 110
Control Processor JIl1Fll CPU to halt and enter ODT.) The prompt is to press the break key
and to observe if the HSC does not halt.
If pressing the break key causes the terminal to print an @ symbol, the secure position of the
SecurelEnable switch is not disabling break from halting the JI11Fll CPU.

All error messages produced by OOCP conform to the HSC diagnostic error message format (refer
to Section 6.2.7). A typical OOCP error message follows:
OOCP>hh:mm T aaa E bbb 0-000

MA -xxxxxxxx
EXP-yyyyyy
ACT-zzzzzz
where:

• Error 000, wrong bit setr-Occurs when the test detects a switch bit other than the switch bit
being tested set in the I/O Control Processor switch/display register. This error can be caused
by:
The operator pressing the wrong switch.
A short causing an additional switch bit to set along with the expected bit.
A wiring error causing the wrong bit to set when a switch is pressed.
The media address (MA) field of the error report gives the address of the 110 Control Processor
switch/display register. The EXPected and ACTual data in the error report show the switch bit
the program expected to find set and the bit or bits that actually were set.
If the EXPected and ACTual data each consist of only one bit, the failure was caused by either
the operator pressing the wrong switch or by a wiring error. If the ACTual data consists of two
or more set bits, a short between switches is likely. Refer to the troubleshooting procedures in
Section 6.11.8.
• Error 001, bit set when Init is pressed - Occurs when the Init switch is pressed while the
HSC is in the secure mode (test 008). This error can be caused by one of the following:
Pressing some switch other than the lnit switch.
Pressing the lnit switch, causing a switch bit in the 110 Control Processor switch/display
register to set.
The media address (MA) field of the error report gives the address of the 110 Control Processor
switch display register. The EXPected data is always 0 (no bit is expected to set). The ACTual
data shows the bit or bits that read as a 1 when the lnit switch was pressed. Refer to the
troubleshooting procedures in Section 6.11.8.
6.11.8 Troubleshooting Registers and Displays through CDT

The following paragraphs and layouts are included to assist you with troubleshooting.
6.11.8.1 Switch Check through ODT

'lb check the operation of an HSC switch, follow this procedure:
1. With the SecurelEnable switch in the enable position, press the tenninal break key. The 110
Control Processor J111F11 CPU halts and displays an @ symbol.
2. Type: 177700421.
The contents of address 17770042 (the 110 Control Processor switch display register) are
displayed in octal. Refer to the layout of the switch display register in Figure 6-1 to locate the
switch bits.
Each bit is in the 1 state when the associated switch is ON (pressed in).
3. Type a carriage return.
4. Type a slash (I) to re-examine the switch display register.
5. To restart the off-line loader (or the diagnostic that was interrupted), type RETURN, then type
p.
ADDRESS 17770042 VIA ODT
. - - - - - - - - 4000(8) FAULT SWITCH
, . . . - - - - - 2000(8) ON-LINE SWITCH
.....----1000(8) FIRST UNMARKED SWITCH
.----400(8) SECOND UNMARKED SWITCH
,It , ,
'_____-....y,._-_____J I~ I~
(UNUSED)
200(8) GREEN LED - - _.....
100(8) CHEM/DMEM NXM-----'
40(8) INH PARITY TRAP--------I
20(8) INIT LAMP - - - - - - - -......
10(8) FAULT LAMP - - - - - - - - -......
4(8) ON-LINE L A M P - - - - - - - - - - - - '
2(8) FIRST UNMARKED L A M P - - - - - - - -......
1(8) SECOND UNMARKED LAMP----------.....I

CXO-1119A
Figure 6-1 P.loj Switch Display Register Layout

Using this method, the switch bits of the switch/display register can be monitored when various
switches are in the ON or OFF position.
6.11.8.2 Lamp Bit Check

To check the operation of the lamp control bits in the 110 Control Processor switch/display register,
use the following method:
1. With the SecurelEnable switch in the enable position, press the terminal break key.
The 110 Control Processor J111F11 CPU halts and displays an @ symbol.
2. Type 17770042 RETURN.
The contents of the switch/display register are displayed in octal.
3. Use Figure 6-1 to locate the bits controlling the OCP lamps.
When a lamp bit is set, the corresponding lamp lights.
4. To light a lamp, type the octal value that corresponds to the proper lamp, then type RETURN.
The lamp lights.
5. Type a slash (I) to re-examine the contents of the switch/display register.
6. Type RETURN to restart the off-line loader (or the diagnostic that was interrupted), then type
aP.
Using this method, various lamps can be manually enabled or disabled.
6.11.8.3 Secure/Enable Switch Check

To manually check the operation of the secure/enable bit in the 110 Control Processor control and
status register, use the following procedure. Using this method, the secure/enable bit in the 110
Control Processor CSR can be checked with the SecurelEnable switch in both positions.
1. With the SecurelEnable switch in the enable position, press the terminal break key. (If the HSC
is stuck in the secure mode, this method cannot be used because break is disabled.)
2. The 110 Control Processor J111F11 CPU halts and displays an @ symbol.
3. Type 17770040.
4. The content of the I/O Control Processor control and status register is displayed in octal.
Figure 6-2 identifies the various bits of this register.
When the SecurelEnable switch is in the enable position, the contents of the register is 1xxxxx.
When in the secure position, the contents is Oxxxxx.
5. Type RETURN and then a slash (/) to re-examine the register.
6. Type RETURN, then type P to restart the off-line loader (or the diagnostic that was
interrupted).
ADDRESS 17770040 VIA ODT
, . . . . - - - - - - - - - - - - - - 1 0 0 0 0 0 ( 8 ) 0 WHEN SECURE
, . . . . - - - - - - - - - - - - 40000(8) ALWAYS 0
,....-----------20000(8) ALWAYS 0
~--------10000(8) ALWAYS 0
...-------4000(8) SWAP BOARD
...------2000(8) SWAP BANK
...----1000(8) ALWAYS 0
r----- 400(8) SELECT BT PG2
, It , ,
115114113112111110J 9 I 8 I 7 I 6 I 5 1 4 I 3 I 2 11 J 0 J
h 'I' " 'I'
200(8) ENA CMEM ARB - - - -....
100(8) ALWAYS 0 - - - - - - - - - '
40(8) HI BYTE PARITY TEST - - - - - - - '
20(8) LO BYTE PARITY TEST - - - - - - - - '
10(8) STATE LED - - - - - - - - - - - - -
4(8) NON-MEMORY-ACCESS (NMA) - - - - - - - - - '
2(8) CONTROL MEMORY INTERRUPT ENABLE - - - - -.....
1(8) CONTROL MEMORY LOCK CYCLE ENABLE - - - - - - - - '

CXO-1120A
Figure 6-2 P.ioj Control and Status Register Layout
6.11.8.4 State LED Check

There are two State LEDs in the HSC. One is on the OCP, far left. The other State LED is on the
110 Control Processor module (rightmost module in the HSC card cage, fourth LED from the bottom
of the module). Both LEDs are controlled by a bit in the I/O Control Processor control and status
register (refer to Figure 6-2 for a layout of this register). To.manually control the State LED, use
the following procedure:
1. With the SecurelEnable switch in the enable position, press the terminal break key.
The 110 Control Processor J111F11 CPU halts and displays an @ symbol.
2. Type 17770040/.
The contents of the control and status register are then displayed in octal.
3. Use Figure 6-2 to find the octal value corresponding to the State-LED.
4. To light the State LED, type the octal value corresponding to the State LED, followed by
RETURN. To extinguish the State LED, put a 0 in the same bit position and press RETURN.
CAUTION
Bit 7 of the 110 Control Processor CSR must he set to allow the HSC Ks to access
Control memory. The setting of other hits in the CSR can result in side effects. Be
careful not to set any hits except the State LED hit and leave hit 7 set when done.
5. 1YPe a slash (I) to re-examine the contents of the I/O Control Processor CSR.
6. To restart the off-line loader (or the diagnostic that was interrupted), type RETURN, then type
p.
Utilities 7-1
7
Utilities
7.1 Introduction
This chapter contains the information required to run the following off-line utilities:
• DKUTIL - Off-line Disk Utility
• VERIFY - Off-line Disk Verify Utility
• FORMAT - Off-line Disk Formatter Utility
• PATCH - Off-line Load Media Modification Utility
The HSC must be in the command mode before running the off-line utilities. Type CTRLIY to get
the command prompt.
Topics covered in this chapter include initiating the utility, using commands, and interpreting error
messages. These HSC utilities are interactive and therefore are prompt-oriented. Note that prompt
information displayed in square brackets is the default.
Refer to the HSC User Guide for information on other HSC utilities that are not documented in this
manual.
7.2 DKUTIL - Off-line Disk Utility

DKUTlL is a general utility for displaying disk structures and disk data. Unlike some other
utilities, DKUTIL is a command language interpreter. It is intended for use in debugging utilities,
diagnostics, error recovery, and bad block replacement. DKUTIL has become a general utility for
displaying disk structure and data.
Initially, the program goes into command mode. The user must then issue a GET command to
obtain the unit to which other commands are to be applied. DKUTIL then returns to the command
mode, prompting for a command, executing it, and prompting for another command. DKUTIL is
terminated by CTRIJC, CTRIJY, CTRLIZ, or the EXIT command.
7.2.1 Starting DKUTIL

DKUTlL is started with the standard CRONIC command syntax RUN DKUTIL.UTL. It
immediately enters the command mode. A drive must be acquired and brought on line before
any other commands can be executed. Type GET Dnnn to acquire the drive and bring it on line.
DKOTIL> GBT Dnnn
The format for entering the drive is a D followed by the unit number. If the drive parameter is
omitted, DKUTIL defaults to DOOO (unit 0).
7-1
7-2 Utilities
The first block of the Format Control Table (FCT) is read, if possible, and dumped in a format
similar to a VERIFY printout. The unit is brought on line with the ignore media format error
modifier so drives improperly or not completely formatted can be examined. If the FCT cannot be
read or the mode is invalid, the program prompts for the sector size.
DKUTIL-Q Enter sector size (512/576) [512]?
The program places the unit in diagnostic mode to access the DBN area. The program retmns to
the command mode and prompts for a command.
DKUTIL>
Comment lines can be entered by prefixing them with an exclamation point (D. A null line is
ignored. Entering CTRUZ terminates the program. Commands are executed immediately and take
only the time necessary to print the results. Entering CTRUY or CTRUC at any time aborts the
program and releases the drive.
7.2.2 Command Syntax

The DKUTIL commands are:
• DEFAULT
• DISPLAY
• DUMP
• EXIT
• GET
• POP
• PUSH
• REVECTOR
• SET
Any initial substring recognizes commands, command options, and modifiers. For example, DUMP
can be entered as DUM, DU~ or D. In cases where the initial substring can indicate one of several
commands, the match depends on an order based on history and expected frequency of usage. Thus,
D specifies DUMp, DI specifies DISPLAY, and DE specifies DEFAULT. In the descriptions explained
in this chapter, only the command (or part of the command) in bold print must be specified.
Some command options take optional parameters which, if omitted, default.
7.2.3 Command Modifiers

Modifiers, specified only for commands that allow them, can occur anywhere after the command
itself. They are preceded by a slash (one slash for each modifier). The following are equivalent:
DUMP /NOEDC RBN 0
DUMP RBN/NOEDC 0
DUMP RBN O/NOEDC
DUMP RBN 0 /NOEDC
Modifiers are processed left to right and applied to the current default modifiers. The DUMP
command is the exception. The default modifiers for DUMP can be changed through the DEFAULT
command. The initial default modifiers for DUMP are !DATA, IEDe, and IIFERROR.
Utilities 7-3
7.2.4 Command Descriptions

This section contains the DKUTIL command descriptions. Command options are shown by separate
lines in the syntax specification. Parameters are indicated lowercase in the syntax by braces ({}).
Options indicated by brackets ([]) can be omitted.
7.2.4.1 Command Summary

A summary of all DKUTIL commands follows:
• DEFAULT-Change default modifiers for DUMP command.
DEFAULT
• DISPLAY-Display characteristics, error history, RCT, or FCT.
DISPLAY ALL
DISPLAY CHARACTERISTICS DBN {block}
DISPLAY CHARACTERISTICS DISK
DISPLAY CHARACTERISTICS LBN {block}
DISPLAY CHARACTERISTICS PBN {block}
DISPLAY CHARACTERISTICS RBN {block}
DISPLAY CHARACTERISTICS XBN {block}
DISPLAY ERRORS
DISPLAYFCT
DISPLAYRCT
• DUMP-Dump given block or table of blocks.
DUMP [BUFFER]
DUMP DBN [{block}]
DUMP FCT [BLOCK {number}] [COpy {copy}]
DUMP LBN [{block}]
DUMP RBN [{block}]
DUMP RCT [BLOCK {number}] [COpy {copy}]
DUMP XBN [{block}]
• EXIT-Terminate execution of the program.
EXIT
• GET-Acquire or change the current drive.
GET [{drive}]
• POP-Restore save buffer to current buffer.
POP
• PUSH-Save current buffer in save buffer.
PUSH
• REVECTOR-Force bad block replacement for the given LBN(s).
REVECTOR {block} [block]
• SET-Change various program parameters.
SET [SIZE {size}]
7-4 Utilities
7.2.4.2 DEFAULT Command

The DEFAULT command is outlined as follows:
• Purpose: To change the default modifiers for the DUMP command.
• Syntax: DEFAULT.
• Parameters: None.
• Modifiers: Shown in the following list.
IIFERROR (NOIFERROR) (defaults ON~Dumps the error, header, and ECC fields in the
buffer if an error occurs when reading the block. When this modifier is used in conjunction
with the !RAW modifier, the error must occur on the reread of the block with the header
code extracted from the first read.
!ERRORS (NOERRORS) (defaults OFF~Dumps the error fields in the buffer.
!EDC (NOEDC) (defaults ON}-Dumps the EDC and calculated EDC fields in the buffer.
!EeC (NOECC) (defaults OFF~Dumps the ECC fields in the buffer.
/DATA (NODATA) (defaults ON~Displays the data in the buffer unless the /NZ modifier is
also specified.
!HEADERS (NOHEADERS) (defaults OFF~Displays the header fields in the buffer.
fALL (NONE}-The same as IERRORSIEDCIECC/DATAlHEADERS). Requests all fields be
displayed. Its opposite, INONE, requests no fields be displayed. When using the INONE
qualifier, only the MSCP status line prints.
/RAW (NORAW}-Allows reading the original LBN that was revectored rather than the
RBN that would be read without the !RAW qualifier. /RAW only affects revectored (primary
or non-primary) LBNs. If IIFERROR is in effect, this modifier applies only to dumping a
revectored LBN.
INZ (NONZ) (defaults OFF}-Prevents the data from being displayed if it is all O's. Instead,
a single line indicating the data is 0 is printed. It has no effect if the /DATA modifier is not
specified or if it is defaulted OFF.
IBBR (NOBBR) (defaults OFF~Usually inhibited when a block is accessed. If this modifier
is specified, bad block replacement can occur. It only occurs, however, if the error recovery
code detects the block being accessed as bad and the block is an LBN in the host area.
fORIGINAL (NOORIGINAL) (defaults OFF~aves the first data seen for display. When
a block is accessed for dumping, the data is seen twice by the program if an error occurs.
It is seen first just after the K detects the error and sends it to error recovery. It is seen
again after error recovery takes place and the data has been corrected or reread. Usually,
the data is saved for displaying when it is last seen.
• Usage: The modifiers specified are applied to the current default modifiers for the DUMP
command. The result becomes the new default. Examples are:
DEFAULTINONE
DEFIRAW/NODATA
DE/A/ORINZ
Utilities 7-5
7.2.4.3 DISPLAY Command

The DISPLAY command is outlined as follows:
• Purpose: To display the disk characteristics, the characteristics of a given block, the error
history in the drive, the FCT, and/or the RCT.
• Syntax:
DISPLAY ALL
DISPLAY CHARACTERISTICS DBN {block}
DISPLAY CHARACTERISTICS DISK
DISPLAY CHARACTERISTICS LBN {block}
DISPLAY CHARACTERISTICS PBN {block}
DISPLAY CHARACTERISTICS RBN {block}
DISPLAY CHARACTERISTICS XBN {block}
DISPLAY ERRORS
DISPLAYFCT
DISPLAYRCT
• Parameters: Block is a number specifying the DBN, LBN, PBN, RBN, or XBN whose
characteristics are displayed. The default radix is decimal, and can be changed to octal by
prefixing the number with the letter O.
• Modifiers:
/FULL-Displays all defined fields in xCT block o. !FULL applies only to the RCT and FCT
command options. For the RCT option, the bad block replacement and write back caching
fields in RCT block 0 are only displayed if the appropriate flags in the flags field are set.
These flags indicate they are currently in use (BBR or caching in progress). This modifier
forces all fields to be displayed regardless of the flags' settings. For the FCT option, the
number of bad PBNs field is normally displayed only if the FCT is valid. Also, the scratch
area parameters, format version, and format flags are normally not displayed. This modifier
forces all fields in FCT block 0 to be displayed.
INOITEMS-Does not display the individual items in the FCT or RCT. It applies only to the
FCT and RCT command options. If given, only the block 0 information is displayed.
• Usage:
DISPLAY ALL-Displays FCT, RCT, disk characteristics, and error history. Because the
error history in the drive is dumped by this option, it should not be used for RA60 drives.
Using the SDI command to read RA60 error history is illegal and causes the drive to become
inoperative.
DISPLAY CHARACTERISTICS DISK-Displays the drive type, media, cylinders, geometry,
group offsets, number of LBNs, number of RBNs, number of XBNs, numbers of DBNs,
number of PBNs, RCT parameters, FCT parameters, SDI version, transfer rate, SDI
timeouts, SDI retry limit, error resume recovery command levels, ECC threshold, revision
levels, drive ID, drive type ID, DBN Read/Only groups, and preamble sizes.
DISPLAY CHARACTERISTICS xBN {blockl-Displays the characteristics of the given
block. For DBNs and XBNs, these are the block numbers in decimal and octal, cylinder,
group, track, position, and PBN in decimal and octal. For RBNs, the RCT block numbers
and offset also are displayed. For LBNs, the primary RBN number and its RCT block
number and offset also are displayed. For PBNs, the display depends on the type of block:
DBN, LBN, RBN, or XBN.
7-6 Utilities
DISPLAY ERRORS-Reads the error history in the drive. The error history in the drive is
read from region 2, offset 0, and dumped in hexadecimal. This option should not be used for
RA60 drives because it causes them to become inoperative. Current drives display only 16
bytes of error log data. Succeeding drives display the error log header and all selected error
log entries.
DISPLAY FCT-Displays the information in FCT block O. Certain fields are not displayed
unless the /FULL modifier is given. The list of bad PBNs is displayed unless the fNOITEMS
modifier is given. For each item in the list, the header bits, PBN number, type (DBN, LBN,
RBN, or XBN), and XBN number are displayed.
DISPLAY RCT-Displays the information in RCT block O. Certain fields are not displayed
unless the /FULL modifier is given. The list of revectors, bad RBN s, and probationary
RBNs are displayed unless the /NOITEMS modifier is given. For bad and probationary
RBNs, just the RBN number is displayed (in decimal). For revectors, the LBN number and
RBN number to which it is revectored are displayed (in decimal). A primary revector is
distinguished by the character sequence "->". A non-primary revector is distinguished by
the character sequence "*->".
Examples are:
DISPLAY/FULL ALL
DIIF A
DI CD
DIS CHAR LBN 1000
DIINOIRCT
7.2.4.4 DUMP Command

The DUMP command is outlined as follows:
• Purpose: To dump the given block or table of blocks.
• Syntax:
DUMP [BUFFER]
DUMP DBN [{block}]
DUMP FCT [BLOCK {number}] [COPY {copy}]
DUMP LBN [{block}]
DUMP RBN [{block}]
DUMP RCT [BLOCK {number}] [COpy {copy}]
DUMP XBN [{block}]
• Parameters:
Block is a number specifying the DBN, LBN, RBN, or XBN to be dumped. The default radix
is decimal. It can be changed to octal by prefixing the number with the letter O.
Number is the relative block number in the FCT or RCT to be dumped. The default radix
is decimal and can be changed to octal by prefixing the number with the letter D. The value
must be in the range 1 through nonpad area of the FCT or RCT size. That is, the first block
is number 1 (not 0) and the block must lie in the nonpad area.
Copy specifies which copy of the given block in the FCT or RCT is to be dumped. The first
copy is number 1. The value must not exceed the number of copies.
DUMP XBN [{block}]-The specified DBN, LBN, RBN, or XBN is read in and dumped
subject to the given modifiers. If the block number is not specified, it defaults to O.
Utilities 7-7
DUMP xCT [BLOCK {number}] [COpy {copy}]-If a BLOCK number is given, that
block in the FCT or RCT is read in and dumped. If none is specified, every block in the
nonpad area of the FCT or RCT is read in and dumped. If COPY is not specified, it defaults
to copy l.
Examples of DUMP command parameters are:
DUMP RCT BLOCK 3 COPY 4
DU/NZRCT C2
DU LBN 1000
DFB2
DX
DIDATA
• Modifiers:
IIFERROR (NOIFERROR) (defaults ON)-Dumps the error, header, and ECC fields in the
buffer when an error occurs while reading the block. When used in conjunction with the
/RAW modifier, the error must occur on the read of the LBN (reread) with the header code
extracted from the RBN (first read). Refer to Section 7.2.4.2.
IERRORS (NOERRORS) (defaults OFF)-Dumps the error fields in the buffer.
IEDC (NOEDC) (defaults ON)-Dumps the EDC and calculated EDC fields in the buffer.
IECC (NOECC) (defaults OFF)-Dumps the ECC fields in the buffer.
/DATA (NODATA) (defaults ON)-Displays the data in the buffer unless the /NZ modifier is
also specified.
/HEADERS (NOHEADERS) (defaults OFF)-Displays the header fields in the buffer.
fALL (NONE)-The same as IERRORSIEDCIECCIDATAlHEADERS. It requests display
of all fields. Its opposite, INONE, requests display of no fields. When using the INONE
qualifier, only the MSCP status line prints.
/RAW (NORAW)-Allows a read of the original revectored LBN (rather than the RBN that
would be read without the /RAW qualifier). !RAW only affects revectored (primary or non-
primary) LBNs. If in effect, the IIFERROR modifier applies only to dumping a revectored
LBN.
/NZ (NONZ)-Prevents data from being displayed when it is all O's. Instead, a single line
prints indicating the data is O's. INZ has no effect unless the /DATA modifier is specified. It
also has no effect if /DATA is not specified (or is defaulted OFF).
IBBR (NOBBR) (defaults OFF)-Permits bad block replacement. Normally, bad block
replacement is inhibited when a block is accessed. BBR occurs if the block being accessed is
detected as bad by the error recovery code and is an LBN in the host area.
fORIGINAL (NOORIGINAL)-Saves the first data seen for display. When a block is
accessed for dumping, the data is seen twice by the program when an error occurs. It is
seen first just after the K detects the error and sends it to error recovery. It is seen again
after error recovery takes place and the data has been corrected or reread. Normally, the
data is saved for displaying when it is last seen.
7.2.4.5 EXIT Command

The EXIT command is outlined as follows:
• Purpose: To terminate execution of the program.
• Syntax: EXIT.
• Modifiers: None.
7-8 Utilities
• Usage: The current drive is released, all resources are returned, and the program exits.
Examples are:
EXIT
E
7.2.4.6 GET Command

The GET command is outlined as follows:
• Purpose: To obtain a drive or change the current drive.
• Syntax: GET [{drive}].
• Parameters: Drive is a valid drive unit specification of the form Dnnn. If this parameter is
omitted, GET defaults to DOOO (unit 0).
• Modifiers:
/NOIMF-Allows the reading of FCT block 0 to determine the mode and the reading and
writing of RCT block 0 to verify the RCT is valid. If this modifier is specified, the IMF
MSCP modifier is not used in the on-line mode and these actions take place. By default, a
new drive is brought on line with the IMF (MD.IMF) MSCP modifier.
IWP-Brings the drive on line with the MSCP SET WRITE PROTECT modifier (MD.SWP)
and WRITE PROTECT unit flag (UF.WPS). The drive is then software or volume write-
protected.
/NOWP-Brings the drive on line with the MSCP SET WRITE PROTECT modifier. The
drive is not software or volume write-protected.
NOONLINE-The drive is acquired but not brought on line with the MSCP on-line
command. Only the display characteristics, display errors, and the set size commands
can be executed on a drive in this state.
• Usage: The current drive is released. The new drive is acquired and then brought on line with
the requested modifiers and unit flags. If the drive is nonexistent, in use, or inoperative, the
user is put back in command mode. The modifiers cannot be changed for this other unit. If the
mode word in FCT block 0 is invalid or all copies of FCT block 0 are bad, the program prompts
for the sector size to use. Examples are:
GET D133
GIWPD64
G
7.2.4.7 POP Command

The POP command is outlined as follows:
• Purpose: To restore the data in the current buffer from the save buffer.
• Syntax: POp.
• Usage: The data in the save buffer is restored to the current buffer. The data in the current
buffer is lost. Examples are:
POP
P
Utilities 7-9
7.2.4.8 PUSH Command

The PUSH command is outlined as follows:
• Purpose: To save the data in the current buffer in the save buffer.
• Syntax: PUSH.
• Usage: The data previously in the current buffer is saved in the save buffer. The data in the
save buffer is lost. Examples are:
PUSH
PU
7.2.4.9 REVECTOR Command

The REVECTOR command is outlined as follows:
• Purpose: To force bad block replacement for one or more given LBNs.
• Syntax: REVECTOR {block} [{block}].
• Parameters: Block is a number specifying the LBN to be replaced. The default radix is decimal.
It can be changed to octal by prefixing the number with the letter O.
• Usage: The specified LBNs are sEmt to the bad block replacement module to be revectored. Ifit
is not a valid LBN or in the RCT, the revector fails and an error message prints. Otherwise, the
result of the replace attempt shows in the error log produced (if the appropriate level message
level is enabled [INFO]). The data in the replacement RBN is read from the specified LBN.
Examples are:
REVECTOR 1000
R 100
R 200 210
7.2.4.10 SET Command

The SET command is outlined as follows:
• Purpose: To change various program parameters.
• Syntax: SET [SIZE {size}].
• Parameters: The size parameter specifies the new sector size to be used for the current drive.
It must be either 512 or 576 bytes.
• Modifier: None.
• Usage: SET SIZE {size}.
The sector size is changed to the given value and the disk parameters are recomputed. This
new sector size is used when doing 110 to the LBN area and is also reflected in the parameters
printed by the DISPLAY CHARACTERISTICS DISK command. Examples are:
SET SIZE 576
S S 512
7-10 Utilities
7.2.5 Sample Session

The following is a sample session using DKUTIL. User input is indicated in bold print. Enter
CTRLIY to get the HSC> command prompt.
"Y
HSC> RUN OJroTIL
DKUTIL> GET 0133
Serial Number: 0000000004
Mode: 512
First Formatted: 17-Nov-1858 00:35:47.48
Date Formatted: 04-Apr-1984 00:05:09.20
Format Instance: 6
FCT: VALID
DKUTIL> OIS/F peT
Factory Control Table for D133 (RA80)
Mode: 512
Format Instance: 6
FCT: VALID
Bad PBNs in FCT: 1 (512), 0 (576)
Scratch Area Offset: 63
Size (Not Last): 417
Size (Last): 289
Flags: 000000
Format Version: 0
PBNs in 512 Byte Subtable
(04) 244865 (LBN 237213),
DKUTIL> REV 1000
ERROR-W Bad Block Replacement (Success) at 04-Apr-1984 17:47:24.20
Command Ref
RA80 Unit * *00000000
133.
Err Seq
Error Flags
* 6.
80
Event 0014
Replace Flags A400
LBN 1000.
Old RBN 32.
New RBN 33.
Cause Event 004A
ERROR-I End of error.
DKUTIL> OIS/F RCT
Revector Control Table for D133 (RA80)
Flags: 000000
LBN Being Replaced: 1000 (000000 001750)
Replacement RBN: 33 (060000 000041)
Bad RBN: 32 (060000 000040)
Cache ID: 0000000000
Cache Incarnation: o
Incarnation Date: 17-Nov-1858 00:00:00.00
Utilities 7-11
Bad RBN: 32, 1000 *-> 33, 25512 --> 822,

139512 --> 4500,
RCT Statistics: 1 Bad RBNs,
3 Bad LBNs,
2 Primary Revectors,
1 Non-Primary Revectors,
o Probationary RBNs.
DKUTIL> DEF/NODATA
DKUTIL> DUMP LBN 1000
****** Buffer for LBN 1000 (000000 001750), MSCP Status: 000000
Error Summary = header compare
Original Error Bits 004000 BN = 1000 (000000 001750)
Error Recovery Flags = 000 ECC Symbols Corrected = 0,0
Error Retry Counts 0,1,0 Error Recovery Command = 000
Header = 001750 030000 001750 030000 001750 030000 001750 030000
EDC 000105 Calculated EDC Difference = 000000
ECC 000000 000000 000000 000000 000000 000000
000000 000000 000000 000003 000000 000000
DKUTIL> DIS CHAR LBN 1000
Characteristics for LBN 1000 (000000 001750)
Cylinder 1, Group 0, Track 4, Position 8
PBN 1032 (000000 002010)
Primary RBN 32 (060000 000040) in RCT Block 3 at Offset 128
DKUTIL> DIS CHAR DISK
Drive Characteristics for D133
Type: RA80 (576 byte mode allowed)
Media: FIXED
Cylinders: 275 LBN, 2 XBN, 2 DBN
Geometry: 14 tracks/group, 2 groups/cylinder, 28 tracks/cylinder
31 LBNs/track, 1 RBNs/track, 32 sectors/track, 32 XBNs
896 XBNs/cylinder, 868 LBNs/cylinder, 28 RBNs/cylinder
Group Offset: 16 (LBN), 16 (XBN)
LBNs: 237212 (host), 238700 (total)
RBNs: 7700
XBNs: 1792
DBNs: 1344 (read/write), 448 (read only)
PBNs: 249984
RCT: 465 (size), 63 (non-pad), 4 (copies)
FCT: 480 (size), 63 (non-pad), 4 (copies)
SOl Version: 3
Transfer Rate: 97
Timeouts: 3 (short) , 7 (long)
Retry Limit: 5
Error Recover: 0 command levels
ECC Threshold: 2 symbols
7-12 Utilities
Revision: 10 (microcode) , 0 (hardware)

Drive ID: OA7AOOOOOOOO
Drive Type ID: 1
DBN RO Groups: 1
Preamble Size: 11 (data) , 4 (header)
DKUTIL> DUMP ReT BLOCK 3
****** RCT Block 3, Copy 1 ******
****** Buffer for LBN 237214 (000003 117236), MSCP Status: 000000
Data = 000000 000000 000000 000000 000000 000000 000000 000000
+16 000000 000000 000000 000000 000000 000000 000000 000000
+32 000000 000000 000000 000000 000000 000000 000000 000000
+48 000000 000000 000000 000000 000000 000000 000000 000000
+64 000000 000000 000000 000000 000000 000000 000000 000000
+80 000000 000000 000000 000000 000000 000000 000000 000000
+96 000000 000000 000000 000000 000000 000000 000000 000000
+112 000000 000000 000000 000000 000000 000000 000000 000000
+128 000000 040000 001750 030000 000000 000000 000000 000000
+144 000000 000000 000000 000000 000000 000000 000000 000000
+1~0 000000 000000 000000 000000 000000 000000 000000 000000
+176 000000 000000 000000 000000 000000 000000 000000 000000
+192 000000 000000 000000 000000 000000 000000 000000 000000
+208 000000 000000 000000 000000 000000 000000 000000 000000
+224 000000 000000 000000 000000 000000 000000 000000 000000
+240 000000 000000 000000 000000 000000 000000 000000 000000
+256 000000 000000 000000 000000 000000 000000 000000 000000
+272 000000 000000 000000 000000 000000 000000 000000 000000
+288 000000 000000 000000 000000 000000 000000 000000 000000
+304 000000 000000 000000 000000 000000 000000 000000 000000
+320 000000 000000 000000 000000 000000 000000 000000 000000
+336 000000 000000 000000 000000 000000 000000 000000 000000
+352 000000 000000 000000 000000 000000 000000 000000 000000
+368 000000 000000 000000 000000 000000 000000 000000 000000
+384 000000 000000 000000 000000 000000 000000 000000 000000
+400 000000 000000 000000 000000 000000 000000 000000 000000
+416 000000 000000 000000 000000 000000 000000 000000 000000
+432 000000 000000 000000 000000 000000 000000 000000 000000
+448 000000 000000 000000 000000 000000 000000 000000 000000
+464 000000 000000 000000 000000 000000 000000 000000 000000
+480 000000 000000 000000 000000 000000 000000 000000 000000
+496 000000 000000 000000 000000 000000 000000 000000 000000
EDC = 023277 Calculated EDC Difference = 000000
DKUTIL> EXIT
7.2.6 Error and Information Messages

DKUTIL error messages conform to the HSC utility error message format.
7.2.6.1 Error Message Variables

Certain portions of the error messages are variable and are shown in bold print. The meanings of
these variables are as follows:
n =A decimal number
par = BLOCK or COpy
parm = The part of the command in error (modifier, and so forth)
status = MSCP status (an octal number)
text = The actual text in error
XBN = DBN, LBN, and so forth
Utilities 7-13
xCT = FCT or RCT
7.2.6.2 Error Message Severity Levels

Each DKUTIL error message contains the utility name at the start of the message followed by a
letter indicating the severity level of the message. These are defined as:
E = Error
F = Fatal
I =Information
7.2.6.3 Fatal Error Messages

The following is a list of the DKUTIL fatal error messages:
• DKUTIL-F Insufficient resources to RUN!-Prints if DKUTIL cannot acquire the necessary
resources to run or if the disk functional code is not loaded. The program terminates after this
message is printed.
• DKUTIL-F I/O request was rejected!-Prints if the diagnostic interface (DDUSUB) rejects a
request to start an 110 operation. It indicates a bug in DKUTIL and should be reported to field
service support. The program terminates after this message is printed.
7.2.6.4 Error Messages

The following is a list of the DKUTIL error messages.
• DKUTIL-E Drive went AVAILABLE-Prints if the unit selected goes available while
DKUTIL is running. DKUTIL then goes into command mode and the user must issue a GET or
EXIT command at the DKUTIL> prompt.
• DKUTIL-E Drive went OFFLINE!-Prints if the selected unit goes off line while DKUTIL
is running. DKUTIL then goes into command mode and the user must issue a GET or EXIT
command at the DKUTIL> prompt.
• DKUTIL-E TIlegaI response to start-up question-Prints if an invalid response to a start-
up question or to a prompt for the GET command is entered. The program reprompts with the
same question.
• DKUTIL-E Nonexistent unit number-Prints if the unit number entered does not
correspond to any known unit. DKUTIL then goes into command mode and the user must
issue a GET or EXIT command at the DKUTIL> prompt.
• DKUTIL-E Unit is not available-Prints if the unit requested is unavailable. The unit may
be in use by a host or another diagnostic or it may be inoperative. DKUTIL then goes into
command mode and the user must issue a GET or EXIT command at the DKUTIL> prompt.
• DKUTIL-E Cannot bring unit ONLINE-Prints if the requested unit is available, but the
ONLINE command failed. The unit is released, and DKUTIL then goes into command mode
and the user must issue a GET or EXIT command at the DKUTIL> prompt.
• DKUTIL-E Invalid decimal number-Prints if an invalid decimal number is entered in a
command line.
• DKUTIL-E Invalid octal number-Prints if the user entered an invalid octal number in a
command line.
• DKUTIL-E Missing parameter-Prints if a command line is entered with a required
parameter missing.
• DKUTIL-E There is no buffer to dump-Prints if the DUMP BUFFER command is entered,
and there is no current buffer. This can only happen if a drive has just been selected.
• DKUTIL-E Missing modifier (only / was specified)-Prints if a command line is entered
with a slash (I) followed by a blank or is entered at the end of the line. A modifier is expected,
but is missing.
7-14 Utilities
• DKUTIL-E SDI command was unsuccessful-Prints when an SDI command is rejected by

the drive. A DISPLAY ERRORS command for an RA60 drive always generates this message.
• DKUTIL-E n is an invalid par number; maximum. is n-Prints if an out-of-range number
is entered for a BLOCK or COPY value for the DUMP command.
• DKUTIL-E xxx is an invalid xxx-Generic error message that prints when an invalid
command, invalid command option, invalid modifier, invalid block type, or invalid SET option is
specified in a command line.
• DKUTIL-E Invalid block number for XBN space-Prints if the block number specified for
a DISPLAY CHARACTERISTICS XBN command is out-of-range for the given space.
• DKUTIL-E Copy n of xCT Block n (XBN n) is bad-Prints when FCT or RCT blocks cannot
be read correctly with error recovery, when the FCT or RCT is being read just after a drive has
been selected. It also occurs when the DISPLAY FCT or DISPLAY RCT command is being used .
.- DKUTIL-E All copies of xCT Block n are bad-Prints when all copies of FCT or RCT blocks
are bad. It occurs when the FCT or RCT is being read just after a drive has been selected, or
when the DISPLAY FCT or DISPLAY RCT command is being used.
• DKUTIL-E Invalid sector size; only 512 and 576 are legal-Prints if the sector size
entered for the SET SIZE command is other than 512 or 576 bytes.
• DKUTIL-E Revector for LBN n failed., MSCP Status: (status)-Prints if a revector (using
the REVECTOR command) fails. If the status indicated that the drive went OFFLINE or
AVAILABLE, DKUTIL goes into command mode.
• DKUTIL-E Error log corrupted., cannot display header-Prints when the DISPLAY
ERRORS command reads a header that does not begin with the standard FFFB code.
• DKUTIL-E Error log corrupted, cannot display entries-Prints when the DISPLAY
ERRORS command is unable to read a valid entry from region FFFB.
• DKUTIL-E Unable to read error log-Prints when the DISPLAY ERRORS command is
unable to execute the read memory command.
• DKUTIL-E Error log not implemented in drive-Prints when the DISPLAY ERRORS
command is executed on an RA60.
• DKUTIL-E Drive must be acquired to execute this command-Prints if the requested
command requires that a drive must first be acquired before the command can be executed. A
drive can be acquired and not brought on line by using the /NOONLINE modifier with the GET
command.
• DKUTIL-E Drive must be on line to execute this command-Prints if the requested
command requires that a drive first be acquired and brought on line before the command can
be executed.
7.2.6.5 Information Messages

DKUTIL has the following information message:
• DKUTIL-I CTRI1Y or CTRIlC Abort!-Termination message that prints if DKUTIL is
aborted by typing CTRIJY or CTRUC.
Utilities 7-15
7.3 VERIFY - Off-line Disk Verifier Utility

VERIFY is a utility that checks the integrity of the disk architectural structure. This utility checks
a disk to ensure it conforms to the DIGITAL Standard Disk Format.
VERIFY has many messages that may print during the course of a disk structure verification.
These messages have significance only when VERIFY reports the drive is bad. At the end of its
run, VERIFY reports the drive is either OK or BAD.
NOTE
The VERIFY utility only reads the disk. It does not destroy user data and does not
perform bad block replacement.
The following steps describe the process by which this utility verifies a disk:
1. The first block of the Factory Control Table (FCT) is read to determine how the disk is
formatted. The serial number, format mode, date first formatted, date last formatted, format
instance, state of the FCT, number of bad PBNs, scratch area parameters (offset, size of not
last, and size of last), flags, and format version are printed.
2. The first block of the Revector Control Table (RCT) is then read. The information in it is
printed, including the serial number, flags, bad block replacement variables (LBN being
replaced, replacement RBN, and bad RBN) , and cache variables (ID, incarnation, and
incarnation date).
3. All copies of the first two blocks in the RCT (used by bad block replacement) are read and
compared. Discrepancies or bad blocks are reported.
4. All copies of the rest of the RCT are read and compared. Any discrepancies or bad blocks are
reported. The information about revectors and bad RBNs is dumped. A summary of the number
of bad blocks and revectors by type is printed.
5. All copies of FCT block 0 are read and compared, and bad blocks or discrepancies are reported.
6. All copies of the appropriate FCT subtable are read (if not null) and bad blocks or discrepancies
are reported.
7. The list of bad PBNs is printed. Each entry is printed with the header bits, PBN number, and
XBN number (in parentheses) as separate fields. If a bad PBN which should be in the RCT but
is not is found, the XBN field is printed in brackets instead of parentheses. If any such PBNs
are found, an error message indicating the total number is printed at the end of the bad PBN
list.
S. Mter reading and dumping the FCT, a quick scan of DBN space is done. Every block is accessed
only once. Counts of various detected errors are recorded for a summary printed at the end of
the scan. If more than nine positioner errors are detected, a message is printed suggesting
DBN space be reformatted. If more than nine EDC errors are detected, a message is printed
suggesting the INITIAL WRITE option should be used when running ILEXER.
9. All LBN space up to the RCT and all RBNs are scanned. Any block with an error is reread five
more times to determine the type of error. Information about bad blocks and revectors collected
in this phase is compared with information collected from reading the RCT. During the scan,
four error classes can be found:
Structure errors
Permanent recoverable errors
Permanent unrecoverable errors
Transient errors
Structure and permanent unrecoverable errors are considered inconsistencies and are always
reported. Permanent recoverable errors, usually ECC errors, are reported if requested. During
the five rereads of a block with an error, a block read at least once with no detected error is
considered to have a transient error. Transient errors are reported if requested.
7-16 Utilities
10. At the end of the scan, certain other errors are reported. Some errors can only be determined
at that time by examining information collected during the scan.
11. Finally, a summary, by type, of the errors detected and certain other information is printed.
If no inconsistencies were discovered, a message prints saying the drive is OK Otherwise, the
message indicates the number of inconsistencies.
7.3.1 Running VERIFY

VERIFY is started with the command:
RON VERIFY
The following prompt asks for the unit number of the disk to verify.
VERIFY-Q Enter unit number to verify (U) [DO]?
It then prompts to determine if the unit was recently formatted.

VERIFY-Q Was this unit just FORMATted (YIN) [Y]?
Enter Y or press RETURN if the disk has not been accessed by a host or diagnostic since it was
formatted.
This question is asked because certain errors are classed as inconsistencies only when the disk has
not undergone bad block replacement after formatting. The next prompt determines whether errors
not considered inconsistencies should be reported.
VERIFY-Q Print informational (non-warning) messages (YIN) [N]?
Enter N or press RETURN if you want VERIFY to only report inconsistencies-but not information
messages.
VERIFY reports the total number of transient errors in its final summary. You can also request
that VERIFY display individual blocks with transient errors. If you answered Y to the above
prompt, VERIFY next prompts:
VERIFY-Q Report transient errors by block (YIN) [N]?
If you enter Y, VERIFY displays a message for each block that contains a transient error. If you
enter N or press RETURN, you will not get this report.
Regardless of the response to this question, the number of transient errors is printed in the final
summary. The response to this question determines whether or not individual blocks with transient
errors should be reported.
A CTRLIZ can be entered at any prompt for the remainder of the responses. CTRUZ forces the
default response (in square brackets). Also, the responses to subsequent questions can be supplied
at any question by typing them separated with commas. For example, if unit D133 which was just
formatted is to be verified and all options are to be selected, the user could type DI33"Y,Y at the
first prompt.
If the unit does not exist or cannot be accessed, notification and reprompt for another unit number
are received. If the unit can be accessed, it is acquired and brought on line. VERIFY runs to
completion, unless aborted by CTRLIY or CTRUC.
Utilities 7-17

The following is a sample session using VERIFY. User input is in bold print.
ICTRL/YI
HSC50> RON VERIFY
VERIFY-Q Enter unit number to verify (U) [DO]? D133
VERIFY-Q Was this unit just FORMATted (YIN) [Y]?~r-==TU=RN~
VERIFY-Q Print informational (non-warning) messages (YIN) [N]? Y
VERIFY-Q Report transient errors by block (YIN) [N]? Y
*** FCT Block 0 Information
Mode: 512
Format Instance: 6
FCT: VALID
Bad PBNs in FCT: 1 (512), 0 (576)
Scratch Area Offset: 63
Size (Not Last): 417
Size (Last): 289
Flags: 000000
Format Version: o
*** RCT Block 0 Information
Flags: 000000
LBN Being Replaced: o (000000 000000)
Replacement RBN: o (060000 000000)
Bad RBN: o (060000 000000)
Cache ID: 0000000000
Cache Incarnation: o
Incarnation Date: 17-Nov-1858 00:00:00.00
*** Revector Control Table for 0133
VERIFY-I Copy 1 of RCT Block 2 (LBN 237213.) is bad.
25512 --> 822, 139512 --> 4500,
RCT Statistics: o Bad RBNs,
2 Bad LBNs,
2 Primary Revectors,
o Non-Primary Revectors,
o Probationary RBNs,
1 Bad RCT Blocks,
1 Bad First Copy RCT Blocks.
*** Factory Control Table for 0133
PBNs in 512 Byte Subtable
(04) 244865 (LBN 237213),
*** Quick Scan of DBN Area
Statistics:
o total blocks with any error.
*** Scan of LBN Area
VERIFY-I LBN 26003. has a 1 symbol correctable ECC error.
VERIFY-I RBN 2471. has a 1 symbol correctable ECC error.
VERIFY-I LBN 139962. has a 1 symbol correctable ECC error.
7-18 Utilities
Statistics:
3 total ECC symbols corrected,
3 blocks with 1 symbol ECC errors,
2 revectors verified,
5 total blocks with any error.
VERIFY-I Drive is OK.
The preceding example is the output of an actual session for an RA80 disk with one bad PBN in
the FCT. Notice this PBN corresponds to copy 1 of RCT block 2. RCT block 2 is used to store the
copy of the user data during bad block replacement. In its scan of the RCT, VERIFY noticed this
block was bad and printed an information message indicating that. If information messages had
been suppressed by responding with N to VERIFY-Q Print information (nonwarning) messages, this
information would show only in the summary of the RCT dump.
In the example, VERIFY also printed information messages for the three blocks it found with solid
one-symbol correctable ECC errors. If information messages had been suppressed, these messages
would not have printed. However, the number of such blocks would show up in the summary
statistics.
No transient errors were detected and, therefore, no count is reported in the summary statistics.
Also note that although no messages were printed for them, the two revectors in the RCT were
verified (as indicated in the summary statistics). Note the odd date for the First Formatted field.
This date is the default when no date is supplied by a host or a human during manufacturing
format. If structure inconsistencies had been found, some of the following VERIFY error messages
would also print.
7.3.3 Error and information Messages

This section describes error and information messages that may be printed out by VERIFY. Error
messages are arranged alphabetically according to the actual message.
7.3.3.1 Variable Output Error Fields

Error message fields with variable output print are in bold print. Definitions for these fields are:
xCT = FCT or RCT
n =A decimal number
n. = A decimal LBN, RBN, or XBN
XBN = LBN, RBN, or XBN
o = An octal number
t = Type code: I or W
x =Error: ECC, EDC, and so forth
7.3.3.2 Error Message Severity Levels

VERIFY error messages conform to the HSC utility error message format. In each case, the utility
name at the start of the message is followed by a letter indicating severity level. These are defined
as:
F = Fatal
I =Information
t = Type: either W or I, depending on the error
W =Warning
Utilities 7-19

Following is a list of the error messages fatal to the VERIFY utility. The program terminates after
printing one of these messages.
• VERIFY-F All copies of xCT block n are bad!-Prints if all copies of some block in either
the RCT or the FCT are bad. The program cannot continue to nm because vital information is
missing. In any case, it has verified that the unit is bad.
• VERIFY-F Current system sector size is 512!-Prints if the mode field in FCT block 0
indicates the unit is formatted in 576-byte mode, but the system sector size is set to 512. In
this case, VERIFY cannot run because it cannot read sectors 576 bytes long.
• VERIFY-F Drive went OFFLINE!-Prints if the unit selected goes off line while VERIFY is
running.
• VERIFY-F Insufficient resources to runt-Prints if VERIFY cannot acquire the necessary
resources to run or the disk functional code is not loaded.
• VERIFY-F 110 request was rejected!-Prints if the diagnostic interface (DDUSUB) rejects
a request to start an 110 operation. This message could be an indication of a problem in the
VERIFY utility, but it is more likely that DDUSUB could not obtain resources to complete the
I/O request.
• VERIFY-F Mode is bad or format is in progress on this unitt-Prints if the mode field in
FCT block 0 of the selected unit is not valid.
• VERIFY-F Too many bad blocks-VERIFY detected more bad blocks than it can fit into its
internal buffers. This may not mean that the disk is bad, but you should not run VERIFY on
this disk.
7.3.3.4 Warning Messages

The following messages are warning messages. In many cases, they are true warnings; in other
cases, they simply precede a reprompt.
• VERIFY-W n bad PBNs (in brackets above) not in the RCT-When VERIFY searches the
RCT for an LBN or RBN corresponding to a known bad PBN, it displays the PBN in brackets
and counts it. This message is displayed if the count is greater than zero.
• VERIFY-W Cannot ONLINE unit-Prints if the unit requested is available but the ONLINE
command failed. The unit is released and the user is reprompted for another unit.
• VERIFY-W Cannot read track with starting XBN n-VERIFY failed to access an LBN
space or RBN space track. This error may be caused by a hardware problem.
• VERIFY-W Copy n of xCT block n (XBN n.) does not compare-Prints whenever a block
is found that does not compare to the first good one.
• VERIFY-W Dlegal response to start-up question!-Prints if an invalid response is entered
for a start-up question. The program reprompts with the same question.
• VERIFY-W LBN n., a non-primary revector, is improper-Prints if an LBN was not a
non-primary revector but was recorded in the RCT as such. When VERIFY reads an LBN with
a header indicating it is a non-primary revector, it looks it up in the collected ReT information
and flags the fact if it was not found to be so.
• VERIFY-W LBN n., a primary revector, is improper-Prints if an LBN was not a primary
revector but was recorded in the ReT as such. When VERIFY reads an LBN with a header
indicating it is primarily revectored, it looks it up in the collected RCT information and flags
the fact that it was not found to be so.
• VERIFY-W LBN n. revectors to RBN n. which is bad-If VERIFY finds an RBN is good
(can be read with error recovery) or only has a forced error (after error recovery), it looks it
up in the collected RCT information. If found, VERIFY marks it as good. If, after the scan is
finished, this flag is not set for an RBN revectored to, this message is printed. .
7-20 Utilities
• VERIFY-W Nonexistent unit number-Prints if the unit number entered does not
correspond to any known unit. The program reprompts for the unit number.
• VERlFY-W Unit is not available-Prints if the unit requested is unavailable. It may be
in use by a host or another diagnostic, or it may be inoperative. The program reprompts for
another unit.
• VERIFY-W XBN n. has a hard EDC error-Prints for LBNs and RBNs found to have
a bad EDC (neither correct nor forced error). This error is classed as an inconsistency.
Only a software error can result in a record with a bad EDC (unless the DKUTIL command
WRITEfBAD is used).
• VERIFY-W XBN n. I/O error in access (MSCP Code: o)-Indicates that an inconsistency
was found in the drive or data channel module. VERIFY provides its own error processing for
records read where the K detects errors. This message is displayed under two conditions:
a. If requests do not return with a SUCCESS code, indicating a problem in the drive or disk
data channel.
b. If the return from the 110 operation is not successful after VERIFY reads the record in error
one more time with error recovery enabled.
• VERlFY-W RBN block is good but not used for a revector-VERIFY found a valid RBN
(with valid EDC) in the verification pass, but it is not recorded in the RCT as being used. This
record should not exist just after FORMAT has been run.
If you answered YES to the Was Unit Formatted prompt, this message is displayed with a
severity of W (warning). If the disk has undergone bad block replacement, this message is
displayed with a severity of I (informational).-If the drive was just formatted, reformat. If
the drive was not recently formatted, no action is necessary. If the message is displayed again,
submit an SPR.
• VERIFY-W RBN block_no marked bad in the RCT was not bad-VERIFY flagged a bad
RBN in the RCT that is not bad. This condition should not exist after FORMAT has been run.
displayed with a severity of I (informational).
• VERIFY-W LBN block has corrupted data (forced error)-A bad block replacement
in which the data could not be recovered produced a revectored LBN with forced error set,
indicating the data is probably bad. No such LBNs should exist after FORMAT has been run.
displayed with a severity of I (informational).
• VERIFY-W LBN n marked primary in RCT, not revectored to its primary-The specified
LBN number that was marked primary in the RCT was not revectored to its primary RBN.
• VERIFY-W xBN block has an uncorrectable ECC error-VERIFY detected an LBN or an
RBN with an uncorrectable ECC error that was not marked bad in the RCT. An LBN with an
uncorrectable ECC error should be revectored by FORMAT or bad block replacement; an RBN
with an uncorrectable ECC error should be marked bad in the RCT. Both of these errors are
classed as inconsistencies.
NOTE
If VERIFY detects an RBN with an uncorrectable ECC error and it IS marked bad in
the ReT, this message is displayed as an informational message.
Utilities 7-21

Following are descriptions of the information messages printed by VERIFY. Note that this type of
message mayor may not need information messages enabled in order to print.
• VERlFY-I CTRI1Y or CTRUC Abort!-Prints if the user aborts VERIFY by typing a CTRLIY
or CTRI.JC.
• VERlFY-I Drive is OK-A termination message which prints at the end of VERIFY if no
inconsistencies were discovered.
• VERIFY-I There were n inconsistencies found for this drive-A termination message
which prints at the end of VERIFY if inconsistencies were discovered.
• VERIFY-I Copy n of xCT Block n (XBN n.) is bad-Prints if information messages are
enabled for RCT or FCT blocks that cannot be read correctly with error recovery.
• VERlFY-I DBN area should probably be reformatted-Prints whether or not information
messages are enabled. If more than nine DBNs were detected with EDC errors (not forced
errors), this message prints after the DBN scan.
• VERIFY-I INITIAL WRITE should be specified for n,EXER-Prints whether or not
information messages are enabled. If more than nine DBNs were detected with EDe errors,
this message prints after the DBN scan.
• VERIFY-I LBN n., a primary has a bad header (is non-primary)-Prints if information
messages are enabled. An LBN is recorded in the RCT as a primary revector, but has a garbled
header. Such a condition is abnormal but not an error.
• VERIFY-I XBN n. has a transient (n out of 6) error type error-Prints if an LBN or RBN
has been read six times with a least one error-free read when information and transient error
messages are enabled. The number of times out of six that errors were detected is indicated in
the message.
• VERIFY-I XBN n. has a n symbol correctable ECC error-Prints for LBNs or RBNs with
solid ECC errors (errors on all six accesses) that are correctable when information messages
are enabled. The highest number of symbols corrected on a seventh access is indicated in the
message.
• VERIFY-I XBN n. has solid errors: error type-Prints for LBNs or RBNs with errors on
all six accesses when information messages are enabled. The errors included those other than
ECC or EDC. The record is read a seventh time with error recovery to determine if the error is
correctable. If it is not, a warning message is printed along with the following:
"NOTE: Tabl.e is nul.l or empty (NO BAD PBNs).
• VERIFY-I RBN block is good but not used for a revector-VERIFY found a valid RBN
(with valid ED C) in the verification pass, but it is not recorded in the RCT as being used. This
record should not exist just after FORMAT has been nIn.
• VERIFY-I RBN block_no marked bad in the RCT was not bad-VERIFY flagged a bad
RBN in the RCT that is not bad. This condition should not exist after FORMAT has been run.
• VERIFY-I LBN block has corrupted data (forced error)-A bad block replacement in
which the data could not be recovered produced a revectored LBN with forced error set,
indicating the data is probably bad. This condition should not exist after FORMAT has been
run.
• VERIFY-I xBN block has an UDcorrectable ECC error - VERIFY detected an RBN with
an uncorrectable ECC error that was marked bad in the RCT.
NOTE
If VERIFY detects an RBN with an uncorrectable ECC error and it is NOT marked
bad in the RCT, this message is displayed with a severity of W (Warning).
7-22 Utilities
• VERIFY-I n Blocks with hard EDC errors-The number of blocks with hard EDC errors.
• VERIFY-I n LBNs with corrupted data-The number of revectored LBNs with forced error
set. This indicates the data is probably bad.
• VERIFY-I n unused RBNs with good EDC-The number of RBNs with valid EDCs that are
recorded in the RCT as being used.
• VERIFY-I n good RBNs marked bad in the RCT-The number of RBNs marked bad in the
RCT that are actually good.
• VERIFY-I n blocks with solid (non-ECC) errors-Indicates the number of blocks with solid
(not ECC or EDC) errors.
• VERIFY-I n total ECC symbols corrected-Indicates the total number of ECC symbols
corrected during the VERIFY operation.
• VERIFY-I n blocks with n symbol ECC errors-Indicates the number of blocks in which a
specified number of ECC symbols were corrected.
• VERIFY-I n blocks with uncorrectable ECC errors-The number of blocks in which
uncorrectable ECC errors were detected.
• VERIFY-I n blocks with transient errors-The number of blocks in which transient errors
were detected.
• VERIFY-I n revectors verified - Indicates the number of revectors that were correctly
verified. Correct verification requires that all blocks that are primary or non-primary revectors
are recorded as such in the ReT.
• VERIFY-I n bad RBNs verified - The number of bad RBNs that were verified as bad in the
RCT.
• VERIFY-I n total blocks with any error-The total number of blocks that contained an
error(s) of any kind.
• VERIFY-I n bad DBNs-The number of bad DBNs VERIFY encountered.
• VERIFY-I n blocks with positioner errors-The number of blocks in which positioner errors
were detected.
• VERIFY-I n blocks with header compare errors-The number of blocks in which header
compare errors were detected.
• VERIFY-I n blocks with EDC errors-The number of blocks in which EDC errors were
detected.
• VERIFY-I n blocks with non-header, non-EDC errors-The number of blocks with in
which non-header, non-EDC errors were detected.
• VERIFY-I Exiting-Has completed and is exiting, or you pressed CTRUC or CTRUY.
7.4 FORMAT - Off-line Disk Formatter Utility

FORMAT is the utility used to format disks. It formats with either a 512- or 576-byte sector size.
It can be used to format only the read-only DBN space or to format both the LBN area and the
read-only DBN space.
CAUTION
The FORMAT utility destroys user data if used by persons not familiar with DSA.
The DBN area is always formatted. If the user requests it, the LBN area also is formatted. When
the LBN area is formatted, there are two modes of operation: the reformat and the best guess
modes.
Utilities 7-23
In reformat mode, the FCT on the disk is used and the XBN area is not formatted. If a reformat is
requested, but the FCT is null or corrupt, a modified best guess mode is used where only the LBN
area is formatted.
The main difference between best guess mode and reformat mode is each track is reread at least
three times during the check pass (best guess mode) instead of once (reformat mode). If any error
is detected, the track is reread 20 times instead of 3 times for reformat mode.
CAUTION
Be careful when using CTRUC or CTRUY to abort the FORMAT utility after formatting
operations begin. Doing this may destroy the contents of the FCT and/or the RCT.
The FORMAT utility should only be aborted under fatal-unrecoverable disk failure
conditions.
7.4.1 Running FORMAT

FORMAT is started with the standard CRONIC command syntax RUN DXO:FORMAT.UTL. Note
the last field in the following prompts (shown in square brackets); this indicates the default for that
prompt.
The program prompts for the unit number of the disk to format with the following:
FORMAT-Q Enter unit number to format (U) [DO]?
The next prompt determines whether the LBN (user data) area should be formatted or whether
only the DBN (diagnostic) area should be formatted. If this prompt is answered with a Y, user data
is destroyed.
FORMAT-Q Format user data area (Y/N) [N]?
If replied with an N or a carriage return only (to obtain the default), the program starts executing
and formatting only the DBN area. If a Y is entered, the program prompts for the sector size to use
when formatting the disk.
FORMAT-Q Enter sector size to be used (512/576) [512]?
If only the carriage return is pressed, the sector size used is 512 bytes. Otherwise, either 512 or
576 should be entered.
FORMAT-Q Continue if bad block information is inaccessible (Y/N) [N]?
If an N is entered, reformat mode is used if the FCT is valid. If it is not valid, the program aborts
with an appropriate error message. If Y is entered, reformat mode is used if the FCT is valid or a
modified best guess mode is used if the FCT is null or corrupt.
If the response to the preceding prompt is Y or the response to the destroy FCT prompt is Y, the
program prompts for a serial number:
FORMAT-Q Enter a non-zero serial number (D)?
This serial number is used when all copies of FCT block 0 are unreadable (in modified best guess
mode). FORMAT allows a number of special options, not only for debugging purposes but also to
increase data reliability. To determine if any of these options are desired, the program prompts
with the following:
FORMAT-Q Do you want special options (YIN) [N]?
If the response is N or a carriage return (the default ofN), FORMAT. starts processing.
7-24 Utilities
If the response is Y, the following three special option prompts appear. The first prompt option: is:
FORMAT-Q Revector blocks with 1 symbol ECC errors (YIN) [N]?
Nonnally, blocks discovered during the check pass of fonnatting with one-symbol ECC errors are
not retired. The program assumes this level of error is tolerable. If the response to this prompt
is Y, all blocks with solid (nontransient) ECC errors are retired. However, in all cases, blocks
with two-symbol (or more) ECC errors are always retired, regardless of the drive's ECC symbol
threshold.
The second special option prompt is:
FORMAT-Q Revector blocks with transient errors (YIN) [N]?
Mter a track is formatted, it is read either once (reformat) or three times (best guess). If an error is
detected, and the mode is reformat, the track is read twice more. If any block not previously retired
shows an error twice, it is retired and the track is reformatted with this check pass done again. If
no block had errors twice, the track is read 3 more times (reformat) or 20 more times (best guess).
Blocks that show an error only once during all of these reads are nonnally not retired. Such errors
are considered tolerable transient errors. If the response to this prompt is Y, blocks that show any
error are retired.
The third and final special option prompt is:
FORMAT-Q Report position of bad blocks (YIN) [N]?
Blocks retired during the format process are reported with a single line printout. The type, block
number, and cause are printed. If the response to this prompt is Y, the PBN number, cylinder,
track, group, and pesition are also printed on a subsequent line.
The user can enter CTRLIZ at any prompt to use the default for the remainder of the responses.
Also, the responses to subsequent questions can be supplied at any question by typing the responses
separated by commas. For example, if unit DI33 has an FCT and is to be formatted in 512-byte
mode with no special options, the user could type D133,Y"" at the first prompt.
7 .4.2 Sample Session

The following is a sample session using FORMAT. User input is in bold print.
ICTRL/YI
HSC70> RON DXO:FO~T
Utilities 7-25
FORMAT-Q Enter unit number to format (U) [DO]? D133

FORMAT-Q Format user data area (Y/N) [N]? Y
FORMAT-Q Enter sector size to be used (512/576) [512]?
FORMAT-Q Use existing bad block information (Y/N) [Y]?
FORMAT-Q Continue if bad block information is inaccessible (Y/N) [N]?
FORMAT-Q Do you want special options (Y/N) [N]? Y
FORMAT-Q Revector blocks with 1 symbol ECC errors (Y/N) [N]?
FORMAT-Q Revector blocks with transient errors (Y/N) [N]?
FORMAT-Q Report position of bad blocks (Y/N) [N]?
FORMAT-S Format begun.
FORMAT-I 2 cylinders left in DBN space at 00:05:34.60.
FORMAT-I 275 cylinders left in LBN space at 00:05:39.60.
FORMAT-I Bad LBN 237213 (FCT) , in the RCT area.

FORMAT-S Format completed.
FORMAT-I Stats: 0 Bad RBNs,
2 Revectored LBNs,
2 Primary Revectored LBNs,
o Non-Primary Revectored LBNs,
1 Bad Blocks in RCT Area,
o Bad Blocks in DBN Area,
o Bad Blocks in XBN Area,
9 Blocks Retried on Check Pass.
FORMAT-I FCT was used successfully.
***********************************************************
* VERIFY must be RUN to complete FORMAT verification!
*
* *
* *
***********************************************************
CAUTION
The message in the BOLD indicates VERIFY must be run to complete verification. This
is an essential step and should not be skipped.
The preceding example is the output for an actual session for an RA80 disk with one bad PBN in
the FCT. Notice the message that indicates it was retired because it was in the FCT and also the
RCT area. Note the information message that is printed every 10 cylinders. This confirms that
progress is actually being made and to show at what rate. Also, note the two LBNs that were
retired because they had two-symbol ECC errors; they became primary revectors. The error log
messages were printed for them because, in the case of an RA.80, two symbols are in excess of the
ECC drive threshold.
NOTE
The final statistics indicate two LBNs were revectored and one bad LBN was found in
the RCT area. The nine Blocks Retried on Check Pass include the two bad LBNs plus
seven other blocks with transient errors only and therefore not retired. The bad block
in the ReT was not retried in the check pass because it was known to be bad from the
FCT. This would be true for any blocks retired due to their location in the FCT. The final
message indicates an FCT was found and was successfully used.
7-26 Utilities

This section describes the error and information messages printed by FORMAT. Error messages are
arranged alphabetically according to the actual message.
7.4.3.1 Error Message Variables

Variable output in the error and information messages is shown in bold print. These fields are
formed as follows:
n = A decimal number
x = The way a block was found bad: FCT or check
XBN = A space: DBN, XBN, or LBN
hh = Hours
mm = Minutes
ss = Seconds
xx = Hundredths of a second
7.4.3.2 Message Severity· Levels

FORMAT error messages conform to the HSC utility error message format. In each case, the utility
name at the start of the message is followed by a letter indicating severity level. These are defined
as:
F = Fatal
I = Information
E = Error
S = Success
W = Warning

This section describes the fatal error messages printed by FORMAT.
• FORMAT-F Cannot position to DBN areal-Attempts to verify it has positioned the heads
to the DBN area before it formats the disk unless FORMAT is running in best guess mode.
FORMAT does this by reading the first sector of every track in the DBN read/write area until a
sector is read without a header error. This fatal error message is printed if no such sector can
be found.
• FORMAT-F Current maximum sector size is 512!-Prints if the user requests a 576-byte
sector size but the system sector size is set to 512. In this case, FORMAT cannot run because
110 cannot be done with sectors that are 576 bytes long.
• FORMAT-F DBN format error (drive FORMAT command Failed)!-Prints if a FORMAT
command fails for five retries when formatting the DBN area.
• FORMAT-F Drive does not support 576 mode on this medial-Prints if the user requests
a 576-byte sector size for a drive that does not support it.
• FORMAT-F Drive is write-protected!-Prints if the requested drive is hardware write-
protected and therefore cannot be formatted.
• FORMAT-F FCT Does not have enough good copies of each block!-Prints if any block
in the FCT does not have two good copies.
• FORMAT-F FCT is improper!-Prints if one or more PBNs remain to be processed. When
the program finishes formatting the LBN area, it checks to ~ee if all PBNs in the FCT have
been processed. It usually indicates an FCT where some PBNs are out of order.
• FORMAT-F FCT nonexistent!-Prints if the FCT is null or clobbered, and the user has
instructed the program not to continue.
• FORMAT-F FCT read error!-Prints if all copies of some given block of the FCT cannot be
successfully read.
Utilities ·7-27
• FORMAT-F FCTwrite error!-Prints if all copies of some given block of the FCT cannot be
successfully written.
• FORMAT-F Formatter initialization error!-Prints if FORMAT cannot acquire enough
Data Buffers or control blocks to start formatting, or if the disk functional code is not loaded.
• FORMAT-F GET STATUS failure!-Prints if the unit requested is not available or cannot be
brought on line.
• FORMAT-F LBN format error (drive FORMAT command failed)!-Prints if a FORMAT
command fails for five retries when formatting the LBN area.
• FORMAT-F Nonexistent unit number!-Prints if the unit requested does not exist.
• FORMAT-F RCT does not have enough good copies of each blockY-Prints if any block
in the RCT does not have two good copies.
• FORMAT-F RCT is fullY-Prints if so many bad blocks are encountered that the RCT
overflows.
• FORMAT-F RCT read error!-Prints if all copies of some given block of the RCT cannot be
successfully read.
• FORMAT-F RCT write error!-Prints if all copies of some given block of the RCT cannot be
successfully written.
• FORMAT-F SDI receive error!-Prints if a track cannot be read at all after it has been
formatted.
• FORMAT-F Too many bad RBNs found before RCT was formatted-Prints if more RBNs
than can be recorded in memory are encountered before the RCT area has been formatted.
• FORMAT-F Unsuccessful SDI command!-Prints if the drive fails to respond to an SDI
command. FORMAT issues SEEK, RECALIBRATE, and DRIVE CLEAR SDI commands.
7.4.3.4 Warning Message

The FORMAT utility prints only one warning message.
• FORMAT-W WARNING: Possible head addressing problem-Prints if no sector was
successfully read from one or more tracks in the XBN area. Note that all cylinders are checked.
This is a simple check for a bad head.

Following are the information messages printed by FORMAT.
• FORMAT-I Bad LBN n (x), a non-primary revector-Prints for LBNs retired by being
revectored to some RBN other than the primary RBN; they are marked in the ReT as
nonprimaries. They are formatted with a header code of non-primary or with a header code
of bad if their header area is bad.
• FORMAT-I Bad LBN n (x), a primary revector to RBN n-Prints for LBNs retired by
being revectored to the first RBN on the same track; they are marked in the RCT as primaries.
They are formatted with a header code of primary.
• FORMAT-I Bad LBN n (x), in the RCT Area-Prints for retired LBNs in the ReT area.
They are formatted with a header code of bad.
• FORMAT-I Bad RBN n (x)-Prints for retired RBNs. They are marked bad in the RCT and
are formatted with a header code of bad.
• Cylinder n, Group n, Track n, Position n, PBN n-Prints following the preceding four
messages, if the user requested the special option to print bad block position.
7-28 Utilities
• FORMAT-I CTRUY or CTRLlC abort!-An information message and prints if the user
aborts FORMAT by typing a CTRLIY or CTRUC. Note, this probably leaves the disk in an
unusable state if the format has begun.
• FORMAT-I FCT was not used-Prints if a null or clobbered FCT was found on the disk or
generated at the request of the user (best guess mode).
• FORMAT-I FCT was used successfully-Prints if a valid FCT was found on the disk and
used.
• FORMAT-I n Cylinders left in XBN space at hh:m.m:ss.xx-Prints after every 10 cylinders
are formatted in order to record the progress of the FORMAT program. .
• FORMAT-I Only DBN area formatted (n bad DBNs)-Prints if the user requested
formatting of the DBN area only. It prints after the format of the DBN area is completed.
Mter this message prints, the program terminates.
7.4.3.6 Error Messages

Following are the error messages printed by FORMAT.
• FORMAT-E Dlegal response to start-up question!-Prints if an invalid input is supplied
for a start-up question. The program reprompts with the same question.
• FORMAT-E Nondefaultable parameter-Prints if the user enters only a carriage return,
requesting the default for the only nondefaultable parameter (the serial number). The program
reprompts for the serial number.
7.4.3.7 Success Messages

Following are the FORMAT success messages.
• FORMAT-S Format begun-Prints when FORMAT actually begins formatting the disk.
• FORMAT-S Format completed-Prints after the format process is done, and all verification
tests are complete.
7.5 PATCH - Off-line Load Media Modification Utility

The PATCH utility is designed to modify files on the HSC load media. You can also use it to
examine locations within a file without modifying the file.
Mter you make changes to the load medium, the system version number is incremented on both the
system and utility software. Because the version number is included in the checksum, make sure
you apply patches in the exact order supplied. All previously shipped patches must be made before
you make a new patch. Make sure you supply a checksum to verify the patch has been applied
properly.
7.5.1 PATCH Commands

Table 7-1 lists the PATCH commands to enter in response to the PATCH prompts. All commands,
except control characters (such as CTRUC or CTRLIY), must be terminated with a RETURN. In
all instances, the current location refers to the file location of the address (Base + Offset) and the
contents displayed under the Old column.
Table 7-1 PATCH Commands

Command Function
RETURN Closes the current location without modifying it and opens and displays the next consecutive
location.
Utilities 7-29
Table 7-1 (Cont.) PATCH Commands

Command Function
n Sets the current location to a value of n, closes it, and opens and displays the next location.
Closes the current location without modifying it and opens and displays the previous
location.
nA Sets the current location to n, closes it, and opens and displays the previous location.
;B Backs up to the previous prompt. If requesting new value, backs up to Offset. If at Offset,
backs up to Base. If at Base, prompts for the checksum.
;C Restores the value originally in the open location. Used when an incorrect location is
modified.
;E Exits to the checksum prompt. If entered at the checksum question, causes PATCH to exit
without making any modifications to the file.
;Q!n Performs a logical inclusive OR of the contents of the open location with n, and opens and
;Q!nA displays the contents of the next (or previous) location.
;Q&n Performs a logical AND of the contents of the open location with n, and opens and displays
;Q&nA the contents of the riext (or previous) location.
;V Displays all changes made to the file during the current session.
CTRUC Exits PATCH without modifying the file.
CTRUY
7.5.2 Running PATCH

The following section describes the steps in the PATCH operation. Each PATCH prompt is
described, as well as the input you must supply.
Before running PATCH, take all disks and tapes off line that are attached the HSC to be patched.
Take the HSC off line, or fail over to an alternate HSC in the cluster.
To initiate PATCH, enter the following command at the HSC> prompt:
HSC> RON PATCH
PATCH prompts for the device and name of the file to be patched.
HSC>PATCH-Q RX33 Unit and fi~ename (dev: fi~e • ext) [SY:] ?dev: fi~e • ext
Enter the file and device name. To use the default system drive, specify only the filename. Refer to
the following table for special considerations.
If Then
The file being modified with Install system and utility software of the same revision level.
PATCH is on the device containing
the system or utility software
The system version numbers do not PATCH displays an error message, releases all acquired resources, and
match terminates.
The file or device name is invalid PATCH displays an error message and prompts for another file and
device name.
7-30 Utilities
H Then
The filename extension is .SAV The load device unit is acquired for exclusive access and all changes by
other programs are locked out.
The filename extension is not .SAV The SYSCOM.INI file on the system medium and the. 2NDTAP.VER
on the utility medium are accessed to compare the system version
numbers.
PATCH locates the file and checks the version numbers if necessary.
PATCH prompts for the base address of the patch.
Base (0) [000034]? IRETURN I
If the base address is odd, PATCH displays all values as bytes, rather an word values, unless no file
was specified.
PATCH prompts for the offset address of the patch.
Offset (0) [OOOOOO]? (100)
The default offset address is zero. If the offset address is odd, PATCH displays all values as bytes,
rather an word values, unless no file was specified.
PATCH prompts for the new contents of the file location.
Bas. Offset Old New?
aaaaaa bbbbbb cccccc ?
Enter the new contents of the file location in the NEW field. Each field in the prompt is described
in the following table:
Field Meaning
aaaaaa Base address as supplied

bbbbbb Current offset
cccccc Current contents of file location (sum of base and offset)
? New contents of the file location, as supplied in the patch documentation
Enter RETURN after typing in the new location contents. PATCH automatically increments
the offset to the next address, displays the current contents of that location, and prompts for a new
location contents. Use;B to enter new addresS/offset values as directed by the patch documentation,
and continue installing the patch data until completed.
Enter ;V to list the patch locations for verification. Check the changes carefully, and when you are
satisfied that all data has been entered correctly, exit the data entry prompt by typing ;E. If you
discover a mistake, enter CTRUC to abort the program without the patch, and start over.
Mter entering ;E, PATCH next prompts for the checksum.
Enter c:h.ckaum?
Enter the checksum given in the patch documentation. If you enter the wrong value, patch gives
you an error message and reprompts for the checksum. PATCH does not make any changes to the
file until you have entered the checksum.
PATCH then proceeds with the patch operation and displays a message indicating whether the
patch was successful and what patches were made.
Utilities 7-31
Mter installing the patch, reboot the HSC and verify that the system version number is
incremented on both the system and utility software. COPY the patched media to any backup
media as required and return. the system to service.

Example 7-1 is a sample PATCH session. The checksum shown is an arbitrary value. This example
is for display only.
HSC>RUN PA'l'CR
PATCH-Q Unit and filename idev:file.ext) [SY: ]? PA'l'CH. O'l'L
Base (0) [000034]?12000 tJ
Offset (0) [OOOOOO]? ~ 6)
Base Offset Old New?
012000 000000 004767 ?1378
012000 000002 010524 ?15000 0
012000 000004 012700 ?;B 0
Offset (0) [000004]?;B 6)
Base (0) [012000]?15000 ~
Offset (0) [000004]?O 0
Base Offset Old New?
015000 000000 000000 ?12701 CD
015000 000002 000000 ?177560
015000 000004 000000 ?137
015000 000006 000000 ?l004
015000 000010 000000 ?;E GD
Enter checksum?12345 ~
PATCH-S Wait ...
PATCH-S 6 changes made
Example 7-1 Example Patch of a File
0 Base is 12000
6) Offset is 0 (default)
8 Replace 4767 with 137

e Replace 10524 with 15000
0 Back up to new Offset
0 Back up to new Base
& New Base of 15000
0 New Offset of 0
0 Insert new values ...
G> Exit command
~ Enter checksum of 12345

This section describes the messages that may be displayed during a PATCH operation. The most
common causes of the errors and the correct actions are included.
7-32 Utilities

The following fatal error messages are issued by PATCH:
• PATCH-F Cannot access PATCH data file - PATCH cannot access the file you specified to
be patched.
• PATCH-F Cannot Access Version On Off-line Diagnostic medium-PATCH could not
access the file OFLLDR.SAV on the off-line diagnostic utility medium.
• PATCH-F Cannot Access Version On System medium - PATCH could not access the file
SYSCOM.INI on the system medium. If this message displays after the Wait ... message is
displayed, a serious bug or problem exists with the load medium. If the message displays just
after the filename has been entered, the system medium is probably not mounted.
• PATCH-F Cannot Access Version On Utility medium-PATCH could not access the file
2NDTAP.VER on the utility medium. If this message displays after the Wait ... message is
displayed, a serious bug or problem exists with the load device. If this message displays just
after the filename has been entered, the utility medium is probably not mounted.
• PATCH-F File Not Found-The user specified a nonexistent file. PATCH exits cleanly.
• PATCH-F Insufficient Resources To Run-The program could not acquire the resources to
run. Sufficient common pool memory is not available to allocate the necessary structures.
• PATCH-F Read Failure: block-number-When PATCH attempted to read the specified
block of the file, a media error occurred that cannot be recovered. PATCH exits cleanly.
• PATCH-F You cannot PATCH this file-The filename you specified cannot be patched.
• PATCH-F Unit(s) write-protected: update was not done-The disk unit on which the load
medium to be patched resides is write-protected.
• PATCH-F Version On System Medium Does Not Match Utility Medium- The version
numbers on the system and utility media do not match, indicating media from two different
revision levels are being used.
• PATCH-F Write Failure: block-number-When PATCH attempted to update the file with
the requested changes, a media error that cannot be recovered occurred in the specified
:file block. This is the most serious of errors because both file and medium integrity are
questionable. The usual recovery is to restore the medium with the backup copy made before
starting the patch.
• PATCH-F Write failure during write check, status: block number-PATCH verifies that
a file can be written before it actually writes to the file. You will get this message if the the
:file cannot be written for some reason other than the file being write-protected or a recoverable
software error.
7.5.4.2 PATCH Error Messages

The following error messages are issued by PATCH:
• PATCH-E Incorrect Checksum-You entered a checksum different from the checksum
PATCH calculated. PATCH repeats the checksum question.
• PATCH-E Invalid Command-You entered information out of context.
• PATCH-E Invalid Device Name Or Switch-You entered an invalid device name (not the
load medium) or the syntax of the filename is incorrect.
Utilities 7-33
7.5.4.3 Warning Messages

• PATCH-W Buffer Space Exhausted-You entered a patch when the internal buffers were
full.
• PATCH-W Nonfile Structured Mode Assumed - You entered a null filename. This message
is a reminder that all patches will now be applied to the load medium as a whole instead of to a
specific file.
7.5.4.4 Informational Messages

(0))
• PATCH-I Checksum = octal-checksum (O)-You entered the;X command, requesting that a

checksum prompt not be displayed.
• PATCH-I CTRUY Or CTRUC Abort-You typed a CTRLIY or CTRUC. No changes are
made to the file unless the abort occurs during the update, after the Wait ... message has been
displayed.
• PATCH-I No patches recorded-In a normal session, PATCH informs you of how many
patches have been made. You will get this message if PATCH records no patches in the file
PATCH.DAT, the file m;ed to keep track of the patches made.
• PATCH-I Patches made:-PATCH completes and displays the patches made following the
colon (:).
7.5.4.5 Success Messages

• PATCH-S patch-count Changes Made-You exited PATCH. The value of patch-count is the
number of changes made to the file. Note that if you modify and then restore a location, it is
not considered a change. The file is only rewritten when patch-count is not a zero.
• PATCH-S Wait•••-PATCH has started writing changes to the :file.
7-34 Utilities
Troubleshooting Techniques 8-1
8
Troubleshooting Techniques
8.1 Introduction
This chapter describes the types of errors occurring during HSC boot and operation. The major
divisions are initialization errors and system-type errors. Initialization errors occur while the HSC
is trying to boot. System-type errors occur while the HSC is running functional code. System-type
errors may be reported to a host node and possibly the HSC console device. Some system errors
may result in the HSC crashing and rebooting. System errors include MSCP, TMSCP, BBR, and
out-of-band errors.
8.2 How To Use This Chapter

Initialization error indications are displayed by the Operator Control Panel (OCP) fault codes and
the module LEDs. In addition, the bootstrap diagnostics may produce error messages printed out
to the console. Read Section 8.3 for an understanding of initialization errors that do not produce a
message. All errors displayed as English messages on the console are listed in alphabetical order in
Section 8.5 and are listed in this manual's index. Section 8.3 divides initialization errors into three
types:
• OCP fault codes
• Module LEDs
• Boot diagnostic messages
HSC console error message descriptions for system-type errors are described in this chapter and
are organized into the following sections:
• MSCPfl'MSCP errors, Section 8.4.1
Controller errors, Section 8.4.2.5
MSCP SDI errors, Section 8.4.2.6
Disk transfer errors, Section 8.4.2.7
• BBR errors, Section 8.4.3
• TMSCP errors, Section 8.4.4
STI communication or command errors, Section 8.4.4.1
STI formatter error log, Section 8.4.4.2
STI drive error log, Section 8.4.4.3
• Out-of-band errors, Section 8.4.5
8-1
8-2 Troubleshooting Techniques
8.3 Initialization Error Indications

Initialization errors are indicated by:
• OCP fault code displays
• Module LEDs
• Boot diagnostic messages
8.3.1 OCP Fault Code Displays

OCP fault codes are divided into two categories, hard fault codes and soft fault codes. Soft fault
codes are also called nonfatal fault codes. Soft faults impede HSC operation, but the fault does not
hinder the boot process. Hard fault codes are fatal to the HSC and prevent further operation of the
HSC subsystem until the condition is remedied.
Figure 8-1 shows the possible displays available on the OCP in the event of errors during
initialization or operation.
DESCRIPTION HEX OCT BINARY
K.PLI ERROR **o, 01 01 00001 OFF OFF OFF
K.SDIIK.SI INCORRECT
VERSION OF MICROCODE **o, 02 02 00010 OFF OFF OFF
K.STI/K.SI INCORRECT
03 03 00011 OFF OFF
VERSION OF MICROCODE **o,
P.IOJ CACHE FAILURE '" 08 10 01000 OFF
K.CI FAILURE * 09 11 01001 OFF
DATA CHANNEL MODULE ERROR * OA 12 01010 OFF
P.lOJ/C MODULE FAILURE 11 21 10001
M.STD2 MODULE FAILURE **o,** 12 22 10010 OFF OFF
BOOT DEVICE FAILURE ** 13 23 10011
PORT LINK NODE ADDRESS

15 25 10101
SWITCHES OUT OF RANGE
MISSING FILES REQUIRED **** 16 26 10110
NO WORKING K.CI, K.SDI,

18 30 11000
K.STI, OR K.SIIN SUBSYSTEM
INITIALIZATION FAILURE 19 31 11001
SOFTWARE INCONSISTENCY 1A 32 11010
ILLEGAL CONFIGURATION 1B 33 11011
THESE ARE THE SO-CALLED SOFT OR NONFATAL ERRORS.

POSSIBLE MEMORY MODULE/CONTROLLER ON HSC70.
INCORRECT VERSION OF MICROCODE.
**** THIS FAULT CODE WILL ALSO BE DISPLAYED IF THE L0105 MODULE IS NOT AT THE MINIMUM
REV LEVEL.
o,"**SWAP MEMORY MODULE FIRST. IF PROBLEM PERSISTS, TRY THE P.IO MODULE. CXO-905D
Figure 8-1 Operator Control Panel Fault Codes
8.3.1.1 Fault Code Interpretation

All failures occurring during the lnit P.io test are reported on the OCP LEDs. When the Fault lamp
is lit, pressing the Fault switch results in the display of a failure code in the OCP LEDs. This code
indicates which HSC module is the most probable cause of the detected failure. The failure code
blinks on and off at I-second intervals until the HSC is rebooted if the fault code represents a fatal
fault. A soft fault code is cleared in the OCP by pressing the Fault switch a second time. To restart
the boot procedure, press the Init switch. To identify the probable failing module, see Figure 8-1.
The following paragraphs describe specific fault codes displayed in the OCP lamps. All fault codes
are indicated with octal values.
• Fault Code 1, K.pli error
CODE VALUE OCP INDICATORS
HEX OCT BINARY INIT FAULT ONLINE
01 01 00001 OFF OFF OFF OFF ON
CXO-2666A
Figure 8-2 OCP Fault Code 1

Indicates the CIMGR initialization routine discovered bad requestor status from a previously
tested good requestor module in requestor slot 1. The expected requestor status should be 001.
The FRU is the Kpli.
During CIMGR initialization, the K.ci is directed to set the HSC node address into its own
control structure. If the Kci failed to modify this node address field after one-half second
from K.ci requestor initialization, this fault code is displayed. In addition, the Kpli microcode
version is checked to ensure it is compatible with this functional version. If compatibility checks
fail, this is the fault code displayed.
Run off-line diagnostics to test the K.ci requestor. Replace the Kpli module on failure. If the
fault code persists, refer to the HSC revision control document to verify all HSC components
are at the current revision.
• Fault Code 2, K.sdilK.si incorre~t version of microcode
02 02 00010 OFF OFF OFF ON OFF
CXO-2667A

All K.sdilK.si modules are initialized during the Disk Server functional code initialization.
If a KsdilK.si passes initialization, the Disk Server initialization code checks the K.sdilK..si
microcode version number to ensure it is compatible with this version of functional code. If code
versions are not compatible, this fault code is displayed. The FRU is the K.sdilk.si.
• Fault Code 3, K.stilK.si incorrect version of microcode
03 03 00011 OFF OFF OFF ON ON
CXO-2668A

Indicates tape data channel microcode is incompatible.
• Fault Codes 10, 11, and 12, Soft errors
These are the soft or nonfatal errors related to the data channels, the K.ci host interface, and
the P.ioj cache. None of these errors causes the HSC functional operation to suspend when the
fault is reported. Once displayed, soft error indicators cannot be recalled. The HSC may buffer
up to eight soft fault codes. Subsequent toggling of the Fault switch displays all remaining soft
fault codes until the buffer is empty.
- Fault Code 10,P.ioj cache failure
HEX OCT BINARY IN IT FAULT ONLINE
08 10 01000 OFF ON OFF OFF OFF
CXO-2669A

Results in disabling the cache and displaying this soft fault code for any failure detected in
the J-ll instruction cache during HSe subsystem initialization while the HSe continues
operation. Replace the P.ioj module (LOIIl1LOl11-YA) and reboot.
Fault Code 11, Kci failure
09 11 01001 OFF ON OFF OFF ON
CXO-2670A

One or more modules of the Kci set is not present or has failed its initialization tests. This
soft fault is displayed while the HSe continues to operate. The most probable FRU is the
LINK module (LOIOOIL0118).
- Fault Code 12, Data channel module failure

-
OA 12 01010 OFF ON OFF ON OFF
CXO-2671A

Reports an unknown requestor type was found in a requestor slot other than 0 or l.
Expected valid requestor types for requestor slots 2 through 8 are either 002 for a K.sdi
(L0108-YA) or 203 for a K.sti (L0108-YB). The Ksi (L0119-YA) module will answer with 002
if it is initiated as a disk data channel, or with 203 if it is initiated as a tape data channel.
The data channel with the red LED on is the failing module.
• Fault Code 21, P.ioj/c module failure
11 21 10001 ON OFF OFF OFF ON
CXO-2672A

Indicates the P.ioj/c module is the most probable cause of the failure detected by the Init P.io
test. If possible, run the off-line P.io test for a more definitive report on the error. Otherwise,
replace the P.ioj/c module and run the Init P.io test again. If the test still fails, run the off-line
P.io test to help further isolate the failure.
• Fault Code 22, M.std.2 module failure
12 22 10010 ON OFF OFF ON OFF
CXO-2673A

Indicates the M.std2 module (L0117) is the most probable cause of this bootstrap failure.
Possible causes include:
The failure of the memory test of the first 1 Kword (vector area) of Program memory as well
as the use of the Swap Banks bit in the P.ioj/c in trying to correct the problem (test 2).
A contiguous 8-Kword partition not found in Program memory below address 00160000 (test
3).
A hard fault detected in the RX33 controller logic (test 4).
Determine the error that occurred by examining physical location 17772340, which contains the
number of the failing boot ROM test. In each of these cases, replace the M.std2 module, and
run the initialization tests again. If the module still fails, run the off-line P.io test.
Enter the SETSHO utility and execute the SHO MEM command. If any memory locations
appear in the suspect or disabled memory locations list, set the SecurelEnable switch to enable
and execute the SET MEM ENABLE/ALL command.
• Fault Code 23, Boot device failure
13 23 10011 ON OFF OFF ON ON
CXO-2674A

Indicates a problem with an HSC boot device, the system media, the boot device controller, or
the read/write logic on the memory module. This fault can be any of the following, in order of
probability:
A failure in the P.ioc.
A failure in the read/write logic of the M.std2 module. Replace M.std2 (LOl17).
A faulty boot device controller/drive interface cable. Replace the cable.
Diskettes or tapes not installed in the drives.
Doors left open on the RX33 drives
Tape improperly inserted in the TU58 drive.
No bootable image in the system device media.
Ensure a known good HSC bootable media is properly loaded in the system boot device.
If checking the obvious (doors, diskettes, or tapes) does not remedy the situation, refer to
Chapter 6 for more information before beginning repair. Running the off-line P.io and off-line
RX33 or TU58 tests (if possible) is strongly recommended before modules are replaced. These
tests may help further isolate or define the problem.
• Fault Code 25, port link node address switches out of range
15 25 10101 ON OFF ON OFF ON
CXO-2675A

Indicates the LINK (L0100ILOl18) module node address switches are set to a value outside the
currently suggested range of 32 decimal (HSC software V3.90).
• Fault Code 26, missing files required
16 26 10110 ON OFF ON ON OFF
CXO-2676A

Indicates the system diskette does not contain one of the files necessary for operation of the
HSC control program. This failure should occur only if one of the required files is inadvertently
deleted from the HSC system media.
Note that the condition of the State light must be observed prior to the fault occurrence. The
State light is always steady (either ON or OFF) when the fault light is lit during boot faults.
If the State light is steady (ON), it can mean:
SYSCOM.lNI is not present on the load device.
EXEC.INI is not present on the load device.
A version mismatch exists between either EXEC, SUBLIB, or 8YSCOM and OLBV8N
(Object Library Version Number).
If the State light was blinking before the fault, it can mean:
Any of the the normally loaded programs (8INI, CERF, DEMON, etc.) are not present on
the load device.
A version mismatch exists on anyone of the normally-loaded programs.
Replace the system media with a backup copy.
• Fault Code 30, No working K.ci, Ksdi, Ksti, or Ksi in subsystem
18 30 11000 ON ON OFF OFF OFF
CXO-2677A

Indicates the HSC does not contain any working K.ci, K.sti, K.sdi, or K.si modules. Either none
is installed in the HSC, or all of those installed failed their initialization diagnostics. Also, if
the Disk Server code is loaded and no working K.sdi is found, this fault code is displayed.
Insert the HSC Off-line Diagnostic media into the appropriate system drive and reboot the
HSC. When the off-line loader prompts with ODL>, type SIZE. The SIZE command displays
the status of all the Ks. This status indicates whether the modules are missing or are failing
initialization diagnostics.
If all else fails, replace the P.ioj (LOlll!LOlll-YA) or P.ioc (LOI05) and check subsystem power
for proper operation.
• OCP error code of 31, Initialization failure
19 31 11001 ON ON OFF OFF ON
CXO-2678A

Indicates a crash occurred while the HSC was attempting to load and initialize its control
program.
Use micro-ODT to diagnose these initialization crashes, as follows:
1. Press the break key on the local console terminal.
2. Type 17 777 656/.
This is the address of the UPAR7 register. The reasons for reboot codes are stored in
UPAR7 bits 8 to 11 when an OCP code of 31 has been detected. The other UPAR registers
store useful information for some of the errors related to an OCP fault code of 31. Refer to
the fault code 31 reasons in the following paragraphs for UPAR content usage. Table 8-1
shows the addresses of the UPAR registers.
Table 8-1 UPAR Register Addresses

Register Address
UPARO 17777640
UPAR1 17777642
UPAR2 17777644
UPAR3 17777646
UPAR4 17777650
UPAR5 17777652
UPAR6 17777654
UPAR7 17777656
3. Analyze bits 8 to 11 of the 16-bit message displayed by examining UPAR7. Table 8-2 shows
the bit/error relationship.
Table 8-2 Control Program Bits

16 Bit Message Meaning FRUs
x XXX XXX 1XX.:XXX XXX NXM P.ioj (L01111L0111-YA)

M.std2 (LOll7)
Software
X XXX XXI OXXXXXXXX illegal inst. P.ioj (LOl111L0111-YA)
M.std2 (L0117)
Software
Table 8-2 (Cont.) Control Program Bits

16 Bit Message Meaning FRUs
X XXX XXI 1XX XXX XXX Parity trap M.std2 (LOl17)

P.ioj (LOllllLOlll-YA)
XXXXXIO OXX:XXXXXX Level 7 interrupt K.xx (LOl08!L0119-YA)
K.pli (LOl07)
X XXX XIO 1XX:XXX XXX MMUtrap P.ioj (LOlll-LOlll-YA)
Software
X XXX XII OXX:XXX XXX Software crash Software
X XXX XII 1XX:XXX XXX K.ci host reset M.std2 (LOl17)
X XXX 100 OXX XXX XXX User requested N/A
reboot
4. If this error occurs repeatedly, it indicates an intermittent hardware error or degraded

diskette media. The boot-in-progress flag is indicated by KPDR7 bit 3 set. The KPDR7
register address is 17 772 316. Use micro-ODT to examine bit 3 (it can be reset).
The following list describes actions to be taken for each type of error related to an OCP fault
code of 31 as pointed out by examining UPAR7:
NXM trap: Examine UPARI to find the lower 16 bits of the failing memory address by
typing 17 777 6421. Examine UPAR2's lower byte for the high 6 bits of the failing memory
address by typing 17 777 644/.
Dlegal inst: Replace the P.iojlP.ioc module.
Parity trap: Use the same method for parity traps as for NXM traps to determine the
failing address.
Level 7 interrupt: Determine which K UPAR4. Refer to Table S-1 for the address of each
UPAR register. Each byte of each register contains module status for each requester (K) in
the HSC. Refer to Appendix C to determine a failing status code. Refer to Table S-3 for the
designation of requestors to UPAR registers for a level 7 interrupt.
Table 8-3 Status of Requestors for Level 7 Interrupt

Register High Byte Low Byte
UPARO REQ2 REQ 1

UPARI REQ4 REQ3
UPAR2 REQ6 REQ5
UPAR3 REQ8 REQ7
UPAR4 N/A REQ9
Memory Management Unit (MMU) trap: Examine UPAR1, UPAR2, and UPAR3 to
determine the status of the :MMU at the time of the OCP fault code of 31. When an MMU
trap occurs, status of the MMU is found in these registers.
Software crash: Try using another copy of the boot media. If the problem is not corrected,
replace the P.iojlP.ioc module. If the problem still persists, replace the memory module.
K.ci host reset: Press the break key again and at the @ symbol type 17 770 000/ when a
host reset is known as the reason for an OCP fault code of 31. This is the address of Control
memory window o. When the / is pressed, the contents of control window 0 are displayed.
Enter a 0 into this location followed by a carriage return. Then type 17 760 0021. This is
the second location in Control memory. The number displayed as the contents of 17 600 002
is the number of the host that issued the HOST RESET command.
• OCP error code 32, Software inconsistency
1A 32 11010 ON ON OFF ON OFF
CXO-2679A

Indicates an inconsistency in the software. Reboot the HSC. If this failure persists, use a
backup copy of the system media. If the failure still persists, use the Off-line diagnostics to help
isolate any hardware failures in the subsystem. Also, try using an earlier version of the HSC
operating software.
• OCP error code 33, Dlegal configuration
1B 33 11011 ON ON OFF ON ON
CXO-2906A

Indicates that you have installed an illegal configuration of modules in the HSC40 backplane.
Check the configuration of modules in the backplane and install the modules according to the
following rules:
1. Install any combination ofK.si, K.sdi, or K.sti data channel modules in requestors 2 through
4 (backplane slots 8 through 10).
2. Leave requestors 5 through 9 (backplane slots 3 through 7) unoccupied.
8.3.2 Module LEDs

HSC modules contain LEDs used as State indicators for each module. Descriptions of these LEDs
follow in the next sections. Also, refer to Chapter 2 for the locations of the module LEDs.
8.3.2.1 P.ioj/c LEOs

Table 8-4 shows the P.ioj/c (LOlll, LOlll-YA, or LOl05) LEDs and their functions.
Table 8-4 P.ioj/c LEOs

LED Color Meaning
DI Yellow Micro-ODT-Used during J-II power-up microdiagnostics. ON when J-II

is executing micro-ODT.
D2 Yellow Terminal Port OK-Used during J-II power-up microdiagnostics. Serial
Line Unit (SLU) output ofUART.
D3 Yellow Memory OK-Used during J-II power-up microdiagnostics. 'fumed OFF
as J-II successfully accesses Program memory.
D4 Yellow Sequencing indicator-Used during J-II power-up microdiagnostics.
Turned OFF as J-II verifies proper functioning of its sequencers for
control store.
D5 Yellow State indicator-Mirrors the OCP State indicator (under software
control).
D6 Yellow Run indicator-Pulses at the on-board microprocessor run rate. Blinks
once for every PDP-II instruction fetched (J-II run LED).
D7 Red Board- status-Indicates an inoperable module except during initialization
when it comes on during module testing.
DB Green Board status-Indicates the module has passed all applicable diagnostics.
8.3.2.2 Power-up Sequence of 1/0 Control Processor LEOs

This section defines the power-up sequence of the LEDs shown in Table 8-4. First, LED numbers
D8 and D7 are used to indicate whether the P.ioj/c module has successfully completed all of its
initialization diagnostics. The module powers up with the red (D7) LED ON and the green (D8)
LED OFF. Dl through D4 (yellow) are initially ON. As soon as the J-ll starts operating, Dl
(micro-ODT LED) turns OFF.
Several microcode steps later, D4 (sequence LED) is turned OFF, indicating the J-ll is sequencing
and succeeded in reaching this point in its microcode. The J-ll performs several Program memory
operations and, if successful, turns OFF D3 (memory OK LED). Finally, the J-ll accesses the
console terminal port of the UART (universal asynchronous receiver/transmitter) and turns OFF D2
(SLU or Serial Line Unit LED).
Upon successful completion of the boot time initialization diagnostics, D8 (module OK LED) turns
ON, and D7 (module failure LED) turns OFF. The J-ll then proceeds to the software initialization
programs.
In addition to being initially ON, the D1 (micro-ODT run LED) is ON any time the J-1l is executing
micro-ODT. D6 (the fetch LED, sometimes referred to as the run LED) blinks once for every PDP-ll
instruction fetch cycle. When the J-1l is running, D6 is illuminated at half-brilliance compared to
the other yellow LEDs.
8.3.2.3 Memory Module LEOs

Table 8--5 shows the M.std2 (LOll7) and M.std (L0106) module LEDs and their functions. These
LEDs are controlled by a bit in the system boot device FDC MAR02 register. The green LED is
set to ON by the P.ioj/c bootIROM self-test diagnostics after the system boot device has passed its
self-tests, and Program memory has found 8-Kwords to load INIPIO/OFLPIO.
Table 8-5 M.std2 and M.std LEDs

LED Color Meaning
D2 Red Module not OK

D2 Green Module OK
D2 Yellow Memory active
NOTE
The entire LED package on the M.std2 or M.stel is called D2. All three LEDs are contained
in the D2 package.
8.3.2.4 Data Channel LEOs

Table 8-6 shows the KsdilK.sti (LOIOS-YAIYB) and K.si (LOl19-YA) data channel module LEOs and
their functions with the system software.
Table 8-6 K.sdi/K.sti and K.si LEOs

LED Color Meaning
Red Module failure-Indicates a module microdiagnostic failed to successfully

complete or this module is still under initialization by the subsystem.
J
Green Module OK-Turned on by the lnitIFunc Flag signal in the K functional microcode.
The green LED comes ON after successful initialization or while the data channel
is running functional microcode.
LED pack Amber DI-OFF for PROM 10adJ ON for RAM load.
(K.si (eight D2 through DB-Upper register #2 contents.
only) LEDs) The LEDs reflect the implemented bits of the upper error register #2. When a
microinstruction parity error is detected, the module clocks are inhibited, stopping
the module. The bit content of the upper error register #2 is displayed on the
LEDs.
8.3.2.5 Host Interface LED

Table 8-7 shows the three modules in the K.ci set, their LEOs, and the functions of the LEOs with
the system software.
Table 8-7 K.ci (LINK, PILA, K.pll) LEOs

Module LED Color Meaning
Kpli D2 Red ON when P.io has booted or rebooted, but K. pli module has not yet passed its
self-test.
Kpli Dl Green ON when Kpli has passed its self-test.
PlLA D2 Red ON when PILA module has not yet passed the test performed by the K.pli.
PlLA Dl Green ON when the PlLA module has pas~d the test performed by the K.pli. LED
is controlled by the port processor.
PlLA D3 Yellow Not found on all module revisions. ON when K. pli is asserting Init. When
lnit is true, both the red and the green PILA LEDs are forced OFF.
Table 8-7 (Cont.) K.ci (LINK, PILA, K.pli) LEOs

Module LED Color Meaning
LINK D998 Green ON when local activity is present on the LINK. module. Whenever the LINK.
module detects a message directed to its node or when it detects an outgoing
message.
LINK D999 Red ON during the CI maintenance loop test.
8.3.3 Communication Errors

It is possible for the HSC to complete its initialization and not report the fact on the local console
tenninal. This is an indication of a failure in the serial communication path between the UART
chip on the P.ioj/c and the local console terminal.
As a method of testing this serial path, the HSC echoes the characters typed on the local console
tenninal as if the terminal were in local mode. Use the following procedure to test the serial path:
1. Place the SecurelEnable switch in the enable position.
2. With power on, push in and hold the OCP Init switch.
3. Type a series of characters on the terminal keyboard.
4. Check to see that the series of characters is echoed correctly on the terminal.
NOTE
When the Init switch is released, the HSC reboots.
If this procedure fails to echo characters typed at the keyboard, the failure is either a terminal
to P.ioclj baud-rate mismatch (default is 9600), a P.ioj/c module failure, or a problem within the
terminal-cabling subsystem. Ensure the terminal setup parameters are correct. Refer to theHSC
Installation Manual for the proper terminal configuration, the VTxxx Owner's Manual for problem-
solving techniques related to the VTxxx, and the DECwriter Correspondent Technical Manual for
problem-solving techniques related to the LA12.
8.3.4 Requestor Status for Nonfailing Requestors

When a requestor successfully completes all internal microdiagnostics, the requestor status
(Status=) contains the following codes defining module types:
• Code 001 represents a properly functioning host interface module set (Kci).
• Code 002 represents a properly functioning disk data channel module (K.sdilK.si with the disk
channel microcode loaded).
• Code 004 represents a Ksi with no microcode loaded.
• Code 203 represents a properly functioning tape data channel module (KstilKsi with the tape
channel microcode loaded).
• Code 377 indicates the requestor slot does not contain a module.
NOTE
When a module fails internal microdiagnostics or its functional code, the status byte
reflects the failure. See Appendix D for a complete list of Kci, Ksdi, Ksti, and K.si
detected failures.
8.3.5 HSC Boot Flow and Troubleshooting Chart

The HSC boot flow and troubleshooting chart calls out useful visual milestones that aid in
troubleshooting problems which can occur during initialization.
The flowchart has three main divisions:
1. Information on activity common to both the system and off-line diskettes is contained in boxes
A through 0
2. Information on activity specific to the system diskette is contained in boxes SA through SJ
3. Information on activity specific to the off-line diskette is contained in boxes OA through OG
The flowchart begins when one of the following occurs:
• lnit button is pushed.
• Powerup has started.
• Other software caused reboot.
Figure 8-17 rna ps the entire HSC boot sequence.
INTERNAUEXTERNAL
INITIALIZATION
ENTRY POINT
TIME = 0
NO FAULT CODE
J-11 PERFORMS INTERNAL FAIL STATE INIT FAULT
MICRO TEST ... A THROUGH C
--- 07 (RED LED) ON
? ? ?
NO FAULT CODE
A TEST INTERNAL J-11
SEQUENCER; TURN OFF FAIL STATE INIT FAULT
01 (MICRO-ODT) IF NOT
---
? ? ?
IN ODT; TURN OFF 04
NO FAULT CODE
B TEST MEMORY: LOC 0
RESPOND (NO NXM?); FAIL STATE INIT FAULT
LOC 1777700 SHOULD
---
? ? ?
NXM; TURN OFF 03
NO FAULT CODE
C TEST FOR SLU, CHECK FAIL STATE INIT FAULT
177560 FOR RESPONSE;
?
-? - ?-
TURN OFF 02
NO FAULT CODE
0 BEGIN EXECUTION OF FAIL STATE INIT FAULT
BOOT ROM; TURN OFF ---
ALL OCP INDICATORS 0 0 0
NO FAULT CODE
E STATE INIT FAULT
FAIL
TEST J-11 BASIC
INSTRUCTIONS TEST 0
---
0 0 0
FAUL T = 21 OCTAL
F 07 STILL ON;
TEST J-11 ADC, DIV, FAIL STATE INIT FAULT
REGISTERS RO:R7
--- OCP INDICATORS
0 0 NOT RELIABLE
TIME G
<1/2 TURN ON INIT INDICATOR
SECOND
NOTES:
1. LEOs 01-04 AND 07 ARE ON THE P.IOJ MODULE.
2. ? MEANS OCP LEOs ARE INDETERMINATE AND HAVE NO MEANING AT THIS TIME.
CXO-945C
Sheet 1 of 5
Figure 8-17 (Cont.) HSC Boot Flow and Troubleshooting Chart

FAULT = 21 OCTAL IF MEMORY FAILS

H TEST 1 WITH NXM OR
FAIL STATE INIT FAULT
TEST BANK SWAP --- PARITY ERROR,
BITS IN P.IOJ CSR 0 FAULT WILL NOT
BE SET
NO
I TEST 2 FAIL STATE INIT FAULT

TEST FIRST 1 KW OF GOOD ---
PROGRAM MEMORY 0 X
YES
IF MEMORY
J TEST 3 DATA ERROR IS
FAIL STATE INIT FAULT
FIND 8 KW OF GOOD --- DETECTED,FAULT
PROGRAM MEMORY 0
IS 22 AND FAULT
LED WILL BE ON
FAULT = 22 OCTAL
K TEST 4 FAIL STATE INIT FAULT
TEST RX33 CONTROLLER ---
HARDWARE 0
TURN ON GREEN
LED (D2) ON
M.STD2 MODULE
FAULT = 23 OCTAL
M EXECUTE READ! FAIL STATE INIT FAULT FAULT = 23; OCCURS ONLY IF
CALIBRATE TEST ---
ON RX33 DRIVE 0 BOTH DRIVES FAIL
FAULT = 23 OCTAL
N READ FIRST 8 BLOCKS FAIL STATE INIT FAULT
--- FAULT = 23; OCCURS ONLY IF
FROM RX33 (BOOT BOTH DRIVES FAIL
BLOCKS) 0
0 TRANSFER CONTROL
TO IMAGE JUST
LOADED
CXO-945C
Sheet 2 of 5

SYSTEM
DISKETTE
INIT LED TURNED OFF,

STATE LED TURNED ON
SOLID, HSC CONSOLE O/P
INIPIO-I-BOOTING;
LOAD REMAINDER OF
INIPIO.INI
FAULT = 21 OCTAL
INIPIO PERFORMS FAIL STATE INIT FAULT
INSTRUCTION TESTS
ON OFF ON
AND MMU TESTS
INIPIO LOADS
INICAC AND
TRANSFERS CONTROL
INICAC TESTS CACHE;

IF CACHE FAILS, FLAGS
FAILURE TO INIPIO
INIPIO INITS ALL

REQUESTORS AND GETS
THEIR STATUS
FAULT = 22 OCTAL
INIPIO TESTS PROG MEM;
HIGHEST REQUESTOR FAIL STATE INIT FAULT TOTAL MEMORY FAILURE
NUMBER TESTS CONTROL ON
---
OFF ON IN CONTROL OR DATA
AND DATA MEMORY
FAULT = 23 OCTAL
INIPIO LOADS EXEC;
FAIL STATE INIT FAULT FAULT OCCURS IF BOOT
INIPIO TURNS ON
DEVICE HAS ERROR WHEN
GREEN LED ON ON OFF ON LOADING EXEC
P.IOJ MODULE
CXO-945C
Sheet 3 of 5

~
~
INIPIO TRANSFERS TO
FAIL ..
STATE INIT FAULT
- - - -ON -
EXEC, STARTS STATE ,. SOLID OFF MOST REMAINING FAULTS
LIGHT BLINKING AT 1/2- ONOR INDICATE SOFT FAULTS
SECOND INTERVALS OFF
t FAULT CODE DEPENDE NT ON FAILURE

~ EXEC RUNS SINI; SINI
LOADS AND INITIALIZES FAIL - - - -ON -
STATE INIT FAULT
SOLID OFF
REMAINING SIW ONOR
MODULES OFF
t
~ SIN I TRANSFERS
COMPLETELY TO EXEC,
STATE LIGHT BLINKS FAIL
SAM E AS ABOVE
AT 1-SECOND INTERVALS; I
OUTPUT OPERATING
SOFTWARE HERALD
NOTE: AFTER THE OPERATING SOFTWARE HERALD, OTHER INITIALIZATION MESSAGES

MAY BE REPORTED.
CXO-945C
Sheet 4 of 5

OFF-LINE
DISKETTE
TURNS INIT INDICATOR

OFF; TURNS STATE
INDICATOR ON SOLID
LOADS REST OF OFF-LINE

P.IOJ TEST (OFLPIO)
STATE INIT FAULT

FAIL ON ON OFF
ERROR TYPEOUT
OR HALT AT 400
LOADS OFF-LINE
DIAGNOSTIC LOADER
(ODL)
STARTS ODL, BLINKS

STATE INDICATOR; ODL
HERALD TO TERMINAL
OG OOL PROMPT WAITS FOR

OPERATOR COMMAND,
ROTATES OCP LAMPS
FOR TEST
ODL FEATURES
8 TESTS 11 CONVENIENCES
BUS SIZE
MEM HELP
MEM BY K @
K TEST SEL LOAD
OCP START
REFRESH SET DEFAULT
CACHE SHOW OEFAULT
RX33 SET RELOCATION
EXAMINE
DEPOSIT
REPEAT
NOTE: FIRST PORTION OF THE OFLPIO TESTS WAS LOADED WITH

THE PREVIOUS LOAD OF EIGHT BOOT BLOCKS.
CXO-945C
Sheet 5 of 5
Figure 8-17 HSC Boot Flow and Troubleshooting Chart

8.3.6 HSC50 Flow and Troubleshooting Chart

The HSC50 boot flow and troubleshooting chart calls out useful visual milestones that aid in
troubleshooting the problems which can occur during initialization.
The flowchart has three main divisions:
1. Information on activity common to both the system and off-line media.
2. Information on activity specific to the system media.
3. Informa tion on activity specific to the off-line media.
The flowchart begins when one of the following occurs:
• Init button is pushed.
• Powerup has started.
• Other software caused reboot.
Figure 8-18 maps the entire HSC50 boot sequence.
FAULT CODE CHART 1

OCTAL
CODE FAILURE
01 K.PLI ERROR
02 K.SDI/K.SI INCORRECT VERSION
OF MICROCODE
03 K.STI/K.SI INCORRECT VERSION
OF MICROCODE
21 P.IOJ/C MODULE FAILURE
22 M.STD2 MODULE FAILURE
23 BOOT DEVICE FAILURE
25 PORT LINK NODE ADDRESS
EXTERNAL EVENTS SWITCHES OUT OF RANGE
INTERNAL EVENTS 26 MISSING FILES REQUIRED
- POWER UP
- HSC S/W CRASH 30 NO WORKING K.CI, K.SDI, K.STI,
- INIT SWITCH
- MODULE FAILURE OR K.SI IN SUBSYSTEM
- HOST RESET REQUEST
31 INITIALIZATION FAILURE
32 SOFTWARE INCONSISTENCY
TIME
TURN ON P.IOC
TO - CHECK POWER (4)
RED LED (3)
TURN OFF ALL 1 - - _..... FAULTY P.IOC OR

OCP INDICATORS POWER SUBSYSTEM
FAULT CODE = NONE

TEST 0 - TEST F11 ,"-_~...I _ST_A_T_E _IN_IT _FA_U_L_T POSSIBLE FRUs
BASIC INSTRUCTION SET OFF OFF OFF P.IOC OR POWER
~------------~
FAULT CODE = 21 OCTAL
TEST 0 - TEST MORE STATE INIT FAULT
F11 INSTRUCTION SET
~-~~ ---- -- ----
OFF OFF ON
P.IOC TURNS ON
T <112 SECOND INIT INDICATOR

TEST 1 - TEST BOARD FAIL STATE INIT FAULT
T>1/2 SECOND AND BANK SWAP BITS
IN P.IOC CSR OFF ON ON

TEST 2 - TEST FIRST 1 KW STATE INIT FAULT
OF PROGRAM MEMORY
~--.....
OFF
-----
- ----
ON ON
NOTES:
1. Ks (REQUESTERS) = K.SDI, K.STI, K.PLI.
2. ALL MODULE RED INDICATORS ON EXCEPT MEMORY (NO RED LED ON IT).
3. REFER TO POWER SUPPLY TROUBLESHOOTING FLOWCHART.
CXO-OS2C
Sheet 1 of 4
Figure 8-18 (Cont.) HSC50 Boot Flow and Troubleshooting Chart


TEST 3 - FIND GOOD 8 KW FAIL STATE INIT FAULT
CHUNK OF PROGRAM
MEMORY OFF ON ON

TEST 4 - LOOK FOR LOAD FAIL STATE INIT FAULT
DEVICE; LOAD 8 BOOT
BLOCKS OFF ON ON
OFF-LINE SYSTEM
3 ~----------~--------~
TAPE TAPE
TURNS INIT INDICATOR OFF,

TURNS STATE INDICATOR
ON SOLID
OUTPUTS TO TERMINAL,
INIPIO-I-BOOTING,
LOADS REST OF INIPIO (1)
FAULT CODE DEPENDENT
ON FAILURE (2)
RUNS INIPIO (INITS
REQUESTORS AND GETS
THEIR STATUS)
FAIL
~--~ ----
ON
-- ----
STATE INIT FAULT
OFF ON
LOADS OPERATING SOFTWARE

WHILE CONTROL AND DATA
MEMORY ARE TESTED BY
HIGHEST NUMBER REQUESTOR;
STATE INDICATOR BLINKS AT
1/2-SECOND INTERVALS
P.IOC TESTS REMAINDER OF

PROGRAM MEMORY AND
CREATES BAD MEMORY
LINKED LIST
NOTES:
1. FIRST PORTION OF INIT P.lOC TESTS (INIPIO) WAS LOADED WITH PREVIOUS
LOAD OF EIGHT BOOT BLOCKS.
2. FOR DETAILED INFORMATION ON INIPIO TESTS AND ERROR REPORTS,
REFER TO HSCSO IN-LINE DIAGNOSTICS USER DOCUMENTATION.
CXO-OS2C
Sheet 2 of 4
Figure 8-18 (Cont.) HSCSO Boot Flow and Troubleshooting Chart

TURNS ON P.IOC GREEN LED
FAULT CODE DEPENDENT

ON FAILURE
STARTS OPERATING SOFTWARE; STATE INIT FAULT
BLINKS STATE INDICATOR AT 1- FAIL
ON OR OFF ON
SECOND INTERVALS; OPERATING
OFF
SOFTWARE HERALD (1)
OPERATING SOFTWARE WILL

NOT PROMPT UNTIL REQUESTED
BY CTRUY; THE PROMPT IS:
HSCSO>
NOTE: AFTER THE OPERATING SOFTWARE HERALD, ONE OF SEVERAL NONFATAL

INITIALIZATION FAILURE MESSAGES MAY BE PRINTED:
- REQUESTOR n FAILED TO INIT STATUS = XXX

THIS MESSAGE INDICATES SPECIFIED REQUESTOR FAILED ITS INTERNAL
INITIALIZATION SELF-TEST OR COULD NOT BE LOADED WITH MICROCODE (K.SI).
- SWAP BANK BIT SET
THIS MESSAGE INDICATES THAT A GOOD CONTIGUOUS SECTION OF 8 KW

PROGRAM MEMORY COULD NOT BE FOUND WITHOUT USING THE SWAP
BANK BIT IN THE P,.IOC CSR. THE MEMORY MODULE IS SUSPECT.
CXO-OS2C
Sheet 3 of 4
Figure 8-18 (Cont.) HSC50 Boot Flow and Troubleshooting Chart

OFF-LINE
TAPE
TURNS INIT INDICATOR OFF,

TURNS STATE INDICATOR
ON SOLID
NO FAULT CODE (2)
STATE INIT FAULT
LOADS REST OF OFF-LINE FAIL
P.IOC TESTS (OFLPIO) (1) ON ON OFF
ERROR TYPEOUT OR
HALT AT 400
RUNS OFF-LINE P.lOC TESTS

(OFLPIO)
LOADS OFF-LINE DIAGNOSTIC

LOADER (ODL)
TURNS ON P.IOC GREEN LED
STARTS ODL,
BLINKS STATE INDICATOR,
ODL HERALD TO TERMINAL
ODL PROMPT WAITS FOR

OPERATOR COMMAND,
ROTATES OCP LAMPS FOR TEST
ODL FEATURES
6 TESTS 11 CONVENIENCES
BUS SIZE
MEM HELP
MEM BY K @
K TEST SEL LOAD
OCP START
REFRESH SET DEFAULT
SHOW DEFAULT
SET RELOCATION
EXAMINE
DEPOSIT
REPEAT
NOTES:
1. FIRST PORTION OF THE OFLPIO TESTS WAS LOADED WITH PREVIOUS LOAD OF
EIGHT BOOT BLOCKS.
2. REFER TO FAULT CODE CHART. FOR DETAILED INFORMATION ON INITPIOC TESTS,
REFER TO THE HSCSO IN-LINE DIAGNOSTICS USER DOCUMENTATION.
CXO-OS2C
Sheet 4 of 4
Figure 8-18 HSC50 Boot Flow and Troubleshooting Chart

8.3.7 Boot Diagnostic Indications

The HSC can pass boot diagnostics with a failing requestor. Although the HSC passed the boot, the
failure associated with the requestor is considered an initialization error.
Following is an example of an error message displayed when a requestor fails on initialization
of the operating software. The HSC has passed most of the initialization/boot diagnostics, but a
requestor has failed.
SINI-E ERROR SEQUENCE 2. AT 20-SEPT-1985 00:00:02.80
REQUESTOR 2 FAILED INIT DIAGS, STATUS = 107
The requestor with the red LED ON is the failing requestor. In this case, the diagnostic identifies
requestor 2 as failing its internal self-test number 7. Additionally, the fault indicator turns on,
and a soft fault code of octal 12 is displayed on the OCP after the Fault switch is pressed. Refer to
Appendix C for a listing of STATUS = nnn codes.
See Section 8.3.1 for more information on errors indicated by the OCP.
8.4 Software Error Messages

Software error messages are classified into three categories:
_ 1. MSCPtrMSCP errors
2. Bad block replacement errors (BBR)
3. Out-of-band errors
This section explains the different error types in each category, and shows examples of console error
formats for each type in a category.
8.4.1 Mass Storage Control Protocol Errors

~e Mass Storage Control Protocol (MSCP) or Tape Mass Storage Control Protocol (TMSCP) errors
pnnted out at the console terminal and reported to a host can be one of the following types:
1. Controller errors
2. SDI errors
3. Disk transfer errors
4. STI communication errors
5. STI formatter errors
6. STI drive errors
8.4.2 MSCpnMSCP Error Format, Description, and Flags

Error formats, descriptions of the fields within the error format, and error flags are nearly identical
for MSCP and TMSCP errors. Differences are noted where they exist. See Section 8.5 for listings
and explanations of controller errors.
8.4.2.1 Error Format

Example 8-1 shows the generic error format for all MSCPII'MSCP errors. Optional lines may be
used with some errors to display additional information.
ERROR-X Text of message at (date) (time)

Command Ref # xxxxxxxx
Err Seq # x.
Format Type xx
Error Flags xx
Event xxxx
(Optional line)
(Optional line)
(Optional line)
ERROR-X End of error.
Example 8-1 MSCPITMSCP Error Message Format
8.4.2.2 Error Message Fields

Table 8-8 describes the various fields found in an MSCPITMSCP error message. These are common
fields to all error messages of this type.
Table 8-8 MSCPfTMSCP Error Message Field Descriptions

Field Description
ERROR-X The X is a code indicating the severity level of an error. The codes are: E for non-
fatal, Q for inquiry, I for informational, F for fatal, W for warning, and S for success.
NOTE
Only severity levels E and Q require user action.
Information following the severity level code is a textual version of the error message
describing the event code, followed by the date and time.
Command Ref # This nwnber (in hexadecimal) is the MSCPfI'MSCP command number which caused
the reported error. It is zero if the error does not correspond to a specific outstanding
command. This number is normally assigned by the issuing host CPU.
Err Seq # This number (in decimal) is a sequential number which counts error log messages
since the MSCPtrMSCP server established a connection with the host. It is zero if
the MSCPfI'MSCP server does not implement error log sequence numbers.
Format Type This number (in hexadecimal) is the byte that describes the detailed format of the
error log message. Table 8-9 defines the format type codes. Format Type xx basically
defines the type of error packet.
Error Flags This number (in hexadecimal) indicates bit :flags, collectively called error log message
11ags, used to report various attributes of the error. Refer to Table 8-10.
Event This number (in hexadecimal) identifies the specific error or event being reported by
this error log message. This code consists of a 5-bit major event code and an 11-bit
subcode. The event codes and their meanings are listed in Appendix C.
8.4.2.3 Format Type Codes

Table 8-9 defines the format type code numbers. The format type code numbers are in
hexadecimal.
Table 8-9 MSCPITMSCP Error Message Format Type Code Numbers

Number Definition
00 Controller errors
01 Host memory access errors with memory address
02 Disk transfer errors
03 SDI errors
04 Small disk errors
05 Tape transfer errors
06 STI errors
07 STI drive error log
08 STI formatter error log
09 Bad block replacement
8.4.2.4 Error Flags

Table ~10 defines the MSCPtrMSCP error flags.
Table 8-10 MSCPITMSCP Error Flags

Bit
Bit Mask
Number Hex Format Description
7 80 If set, the operation causing this error log message has successfully completed.
The error log message summarizes the retry sequence necessary to successfully
complete the operation.
6 40 If set, the retry sequence for this operation continues. This error log message
reports the unsuccessful completion of one or more retries.
5 20 This is MSCP-specific. If set, the identified logical block number (LBN) needs
replacement.
4 10 This is MSCP-specific. If set, the reported error occurred during a disk access
initiated by the controller bad block replacement process.
o 1 If set, the error log sequence number has been reset by the MSCP server since the
last error log message was sent to the receiving class driver.
8.4.2.5 Controller Errors

Example 8-2 is a example of a typical MSCPITMSCP controller error.
ERROR-E Data memory error (NXM or parity) at 5-Mar-1985 12:52:14.43

Command Ref # 1C430008
Err Seq
Error Flags
* 1.
41
Format Type 00
Event 012A
Buffer Addr 143611
Source Req. O.
Detecting Req. 3.
Example 8-2 Controller Error Message Example

The direction of data transfer may be deduced from the types of requestors identified in the Source
Requestor and Detecting Requestor fields of the error message. These fields correspond to the
requestor slot in the HSC backplane. In this example, the source requestor number shows it to be
a P.ioj/c, which filled the buffer. The detecting requestor number is 3, which is reading the buffer.
This section lists controller and compare errors together because their format and fields are the
same. These errors contain three optional fields in addition to those described in Table 8-8. The
controller/compare specific fields are shown in Table 8-11.
Table 8-11 MSCPITMSCP Controller Error Message Field Descriptions

Field Description
Buffer Addr This number (in octal) is the starting address of the HSC Data Buffer where
the error occurred.
Source Req. This is the number (in decimal) of the requestor that originally filled the
buffer with data.
Detecting Req. This is the number (in decimal) of the requestor that detected the error.
8.4.2.6 MSCP SOl Errors

The SDI-type errors total 15. Example ~3 shows a typical SDI error message. Table ~12
describes the fields specific to SDI errors. Table ~13, Table ~14, Table ~15, and Table ~16
further define the fields in Table ~12. For the remaining fields, refer to Table 8-8. For listings
and explanations of SDI type errors, see Section 8.5.
ERROR-E Drive Detected Error at 5-Mar-1985 12:52:14.43

Command Ref 00000000 '*
RA81 unit '*
124.
Err Seq '* 4.
Error Flags 40
Format Type 03
Event OOEB
Request 1B
Mode 00
Error 80
Controller 00
Retry/Fail 00
Extended Status 88
00
03
00
07
4B
1A
Requestor 6.'*
Drive port 2. '*
Example 8-3 MSCP SOl Error Example

Table ~12 describes the SDI error example fields.
Table 8-12 MSCP SOl Error Field Descriptions

Field Description
RA81 unit # This is the number of the unit the error log message relates to, or is 4095 if the unit
number is unknown. In this example, the RA81 indicates the drive is an RA81 and
is unit 124.
Request This number (in hexadecimal) is a byte describing the various requests from the drive
for controller action. Figure 8-19 shows the bits of this byte field, and Table 8-13
describes the bits. In this example, the 1B indicates:
• RUN/STOP switch in
• Port switch in
• Logable information in extended area
• Spindle ready
Table 8-12 (Cont.) MSCP SO. Error Field Descriptions

Field Description
Mode This number (in hexadecimal) is a byte describing the mode of the unit. These modes
can be altered by the controller. Figure 8-20 shows the bits of this byte field, and
Table 8-14 describes the bits. In this example, the 00 indicates:
• No subunits are write-protected.

• The disk is in 512-byte sector format.
Error This number (in hexadecimal) is a byte describing the the current drive error
conditions that prevent normal drive operations. Figure 8-21 shows the bits of
this byte field, and Table 8-15 describes the bits. In this example, the 80 indicates a
drive error has occurred, and the drive Fault lamp may be on.
Controller This· number (in hexadecimal) is a byte describing the subunits with attention
available messages suppressed in the controller and a status code indicating
various states of drive operation. Figure 8-22 shows the bits of this byte field,
and Table 8-16 describes the bits. In this example, the 00 indicates:
• No subunits with attention available message suppressed in the controller.

• Drive normal operation.
RetrylFail This number (in hexadecimal) is a byte containing one of two types of information
depending upon the status of the DF bit in the error field. The DF bit describes
the drive initialization process. The DF hit is a zero if the drive initialization
was successful. In this case, the RetrylFail field contains the retry count from
the previous operation. For example, a Seek operation required 14 retries to be
successful. If a GET STATUS command is initiated, the RetrylFail field contains the
number 14.
The DF bit set indicates the drive initialization failed, and therefore, the RetrylFail
contains a specific drive error code. This error code is defined in the appropriate
drive service manual.
In this example, 00 indicates no retry count exists for the previous operation. (The
DF bit is zero in the Error field.)
Extended status These bytes (in hexadecimal) contain the extended status of the particular drive. (In
this example it is an RA81.) Refer to the appropriate drive service manual for the
meaning of these bytes.
Table 8-12 (Cont.) MSCP SOl Error Field Descriptions

Field Description
In this example, the extended status is:
• 8~Controner command functional code last executed by the drive. (In this
case, a GET SUBUNIT CHARACTERISTICS command.)
• OO--Interface error status bits which are all reset.
• 03-Low-order cylinder address bits of the last Seek operation.
• OO-High-order cylinder address bits of the last Seek operation.
• 07-The present group address.
• 4B-Error code (index pulse error) displayed by the drive LEDs during the
execution of a drive-resident diagnostic.
• lA-Error code (servo fine positioning error) displayed on the OCP of the RABI.
Requestor # This number (in decimal) is the number of the requestor connected to the drive.
Drive port # This number (in decimal) is the number of the port on the requestor. (The ports are
numbered 0 through 3.)
OA RR DR SR EL PB PS RU
CXO-1121A
Figure 8-19 Request Byte Field
Table 8-13 Request Byte Field Descriptions

Bit Description
OA A logical 1 in this position indicates the drive is unavailable to the controller. A logical 0
indicates the drive is available to the controller.
RR A logical 1 in this position indicates the drive requires an internal readjustment. Some drives
do not use this bit.
DR A logical 1 in this position indicates a request is outstanding to load a diagnostic in the drive
microprocessor memory. A logical 0 indicates no diagnostic is being requested of the host
system.
SR A logical 1 in this position indicates the drive spindle is up to speed. A logical 0 indicates the
drive spindle is not up to speed.
EL A logical 1 in this position indicates usable information in the extended status area. A logical
o indicates no information is available in the extended status area.
PB A logical 1 in this bit position indicates the drive is connected to the controller through Port B.
A logical 0 indicates the drive is connected through Port A.
Table 8-13 (Cont.) Request Byte Field Descriptions

Bit Description
PS A logical 1 in this bit position indicates the drive port select switch for this controller is pushed
in (selected). A logical 0 indicates the switch is out.
RU A logical 1 in this position indicates the RUN/STOP switch is pushed in (RUN). A logical 0
indicates the switch is out (STOP).
W4 W3 W2 W1 DD FO DB 57
CXO-1122A
Figure 8-20 Mode Byte Field
Table 8-14 Mode Byte Field Descriptions

Bit Description
W4-W1 A logical 1 in any of these four bit positions represents the write-protect status for the subunit.
(For example, a 0001 indicates subunit 0 within the selected drive is write-protected.)
DD A logical 1 in this position indicates the drive was disabled by a controller error routine or
diagnostic. The fault light is on when this bit is set. A logical 0 indicates the drive is enabled
for communication with a controller.
FO A logical 1 in this position indicates the drive can be formatted.
DB A logical 1 in this position indicates the diagnostic cylinders on the drive can be accessed.
S7 A logical 1 in this position indicates the 576-byte sector format is selected. A logical 0 indicates
that the 512-byte sector format is selected.
DE RE PE OF WE
CXO-1123A
Figure 8-21 Error Byte Field
Table 8-15 Error Byte Field Descriptions

Bit Description
DE A logical 1 in this position indicates a drive error has occurred and the drive Fault lamp may
be on.
RE A logical 1 in this position indicates an error occurred in the transmission of a command
between the drive and the controller. The error could be a checksum error or an incorrectly
formatted command string.
Table 8-15 (Cont.) Error Byte Field Descriptions

Bit Description
PE A logical 1 in this position indicates improper command codes or parameters were issued to the
drive.
DF A logical 1 in this position indicates a failure in the initialization routine of the drive.
WE A logical 1 in this position indicates a write-lock error has occurred.
84 83 82 81 C1 C2 C3 C4
CXO-1124A
Figure 8-22 Controller Byte Field

Table 8-16 Controller Byte Field Descriptions

Bit Description
S4-S1 This is a 4-bit representation of the subUnits with attention available messages suppressed
in the controller. The right-most bit (Sl) represents subunit 1. The left-most bit (84)
represents subunit 4.
If one of the bits is set, it indicates the controller is not to interrupt the host CPU with
an attention available message when the specified subunit raises its available real-time
drive status line to the controller. The S4 through Sl bits reflect the results of a CHANGE
CONTROLLER FLAGS command in which attention available messages are not desired for
certain subunits.
C4-C1 This is a 4-bit drive status code indicating various states of drive operation.
NOTE
When the HSC marks the drive as inoperative, it places the drive in a state of Unit.-Off-
line with a substate of Unit-Inoperative relative to this HSC.
8.4.2.7 Disk Transfer Errors

Disk transfer errors are either data or media format type errors. Example 8-4 shows a disk
transfer error example, and Table 8-17 describes the various fields of the example. See Section 8.5
for listings and explanations of the disk transfer errors.
ERROR-E SEVEN Symbol ECe Error at 27-Mar-1ge5 12:15:15.00
**
Command Ref 50400015
RAe1 unit 120.
Err Seq .f 9.
Format Type 02
Error Flags EO
Event 01Ce
Recovery level O.
Recovery count O.
LBN 426978
Orig err flags 100020
Recovery Flags 000003
LvI A retry cnt 1.
LvI B retry cnt o.
Buffer addrs 143022
Source Req. 5.
Detecting Req. 5.
Error-I End of error.
Example 8-4 Disk Transfer Error Example

Table 8-17 describes the fields in a disk transfer error message not described in Table 8-8. Unless
otherwise specified, all fields in this table are shown in decimal numbers. These fields are specific
to an RA81 disk and may not be the same for other RAxx type drives.
Table 8-17 Disk Transfer Error Field Descriptions

Field Description
RA81 unit # This is the number of the -iInit the error log message relates to, or is 4095 if the unit
number is unknown. In this example, the RA81 indicates the drive is an RA81 and
is unit 120.
Recovery level This number indicates the drive error recovery level used for the most recent transfer
attempt by the unit. In this example, the 0 indicates it used error recovery level o.
An RA81 only has a recovery level of 0 (recalibration).
Recovery count This number indicates the number of times the drive recovery level was tried. In this
example, the 0 indicates the recovery level was not retried.
LBN This number indicates the logical block number. In this example, the LBN is 426978.
Original This number (octal) indicates the original errors associated with this error.
error flags Table 8-18 describes the bits associated with this field. In this example, the 100020
indicates:
• ECC error
• EDC error
Recovery flags This number (octal) indicates the recovery flags the software processes should take
to recover from this error. Table 8-19 describes the bits associated with this field. In
this example, the 000003 indicates:
• An LBN should be replaced.

• The current error should be logged on the console and to the host if a connection
is present.
LvI A retry This number indicates the number· of times the HSC attempted the level A recovery
count routines. These routines are those not requiring any exhaustive SI exchanges as part
of the recovery sequence. In this example, the 1 indicates the ECC error correction
was completed in the HSC without going over the SI.
LvI B retry This number indicates the number of times the HSC attempted the level B recovery
count routines. These routines require extensive SDI exchanges as part of the recovery
sequence. In this example, the 0 indicates no level B recovery was attempted.
Buffer address This number (octal) is the address of the HSC internal Data Buffer associated with
this error. In this example, the buffer address is 143022.
Source This number is the requestor that filled the buffer with data. In this example, the 5
Requestor indicates the source requestor was requestor number 5. A requestor of 1 in this field
would indicate a disk Write operation. All other values would indicate a disk Read
operation.
Detecting This number is the requestor that detected that error. In this example, the 5
Requestor indicates requestor number 5 detected the ECC error.
Table S-18 shows definitions of the original error £lags and Table S-19 defines the recovery flags.
Table 8-18 Original Error Flags Field Descriptions

Bit Mask (Octal) Definition
15 100000 ECC error

14 040000 SERDES overrun error
13 020000 SDI ResponselData line pulse error
12 and 11 014000 Suspected position error-low header mismatch
12 010000 Header sync timeout
11 004000 Header compare error--compare-64 performed (high header
mismatch)
10 002000 Data sync timeout
09 001000 Drive clock timeout
08 000400 SDI State line pulse or parity error
07 000200 Data bus overrun
06 000100 Data memory parity error
05 000040 Data memory NXM
04 000020 EDC error
03 and 02 000014 ReadIWrite Ready down at end of sector
03 000010 Lost Read/Write Ready before transfer began
02 000004 Lost Receiver Ready before transfer began
01 000002 Forced error (EDe = 1's complement of correct EDC)
00 000001 Drive inoperative
Table 8-19 Recovery Flags Field Definitions

Bit Mask (Octal) Definition
07 000200 Indicates a revector was done for this LBN.

06 000100 Indicates a positioner error was detected on this block.
05 000040 Indicates the error count reported by the ILEXER should be updated.
04 000020 Indicates an error log message has already been generated for the
current error.
03 000010 Indicates an RCT entry for the desired logical block number was
found.
02 000004 Indicates revectoring and replacement should be suppressed.
01 000002 Indicates the current error should be logged on the console and to the
host if a connection is present.
00 000001 Indicates the logical block should be replaced.
8.4.3 Bad Block Replacement Errors (BBR)

Another type of error displayed on the console terminal is for a bad block replacement request. The
bad block replacement request is a result of the one of the following errors:
• Data sync timeout
• ECC symbol error above the threshold
• Header compare error
• Header sync timeout
• Loss of RIW Ready at end of read from disk (SERDES read)
• Uncorrectable ECC
Example 8-5 shows a bad block replacement message. This message reports completion, successful
or unsuccessful, of a bad block replacement attempt. A message is generated regardless of the
success or failure of the replacement attempt. See Section 8.5 for listings and explanations of BBR
errors.
ERROR-W Bad Block Replacement (Success) at 18-Dec-1985 18:05:37.1

Command Ref
RA60 Unit **
B8590012
251
Err Seq
Format Type
* 2
09
Error Flags 80
Event 0014
Replace Flags 8000
LBN 205
Old RBN 0
New RBN 5
Cause Event 00E8
ERROR-I End of error
Example 8-5 Bad Block Replacement Error Example

Table 8-20 defines BBR error fields not previously described in Table 8-8.
Table 8-20 Bad Block Replacement Error Field Definitions

The replace flags field bits are defined in Table 8-21.
Field Description
Replace Flags This number (in hexadecimal) indicates bit flagS used to report in detail the outcome
of the bad block replacement attempt. In this example, the 8000 indicates the block
was verified as bad.
LBN This number (in decimal) is the logical block number that is the target of the
replacement. In this example, the LBN is 205.
Old RBN This number (in decimal) indicates the RBN the bad LBN was formerly replaced
with, or zero if it was not formerly replaced. In this example, the 0 indicates it was
not formerly replaced.
NewRBN This number (in decimal) indicates the RBN the bad LBN was replaced with, or is
zero if no actual replacement was attempted. In this example the new RBN is 5.
Table 8-20 (Cont.) Bad Block Replacement Error Field Definitions

Field Description
Cause Event This number (in hexadecimal) is the event code from the original error that caused
the replacement to be attempted. The number is zero if that event code not available.
Refer to Appendix C for a listing of generic error log fields. In this example, the OOE8
indicates an uncorrectable ECC error caused the bad block replacement.
Table 8-21 Replace Flags Field Bit Descriptions

Bit Bit Mask
Number (Hex) Flag Bit Definition
15 8000 Replacement attempted-This bit is set if the suspect bad block indeed tested
bad during the initial stages of the replacement process. If not set, the suspect
block did not check bad and no replacement was completed.
14 4000 Forced error-The data from the suspect bad block could not be corrected
or obtained without error. The Forced Error Indicator will be written to the
replacement block along with the bad data from the block that was replaced.
The user data from the bad block is read with a forced error when accessed. If
this condition occurs frequently on a specific drive, then a closer analysis of the
drive for possible problems is recommended.
13 2000 Nonprimary revector-This bit is set if the replacement process was
accomplished and required putting the bad block data into a replacement
block that is not the bad block's primary RBN.
12 1000 Reformat error-This bit is set during the replacement process if the status
coming back from the execution of the MSCP REPLACE command is not
successful. If this occurs, the drive should not be used until it is reformatted.
NOTE: The HSC does not use the REPLACE command as it initiates its
own BBR. This message is printed for the HSC equivalent of the REPLACE
command such as FORMAT SECTOR.
11 800 RCT inconsistent-This bit is set if the Replacement Control Tables are not
usable. The drive should not be used until it can be reformatted.
10 400 Bad replacement block-This bit is set if the bad block reported is a
replacement block. The replacement block can be replaced just like any other
LBN.
8.4.4 TMSCP Errors

The Tape Mass Storage Control Protocol (TMSCP) error messages printed out at the console
terminal are one of the following types:
• STI communication or command errors
• STI formatter error log errors
• STI drive error log errors
• Controller errors. (Refer to Section 8.4.1)
See Section 8.5 for listings and explanations of tape errors.
8.4.4.1 STI Communication or Command Errors

Example S-6 is a sample console printout of an STI communication or command error. Table 8-22
explains the fields not previously defined in Table 8-8.
ERROR-E Drive detected error at 6-Mar-1985 09:51:11.88

Conunand Ref '*
864EOO04
TA78 unit '*
0
Err Seq '* 12
Error Flags 40
Event OOEB
Position 13026
GSS Text 02 00 00 00
05 00 00 00 00 00 00 00
Error-I End of error
Example 8-6 STI Communication or Command Error Example

The following table explains the error fields:
Table 8-22 STI Communication or Command Error Printout Field Descriptions

Field Description
Event The number (in hexadecimal) identifies the specific error or event reported by this error log
message. The event codes and their meanings are shown in Appendix C. In this example, the
OOEB means drive-detected error.
Position This is the last known tape position the formatter received. This is given in gap counts from
BOT. In this example, the number 13026 means 13026 gaps from BOT.
GSS Text The GSS Text field is the response received by the HSC from the formatter when the HSC
issues the GET SUMMARY STATUS (GSS) and TOPOLOGY commands. The GSS text in this
example is 02 00 00 00 05 00 00 00 00 00 00 00. This means level 2 protocol error, Speed
Management Enabled, and Zero Threshold. See Section 8.4.4.5 for details on field definitions
and bit decoding.
8.4.4.2 STI Formatter Error Log

The following is an example of the console printout of an STI formatter error log. Example 8-7
shows the example, and Table 8-23 explains the fields not previously defined in Table 8-8.
ERROR-E Tape Formatter Requested Error Log at 30-Jan-1986 11:20:09.31

Command Ref # 43900012
TA81 unit :# 95
Err Seq :# 47
Format Type 08
Error Flags 40
Event FF6C
Position 1057
Formatter E Log 40 00 00 81 00 00 00
01 98 72 00 00 00 00
C4 48 00 00
Example 8-7 STI Formatter Error Log Example

The following table explains the error log fields:
Table 8-23 STI Formatter Error Log Field Descriptions

Field Description
Position The last known tape position the formatter received. This is given in gap counts
from BOT. In this example, the number 1057 means 1057 gaps from BOT.
STI Formatter Error See Table 8-24.
Log
Table 8-24 STI Formatter E Log

Byte Byte
No. Data Description
1 40 Formatter error
2 00 Not set for this example
3 00 Not set for this example
4 81 Data pulse parity error during data transfer
The information contained in these fields is product specific. Refer to the
appropriate drive manual for a description of the remainder of the bytes.
8.4.4.3 STI Drive Error Log

The following is an example of a console printout of an STI drive error log. Example 8-8 shows the
example. Table 8-26 describes GEDS Text field, and Table 8-27 describes the Drive Error Log field.
ERROR-E Tape Drive Requested Error Log at 5-Mar-1985 14:43:31.15

Command Ref
TA78 unit * *D6300023
528
Err Seq
Error Flags
* 210
40
Event FF6B
Position 1
GEDS Text 7D 04 5000 01000000
Drive Error Log 00 00 00 00 50 3B 04 00
46 FF 07 FF 00 00 00 00
81 00 00 00 FF 22 04 C4
00 00 80 FF 17 94 00 08
00 00 D9 FF FF FF FF FF
FF 47 E6 EO 00 16 25 97
A2 00 00
ERROR-I End of error
Example 8-8 STI Drive Error Log Example

The following table explains the error log fields:
Table 8-25 STI Drive Error Log Field Descriptions

Field Description
Position The last known tape position where the HSC believes the tape drive is, upon
successful completion of all outstanding commands. This is given in gap counts
from BOT. In this example the number 1 means 1 gap from BOT.
GEDSText See Table 8-26.
Drive Error Log See Table 8-27.
Refer also to Section 8.4.4.4 for field definitions and bit decoding.
Table 8-26 GEDSText

Byte No. Byte Data Description
1 7D 125 ips tape drive

2 04 6250 bpi GCR encoding
3 50
4 00 MSCP unit number =80
5 01 Gap count =1
6 00
7 00
8 00
The infonnation shown in Table 8-27 is product specific to the TA78. See the TA78 Service Manual
for details.
Table 8-27 STI Drive Error Log (TA78 Drive Product Specific)
1,2 00 No soft error

3 00
4 00
5 50 Set byte count
6, 7 3B,04 Operational error
Error ID 59 =
CRC error
ACRC error
Pointer mismatch
U :n.correctable or two-track error set in ECCSTA register
Unknown fault number
8 OD RMC write fail bits
9 46 Statistics select clock stopped
STATUS VALID
Table 8-27 (Cont.) STI Drive Error Log (TA78 Drive Product Specific)
10 FF Non-BOT command status is OK

11 07 Last cmd sent to M8953 through
RCMD =normal NON-BOT read
12 FF Read channel AMTIE status (CH 7:0)
13 00
14 00 Read channel illegal status (CH 7:0)
15 00
16 00 End mark for read channels 7:0
17 81 Weak amplitude on parity bit
ECC corrected output (parity bit)
18 00 Read channel PE postamble detect
19 00 Data from read channels to ECC
20 00 CRC checker output bits
21 FF Corrected data (ECC to CRC)
22 22 Two-track ECC performed on data
AMTIE during data of record
23 04 Channel 0 tie bus 2
Amplitude track in error AMTIE
24 C4 Channel 3 tie bus 3
25 00
26 00
27 80 Tie bus =OF(X)
28 FF Tape unit bus line AMTIE 7:0
29 17 AMTIE parity
READ parity
WCS parity
Tape unit present
30 94 TU bus line read data 7:0
31 00 STI bus error byte
32 08 CRC to WMC DR bus
33,34 00 Tape unit selected =0
35 D9 R!W Data, intermediate DRD bus
36,37 FF Byte count = 65535
38,39 FF PAD counter = 65535
40,41 FF Unknown error code
42 47 DR MBD parity error
43 E6 PE write parity error
POWER OK
44 EO Tape unit ready and on line
Table 8-27 (Cont.) STI Drive Error Log (TA78 Drive Product Specific)
45 00
46 16 125 ips tape drive
47,48 25,97 Tape unit serial #2597
49 A2 AMTIE threshold field =2
READ ENABLE
Write BIT 4
50 00
51 00
8.4.4.4 Breakdown of GEDS Text Field

The following is an example of a tape drive-related error message printed on the HSC terminal:
ERROR-W Tape Drive Requested Error Log at 15-Aug-1984 18:43:05.80

Command Ref
TA78 unit * *
00001D8E
20.
Err Seq
Error Flags
* 1.
40
Event FF6B
Position 2.
GEDS Text 7D 02 0014 00000002
Drive Error Log 00 00 00 00 C5 38 04 04
46 FF 07 FF 00 00 00 00
81 00 00 21 FF BO 00 04
00 00 80 FF 17 DE 00 08
00 00 21 FF FF 00 00 99
99 47 F4 E8 00 56 85 19
A2 OA 80 FF 17 DE
Example 8-9 Tape Drive Related Error Message

Both the GEDS Text and Drive Error Log portions of this message result from a GET EXTENDED
DRIVE STATUS command to the drive from the HSC. The Drive Error Log portion can be
interpreted by referencing the service manual for the appropriate tape drive. (The preceding
example is for a TA78 drive.)
Following is a breakdown of the information contained in the GEDS Text field. The left-most byte
is referenced as the first byte and the right-most byte as the eighth byte.
Bytes in the GEDS Text field are described in the following list:
• First byte =Speed-Currently set speed of the drive; drive speed is defined as an integer
value (in hex) in inches per second (ips) rounded down to the nearest integer. For a totally
variable speed drive, the speed returned is the lower bound on the range of permissible speeds.
In the example shown, this field contains a value of 7D, which corresponds to 125 ips.
• =
Second byte Density-The current operating density of the tape unit. Only one bit is set to
indicate the current operating density.
04 = 6250 bpi
02 = 1600 bpi
01 =800 bpi
• Third and fourth bytes =Unit nUDlber-Contain the drive unit number (in hex).
• Fifth through eighth bytes = Gap count-The formatter's gap count is from the beginning
of the tape to where the tape drive is. The contents of this field may differ from the Position
field in this error message. The HSC's gap count is contained in the Position field at the end of
successful completion of all outstanding commands.
8.4.4.5 Breakdown of GSS Text Field

Following is another example of a tape drive-related error message printed at the HSC console:
ERROR-E Drive detected error at 18-Aug-1984 12:05:34.82

Command Ref # 0346003
TA78 unit # 3.
Err Seq # 7.
Error Flags 40
Event OOEB
Position o.
GSS Text 02 20 00 00
28 00 00 00 00 00 14 00
Example 8-10 Additional Tape Drive-Related Error Message

The HSC receives the GSS Text field form of this error message from the tape formatter when the
HSC issues the GET SUMMARY STATUS (GSS) and TOPOLOGY commands. The field is also the
unsuccessful response for all Level 2 commands. Figure ~23 is a breakdown of this response.
AF A3 A2 A1 AO OA PS DR SUMMARY MODE BYTE 1

FE TE PE DF CE SUMMARY ERROR BYTE
AC PB EL RP RT FD SUMMARY MODE BYTE 2
C1 C2 C3 C4 C5 C6 C7 C8 CONTROLLER BYTE
TM EOT BOT WL OL AV MR EL DRIVE 0 MODE BYTE
DE LP PL EX DTE SME DI ZT DRIVE 0 ERROR BYTE
CXO-2116A
Figure 8-23 GSS Text Field Bits Summary Breakdown
8.4.4.6 GSS Text Field Bit Interpretation

An interpretation of the GSS text field bits follows:
• AC: Cache attention
• AF: Formatter attention asserted
• AS: Drive 3 attention asserted
• A2: Drive 2 attention asserted

• AI: Drive 1 attention asserted
• AO: Drive 0 attention asserted
• ASM: Automatic speed management
• AV: Drive available to formatter
• BM: Block mode
• BOT: Beginning of tape
• CB: Cache busy
• CC: Cache capable
• CDL: Cache data list
• CE: Cache error
• CF: Cache full
• CMT: Cache empty
• Cn: Controller flags (CO - C8)-The following combinations are implemented. All other
combinations are reserved.
co: Normal operation
Cl: Formatter off line-Formatter is off line to hosts due to being under diagnostic
control.
• DE: Drive error-Asserted when any drive error not covered by other status bits is detected.
• DF: Formatter diagnostic failed
• DI: Diagnostic mode-When set, instructs the formatter to use special internal algorithms to
report imperfect performance.
• Dm: Direction-When clear, indicates the tape will be positioned in the forward direction.
• DR: Diagnostic requested-Asserted when the formatter is requesting permission to execute
a diagnostic.
• DTE: Data transfer error-Asserted when any error occurs which prevents a data transfer
from completing successfully.
• EL: Error logging request-Asserted by either the drive or formatter when error logging
information is available.
• EOT: End of tape-Asserted when the tape is positioned at or past the end of tape marker.
• ER: Erase-When set, indicates that a Rewind operation will erase the tape from the current
position forward to EOT before rewinding the tape.
• EX: Exception condition-Asserted whenever the formatter encounters TM, BOT, or EOT
during a data Transfer operation.
• FD: Retry Bit, failure/direction - Asserted during error recovery to indicate the direction of
a retry or to indicate a failing operation. If RP = 0 and RT = 1, then FD = direction to transfer.
FD = 0 means transfer in the same direction as original operation; FD = 1 means transfer in
the opposite direction of original operation. If RP = 1 and RT = 0, then FD indicates success
or failure of operation. FD = 0 means the retry sequence succeeded; FD = 1 means the retry
sequence failed.
• FE: Formatter error-Asserted on formatter errors not covered by the TE, PE, or DF bits.
• LP: Lengthy operation in progress-Asserted when a Rewind operation (including the

optional data security erase portion of a rewind) is in progress.
• LS: Long/short success time-When a formatter rejects a command with cache busy and
cache full, it also appropriately sets the LS bit. If the formatter thinks the rejected command
can be accepted if immediately issued, LS is clear. If not, LS is set. LS is clear if CB is cleared,
and therefore if CB is clear, LS must be clear.
• MR: Maintenance mode request-Asserted when the drive is put into maintenance mode.
On the TA78, this is accomplished through a thumbwheel switch on the operator panel.
• NR: No read-ahead-Set if read-ahead caching is disabled on this unit.
• OA: Formatter on line or available (for the TOPOLOGY command)
• OL: Drive on line to formatter
• PB: Active port button-PB = 0 if the formatter is connected to the controller through port A:,
PB = 1 if the formatter is connected to the controller through port B.
• PE: Level 2 protocol error-Asserted when a protocol error is detected while processing a
Level 2 command.·
• PL: Position lost-Asserted when the formatter is not certain of the current tape position.
• PR: Position for retry
• PS: Port switch-Asserted when the port switch is enabled.
• PI': Position for termination-Positions the tape to where it would have been had there been
no error and exits the error recovery state.
• RP: Request position-Used by the formatter along with RT to inform the controller of the
next step in the error recovery sequence.
Retryable RP = 1, RT =1
Transfer RP = 0, RT = 1
Done RP = 1, RT = 0
No Error RP = 0, RT =0
• RR: Read reverse is supported
• RT: Request transfer-Refer to the explanation for RP.
• RWC: Rewrite capable-This bit must not be set if CC is not set.
• RWE: Rewrite error recovery-Can only be set if CC is set.
• SG: Space gaps-Indicates a location where the tape operation will position the tape. This is
determined by the number of gaps specified in the gap field.
• SM: Speed mask-SM =0, supports up to four fixed speeds. 8M = 1, supports totally variable
speeds.
• SME: Speed management enabled-Asserted whenever the formatter may change the
current operating speed of a particular drive at any time (provided the changing of the drive
operating speed is transparent to the controller).
• SR: Space records-Positions the tape according to the number of records in the count field.
• TE: Transmission error-Used by the formatter to report level 0 and level 1 STI errors. The
formatter only reports level 0 real-time state parity errors and WritelCmd Data Line pulse
errors when a transfer is in progress. Level 1 errors are framing errors, checksum errors,
inappropriate value in data field of real-time command, or a real-time command occurring in an
invalid context.
• TM: Tape mark

• UO: Low order drive number bit-The drive to which a command applies.
• U1: High order drive number bit-The drive to which a command applies.
• UN: Unload-Unloads a tape after rewind.
• WB: Write back-Set if write back caching is enabled on this unit. CC also must be set.
• ~:Writelocked
• WP: Write protect-Set when the controller desires to illuminate the write protect light on the
selected unit.
• ZT: Zero threshold-Instructs the formatter to change all error thresholds from their default
values to zero.
NOTE
Always verify proper dc voltage levels if the indicated possible FRUs do not rectify
failure.
8.4.5 Out-of-Band Errors

The out-of-band errors are those not conforming to a specific template format, as the MSCP and
TMSCP errors do. The method of reporting differs for each of these errors.
The HSC operating software allows the setting of different levels of error reporting for out-of-band
type errors using the SETSHO utility. These message error levels are Informational, Warning,
Fatal, Error, and Success. The identifiers for the out-of-band errors are followed by an I, W, F, E, or
S, depending on the SETSHO value. The x in the following list represents the message error level.
Out-of-band errors are classified into the five categories listed below:
1. CI errors-Identified by HOST-x identifier printed prior to message
2. Load device errors-Identified by SYSDEV-x identifier prior to message
3. Disk functional errors-Identified by DISK-x identifier prior to message
4. Tape functional errors-Identified by TAPE-x identifier prior to message
5. Miscellaneous (software inconsistencies)-Identified by SINI-x identifier prior to message
See Section 8.5 for listings and explanations of the above categories of out-of-band errors.
NOTE
Some out-of-band errors report microcode-detected error status codes within the
printout. Refer to Appendix D for a full list of all K.ci, K.sti, K.sdi, K.si, and microcode-
detected errors.
NOTE
When replacing indicated FRUs, always verify correct dc voltage levels before and after
replacing a module.
8.4.5.1 RX33 Errors

Detected errors from the RX33 load device are classified in the out-of-band error category. The
following is an example printout of a detected RX33 error:
SYSDEV-S Seq 104. at 6-JAN-1986 10:12:00.76
DX1: LBN 1488. (49,0,02), Status 001
Seek 000, 000000
Tran 003, 021404
T.O. 000
87 3 1485 -7680 1 49 1 4
The -S following the SYSDEV prompt and before the Seq. number indicates the severity level. The
RX33 has three severity levels:
1. Success (S): 1\vo or less errors during a command/retry.
2. Informational #(1): More than two errors.
3. Error (E): Unrecoverable error.
The status field is most important and is a direct indication of the error. Following is a list of the
RX33 status codes:
• 000: Success.
• 001: Success with retries.
• 002: Software version mismatch (driver versus operating code).
• 200: Command aborted through a CTRLIY or exception operation.
• 201: Illegal file name.
• 202: File not found.
• 203: File is not in a loadable image format.
• 204: Insufficient memory to load image.
• 205: No free partition to load image into.
• 206: Unit is software-disabled.
• 365: Unit is write-protected.
• 367: No media mounted.
• 375: EOF detected during read or write.
• 376: Hard disk error, other than the following:
370: Bad unit number.
357: Data check error.
343: Motor broken (would not spin up).
340: Un correctable seek error (desired cylinder not found).
311: Bad record (LBN) number (not on media).
272: Parity error in controller on M.std2 module.
In the example, the failing diskette drive is indicated by DX1:. The logical block number where
the failure occurred is displayed by LBN 1488. The three numbers in parentheses, separated by
commas after the logical block number indicate in order the cylinder, the media surface, and the
drive sector.
The Seek entry's first group of zeros shows the retry count for seeklrecal errors or the number of
times the command was issued but not completed. The second group of zeros shows an inclusive
OR of the control and status registers CSR bits set during seek error retries. The important bit in
a seek error is bit 4.
The Tran (transfers) entry's :first group of zeros shows the retry count for read, write, and format
errors, or the number of times the command was issued and not completed. The second group of
zeros shows an inclusive OR of the CSR bits set during read, write, and format error retries. A
breakdown of the upper CSR bits is shown in Figure ~24.
PAR NXM DMA TST MOTR DRV

ERR ERR
HI
ENA SEL
S7 86 S5 S4 S3 S2 S1 so
DIS PAR
~-----------------yr------------------I
STATUS REGISTER BITS
CXO-1125B
Figure 8-24 RX33 Floppy Controller CSR Breakdown

Table 8-28 shows the status of the lower C8R bits 87 through SO.
Table 8-28 Status Register Summary

All Type I Read Read Read Write Write
Bit Commands Address Sector Track Sector 'lrack
87 Not Ready Not Ready Not Ready Not Ready Not Ready Not
Ready
86 Write 0 0 0 Write Write
Protect Protect Protect
S5 Head Loaded 0 Record Type 0 0 0
S4 Seek Error RNF RNF 0 RNF 0
83 eRe Error eRe Error eRe Error 0 eRe Error 0
82 Track 0 Lost Data Lost Data Lost Data Lost Data Lost
Data
81 Index Pulse DRQ DRQ DRQ DRQ DRQ
80 Busy Busy Busy Busy Busy Busy
The T.O. entry line is a timeout recording for each command type. This counter reflects the total
number of timeouts for the command in error. All commands (Read, Write, Recal, 8pinup, and
Format Track) time out in one second.
The last line in the error message is more complicated to break down. Figure 8-25 shows the
breakdown of the last line of the example RX33 error message.
87 3 1485 -7680 49 1 4
1t SECTOR NUMBER
SURFACE NUMBER
' - - - - - - - CYLINDER NUMBER
' - - - - - - - - U N I T NUMBER
' - - - - - - - - - - BYTE COUNT (NEGATIVE IMPLIES WRITE)
~-----------------------LBN
~----------------------------SUCCESSCOUNT
~---------------------------- ERR COUNT NUMBER

CXO-2117A
Figure 8-25 RX33 Error Message Last Line Breakdown

Most information in the error printout is reiterated in the last line. Starting from the right,
sector, surface, cylinder number, and unit number are displayed as in the main body of the error
message. The byte count has an indicator for write and read commands; the negative indicates a
Write operation. The LBN in this field is the starting LBN for this transfer. The LBN in the main
message body is the failing LBN. The success count and error count are for informational purposes.
8.4.5.2 Disk Functional Errors

Although most disk drive-related errors are MSCP errors, several disk functional errors fall into
the out-of-band error category. They are identified by the DISK-E identifier printed on the terminal
display prior to the error.
The message, message description, action, and probable FRUs for the disk functional out-of-band
errors are listed in Section 8.5.
8.4.5.3 Tape Functional Errors

Although most tape errors are covered under TMSCP errors, certain tape functional errors
are classified in the out-of-band error category. They are identified by the TAPE-E identifier
printed prior to the error printout on the local console terminal. See Section 8.5 for listings and
explanations of out-of-band tape functional errors.
8.4.5.4 Miscellaneous Errors

Miscellaneous errors are identified by the SINI-E identifier printed on the local console
terminal. Many of these messages are one-line or two-line messages, but some have several
lines of informational text that result from subsystem exceptions. Subsystem exceptions detect
inconsistencies in the operating software. Listings and explanations of SINI errors are located in
Section 8.5.
The SINI error messages are a result of the operating software performing a consistency check
which failed. When consistency checks fail, the HSC performs a soft initialization causing it to
crash and reboot. This is known as a subsystem exception. Upon successful completion of the
reboot, the subsystem exception printout displays the contents of several HSC registers as well
as the status of all requestors. As a result of the subsystem exception, the SINI error message is
printed. This message tells why the last soft Init happened.
, Troubleshooting Techniques 8-53
The actual sequence of events for a SINI-E out-of-band error printout is as follows:
1. When the HSC detects an unrecoverable problem, a soft lnit or crash occurs. A system dump is
perfonned under the heading SUBSYSTEM EXCEPTION. The HSe then reboots.
2. When the HSe reboots, a message indicating it has rebooted, followed by the multiline SINI
message, gives the reason for the last soft Init (crash).
3. The same message is written on the system diskette and can be examined with the SHO
EXCEPTION command. A host error message log is also filed in host memory as an HSe
datagram, storing the out-of-band error SINI message.
8.4.6 Traps
The four traps described in the following sections (Trap through 4, Trap through 10, Trap through
114, and Trap through 134) are the same as are found in the 11170 CPU.
8.4.6.1 NXM (Trap through 4)

If the error registers in the NXM: printout equal 170024 000077, the error is not a Nonexistent
Memory (NXM) error. Instead, it is a stack overflow or some illegal instruction. When the error
register is any number other than 170024 000077, the number represents the unresponsive address.
The NXM trap produces a subsystem exception printout similar to the example in Section 8.4.6.6.
If the error register equals 1Bxxxx, the Window Bus register equals the Control memory address
causing the NXM: error. If the failing address is in Control memory and shows an NXM error, it is
definitely a hardware problem. Otherwise, it can be either a software or a hardware problem.
8.4.6.2 Reserved Instruction (Trap through 10)

The subsystem exception message for this trap indicates the vector number is 10 and identifies
the trap as ILOP (an illegal Opcode). Refer to the (PC-B) to (PC): field in the Level 7K interrupt
example of Section 8.4.B.B.
With a Trap through 10, the third word from the left is the instruction causing the trap. If this is a
valid PDP-11 instruction, it is definitely a hardware problem. Otherwise, the program may not be
executing in the right place, indicating the problem could be either hardware or software.
8.4.6.3 Parity Error (Trap through 114)

This error, caused by hardware, does not crash the HSC but causes a reboot and SINI error
message. The error message shows the last reboot caused by the Trap through 114 and the address
that caused the trap.
Determine if the error occurred in memory or in cache memory by reading the contents of the low
error address displayed in the error printout. If the content is the address of the low error address
register (170024), the error is in cache memory. Any other address indicates the error is in memory.
In the following example, note the low error address and the high error address fields. When these
fields contain the exact addresses as shown in this example, the error is from the P.ioj cache.
SINI-E Seq 1. at 17-Nov-1858 00:00:01.60
Parity Error (Trap through 114)
Process PSCHED
PC 111022
PSW 140000
Lo err adr 170024
Hi err adr 000077
WBUSR 020633
8.4.6.4 Level 7 K Interrupt (Trap through 134)

A level 7 K interrupt, detected by hardware or microcode, occurs when one or more requestors
detect a fatal error condition while executing functional code. The microcode-detected errors
causing level 7 K interrupts result from a microcode consistency check failure in either Ksdi, K.sti,
K.si, or K.ci microcode. Requestor hardware-detected errors are the result of errors detected on the
Control bus, scratchpad RAM parity errorslData bus parity errors, or host clears, or Control bus
NXM:s (not related to data transfers). The requestor, upon detecting the error, generates a level 7
interrupt to the P.ioj/c. The P.ioj/c traps through location 134, causing a reboot.
8.4.6.5 Control Bus Error Conditions (Hardware-Detected)

The hardware-detected Control bus errors causing level 7 K interrupts are:
• Control bus error-The requestor was in the process of executing a Control bus cycle and
received CERR L (Control bus error low) from the P.ioj/c. The P.ioj/c had detected an illegal
Control bus cycle type.
• Control bus parity error-The requestor detected bad parity on the data it read off the
Control bus.
• Control bus NXM-The requestor tried to reference Control memory and did not receive an
acknowledgment (CACK L) from the M.std2 within the timeout period.
8.4.6.6 Level 7 K Interrupt Example

An example of a detected level 7 K interrupt follows.
SUBSYSTEM EXCEPTION *- Vi 250 HSC LONDON

at 25 Oct 1985 00:08:46.64 o 23:23:21.40
User PC: 110574 caused by (134 Kint
PSW: 140011
PSCHED active PCB addr = 054536
RO-R5:
000000 000024 000000 000000 000000 000000
Kernel SP: 000774
Kernel Stack
005046 000004 052744 045412 001012 000000 045644 000000
052136 000000 047260 000000 051300 000000 054742 000000
User SP: 000774
User Stack:
150042 147502 147516 000000 10214~ 000000 000000 000000
000000 000000 000000 000000 000000 000000 000000 000000
KPAR(0-7) :
000440 000640 001040 1577770 001440 001240 000240 177600
KPDR(O-7) :
077506 077506 077506 077406 077506 077406 077506 077506
UPAR(0-7) :
000000 000000 000000 000000 002204 001240 000240 177600
UPDR(0-7) :
077406 077406 077406 077406 063406 077406 077406 000116
MMSR(0-2): 000017 000000 037260
Window Index Reg: 000026
Window Bus Reg: 001431
WADR(0-7) :
Troubleshooting Techni~ues 8-55
160004 161004 162004 163004 164004 165004 166004 167004

Translated WADR(0-7) :
001401 001401 001401 001401 001401 001401 001401 001401
Error Regs: 170024 000077
Status of Requestors (1-9) :
000001 000377 000377 000377 000377 000175 000377 000377 000377
(PC-6) to (PC) :
013737 141020 110560 013701
Control area for slot *000006
Control area address: 017660:
Register area contents:
000000
000000
000011
021154
102557
000770
000000
000000
017650
000000
057502
005317
002224
001000
000000
000671
000000
143444
107001
001000
005317
002212
000671
001000
000000
000000
000000
040506
000010
000374
043520
005400
001000
Booting
INIPIO-I Booting
Requestor 6 has failed with a status of 175. Refer to Appendix D to determine if the failure was a
Control bus error.
At this time the HSC reboots. A message is displayed on the local console terminal stating the HSC
has rebooted.
HSC Version 200 29-Sept-1985 23:17:28 System LONDON

The actual SIN! error message is printed on the local console terminal after the HSC has rebooted.
SINI-E Error sequence 1. at 17-Nov-1858 00:00:03.00

Last soft Init caused by level 7 K interrupt
From process PSCHED
PC 110574
Status: 001 377 377 377 377 175 377 377 377
The resulting 134 trap information is printed on the local console terminal. The PSCHED
statement indicates PSCHED was the active process when the error occurred. The status statement
shows requestor 6 failed with a status of 175. Also, three lines after the status line is a message
line indicating the control area for slot six and slot six control address. This indicates requestor 6
is the failing requestor. The INIPIO-I Booting statement indicates the HSC is attempting to reboot.
When the HSC completes the initialization, the Last soft Init caused by level 7 K interrupt failure
is printed on the local console terminal identified by SINI-E. The active process at time of failure
is identified. In this case, the active process was PSCHED. If the failure is a hard failure, the
following message may also be displayed on the local console terminal.
SINI-E ERROR SEQUENCE 1. AT 25-0CT-1858 00:00:02.80

REQUESTOR 6 FAILED INIT DIAGS, STATUS 107
This message is also considered an out-of-band error.
8.4.6.7 MMU (Trap through 250)

Following is an sample printout of a detected Memory Management Unit (MMU) failure.
**SUBSYSTEM EXCEPTION** vt Y10B HSC LAYER

at 12-DEC-1985 13:43:40.05 up 2 19:24:07.40
User PC: 004747 caused by (250 MMU
PSW: 140000
SETSHO active, PCB addr = 104116
RO-R5:
000320 000001 100000 100212 000266 000002
Kernel SP: 000774
Kernel Stack:
005046 000004 053314 045762 001012 000000 046214 000000
047022 000000 047426 000000 052052 000000 051042 000000
User SP: 000226
User Stack:
040314 021356 033552 021356 021246 000040 017440 017440
020040 020037 020037 000330 101000 027113 000144 060542
KPAR(0-7) :
000440 000640 001040 001440 002040 001240 000240 177600
KPDR(0-7) :
077506 077506 077506 077506 077506 077506 077506 077506
UPAR(0-7) :
007074 007274 006410 000000 002240 001240 000240 177600
UPDR(0-7) :
077506 077406 013406 077406 077406 077506 077506 000116
MMSR(0-2): 040145 000000 004743
Window index reg: 000002
Window bus reg: 001407

WADR(0-7) :
160000 161004 162440 163000 164004 165004 166220 167034
000000 001401 067510 040000 001401 001401 010444 001407
Error rags: 170024 000077
Status of requestors (1-9) :
000001 000002 000002 000002 000203 000203 000203 000377 000377
(PC-6) to (PC):
027441 067516 051040 071545
Because the trap is a MMU trap, look first at the register contents of MMSRO (memory
management status register 0). Refer to Figure 8-26 for a breakdown of the bits in MMSRO.
115114113112111 1019181716 514131211101

'I'
, ~ I' \... .) ,
~ ,
~ , \.. .)
~ "----- .J ,
~ II' ~
ABORT, NON RES IDENT-
ABORT, PAGE
LENGTH ERROR
ABORT, READ-O NLY
ACCESS VIOLAT ION
TRAP, MEMORY MANAGEMENT
NOT USED - -
ENABLE MEMOR Y MANAGEMENT TRAP
MAINTENANCE MODE
INSTRUCTION CO MPLETED
PAGE MODE-
PAGE ADDRESS SPACE I/D
PAGE NUMBER -
ENABLE RELOCATION
CXO-1126A
Figure 8-26 MMSRO Bit Breakdown

Look at the printout lines for MMSR (0-2). Compare the bits set in MMSRO to the bit breakdown
in Figure 8-26. The example indicates a page length violation on page 2. The page length error bit
is set, and the page number 2 bit is set.
Next, check the PSW line and determine the mode in which the HSC reported this error. A 14xxxx
in the PSW means User mode, a OOxxxx in the PSW means Kernel mode. Also, above the PSW line
the word User or Kernel appears to identify the mode. Our example shows User mode is active.
Therefore, the next register contents of any value are the UPAR and UPDR. If the active mode had
been Kernel, the important registers would have been the KPDR and KPAR registers.
The first group of numbers under the UPAR(0-7) line is for page zero, the second for page one,
the third for page two, and so forth. The third group of numbers in the example are for page two,
the violated page. Note the difference in UPDR contents on page two versus the UPDR contents
on other pages. The UPDR contents on other pages all start with 077 designating a full page of
memory to be allocated for that page. The UPDR contents on page two starts with a 013., indicating
a short page.
Two possible problems cause this error:
1. Memory Management Unit on the P.ioj/c
2. Software
Software inconsistency (Trap through 20) is reported similar to an MMU trap. A subsystem
exception is dumped on the local console terminal with the trap vector reported being a Trap
through 20 (AT). An example printout and explanation are found in Appendix B.
The subsystem exception is followed by the HSe reboot. Upon successful reboot, the following
message is displayed.
HSC Version YI0B 16-Jan-1986 15:30:20.20 System MASTER
Then the SIN! error resulting from the detected subsystem exception is printed.
SINI-E Sequence 1. at 16-Jan-1986 00:00:11.20
Last soft Init caused by software inconsistency
From process HOST
PC 007044
PSW 140001
Stack dump: 000016 006401 015476
8.5 Alphabetical Listing of Software Error Messages

Each message description includes the following:
a. Actual error message-Displayed in English at the HSe console terminal.
b. Error type-The subsystem where the error occurred or was detected.
c. Error message severity level-Included in the error message.
d. Message description-A review of what the error is about.
e. Action-Remedies or troubleshooting paths that the Customer Service representative can take
to correct the problem.
f. Possible FRUs--Suggested components that are likely suspects causing the malfunction.
Aborting Error Recovery Due to Excessive RECALs

Disk Unit xxx.
Requestor xx
Port xx
Error Type: Disk functional out-of-band
Severity: Error
Description: For each group is a transfer, a count of the number of RECALS issued to the
drive is kept. If the count exceeds a hard-coded value, this message is printed. Recovery from
an error is not possible because of excessive RECALS and the drive is declared inoperative.
Action: Refer to the drive service manual and any other type errors being logged to determine
reasons for persistent positioning failures.
Possible FRUs: Drive modules. (Refer to the drive service manual.)
Aborting Error Recovery Due to Excessive Timeouts

Disk Unit xxx.
Severity: Informational
Description: The HSC detects several timeouts on the disk drive. A timeout occurs because
the drive did not complete its expected work in the expected time. All error recovery attempts
will be aborted and the drive will be declared inoperative.
Action: May need to replace the following FRUs. Further testing may be necessary.
Possible FRUs:
1. Drive modules. (Refer to the drive service manual.)
2. KsdilK.si module
Acknowledge Not Asserted at Start of Transfer

Error Type: Tape error
Severity: Error
Description: The HSC is ready to start a transfer by sending the formatter a Level 1 command
and the formatter does not have ACKNOWLEDGE asserted.
Action: Check the formatter. This error may indicate a formatter STI communications error,
or if preceded by tape transport errors, may be a result of a transport failure.
Possible FRUs:
1. Formatter
2. KstilK.si module
3. STI cable set
ATN Message Sent to Node xx, for Unit xx.

Description: An attention condition was found on the indicated drive unit and an attention
message has been sent to the host to notify it of this condition.
Action: None
Possible FRUs: None
Attention Condition serviced for ONLINE disk unit xxx.

Description: An attention condition indicating a state change in the drive needs servicing. A
GET STATUS exchange is invoked to the drive. Note: This may not indicate a failure condition.
Action: Refer to the console printed Get Status response.
Bad Block Replacement (Block OK)

Error Type: BBR error
Severity: Warning
Description: Block tested OK-not replaced.
Action: Monitor drive for the frequency of these reports. If frequency increases, troubleshoot
the error that triggers BBR.
Possible FRUs: Refer to Cause Event error message field in Table 8-20.
Bad Block Replacement (Drive Inoperative)

Severity: Warning
Description: Replacement failure or drive access failure. One or more transfers specified by
the replacement algorithm failed. If necessary and possible, write-protect the drive and perform
a volume backup immediately.
Action: Drive should be tested further. Move the drive to another KsdilK.si (or to just another
KsdilK.si port) if available. If the problem persists, failure is probably in the drive.
Bad Block Replacement (RCT Inconsistent)

Severity: Warning
Description: Replacement failure-the ReT table is not usable.
Action: Drive media should not be used until replaced or verified as good. If necessary, write-
protect this drive and have the customer perform a volume backup immediately. Further testing
of the drive may be necessary.
Bad Block Replacement (Recursion Failure)

Severity: Warning
Description: Replacement failure-recursive failure. Two successive RBNs were bad.
the error triggering BBR.
Bad Block Replacement (REPLACE Failed)

Severi ty: Warning .
Description: Replacement failure-REPLACE command or its analogue failed. The status
returned from the replacement process indicates the command was not successful.
Action: Drive media should not be used until it is replaced or verified as good. If necessary,
write-protect this drive and have the customer perform a volume backup immediately. Further
testing of the drive may be necessary.
Bad Block Replacement (Success)

Severity: Warning
Description: The bad block was successfully replaced.
the error triggering BBR.
Bad Dispatch State in CB .•.
Error Type: CI-detected out-of-band
Severity: Warning
Description: The CI Manager sends a SCS control message and finds an invalid dispatch state
in the control block. The CI Manager then uses the dispatch state to determine where to send
the proper control message. If this is the only known problem, a software problem could exist
within the HSC. Otherwise, the problem could be caused by a Control bus addressing problem
with the K. pli, M.std2IM.std, or P.ioj/c modules.
Action: Replace the following FRUs.
Possible FRUs:
1. K.pli module
2. M.std2lM.std module
3. P.ioj/c module
Booted from Drive 1. Drive 0 Error (text)

Error Type: SIN! error
Description: The system was booted from system device drive 1. Normal boot is from drive O.
Action: None
Possible FRUs: Drive 0
Buffer EDC Error

Severity: Error
Description: The KstilK.si detected an EDC error on the Data Buffer it read from memory on
a Write operation.
Action: Test the data path from K.stilK.si to HSC Data memory and the Kci.
Possible FRUs:
1. Formatter
2. M.std2!M.std module
3. KstilK.si module
4. Kci module
Cache Disabled Due to Failure

Error Type: SIN! error
Severity: Error
Description: SIN! looks back at the Cache diagnostic and senses the cache is disabled due to
cache failure or manually disabled in the diagnostic. This error also shows as a soft fault code
on the OCP.
Action: Load the Off-line Cache diagnostic and answer the prompt asking to disable or enable
Cache with an enable. Reboot the system diskette and check if the original message is displayed
again.
Possible FRUs:
1. P.ioj module
2. M.std2 module
Clo~k dropout from ONLINE disk unit xx.

Severity: Error
Description: The on-line disk has lost its real-time state clock.
Action: Check the path between the KsdilK.si and the disk drive that was reported.
Determine if the problem is in the HSC or the disk drive. Other disk error reports may precede
this message and provide more detail about this error condition.
Possible FRUs:
1. Drive modules. (Refer to the drive service mahual.)
2. SDI cable
3. KsdilK.si module
Compare Error
Error Type: Controller error
Severity: Error
Description: A compare error occurred during a Read-Compare or a Write-Compare operation.
For the Read-Compare operation, the HSC again obtains the data from the unit or shadow
set and compares it with data obtained from host memory. If the data is not the same, a
compare error results. For the Write-Compare operation, the controller obtains data from each
destination and compares it with data again obtained from host memory. If the data is not the
same, a compare error results.
Action: Isolate the FRU by moving the disk or tape drive to another data channel and retrying
the exact failing operation. Also, check the HSC Data memory buffer address for repetition. If
failure occurs on multiple physical units across multiple data channels and HSC Data memory
buffer address is not repetitive, investigate a possible K.ci problem.
Possible FRUs:
1. Isolated disk (or tape) unit
2. K.sdilK..stilK.si module
4. K.ci module set
5. Host CIImemory
Controller Detected Position Lost
Severity: Error
Description: Information contained in the response from the formatter to the HSC POSITION
command did not match the expected tape drive position.
Action: Check the formatter. If the error persists, run the In-line Tape (ILTAPE) diagnostic to
help isolate to the FRU.
Possible FRUs: Formatter
Controller Transfer Retry Limit Exceeded
Severity: Error
Description: The controller failed to perform the command within the limit of allowable
retries.
Action: Check the formatter and the drive.
Possible FRUs:
2. Formatter
Controller Detected Transmission or Timeout Error

Error Type: SDI
Severity: Error
Description: The controller detected an invalid framing code or a checksum error in a level 2
response from the SDI drive.
Action: Determine if this error is occurring on more than one drive, which may indicate a
K.sdilKsi problem. However, if it is occurring only on one drive, the SDI cable or the drive may
be at fault. Refer to the appropriate drive service manual for assistance with drive FRUs.
Possible FRUs:
1. SDI cable
2. Drive SDI interface module
3. K.sdilK.si module
4. SDI transition bulkheads
Could Not Complete On-line Sequence

Severity: Error
Description: Could not complete on-line sequence due to a condition in the drive.
Action: Check the formatter and the drive.
Possible FRUs:
2. Formatter
Could Not Get Extended Drive Status

Severity: Error
Description: Issued the GET EXTENDED DRIVE STATUS command and the drive did not
respond with the extended drive status.
Action: Check the formatter.
Could Not Get Formatter Summary Status During Transfer Error Recovery
Severity: Error
Description: Issued the command and the formatter did not respond with the formatter
summary.
Could Not Get Formatter Summary Status While Trying to Restore Tape Position
Severity: Error
Description: Issued the command and the formatter did not respond with the formatter
summary status.
Action: Check the fonnatter.
Could Not Position for Formatter Retry

Severity: Error
Description: The HSC issued a command for data recovery with position required, and the
drive could not complete the command.
Action: Check the media, drive, and formatter.
Possible FRUs:
2. Media
3. Formatter
Could Not Set Byte Count

Severity: Error
Description: Issued command to set byte count and could not complete command.
Could Not Set Unit Characteristics

Severity: Error
Description: Issued command to set unit characteristics and could not complete command.
Data Bus Overrun

Severity: Error
Description: The HSC attempted to perform too many concurrent transfers, causing one or
more of them to fail due to a data overrun or underrun. For example, data is sent to a bus
by a data producer and then removed from the bus by a data consumer. If the producer sends
data to the bus more quickly than the consumer can remove it, a data overrun occurs. If the
consumer removes data more quickly than the producer can send it, a data underrun occurs.
Action: Determine which module is the data producer and which module is the consumer for a
given error. Use the requestor number for assistance.
If the problem persists after replacing the suspect module(s), an HSC software problem should
be investigated.
Possible FRUs: Source or detecting requestor modules.
Data Error Flagged in Backup Record

Disk Unit xx LBN xx
Tape Unit xx
Error Type: Tape functional out-of-band
Severity: Warning
Description: During a backup, a data error was encountered. During the BBR, the record was
written with a forced error bit set.
Action: Check BBR history on source drive.
Possible FRUs:
1. Disk unit
2. Media
Data Memory Error (NXM or Parity)

Severity: Error
Description: The HSC detected an error in internal Data memory. The error was either
a parity error, detected through a parity generator/checker (data only-not address) on the
requestor module, or a nonresponding address (the requestor did not receive a DACK from the
memory module).
Action: Determine if this error is repetitive; if so, the problem is probably the M.std2/M.std
module. However, it may be a Data bus problem caused by a number of things, such as failing
bus drivers/receivers on the indicated requestor modules.
Possible FRUs: M.std2lM.std module or a possible Data bus problem.
Data Ready Timeout

Severity: Error
Description: The controller did not detect Data Ready from the formatter within the timeout
interval after sending it a Level 1 command.
Action: Check the STI path.
Possible FRUs:
1. STI cable set
2. KstilK.si module
3. Formatter
Troubleshooting Techni.ques 8-67
Data Sync Not Found

Error Type: Disk transfer error
Severity: Error
Description: This error occurs when the SERDES 16 does not detect the SYNC character
(26BC hex) immediately preceding read data from the disk drive. The K.sdi/K.si has already
read a valid header and is awaiting the data SYNC character.
Action: Determine if additional errors occur from this drive to indicate a drive or media error.
If not, the problem is probably the K.sdi/K..si module.
Possible FRUs:
2. K.sdi/K..si module
3. SDI interface
Datemme Set By Node nn

Error Type: CI-detected out-of-band error
Description: The HSC received either a START or STACK (start acknowledge) message over
the CI, and the date and time was not set.
Action: None. This is a normal message as part of establishing a VC between a host and an
HSC.
Possible FRUs: None
Deferred ATTN. Message for Node xx, Unit xx.
Error Type: Disk functional out-of-band error
Description: An attention message is delayed in process.
Action: None
Possible FRUs: None
Disk unit xx. (Requestor xx., Port xx.) being INITialized.
DeB addr: xxxxxx
Description: A disk is being initialized.
Action: None
Possible FRUs: None
Disk unit xx. ready to transfer.

Retrieval failure or subsystem deadlock probable.
Description: A disk transfer did not complete within the allowable timeout period. The HSC
software cannot detect any problems to account for the failure. Possible problems include:
1. No available buffers
2. Drive problems
3. KsdilK.si problems
Action: Check data transfer path. This error may indicate too many utilities or in-line
diagnostics running simultaneously. The problem might also be an HSC software problem.
Possible FRUs: K.sdi/K.si module
Disk unit xxx. (Requestor xx., Port xx.) declared inoperative.

Intervention required.
Severity: Error
Description: The Disk Path process has concluded that the drive is no longer usable. Any
pending 110 is cleaned up and the drive state is set to either UNDEFINED or OFFLINE. The
HSC ignores the disk until it detects some intervention.
Action: Examine previous error reports to help resolve failure. Toggle port switch on drive.
DRAT/SEEK timeout, disk unit xxx.

Description: A stimulus resulting in error recovery code action is the expiration of the
DRAT/SEEK timer for the drive. A DRAT represents data transfer action with the drive,
whereas the SEEK represents position requests to the drive.
Each drive has a timer (set to three times the SDI drive short timeout value) allocated on its
behalf at subsystem initialization time. This timer, called the DRAT/SEEK timer, is active
whenever data transfer activity to the drive is outstanding.
When the disk transfer code queues transfer work to KsdilK.si on behalf of a previously idle
drive, the timer starts. When it adds transfer work to a drive that already has transfer work,
the timer restarts. When it detects the completion of the last DRAT queued to the drive, the
timer stops. Thus, the timer is running only as long as transfer work is outstanding. A timer
may expire for several reasons:
1. The drive has detected a drive error and has lowered Read/Write Ready.
2. The drive has stopped sending clock signals.
3. A SEEK has timed out.
4. Another element in the subsystem that should have supplied resources to the disk transfer
operation in a reasonable time did not.
Action: Check the drive.
Possible FRUs: Drive modules. (Refer" to the drive service manual.)
DRIVE CLEAR attempt on disk unit xx. (Requestor xx., Port xx.).
DeB addr: xxxxxx Error count ******.
Description: The drive detected some previous error and the HSC is now attempting to clear
that error.
Action: Examine the host error log to determine what error the drive is trying to clear.
Possible FRUs: Drive
Drive Clock Dropout

Error Type: SDI error
Severity: Error
Description: Either data or state clock was missing when it should have been present. This is
detected by the requestors connected to this SDI drive, usually by means of a timeout.
Action: Determine if this error is occurring on more than one drive, which may indicate a
KsdilK.si problem. However, if it is occurring on only one drive, the SDI cable or the drive may
be at fault. If other errors surround or precede this one, those errors may have sequentially
triggered this error. Refer to the appropriate drive service manual for assistance with drive
FRUs.
Possible FRUs:
1. SDI cables
2. Drive SDI interface module
3. KsdilK.si module
Drive Detected Error

Severity: Error
Description: The controller received a GET STATUS command or unsuccessful response with
the EL bit set, or the controller received a response with the DR flag set and does not support
automatic diagnosis for that SDI drive type.
Action: Determine if the drive has a hard fault (fault light on and an error code in the drive
microprocessor LEDs). Refer to the drive service manual for assistance with drive internal
diagnostics and LED error codes. Decode remaining error message bytes for more detailed error
informa tion. If error message decoding does not clearly indicate a drive error, move the drive to
another requestor (or requestor port) to help isolate failure between HSC and drive.
Possible FRUs:
2. SDI cables
3. SDI bulkheads
Drive Inoperative
Severity: Error
Description: The HSC has marked the drive inoperative due to an unrecoverable error in the
previous level 2 exchange, the drive's Cl flag is set, or the drive has a duplicate unit identifier.
Once the HSC reports the drive as inoperative, the drive state clocks must transition to return
the drive to an operational state.
Action: Refer to the drive service manual. Run ILDISK to help isolate failure between HSC
and drive.
Possible FRUs:
2. KsdilK.si module
3. SDI cables
Drive Requested Error Log (EL Bit Set)

Severity: Error
Description: The controller requested a drive error log because the drive returned a status
message with the EL bit set in the request byte field.
Action: Determine what drive-detected error (previous error description) caused the drive to
request a drive error log by finding the error in the error log report. Also decode remaining
fields in the drive status response of this error message and any preceding errors on the unit.
Duplicate Disk Unit xx

Description: Disk unit numbers are duplicated within the system.
Action: Locate the duplicate disks and change the plug number on one.
EDC Error
Severity: Error
Description: The sector was read with correct or correctable ECC and invalid EDC. A fault
probably exists in the logic of either this controller or the controller that last wrote the sector.
Look at the source and detecting requestor fields in the error message to determine which
requestor detected the error and the direction of the transfer (read or write).
Action: Determine if other errors indicate a problem with the data path circuitry on the
indicated requestor modules.
Possible FRUs:
1. KsdilK.si module
2. M.std2lM.std module, if an address parity error on Data memory occurs, as this is checked
by the EDC field.
ERASE Command Failed

Severity: Error
Description: Issued ERASE command and command failed.
ERASE GAP Command Failed

Severity: Error
Description: Issued ERASE GAP command and command failed.
Forced Error
Severity: Error
Description: The sector was written with a Force Error modifier indicating this is a replaced
image and the original data could not be read correctly using retries and the ECC algorithms.
Action: Restore the media from a previous backup. A VMS (HSC) backup and restore of the
current media will clear the forced error condition but will leave the sector corrupt.
Possible FRUs: None
Formatter Detected Position Lost

Severity: Error
Description: The formatter lost track of tape position.
Possible FRUs:
2. Formatter
3. Media
Formatter and HSC Disagree On Tape Position

Severity: Error
Description: The formatter and the HSC disagree on position of the tape.
Possible FRUs:
2. Formatter
3. K.stilK..si module
Formatter Requested Error Log

Severity: Error
Description: The formatter detected an error and set the EL hit to request an error log be
taken.
Formatter Retry Sequence Exhausted

Severity: Error
Description: The formatter failed to complete a command within the retry limit.
Possible FRUs:
2. Formatter
3. Media
FRB Error: K.cl, 1st LBN xx., xx. buffers, FE$SUM xx

Description: An error was detected by the Kci while processing a Fragment Request Block
(FRB) and the FRB has been sent to the disk error process. Example: EDe error.
Action: If excessive, reformat drive.
FRB Error: K.sdi, Unit xx., 1stLBN xxx., xx. buffers, FE$SUM xx
Description: An error was detected by the KsdilK.si while processing a Fragment Request
Block (FRB) and the FRB has been sent to the disk error process. Example: Suspected
Positioner error.
Hard transfer error loading (file) xx:

Error Type: SINI out-of-band
Severity: Error
Description: The P.ioj/c detected a hard error while loading a file from the system media into
Program memory. The particular files that can produce this error are DUP and MIRROR. The
xx field is the error status value from the device driver.
Action: Load the file from the other system load device; load the back-up media.
Possible FRUs:
1. System media
2. System load device
Hard transfer error writing SeT xx

Error Type: SIN! out-of-band error
Severity: Error
Description: The HSC detected an error while attempting to write the SCT on the console
load media. The xx designates the octal byte that is the error status value returned from the
device driver.
Action: Make sure the drive is not write-protected; try the back-up media; try the other system
load device.
Possible FRUs:
1. System media
2. System load device
Header Error
Severity: Error
Description: The subsystem reads an inconsistent or invalid header for the requested sector.
The header is inconsistent if three out of four copies of the high order header word do not
match.
The header is considered invalid if all of the following are true:
• The header is consistent (three out of four copies of the high order header word match).
• Two out of four of the low-word header values match the desired target header low-word
value.
• The high-word header values do not match the respective target header values.
For recoverable errors, this code implies a retry of the transfer to read the valid header. For
unrecoverable errors, this code implies the subsystem attempted nonprimary revectoring and
determined the requested sector is not revectored. Causes of an invalid header include header
mis-sync, header sync timeout, and an unreadable header.
Action: Determine if this error is repetitive on this unit indicating a deteriorating media.
Possible FRUs:
HML$ER set-HM$ERR = nn
Severity: Warning
Description: A Host Memory Block (HMB) operation resulted in an error. A breakdown of
HMB error word (HM$ERR) bits follow.
• 000002 HME$BM-Insufficient BMBs to receive message.
• 000004 HME$N~equenced message received over a connection with 0 in credit field.
• 000010 HME$N~equenced message received over a connection with credit field> 1.
Excess has been added to CB$EM.
• 000020 HME$OV-Oversize message received (>1096. bytes).
• 000040 HME$DN-Data memory NXM during BMB operation.
• 000100 HME$DP-Data memory parity error in BMB operation.
• 000200 HME$DO-Data memory overrun during BMB operation.
• 000400 HME$FP-Reception buffer parity error in packet header. Message not receivable.
• 001000 HME$PL-Reception buffer parity error in body of message.
• 002000 HME$CN-Transmission not attempted because connection not valid.
• 004000 HME$VC-Transmission not attempted because VC closed or connection invalid.
• 010000 HME$TE-Transmission attempted but failed (no ACK).
• 020000 HME$TP-Transmission failed due to transmission buffer parity error.
• 040000 HME$HC-Packet inconsistent with Kci context received from host.
• 100000 HME$IC-Illegal control function Opcode.
Action: Compare the displayed code to the previous list and determine where the problem lies.
For example, a code of 000040 indicates a failure in the M.std2/M.std module, and a code of
002000 indicates a problem in the K.ci module set.
Possible FRU s:
1. PILA module
2. Kpli module
Troubleshooting Techniqu.es 8-75
Host Clear from CI Node

Severity: Error
Description: The host cannot function with the HSC for some reason, such aSa nonresponse
within a certain amount of time or too many errors on the CI.
Action: Check the HSC console messages and the error logs of the systems connected to the
HSC.
Possible FRDs:
1. CI cable
2. HSC/CI interface
3. Host/CI interface
4. HSC software
5. System software
Host interface (K.ci) failed INIT diags, status =xxx

Severity: Error
Description: The failing status indicates which module in the K.ci set has failed. A soft fault
code is generated and may be examined by pressing the fault button on the OCP.
Action: Determine which is the failing module by comparing the failing status value to the
values in Appendix D. This comparison points more directly to the failing module.
Possible FRDs:
1. LINK module
2. PILA module
3. Kpli module
Host Interface (K.cl) Is required but not present

Severity: Error
Description: A Kci module set is absent, or the failure in the K.ci module set was so severe
upon initialization, the initialization diagnostics did not run.
Action: Check for the presence of a K.ci module set. If missing, install the K.ci module set. If
Kci module set is present, determine which module is failing by running Off-line diagnostics.
This error generates a soft fault and is examined by pressing the fault button on the OCP.
Possible FRUs: See list below and error message Last Soft lnit resulted from unknown cause.
1. Kpli module
2. Kci module set (anyone of the three modules in the set)
Host Requested Retry Suppression On A Formatter Detected Error

Severity: Error
Description: The formatter detected an error and the host issued a command to suppress the
retry of the command that failed.
Host Requested Retry Suppression On A K.stilK.si Detected Error

Severity: Error
Description: An error was detected in the K.stiJK.si and the host issued a command to
suppress the retry of the command that failed.
Action: Check the K.stilK.si.
Possible FRUs: K.stiJK.si module
Illegal bit change in status from disk unit xxx.

EL bit forced on so status logged.
Severity: Error
Description: An unsupported bit was received in status returned from the disk unit.
Action: Check the drive and the version of software in HSC.
Possible FRUs:
2. Version of software
Insufficient Control Memory for K.stllK.si In Requestor xx

Severity: Error
Description: Not enough Control memory left in the pool to allocate a control block. A certain
amount of Control memory is needed to set up control blocks. Enough memory was not found to
set up control blocks for turning on the K.stilK.si functional code.
Action: Use the HSC SETSHO utility to show available HSC memory (Control, Data, and
Program). If less than 87.5 percent of available Control memory is usable, replace M.std2lM.std
module. Run Off-line TEST MEM by K diagnostic and test Control memory.
Possible FRUs:
2. P.iojlc module
3. Software
Insufficient Private Memory remaining for TMSCP Server

Severity: Error
Description: In the SCT, a parameter determines the maximum number of supported tape
fonnatters. During initialization, all the working K.sti/K.si modules are counted and a
calculation is done showing the maximum number of possible formatters. These two parameters
are compared. Based on the comparison, a certain amount of private memory is allocated for
the TMSCP Server. If that allocated portion of private memory is not enough, this message is
displayed.
Action: Use HSC SETSHO utility to show available HSC Program memory. Ifless than 87.5
percent of available Program memory is usable, replace M.std2. Run Off-line Test Mem or
Test Refresh to test Program memory. Use the SETSHO SET MAX FORMATTER command to
reduce the maximum number of formatters supported.
Possible FRUs:
1. M.std2IM.std module
2. P.iojlc module
3. Software
Internal ConSistency Error

Severi ty: Error
Description: A high-level check detected an inconsistent data structure. For example, a
reserved field contained a nonzero value, or the value in a field was outside its valid range.
This error is probably caused by the requestor microcode or hardware.
Action: If the error is repetitive, check for consistent requestor numbers in detecting requestor
field of error. Determine if any other surrounding error reports indicate a possible internal
memory error.
Possible FRUs:
1. FRU noted in the detecting requestor field
2. M.std2lM.std memory module
K.ci exception detected, code =nnn

Severity: Warning
Description: The code is composed of the contents of KH$FLG (the second word in the K.ci
Control Area). Below is a breakdown of the bits contained in this word.
000001 KHF$PD-Path(s) disabled by Kci due to a transmit error or VC breakage due to
other Kci-detected errors.
000002 KHF$EQ-Item(s) placed on error queue (KH$EQ).
000004 KHF$BL-Data memory error during BMB list operation.
000010 KHF$UP-Unreceivable packet. K.ci stopped (causes a crash).
000100 KHF$NH-Sequenced message received while reserved-to-receive queue was empty.
040000 KHF$PD-Set by diagnostics to disable interrupts.

Action: Compare the code from the printout to the previous list, and determine whether the
error code points to an HSC module or to the host.
Possible FRUs:
1. Status 1: K.pli module
2. Status 4: M.std2lM.std module
3. Status 10: PILA module, host K.ci set
4. Status 100: Host Kci set
K.ci loopback microcode loaded

Error 'J.Ype: CI-detected out-of-band
Severity: Error
Description: The CIMGR detected Kci loopback microcode was loaded during initialization.
When this message occurs, a problem with the Kpli (L0107) module probably exists.
Action: Replace the following FRU.
Possible FRUs: K.pli module
K.sdllK.sl in slot xx. failed Its Inlt orr, status = xxx

Error 'J.Ype: Disk functional out-of-band
Severity: Error
Description: A requestor fails during boot. The displayed K.sdilK.si has failed with the
displayed status. This message is displayed only at the end of the boot procedure.
Action: Record the status for module repair purposes.
Possible FRUs: The K.sdilK.si module displayed.
K.stI/K.s1 In Requestor xx has microcode Incompatible with this TMSCP Server

Error 'J.Ype: Tape functional out-of-band
Severity: Error
Description: The data structure version within the microcode version residing on the Ksti!K.si
module is a lower version than the TMSCP Server can support.
Action: Use the SET REQUESTOR command to ensure the version of microcode on the
KstilK.si module is up to current revision. If not, replace the microcode or replace the K.sti!K.si
module with a KstilK..si module of the current revision.
Possible FRUs: KstilK.si module
Last soft Inlt resulted from unknown cause

Error 'J.Ype: SIN! out-of-band
Severity: Error
Description: Software has a list of known reasons for reboot (Trap through 134, Trap through
250, CRASH$, SETSHO, etc.). If no reason for reboot is apparent, the software may have failed
to detect where the error came from.

Action: Check the HSC console error messages and the system error logs on all the systems
connected to the HSC. This error indicates a probable software problem.
Possible FRUs: Dependent upon the information obtained from the error logs.
LBN xx. repaired for shadow member unit xx.

Description: A shadow Repair operation was done in which good data was written to bad
members of a shadow set.
Action: An uncorrectable error occurred on an LBN on the subject drive and was successfully
rewritten. If the problem persists, check for other errors that would give information on what
the uncorrectable error was.
Possible FRUs: Drive modules or media. (Refer to the drive service manual.)
LBN Restored with Forced Error in RESTOR operation

Disk Unit xx, LBN xx.
Tape Unit xx.
Severity: Warning
Description: An error was detected in the LBN data during backup. A forced error bit was set
in the LBN.
Less than 87.5 percent of (Control, Data, Program) memory is available

Severity: Error
Description: These three messages are a result of the P.ioj/c polling the memories on
initialization and finding an insufficient amount of working memory in either one. Any
combination of the three messages may appear.
Action: The error printout determines which memory is failing.
Possible FRUs: M.std2lM.std module
Level 7 K Interrupt (Trap through 134)

Error Type: SINI out-of-band, subsystem exception
Severity: Error
Description: A level 7 K interrupt occurs when any requestor detects a fatal error condition
while executing functional code. The requestor, upon detecting the error, generates a level 7 K
interrupt to the P.ioj/c. The P.ioj/c traps through location 134, causing a reboot. The requestor
status and the failing requestors' status value are displayed for all requestors on the last line of
the printout.
Action: In some cases, the error printout shows a failing requestor when the real problem is in
the M.std2lM.std module. This error can also be caused by software problems.
Wait for two or more failures of this type to determine if the real problem is the M.std2lM.std
module. If the M.std2/M.std is at fault, the same requestor is not displayed twice as the failing
requestor. Refer to Appendix D for failing status values and their meanings. Check the status
line message to detennine the failing requestor status. Change the requestor exhibiting the
failing status if the same requestor is displayed more than once.
Possible FRUs:
1. Requestor displaying a continuous failing status value
Lost ReadlWrite Ready

Severity: Error
Description: ReadlWrite Ready drops when the controller attempts to initiate a transfer
or at the completion of a transfer with Read/Write Ready previously asserted. This usually
results from a drive-detected transfer error, where additional error log messages containing the
drive-detected error subcode may be generated.
Action: Look for surrounding drive-detected errors and/or associated disk transfer error log.
Move suspect drive to another port or data channel to help isolate failure, as this error may be
caused by any of several communication components.
Possible FRUs:
2. KsdilK.si module
3. SDI cables
Lost Receiver Ready

Severity: Error
Description: Receiver Ready was negated when the controller attempted to initiate an SDI
disk transfer or did not assert at the completion of a transfer. This includes all cases of the
controller's timeout expiring for a Transfer operation (Levell real-time command).
Action: Look for a probable drive error or a possible SDI cable problem. Move suspect drive
to another port or data channel to help isolate failure, as this error may be caused by any of
several communication components.
Possible FRUs:
2. KsdilK.si module
3. SDI cables
Lower Processor Error

Severi ty: Error
Description: A bit was set in the Lower Processor error register. Bits included in the Lower
Processor error register are Data bus NXM, data SERDES overrun, Data bus overrun, Data bus
parity error, data pulse missing, and sync real-time parity error.
Action: Check the K.sti!K.si and the tape formatter.
Possible FRUs: K.stiIK..si module or tape formatter
Lower Processor timeout

Severity: Error
Description: The Upper Processor in the K.sti!K.si detected the Lower Processor had stopped
and restarted it.
Action: Check the K.sti!K.si and tape formatter.
Possible FRUs: K.stiIK..si module or tape formatter
MMU (Trap through 250)

Error Type: SIN! out-of-band, subsystem exception
Severity: Error
Description: A failure was detected in the Memory Management Unit (MMU) on the P.ioj/c.
The active process is displayed as well as the bit assignments for the memory management
status registers.
Action: Examine the MMSR registers to determine the failure in the MMU.
Possible FRUs: P.ioj/c module or software error
nnn Symbol ECC Error

Severity: Error
Description: If a drive has more symbols in error than a drive-defined threshold, the HSC will
print one of the following error messages, even though the error might have been corrected.
One Symbol ECC Error
Two Symbol ECC Error
Three Symbol ECC Error
Four Symbol ECC Error
Five Symbol ECC Error
Six Symbol ECC Error
Seven Symbol ECC Error
Eight Symbol ECC Error
U ncorrectable ECC Error
The following description covers all of the ECC error types that are printed.
ECC errors occur when the data read from the disk does not agree with the data written. When
data is written to the disk, an ECC is calculated (by the R-S GEN) and appended to the end of
the sector. When the data is subsequently read from the sector, the ECC is revalidated. The
two possible results are:
1. The data error falls within the ECC error correction capability Oess than nine lO-bit
symbols in error) and data correction is performed. In this case, depending on the drive
type, no data errors are shown.
2. The data error does not fall within the error correction capability of the ECC, and the error
is retried according to drive dependent parameters. If all of the retries fail, an uncorrectable
ECC error occurred and a bad block is reported through an end packet.
NOTE
An uncorrectable ECC error is reported when a transfer with the Suppress Error
Correction modifier encounters an ECC error of any severity.
Action: Determine if the ECC errors are just normal events or if a very large number of blocks
is being replaced. The latter indicates the drive may have a read path problem.
Possible FRUs:
2. KsdilK.si module
No control block available to satisfy HMB request.

Error Type: CI -detected out-of-band
Severity: Warning
Description: The CIMGR tried to allocate an Host Memory Block (HMB) from the Free
Control Block Queue when none were available. If a significant amount of Control memory was
removed from use due to errors detected during boot, this message occurs. Otherwise, it may
indicate an internal HSC software problem where control blocks in HSC memory are taken by
some service and never returned to the list of free control blocks.
Action: Type in the SHOW MEMORY command for HSC50 software version V300 and later
and HSC software version VIOO and later to determine how much Control memory is being
used. Compare the amount of Control memory shown on the SHOW MEMORY printout to
the amount contained in the HSC. If more than 12.5% has been disabled from use, replace the
memory module.
For HSC50 software before V300, run the off-line memory test on Control memory to determine
if excessive solid failures are causing removal of a large amount of memory. If memory amount
is adequate, the problem may be caused by a software or microcode problem within the HSC.
Possible FRUs:
2. Software
No -tape drive structures available for Requestor xx Port xx Unit xx

Increase structures through SET MAX_TAPE command
Severity: Error
Description: An additional tape drive has been added to an existing tape formatter, but the
tape structures set up in initialization have been exceeded.
Action: Use the SETSHO utility to increase to the number of tape structures with the SET
MAX_TAPE command.
Possible FRUs: None
No tape formatter structures available for Requestor xx Port xx

Increase structures through SET MAX_FORMATTER command
Severity: Error
Description: An additional tape formatter has been added to the HSC, but since tape
formatter structures are set up during initialization, not enough structure space is available for
this additional tape formatter.
Action: Use the SETSHO utility to set the structure level higher to compensate for the
additional tape formatter with the SET MAX_FORMATrER command.
Possible FRUs: None
No usable K.stilK.si boards were found by the TMSCP Server

Severity: Error
Description: The TMSCP Server polled the HSC and found no working KstiJK.si modules.
This message does not appear frequently because the TMSCP Server software is not usually
loaded if there are no K.stiIK.si modules.
Action: Check for a failed initialization diagnostic error message prior to this message. This
prior message displays the failed requestor slot and failing status.
Possible FRUs: The K.stiIK.si(s) module displaying the failing status.
Node nn cables have gone from crossed to uncrossed

Severity: Error
Description: This message occurs only when check fora crossed path finds a previously
crossed path no longer crossed. More detail is covered in the description of the error message
Node nn Cables have gone from uncrossed to crossed.
Action: Note, if both the "uncrossed to crossed" and "crossed to uncrossed" messages are
occurring, it is probably an indication of failing hardware, not a cable problem. See the Action
in the next message for more detail.
Possible FRUs:
1. CI cables, if a single message is displayed
2. Kci module set, if both messages are displayed
Node nn cables have gone from uncrossed to crossed

Severity: Warning
Description: This message occurs when an IDRSP (ID Response) packet is received by an
HSC in response to an IDREQ (ID Request) message. Upon receiving an IDRSP packet, the
HSC checks two bits in the IDRSP message that indicate which path was used by the sending
node. If these two bits do not indicate the same path the HSC received the message on, this
error .occurs.
Action: Determine if the problem is broken hardware in the HSC CI interface, broken
hardware in the host CI interface, or if the CI cables are crossed. Before replacing any modules
or cables, determine if the HSC is encountering crossed paths to multiple nodes in the cluster
or only to a particular node. If the HSC is encountering crossed paths to all nodes, the problem
is probably in the HSC or the cables. If it is encountering the problem to only one node, it is
likely a problem with that host node's CI module set or the cables running from the host to the
Star Coupler.
Possible FRUs:
1. Cables physically connected wrong at HSC, Star Coupler, or host CI
2. Any of the three K.ci modules in the HSC: LINK (LOlOOILOl18), PILA (LOI09), and K.pli
(LOI07)
3. Host CI module set
4. Duplicate node address settings
Node nn path has gone from bad to good

Severi ty: Warning
Description: A disconnected CI cable has been reconnected, or an intermittent hardware or
cable problem is indicated. More detail is found in the description of the error message Node
nn Path (A or B) has gone from good to bad.
This message also occurs if an open VC node path was previously found to be bad. During
this polling cycle the node sends out ID_REQ (lD Request) packets to all nodes and receives
successful ID_RSP ID Response messages.
Action: If the cable was reconnected, there is no further action. Otherwise, replace the possible
FRUs.
Possible FRUs:
1. CI cable
2. Host
3. CI interface hardware in the host
Node nn path (A or B) has gone from good to bad

Error Type: CI-detected .out-of-band
Severity: Warning
Description: K.ci microcode detects a hard (unrecoverable) transmission error on a previously
good path. Examples of hard transmission errors are:
Transmit Buffer Parity Error
Unrecoverable NACK
Unrecoverable NO_RSP
Transmitter Attention Timeout
Determining the reason for failure using the error message is not possible.
Action: Before replacing any FRU, determine if the message is occurring because of problems
with one host or problems with multiple hosts. If the problem involves one host, it is probably
in the Star Coupler's host side. If the problem involves multiple hosts, it is probably on the
Star Coupler'S HSC side. Also, if the message occurs on both paths to a host, that host may
have been powered down, stopped, or may have crashed. Examine the host console log and the
error log to determine if something did happen to the host.
Determining which error caused the bad path is not possible except with the Transmit Buffer
Parity Error (XBUF PE) which prints as an MSCP type message.
Possible FRUs:
1. CI cable
2. HSC/CI interface
3. Host/CI interface
NXM (Trap through 4)

Severity: Error
Description:
a. A memory location did not respond within the specified timeout period.
b. A stack overflow occurred.
c. An odd address access was attempted. For example, a word access instead of a byte.
d. A halt was executed in User mode.
Action: Determine which memory is failing by examining the low and high error address
registers for module repair.
Possible FRUs:
2. P.iojlc module
Parameter change, process yyy

PC xxx
PSW xxx
Reason xxx
Description: A parameter has been changed through the SET/SHO utility.
Action: None
Possible FRUs: None
Parity Error (Trap through 114)

Severity: Error
Description: This message covers parity errors in memory and in cache. In the case of a
memory parity error, the address of the failing memory is latched into the low error address
register. In the case of a cache parity error, the address is not latched into the low error address
register. Instead, the address of the low error address register is displayed in the error printout.
Action: Determine if the error occurred in memory or in cache memory by reading the contents
of the low error address displayed in the error printout. If the contents is the address of the
low error address register (170024), the error is in cache memory. If the error is in cache, the
probable FRU is the P.ioj.
Possible FRUs:
1. P.ioj/c module
2. M.std2fM:.std module
P.loj/c running with memory bank or board swap enabled

Severity: Error
Description: Upon initialization, an error was detected in the low address space of private
memory. The P.ioj/c asserted the Swap Bank signal, and the second bank of private memory
was enabled. The P.ioj/c and memory combination can still function under limited capabilities.
Action: Replace the M.std2/M.std module. The HSC still functions with limited capabilities.
Possible FRUs: Replace memory module
PLI Receive Buffer Parity Error

Severity: Error
Description: When the data from the packet in a receive buffer on the PILA module was
transferred to the K.pli module, a parity error was detected on the bus. In this case, parity is
generated by the LINK module (L01001L0118) and checked by the K.pli module (L0107). The
PILA module stores the data without checking or generating parity.
Action: If failure is persistent and is accompanied by Kci level 7 K interrupt HSC crashes,
analyze K.ci module status code for more detailed information. Run Off-line Test K diagnostic
to test K.ci. Any error report should more clearly indicate the specific K.ci module failure. For
very intermittent failures follow the sequence of possible FRUs.
Possible FRUs:
1. PILA module
2. Kpli module
3. LINK module
PLI Transmit Buffer Parity Error

Severity: Error
Description: When data was being transferred from the K.pli to the PILA transmit buffer, a
parity error was detected on the bus. In this case, parity is generated by the K.pli module and
checked by the LINK module. The PILA module stores the data without checking or generating
parity.
Action: If failure is persistent and is accompanied by Kci level 7 K interrupt HSC crashes,
analyze K.ci module status code for more detailed information. Run Off-line Test K diagnostic
to test K.ci. Any error report should more clearly indicate specific K.ci module failure. For very
intermittent failures follow the sequence of possible FRUs.
Possible FRUs:
1. PILA module
2. LINK module
3. Kpli module
Position or Unintelligible Header Error

Severity: Error
Description: The drive reported a Seek operation was successful by returning successful
status in response to the INITIATE SEEK SDI command and asserting RJW Ready when on
the desired cylinder. However, the cont:roller determined the drive had positioned itself to an
incorrect cylinder. The header read from the drive is consistent (three out of four header copies
are identical) but does not match the desired target header value. The transfer will be retried
several times and the error is considered recoverable if the error flags bit indicates success or a
subsequent replacement succeeds.
Action: The drive servo system or media is probably at fault in this case. If one is available,
move the drive to a different requestor. A drive failure is indicated if the failure persists on the
new requestor.
Possible FRUs:
2. KsdilK.si module
Positioner error on disk unit xxx. DRAT addr: xxx

Desired hdr (Io,hl):xxx, xxx
Actual hdr (Io,hl):xxx, xxx
Description: The drive positioned the heads in the wrong place or the HSC software is
processing transfers out-of-order.
Action: Check drive modules and the K.sdilK.si module .
. Possible FRUs:
1. Drive modules . (Refer to the drive service manual.)
2. KsdilK.si module
Premature LP flag In RTNDAT sequence from host node xx

Severity: Warning
Description: A violation of packet protocol; the last packet flag was set before all data was
received from a host.
Action: If the problem is transient, monitor error for repetitive node numbers as this may
indicate a host CI problem. If the problem is persistent across all cluster nodes, test the Kci.
Possible FRUs:
1. Kci modules
2. CI cables
Pulse or Parity Error

Severity: Error
Description: The controller detected a pulse error on either the SOl drive state or data line,
or the controller detected a parity error in a drive state frame. The HSC does an SDI GET
STATUS command, reports any errors from it, and then clears those errors, if possible. After
this, the HSC retries the original command up to two more times before considering the error
unrecoverable.
Action: If the error is reported on more than one drive, a K.sdilK.si problem is indicated. If
the error is reported on only one drive, an SDI cable or drive problem is indicated.
Possible FRUs:
2. SDI cable
3. SDI transition bulkhead
RCT Corrupted Error

Severity: Error
Description: The ReT search algorithm encountered an invalid RCT entry. The subcode may
be returned under the following conditions:
During replacement of a block
During nonprimary revectoring of a block
When bringing a unit on line
Action: Determine if this error is repetitive for this unit possibly indicating a defective media
or drive read path failure. Run the HSC utility VERIFY on the drive.
Receiver ready not asserted at start of transfer

Severity: Error
Description: The HSC is ready to start a transfer by sending the formatter a Level 1 command
and the formatter does not have Receiver Ready asserted.
Action: Check the formatter, cable, and K..stiJK.si.
Possible FRUs:
1. Formatter
2. Cable
3. K..stilK.si module
Record EDC error

Severity: Error
Description: On a read from tape operation, the EDC calculated by the K.stiJK.si did not
match the EDC generated by the tape formatter.
Action: Check the formatter, cable, and K.stiIK.si.
Possible FRUs:
1. Formatter
2. Cable
3. K.stilK..si module
Requestor xx failed INIT diags, status = xxx

Error Type: SIN! out-of-band
Severi ty: Error
Description: The data channel in the displayed requestor has failed initialization diagnostics
with the displayed status.
Action: Determine which data channel is in the displayed requestor slot. Make note of the
status value for module repair. Replace the failing data channel.
Possible FRUs: The data channel (K.sdi, K.sti, or K.si) module exhibiting the failing status.
Requestor xx has failed initialization diagnostics with status =xx

Severity: Error
Description: The requestor in slot xx has failed initialization diagnostics with the displayed
status. The message indicates the failed K.stiIK.si module.
Action: Refer to Appendix C to determine what the displayed status indicates the failure to be.
Possible FRUs: The K.stiIK.si module in the indicated slot.
Reserved Instruction (Trap through 10)

From process yyyy
PC xxx
PSW xxx
Severity: Error
Description: The P.ioj/c detected an Opcode, resulting in the execution of an invalid
instruction. The process indicated is the process that executed the nonexistent instruction.
Action: Determine what process was active for module repair.
Possible FRUs:
1. P.ioj/c module
3. Software
Resource lost to K.ci-xxx xxx HMBs

Severity: Error
Description: A Control memory Host Message Block (HMB) data structure was lost. HMBs
were expected in the sequence message ready to receive queue (.KHSRR), but none were found.
Action: Report the error, with frequency of occurrence, to support. Also, note sequence of
events that reproduce this failure. This message indicates a software bug. Verify dc power
levels are correct.
Possible FRUs:
1. Software
2. Main power supply
Retry limit exceeded while attempting to restore tape position

Severity: Error
Description: A command was issued to restore the tape position, and the command failed in
the limit of retries.
Reverse retry currently not supported
NOTE
As of Va.50 and above, Reverse Retry is supported.
Severity: Error
Description: Reverse Retry requests from the formatter were not supported before Version
3.50 ofHSC software.
Action: Update software
Possible FRUs: None
Rewind failure
Severity: Error
Description: A command for a rewind was issued, and the command failed (the controller
received an unsuccessful response from the formatter).
Action: Check the drive and/or formatter.
Possible FRUs:
2. Formatter
SCT read or verification error. Using template SeT.

Error Type: SIN! out-of-band
Severity: Error
Description: An error was detected by the P.ioj/c as it attempted to read the System
Configuration Table (SCT) or as it attempted to verify the SCT. This error message will
occur when new, previously uninitialized system diskette is booted. The default settings
from SYSCOM are used instead of the SCT from the load media. The second sentence in
this message indicates the SCT is new, as derived from the template SCT settings set in the
factory. If the system has been previously booted from the same media, a system load device
failure is indicated.
Action: Reinstall the old system diskette and do a SHO SYSTEM. Install the new diskette
exhibiting the error and set all system diskette fields to the old values using the SET command.
Reboot the HSC to validate these values and ensure system continuity.
Possible FRUs: System diskette
SOl exchange retry on disk unit xxx. (Requestor xx. Port xx.)
DeB addr xx Error count xx.
Description: Retry the SDI command on the drive.
Action: None
Possible FRUs: None
SOl Clock Persisted after INIT
Severity: Error
Description: The drive clock did not cease following a controller attempt to initialize the
SDI drive. This implies the drive did not recognize the initialization attempt. This error
condition causes the HSC to retry the lnit command eight more times before marking the drive
inoperative.
Action: Determine if this drive has encountered any other related problems which may be
entered in an appropriate error log report. Also, this error may be due to an SDI cable problem.
Closely examine error logs for surrounding disk errors, as the error may be a result of a
previously-reported drive error.
Possible FRUs:
2. SDI cable
SI Clock Resumption Failed after INIT

Severity: Error
Description: The drive clock did not resume following a controller attempt to initialize the
SDI drive. This implies the drive encountered a fatal initialization error. Closely examine error
logs for surrounding disk errors, as this error may be the result of a previously-reported drive
error.
found in an appropriate error log report. Also, this error may be due to an SDI cable problem.
Possible FRUs:
2. SOl cable
SI Command Timeout

Severity: Error
Description: The controller timeout expired for either a level 2 exchange or the assertion of
ReadlWrite Ready after an INITIATE SEEK command. The HSC retries the command three
more times, reinitializing the SOl drive each time. If the error persists on a single SDI level 2
exchange, the drive is marked inoperative.
found in an appropriate error log report. Also, this error may be due to an SDI cable problem.
Closely examine error logs for surrounding disk errors, as the error may be a result of a
previously-reported drive error.
Possible FRUs:
2. SDI cable
Ensure the drive and all HSC modules are at the latest revision levels.
SI Receiver Ready Collision

Severity: Error
Description: This error occurs when the drive fails to follow the SDI protocol during SDI
command/reception. For example, the controller sends the drive a command, asserts Controller
Receiver Ready, and waits for the SDI response. The following lists the possible drive operations
that lead to this error:
1. The drive fails to deassert Drive Receiver Ready. In this case, the drive indicates it did not
receive the command.
2. The drive deasserts Drive Receiver Ready and then reasserts it before sending a proper
SDI response. In this case, the drive believes it has sent a response and is indicating so by
reasserting Drive Receiver Ready, yet the controller has never received the response.
The HSC KsdilK.si detects this error. The HSC functional code does an SDI GET STATUS
command and clears the drive of any errors found. The original command is then retried. This
cycle is repeated twice before the drive is initialized by the HSC, and the entire operation is
done two more times. If the failure persists, the drive is marked inoperative.
found in an appropriate error log report. Also, this error may be due to an SDI cable or SDI
transceiver/encoder/decoder problem. Closely examine error logs for surrounding disk errors, as
this error may be the result of a previously-reported drive error.
Possible FRUs:

2. SDI cable
51 Response Length or Opcode ~rror

Severity: Error
Description: A level 2 response from the drive had correct framing codes and checksum but
was not a valid response within the constraints of the SDI protocol. The response had an
invalid Opcode, was an improper length, or was not a possible response in the context of the
exchange.
The HSC K.sdilK.si detects this error. The HSC functional code does an SDI GET STATUS
command and clears the drive of any errors found. The original command is then retried. This
cycle is repeated twice before the drive is initialized by the HSC, and the entire operation is
done two more times. If the failure persists, the drive is marked inoperative.
Action: Determine if the drive has experienced other similar errors. Closely examine error logs
for surrounding disk errors, as this error may be the result of a previously-reported drive error.
Possible FRUs:
51 Response Overflow
Severity: Error
Description: A drive sent back more frames than the reception buffer could hold. This can be
caused by a hung drive microdiagnostic or a malfunctioning K.sdilK.si.
Action: Determine if the drive is failing in other ways, indicating a drive problem. If not, the
KsdilK.si may be the more likely cause.
Possible FRUs:
2. KsdiIK.si module
SERDES Overrun
Severity: Error
Description: This error is either a SERDES overrun or underrun error. Either the drive is
too fast for the controller, or a controller hardware fault prevented controller microcode from
keeping up with data transfer to or from the drive.
Action: Determine if other errors have occurred that may indicate a KsdilK.si problem. Move
the offending drive to another requestor. If the problem persists, test the drive further.
Possible FRUs: K.sdilK.si module
Software inconsistency (Trap through 20)

Severity: Error
Description: During operation, the operating software perfonns numerous consistency checks.
When one of these consistency checks fails, the HSC crashes and reboots. The active process is
displayed, as well as the stack dump.
Action: Submit a Software Problem Report (SPR). (Refer to Appendix B).
Possible FRUs: None
Tape drive requested error log

Severity: Warning
Description: The drive detected an error condition and set the EL bit for an error log to be
taken.
Tape formatter connected to Requestor xx Port xx has been declared inoperative.

Severity: Error
Description: The K.stiIK.si has sent a nondata transfer command over the STI cable to the
displayed tape fonnatter three times and has received back the same error three times. The
HSC then ignores the tape formatter until it detects some intervention such as a change in the
state clock.
Action: Replace the possible FRUs. Deasserting the tape drive's port switches, recycling power,
unplugging the STI cable, or any action causing the state clock to come and go is considered an
intervention. The HSC will not attempt to communicate with the failing tape fonnatter until
it detects this change in state clock. Examine any previous error reports for more specific data
regarding this error message.
Possible FRUs:
1. Tape formatter
2. STI cabling
3. K.stiIK.si module
Tape unit number xx connected to Requestor xx Port xx ceased to exi~ while on line
Severity: Error
Description: This message is similar to the previous error message except in the case where
the HSC was using the tape drive to do data transfers when the tape drive went off line.
Action: Check to see if a breaker has blown. The tape drive may be in testing mode also,
causing the tape drive to go off line.
Possible FRUs:
1. Tape drive
2. Tape formatter
3. STI cable
Tape unit number xx connected to Requestor xx Port xx dropped state clock while on line
Severity: Error
Description: The formatter supplies the state clock over the STI cable. The state bits are
encoded on this state clock waveform such as AVAILABLE and ATTENTION. As long as the
KstilK.si is receiving a state clock, the STI cable must still be plugged in, and the formatter
must be operating correctly. Dropping state clock is equivalent to disconnecting the STI cable
from the HSC.
Action: First isolate the problem to the HSC, STI cable, or tape drive. Next, try replacing or
swapping the KstilK.si module exhibiting the failure. If the problem is not solved, try a known
good tape drive.
Possible FRUs:
1. STI cable
2. Tape drive
3. KstilK.si module
Tape unit number xx connected to Requestor xx Port xx is not asserting available when it should be
Severity: Error
Description: The formatter is not on line and is not asserting its Available signal to the HSC.
The HSC does not detect the Available signal and displays this message on the local console
terminal.
Action: First isolate the problem to either the HSC, the STI cable, or the tape drive. Next, try
replacing or swapping the K.stilK.si module exhibiting the failure. If the problem is not solved,
try a known good tape drive.
Possible FRUs:
1. STI cable
2. Tape drive
3. KstiJK.si module
Tape unit number xx connected to Requestor xx Port xx went available without request

Severity: Error
Description: When the formatter is on line, Available is not normally asserted to the HSC.
When the formatter is on line and doing 110 and an Available is asserted, the HSC detects this
as an error. A formatter does not need to send Available unless the K.stilK.si requests it.
Action: First isolate the error to the formatter or to the active K.stilK.si.
Possible FRUs:
1. KstiJK.si module
2. Formatter
3. STI cable
Tape unit number xx connected to Requestor xx Port xx went off line without request
Severity: Error
Description: The formatter lost contact with one of the tape drives. The HSC detected this
loss of a tape drive and printed this message.
Action: Check to see if a breaker has blown. The tape drive may be in diagnostic mode also,
causing the tape drive go off line.
Possible FRUs:
1. Tape drive
2. Tape formatter
3. STI cable
TMSCP fatal initialization error-TMSCP functionality not available

Severity: Error
Description: Something went wrong during initialization with the tape functional code
(TFUNCT). A routine was called up to initialize some part of the functional code, and that
part failed to initialize. Typically, some other message is displayed prior to this message giving
more detail on the error.
Action: Take action depending on the previously displayed message.
Possible FRUs: Dependent on the previously-displayed error message
TMSCP Server operation limited by insufficient Private memory. Use the SET MAX command to reduce
private memory requirements.
Severity: Error
Description: This message appears before the message Insufficient private memory
remaining for TMSCP Server and indicates the same problem. Private memory has
insufficient space to hold the necessary structures the TMSCP Server needs as dictated by
the number of KstilK.si modules and the number of tape formatters on the HSC.
Action: Use HSC SETSHO utility to decrease maximum number of tape formatters for which
the HSC should reserve memory structures.
Possible FRUs:
2. P.ioj/c module
3. Software
TOPOLOGY command failed

Severity: Error
Description: A TOPOLOGY command was issued and the command failed.
TTRASH fatal initialization error

Severi ty: Error
Description: This message is similar to the message TMSCP fatal initialization error--
TMSCP functionality not available except the process failing to initialize is TTRASH
instead of the tape functional process (TFUNCT).
Action: Check for previous error reports displaying a more specific reason for this error report.
If earlier error messages do not exist, reboot HSC using backup HSC software copy.
Possible FRUs:
1. M.std2 modulefM.std module
2. Software
Unable to position before LEOT

Severity: Error
Description: The command to position the tape was unable to complete before LEOT was
detected.
Unclearable Drive Error

Severi ty: Error
Description: Issued a clear bit three times and the bit does not clear.
Action: Check the formatter and drive. Further analysis of tape drive error log may be
necessary.
Possible FRUs:
2. Formatter
3. STI cable set
4. K.stilK.si module
Unclearable Formatter Error

Error Type: Tape en-or
Severity: Error
Description: Issued a clear bit three times and the error does not clear..
Action: Check the formatter
Possible FRUs:
1. Formatter
2. STI cable set
3. K.stilK.si module
Unexpected AVAILABLE signal from ONLINE disk unit xx.

Description: The disk is asserting AVAILABLE while the drive state is ONLINE. This is not
an expected condition.
Action: Determine why the disk drive is asserting the Available signal.
Unit xx. declared Inoperative because no progress. made on Command Reference xxxxx.
Severity: Error
Description: The HSC Disk Path has made no progress on the host command represented by
the given reference number in an extended time period. This scenario can occur if the drive is
degraded to a point where the Disk Path spends too much time in error recovery and can make
no progress on the host command.
Action: The HSC was unable to complete error recovery on the drive and took it off line. Check
the drive with diagnostics to determine the nature of the problem.
Unknown K.tape error

Error Type: Tape en-or
Severity: Error
Description: The ER bit was set but was undefined.
Unrecoverable error on disk unit xx. Drive appears Inoperative.

Severity: Error
Description: An error log message from the drive caused this message, or the drive may be off
line. The Disk Path has concluded that the drive in unusable.
Action: Check the error log and drive.
Unsuccessful SEEK initiation, disk unit xxx. DCB addr: xxx

Description: The dialog control block sent the SEEK exchange and the DCB was sent to its
error queue by the K.sdilK.si. The SEEK may have been rejected, lost, or completed with an
error.
Action: Check drive.
VC closed due to timeout of RTNDAT/CNF from host node xx

Description: The host issued a request over the CI, and the response timed out.
Action: Determine if the problem lies in the HSC K.ci module set or the host CI module.
Possible FRUs:
1. K.ci module set in the HSC
2. CI module set in the host
VC closed with node nn due to disconnect timeout

Severity: Warning
Description: A second disconnect call for the same connection block has been received by the
CI Manager.
Action: Verify other cluster nodes have not failed or have CI port problems. If the problem
persists, run Off-line Test K diagnostic to test K.ci. If no failures exist, verify SET parameters
are valid, use backup copy of the HSC code, and replace FRUs indicated.
Possible FRUs: Host K.ci module set
VC closed with node nn due to request from K.cl

Severity: Warning
Description: The Kci microcode has detected both CI paths have gone from good to bad during
polling. More details are found under the description for error message Node nn path n has
gone from good to bad.
Action: Set error and outband to info. See the descriptions and action for the following error
messages:
Node nn path (A or B) has gone from bad to good
Node nn path (A or B) has gone from good to bad
Possible FRUs:
1. Kci hardware interface in HSC
2. CI cables
3. Host CI hardware
VC closed with node nn due to START received

Severity: Warning
Description: A start message is received over the CI to an already open Virtual Circuit (VC).
Action: Check for two HSCs with the same ID (not node address) on the cluster. This happens
when a new HSC is installed on the cluster and is given an existing ID.
Possible FRUs: CI cables
VC closed with node nn due to unexpected disconnect

Severity: Warning
Description: The HSC receives a DISCONNECT_REQ packet, and the following conditions
exist inside the HSC.
• A connection is not open.
• The HSC is not in the DISCONNECT_SENT state. (The DISCONNECT_SENT state
indicates the HSC also sent a DISCONNECT_REQ packet.)
Action: Verify no other nodes in the cluster failed and caused sending an unexpected
disconnect to the HSC. If failure persists, the K.ci module set may be causing this error. Run
Off-line Test K diagnostic to test Kci. If no failure, verify no duplicate node addresses exist in
this cluster with the LINK module (LOI00\LOl18) node address switches.
Possible FRUs: K.pli module
VC open with node nn

Description: A Virtual Circuit (VC) has been established with the given node. The Online
lamp on the HSC Operator Control Panel lights the first time a VC is established to an HSC.
Action: None is required; this message is for informational purposes only.
Possible FRUs: None
***WARNING*** K.sti microcode too low for large transfers.

Severity: Warning
Description: The amount of I/O the Ksti can accommodate is restricted. The code still
attempts to do transfers, but a warning has been issued.
Action: Update the microcode version level to the proper revision.
Possible FRUs: Change the level of K.sti microcode to a supported version, or change the Ksti
with the out-of-date code.
Word rate clock timeout

Severity: Error
Description: The KstiJK.si detected the loss of clocks from a drive during a transfer.
Action: Check the formatter and the cable.
Possible FRUs:
1. Formatter
2. Cable
lntemal Cabling Diagrams A-1
A
Internal Cabling Diagrams
A.1 Introduction
This appendix contains diagrams of the internal cabling for the HSC, HSC50 (modified), and
HSC50.
A.2 HSC Internal Cabling

Figure A-l is a diagram of the HSC internal cabling.
A-1
A-2 Internal Cabling Diagrams
FLOPPY SIGNAL CABLE

(1701167-O1)
DRIYEO
I
...
DC ON/OFF
SWITCH
DRIYEl
J
SECURE ENABLE SWITCH
DC ON/OFF CABLE S2
(1701231-02) ~
I ./ OCP/COYER ASSY
I' {7023132-01)
OPERATOR CONTROL PANEL
7020203-01
(7023138-01)
1
STANDARD POWER \
SUPPLY ASSEMBLY
(7020033-03 OR -04)
~ ~ OCPCABLE
70111680-01..... ~ TBl-3 II ~ (1701203-01)
~ 1\ .... ~cl.1 ~
AIR FLOW SENSOR
IT! 2
8 7 6 5 ' 7 ,,\
RELAY
(1220!1!18-01)
r.====I=IIII.fi!ilr;"'I~~B1_1 1\ I r:J:I..
OCP TO ROCKER SWITCH
(1701202-01) K1-1
t ~/K1-3
AC ACC\
70201117-01 • L.LJ
~ IEtff
, ~ AIR FLOW SENSOR CABLE ')
1L1~J7 I~ -...... (1701275-01)
~OCPIBACKPLANE CABLE
~ I~ (701215-01) 12280!12-01 V
TBl-2 r--e-AC-K-P-L-AN-E-{-R-EA-R-YI-E-W)-- ~ AIR FLOW
BLOWER SENSOR
" "~i~~~~~~~~5v~~{120'~IY':;t::%: ~ ~/ ~ / (1701278-020R-03)

70201l1li-01 " ~~~ ~~ I
.~~-;:'!-~
. .~.:-~~-~-,
~" ~.-.~" Y~" 'I ~ ~ ~ /v / ~ 000000 ~ u,.ot=-="""'1I
,......" ,~ r;:: ~ " ~ 00000001 "0'=
I')~~J. ~ ... ~ "·cpe. -::oot=+===4l==u.J..
~~~
I _..-POWER SUPPLY 'FLEXCIRCUIT I /E1ACABLE
" AC LINE CORD (1701265-01) I' (1701267-O1)
/ (1701276-01) +TOFRAME - (/
POWER SU~PLY/ rr=====#--'=; II n ~17~;::'~

~;;~~;~rD r-...I~~'''2_ _...I~~!_ _...I~~1...,70111682-O{ CO~Lpm lED
~:T~~"gNNECT N M L
NOT USED
K J
" t"--REAR SHIELD/
C1 CABLE
(1701268-01) 0 0 0 Cl CJ CJo ASSEMBLY
(7023140-01)
o 0 0 Cl CJ CJ1~
POWER CONTROLLER
ASSEMBLY
(3024374-01 OR 02)
-I I
" ' 3 PHASElNEUTRALlGND

AC POWER CORD
RELAYTOPCA/FSENSOR~~===================================-======--~
(1701231-01)
CXO-944A
Sheet 1 of 5
Figure A-1 (Cont.) HSC Internal Cabling

Internal Cabling Diagrams A-3
WIRE TABLE FOR 1701202-01 OCP TO ROCKER SWITCH

COLOR FROM TO SIGNAL REMARKS
P4-01 NO CONNECTION KEYING PLUG
P4-02 NO CONNECTION
RED P4-03 S1-3 . +5 VOLT
BLACK P4-04 S1-6 GND (-5 VOLT)
P4-05 NO CONNECTION SPARE
YELLOW P4-06 S1-4 GND
YELLOW P4-07 S1-5 TERM ENABLE
WHITE P4-09 S1-1 INIT SWL
WHITE P4-10 S1-2 INIT L
WIRE TABLE FOR 1701203-01 OCP CABLE

COLOR FROM TO OCP SIGNAL REMARKS
YELLOW J40-1 P3-1 STATE LAMP L
YEUORG J40-2 P3-2 POWER LAMP L
YEUBLU J40-3 P3-4 LAMP ENA 0 L
YEUGRN J40-4 P3-3 TERM ENA L
YEUBLK J40-5 P3-6 LAMP ENA 2 L
YEUVIO J40-6 P3-5 LAMP ENA 1 L
YEUGRY J40-7 P3-8 LAMP ENA 4 L
YEL/WHT J40-8 P3-7 LAMP ENA 3 L
YEL/RED J40-9 P3-10 PANEL SWITCH 1 L
YEUBRN J40-10 P3-9 PANEL SWITCH 0 L
YEL/BLK/GRY J40-11 P3-12 PANEL SWITCH 3 L
YEUGRN/ORG J40-12 P3-11 PANEL SWITCH 2 L
YEL/RED/WHT J40-13 P3-15 BDCOKH (lNIT L}
BLACK J40-14 P3-14 GND
RED J40-15 P3-16 +5V
P3-20 KEYING PLUG (OCP)
WIRE TABLE FOR 1701215-01 OCP/BACKPLANE

COLOR FROM TO OCP SIGNAL REMARKS
YELLOW J12-1 P40-01 STATE LAMP L
YELLOW/ORG J12-2 P40-02 POWER LAMP L
YELLOW/BLUE J12-3 P40-03 LAMP ENA 0 L
YELLOW/GRN J12-4 P40-04 TERM ENA L
YELLOW/BLACK J12-5 P40-05 LAMP ENA 2 L
YELLOW/VIOLET J12-6 P40-06 LAMP ENA 1 L
YELLOW/GRAY J12-7 P40-07 LAMP ENA 4 L
YELLOW/WHITE J12-8 P40-08 LAMP ENA 3 L
YELLOW/RED J12-9 P40-09 PANEL SWITCH 1 L
YELLOW/BRN J12-10 P40-10 PANEL SWITCH 0 L
YEL/BLKlGRY J12-11 P40-11 PANEL SWITCH 3 L
YEUGRN/ORG J12-12 P40-12 PANEL SWITCH 2 L
YEL/RE D/WHT J12-13 P40-13 BDCOK H (lNIT l)
RED J12-16 P40-15 +5 VOLTS
RED J12-19 P41-04 +5 VOLTS
RED J12-20 P42-04 +5 VOLTS
VIOLET J12-25 P41-01 +12 VOLTS
VIOLET J12-26 P42-01 +12 VOLTS
CXO-944A
Sheet 2 of 5

A-4 Intemal Cabling Diagrams
WIRE TABLE FOR 1701231-01 RELAY TO PC A/F SENSOR

WHITE K1-3 PS-1 TRIP
WHITE K1-S PS-2 RETURN
WIRE TABLE FOR 1701231·02 DC ON/OFF

YELLOW S2-2 P33-4 ON/OFF (-S.2V)
ORANGE S2-1 P33-3 S2-
BLUE S2-4 P33-2 ON/OFF (+S.2V)
BLACK S2-3 P33-1 S1-
WIRE TABLE FOR 1701266-01 BP TO PS

VIOLET J13-1 P31-1 +12V
VIOLET J13-2 P31-3 +12V
VIOLET J13-3 P31-S +12V
VIOLET J13-4 P31-7 +12V
BLACK J13-S P31-9 GND (+12V)
BLACK J13-6 P31-2 GND (+12V)
BLACK J13-7 GND (+12V) DOUBLE
P31-4 STANDARD
BLACK J13-S GND (+12V) CRIMPED
POWER
ORANGE J13-9 P31-6 -S.2V SENSE TWISTED SUPPLY
BLACK J13-10 P31-S GND (-SV SENSE) PAIR
BROWN J13-11 P31-10 POWER FAIL L
BLACK J13-14 J32-1 GND (+SV SENSE) TWISTED
RED J13-13 J32-2 +SV SENSE PAIR
BLACK J13-16 J32-3 GND (+12V SENSE) TWISTED
VIOLET J13-1S J32-4 +12V SENSE PAIR
BLACK J13-17 PSO-2 GND (+SV SENSE) TWISTED OPTIONAL
RED J13-1S PSO-1 +SV SENSE PAIR POWER
BROWN J13-20 PSO-3 POWER FAIL L SUPPLY
WIRE TABLE FOR 1701267-01 EIA

COLOR FROM TO BACKPLANE SIGNAL REMARKS
WHITE J11-1 J60-20 HSC RDY+
WHITE/BLK J11-2 J60-6 TERM PRES L
WHITE/BLU J11-3 J60-1 TERM XMT-
WHITE/ORG J11-4 J60-2 TERM XMT+
WHITE/RED J11-S J60-3 TERM RCV+
WHITEIVIO J11-6 J60-7 TERM RCV-
WHITE/BLK J11-10 J61-6 AUX1 PRES L
WHITE/BLU J11-11 J61-1 AUX1 XMT-
WHITE/ORG J11-12 J61-2 AUX1 XMT+
WHITE/RED J11-13 J61-3 AUX1 RCV+
WHITE/V 10 J11-14 J61-7 AUX1 RCV-
WHITE/BLK J11-1S J62-6 AUX2 PRES L
WHITE/BLU J11-19 J62-1 AUX2 XMT-
WHITE/ORG J11-20 J62-2 AUX2 XMT+
WHITE/RED J11-21 J62-3 AUX2 RCV+
WHITEIVIO J11-22 J62-7 AUX2 RCV-
CXO-944A
Sheet 3 of S

WIRE TABLE FOR 1226092-01 A/F SENSOR

RED A1-+ J70-1 ;'
BLACK A1-GND J70-3
I
WHITE A1-LOAD J70-2 I /
WIRE TABLE FOR 1701275-01 A/F SENSOR CABLE

~V.;..;;IO~L~ET~+--....;.P....;.7..;..O-...;..1_-I K 1 -1 DOUBLE
VIOLET P3S CRIMP
ORANG E P70-2 K 1 06 LOAD (-SV)
ORANGE P70-3 -S.2V BUSBAR 0 BACKPLANE -S.2V
WIRE TABLE FOR 1701276-01 STD POWER SUPPLY

GRN/YEL GND STUD I 2" GND
BLUE TB1-1-7 I 2" ACC ·POWER CONTROLLER, 2
BROWN TB1-1-6 4 2" AC
WIRE TABLE FOR 1701276-01 OPT POWER SUPPLY

GRN/YEL GND STUD 3 .. GND
BLUE TB1-1-7 3 .. ACC ·POWER CONTROLLER, 3
Ia1-J-B 3" AC
WIRE TABLE FOR 1701276-02 BLOWER AC LINE CORD

BLUE P80-1 AC NEUTRAL
BROWN IN MOLDED PLUG P80-2 AC LINE
GREEN P80-3 GND
BLACK P80-S P80-4 JUMPER

GREEN P80-3 GND
BLACK P80-7 P80-4 JUMPER
BLACK P80-8 P80-S JUMPER
WIRE TABLE FOR 7019680-01

VIOLET J31-1 DOUBLE
TB1-3-S +12V
VIOLET J31-3 CRIMP
VIOLET J31-S DOUBLE
TB1-3-6 +12V
VIOLET J31-7 CRIMP
BLACK J31-9 DOUBLE
TB1-3-3 GND (+12V)
BLACK J31-2 CRIMP
BLACK J31-4 TB1-3-3 GND (+12V)
ORANGE J31-6 TB1-2-2 +SV SENSE TWISTED
BLACK J31-8 TB1-2-1 GND (-SV SENSE) PAIR
BROWN J31-10 TB1-:1-4 POWER FAIL
CXO-944A
Sheet 4 of S


BLACK P32-1 TB1-1-2 GROUND TWISTED
RED P32-2 TB1-1-1 +5V SENSE PAIR
VIOLET P32-4 TB1-3-1 +12V SENSE PAIR

RED J50-1 TB1-1 +5V SENSE TWISTED
BLACK J50-2 TB1-2 GND (+5V SENSE) PAIR
BROWN J50-3 TB1-3 POWER FAIL

YELLOW J33-4 TB1-2-3 ON/OFF (-5.3V)
ORANGE J33-3 TB1-2-2 S2-
BLUE J33-2
TB1-1-3 ON/OFF (+5V) DOUBLE CRIMP
BLUE J34-2
BLACK J33-1
TB1-1-2 SI- DOUBLE CRIMP
BLACK J34-1

BLUE J51-2 TB1-3 ON/OFF (+5V)
BLACK J51-1 TB1-2 S-

BLUE P34-1 J51-1 S-
BLACK P34-2 P51-2 ON/OFF (+5V)

COLOR I FROM I TO I SIGNAL I REMARKS
VIOLET I J35 I TB1-3-2 J +12V 1
CXO-944A
Sheet 5 of 5
Figure A-1 HSC Internal Cabling

A.3 HSC50 Internal Cabling

Figure A-2 is a diagram of the HSC50 internal cabling.
011677-01 .....
r--------L~--------------Lm~
DRIVE 1 DRIVE 0
lSI ~HI5
V 7020116-01
TU:~
iI~, S2 / IIRRN R!;/')
IQTU::J
.BRN . RED
lILT< ...JID;
7020204-0~ 7020204-01/'"
~~
BULKHEAD
7428570-01
(REAR) TU68 CONTROL A",~
01 81
r:m U
N1
I
Em
Oc
MODULE
(TU68-XB)
Jl J3 J4
c:::Il::3
J2
Q
~
BEZEL ASSY
7 018676-01",,- I:cJ
70207S::~~3-01 _ _ 11 m ~lT P2 7020186-01
/70111705-01
STANDARD POWER ~
----------~--~-----------
SUPPLY ASSEMBLY
LiiJ
011680-01,
(7020033-01 OR -02)
r---. Gt:llru ....

~
70201117-01 P3 P4 RELAY
OR "OW . . .""
<Ji
J3 J4 (1218828-01)
~:'
I . 1 1 JlI
011681-01 " " - IT! 876 S 4 3 2/,
~ PIiRATOR CONTROL PANE
r-!"'" r!l "....." ...... F 1-5 K1-2
I" ffi~T8'·' V702011111-01 7020201-01~
, ~~
K1-4 3
~ '~67 7
I:C r==='
h "'- BACKPLANE (REAR VIEW)
(5414046-00) """- r-7020200-01
BLOWER
TBl-2
><Pl 1In'
~~
1 2 3. 5 6 7 /7011685-02 II~TB1
~"
BB ... RED
II!TB2
.;, ~/ f/ ,~
3
'- ~ 7011684-01 I~TB3

~~
BLACK i'AIR FLOW
TBQ
U
OPT POWER SUPPLY ASSEMBLY
'--"',~
7011685-01
~ f "I~
~ 1m !TB4
~. 1113
b~ II
TAQ
RBQ
SENSOR
(PART OF
OUTLET DU CT
ASSy)
70201118-01
~
!TBS RAQ i'
124567 ~3
i'~~
BOU~~U'
.1 r"-70111685-02
TBl
1/
701~V~ I"~~
'70111685-02
=
GANIYEL
,
PSO
+ BLACK
1r-m-1 rm1
r-- 1213756-28---..,
I
7 7011682-0"~ ~.' ~
~ ~ ~701852
7011686-01 OA-02 =
~701 _ _ 01 OR-02 ./
011678-(1)" r-;::!,
P8
" 11 " . / ..
PIN ' .
I/O BULKHEAD
ASSEMBLY
biJ cf'cf'cf'~ 6-1A
4 PLAC ES
(7020024-01 ) TA RA TB RB
J12 (60 HZ) J11 (60HZ)
J5 (50 HZ) J4 (50 HZ)
0
.18 (60 HZ) J13 (60 HZ) 1
.13 (50 HZ) .16 (50 HZ)
2
POWER CONTROLLER ASSEMBLY
(7018122-00) (60 HZ)
OR
(7020613-01) (50 HZ) F E D C
.
B\
3
7019~06-OM
11:::"- Jl
-
J3 (80 HZ) 6 PLACES
J1 J2 (50 HZ)
RJ
3-PHASE/
NEUTRAU
GROUND
AC POWER
CORD
b~ b~
7020205-01~ !Q " ' 7020206-01 OR -02
.. DEC POWER
CONTROL BUS
DELAYED
OUTPUT
#'" 7020202-01
CXO-OS1A
Sheet 1 of 6
Figure A-2 (Cont.) HSC50 Internal Cabling

WIRE TABLE FOR LINE CORD OF 7019122-00

COLOR FROM TO REMARKS
BLUE W LF1-N
BLACK Z LF1-L3
GRNIYEL GND LF1-GN15 LINE SIDE
BLACK Y LF1-L2
BROWN X LF1-L 1

VIOLET J11-1 J41-01 +12V
VIOLET J11-2 J44-01 +12V
BLACK J11-3 J41-02 GROUND
RED J11-7 J41':04 +SV
RED J11-8 J44-04 +SV
WHITE J11-12 J60-20 TERM PRESS L
WHITE J11-13 J60-01 XMT-
WHITE J11-14 J60-02 XMT+
WHITE J11-15 J60-03 RCV+
,.
I , J60-20 J60-06 DATA SET READY COMMONING STRIP SEE NOTE *
I
J11-16 J60-07 ACV- * PINS AT J43-06 AND

WHITE J60-06 ARE WIRELESS
WHITE J11-18 J43-20 TERM PRESS L'
PINS. THEY ARE TIED
WHITE J11-19 J43-01 XMT- TO J43-20 AND J60-20
WHITE J11-20 J43-02 XMT+ BY COMMONING STRIPS.
WHITE J 11-21 J43-03 RCV+
,.
I ,
I J43-20 J43-06 DATA SET READY COMMONING STRIP SEE NOTE *
WHITE J11-22 J43-07 RCV-
WHITE J11-26 J42-01 TUO/1 PRESS L
WHITE J11-27 J42-02 TUO/1 XMT-
WHITE J11-28 J42-03 TUO/1 XMT+
WHITE J11-29 J42-04 TUO/1 REV+
WHITE J11-30 J42-0S TUO/1 RCV+
WHITE J11-34 J4S-01 TU2I3 PRESS L
WHITE J11-35 J4S-02 TU2I3 XMT-
WHITE J11-36 J4S-03 TU2I3 XMT+
WHITE J11-37 J4S-04 TU2I3 RCV+
WHITE J11-38 J4S-0S TU2I3 RCV-
YELLOW J11-45 J40-01 STATE LAMP L
YELLOW J11-46 J40-02 POWER ON L
YELLOW J11-47 J40-03 LAMP 0 L
YELLOW J11-48 J40-04 TERM ENA L
YELLOW J11-S2 J40-08 LAMP 3 L
YELLOW J11-53 J40-09 SWITCH 1 L
YELLOW J11-S4 J40-10 SWITCH 0 L
YELLOW J11-57 J40-13 BDCOK H (INT L)
RED J11-60 J40-15 +5V
CXO-051A
Sheet 2 of 6
Figure A-2 (ConI.) HSC50 Internal Cabling


VIOLET P41-1 P1-1 +12V TU POWER
BLACK P41-2 P1-3 GND (+12V) TU POWER
BLACK P41-3 P1-S GND (+SV) TU POWER
READ P41-4 P1-S +SV TU POWER
WHITE P42-1 P2-F GND TU SIGNAL
WHITE P42-2 P2-D RCV- TU SIGNAL
WHITE P42-3 P2-C RCV+ TU SIGNAL
WHITE P42-4 P2-J XMT+ TU SIGNAL
WHITE P42-S P2-H XMT- TU SIGNAL
7
I
7
I I
7
,
7 P2-E ,
7
I
7 KEYING PLUG (TU SIG)
YELLOW P40-1 P3-1 STATE LAMP L OCP
YELLOW P40-2 P3-2 POWER ON L OCP
YELLOW P40-3 P3-4 LAMP 0 L OCP
YELLOW P40-4 P3-3 TERM ENA L OCP
YELLOW P40-S P3-S LAMP 2 L OCP
YELLOW P40-S P3-S LAMP 1 L OCP
YELLOW P40-7 P3-S LAMP 4 L OCP
YELLOW P40-S P3-7 LAMP 3 L OCP
YELLOW P40-9 P3-10 SWITCH 1 L OCP
YELLOW P40-13 P3-1S BDCOKH (INIT L) OCP
BLACK P40-14 P3-14 GND OCP
RED P40-1S P3-1S +SV OCP
P3-20 KEYING PLUG (OCP)

VIOLET J12-1 P31-1 +12V
VIOLET J12-2 P31-3 +12V
VIOLET J12-4 P31-7 +12V
BLACK J12-S P31-9 GNDj +12V
BLACK J12-7 P31-2 GND +12V
ORANGE J12-11 P31-S -SV SENSE TWISTED
BLACK J12-12 P31-S GND -SV SENSE) PAIR
BROWN J12-13 P31-10 POWER FAIL
I
J12-1S NO CONNECTION KEYING PLUG
BLACK J12-17 J32-1 GND +SV SENSE) TWISTED
RED J12-1S J32-2 +SV SENSE PAIR
BLACK J12-19 J32-3 GND + 12V SENSE} TWISTED
VIOLET J12-20 J32-4 +12V SENSE PAIR
CXO-OS1A
Sheet 3 of S

Intemal Cabling Diagrams A-11

VIOLET J31-1 DOUBLE
TB1-3-5 "+12V
VIOLET J31-3 CRIMP
VIOLET J31-5 DOUBLE
TB1-3-6 +12V
VIOLET J31-7 CRIMP
BLACK J31-9 DOUBLE
TB1-3-3 GND (+12V)
BLACK J31-2 CRIMP
ORANGE J31-6 TB1-2-2 +5V SENSE TWISTED
BLACK J31-8 TB1-2-1 GND (-5V SENSE) PAIR
BROWN J31-10 TB1-1-4 POWER FAIL

BLACK P32-3 "TB1-3-4 GROUND TWISTED
WIRE TABLE FOR 7019686-02 STD PWR SUPPLY (50 HZ)

GRN/YEL GND STUD J5* GND
BLUE TB1-2-7 J5* ACC *POWER CONTROLLER J5
BROWN TB1-2-6 J5* AC
WIRE TABLE FOR 7019686-01 AUX PWR SUPPLY (60 HZ)

GRN/YEL GND STUD J13" GND
BLUE TB1-7 J13'* ACC "POWER CONTROLLER J13
BROWN TB1-6 J13* AC
WIRE TABLE FOR 7019686-02 AUX PWR SUPPLY (50 HZ)

GRN/YEL GND STUD J6* GND
BLUE TB1-7 J6* ACC "POWER CONTROLLER J6
BROWN TB1-6 J6* AC

I
f P4-01 , l , NO CONNECTION SPARE
7
7I "}
£
P4-02 ,t.
1
1
, NO CONNECTION KEYING PLUG
RED P4-03 D1-1 +5V
BLACK P4-04 D1-2 GND (+5V)
,I "]
, P4-05 7
t. i
J
NO CONNECTION SPARE
I P4-08 7l 1
, NO CONNECTION SPARE
CXO-051A
Sheet 4 of 6


YELLOW S2-2 P33-4 ON/OFF (-S.3V)
ORANGE 82-1 P33-3 82-
BLUE 82-4 P33-2 ON/OFF (+SV)
BLACK 82-S P33-1 81-

,, ,
,, J13-2 ,t. ,£ 7 ,
£ KEYING PLUG
RED J13-3 PSO-1 +SV SENSE TWISTED
BLACK J13-4 PSO-2 GND (+5V SENSE) PAIR
BROWN J13-S PSO-3 POWER FAIL

RED JSO-1 . TB1-1 +5V SENSE TWISTED
BLACK JSO-2 . TB1-2 GND (+5V SENSE) PAIR

BLACK +V2 TB1-1 GROUND (-SV)
BLACK +V2 TB1-1 GROUN D (-5V)
BLACK +V2 TB3-1 GROUND (-5V)
BLACK +V2 TB3-3 GROUND (-5V)
ORANGE -V2 TB2-3 -5V
ORANGE -V2 TBS-3 -5V
WIRE TABLE FOR 7019686-01 STD PWR SUPPLY (60 HZ)

GRN/YEL GND STUD J12" GND
BLUE TB1-2-7 J12" ACC "POWER CONTROLLER J12
BROWN T81-2-6 J12" AC

YELLOW J33-4 TB1-2-3 ON/OFF (-5.3V)
BLUE J33-2
TB1-1-3 ON/OFF (+5V) DOUBLE CRIMP
BLUE J34-2
BLACK J33-1
TB1-1-2 81- DOUBLE CRIMP
BLACK J34-1

BLUE JS1-2 TB1-3 ON/OFF (+5V)
BLACK J51-1 TB1-2 S-

BLUE P34-1 JS1-1 S-
BLACK P34-2 P51-2 ON/OFF (+SV)
CXO-OS1A
Sheet S of 6


RED A1-+ J70-1 I
BLACK A1-GND J70-3 j

(
/
WHITE A1-LOAD J70-2 ;
(
/

VIOLET P70-1
K1-1 +12V DOUBLE CRIMP
VIOLET P35
ORANGE P70-2 Kl-S LOAD (-5V)
ORANGE P70-3 TB5-3 -5V

WHITE K1-3 P8-1 TRIP
WHITE K1-5 P8-2 RETURN

VIOLET J35 TBl-3-2 +12V
WIRE TABLE FOR 7020206-01 (60 HZ)

BLUE P80-1 AC
BROWN J11* P80-2 AC .. POWER CONTROLLER J11
GRN/YEL P80-3 GROUND
BLACK P80-7 P80-4 I I JUMPER
BLACK P80-S P80-5 j
I
;' JUMPER
WIRE TABLE FOR 7020206-02 (50 HZ)

BLUE P80-1 AC
BROWN S2-2 P80-2 AC .. POWER CONTROLLER J4
GRNIYEL P80-3 GROUND
BLACK P80-5 P80-4 / JUMPER

RED TB1-2 J4S-2 +5V FROM BACKPLANE
BLACK TB2-1 J4S-1 GROUND FROM BACKPLANE
WIRE TABLE FOR LINE CORD OF 7020613-01

BROWN x LF1-PH3 j )
BLACK y LF1-PH2 I ,
.BLACK (2\ z LINE SIDE
LF1-PH1 f
I
/
OF LF1
BLUE N LF1-N I
GRN/YEL GND LF1-GND j

(
/
CXO-051A
Sheet Sot S
Figure A-2 HSC50 Internal Cabling

A.4 HSC50 (Modified) Internal Cabling

Figure A-3 is a diagram of the HSC50 (modified) internal cabling.
7rO='11=6=77=-O=',.~!I="iF==========9 r--------"LEo----- ---------LEDI

I I DRIVE 1 DRIVE 0 I
IS II If¥I is ~IijjPI.N
1===....!.=-=l1 S2
1
/
V 70201116-01 TU::J
II BAN R!;D
I~
~.
TU:-:j
II BAN A!;D
I
I
~ __.§~~______~ mK BLK
~
FRONT 7020204.01............... 7020204-01 ...............
BULKHEAD
7426570-01 PIN ',~
(REAR) TUS8 CONTROL 01 S1
.......,.!,J,N 1 _ MODULE 0
I.1IIWooo;.I .... (TU58-XB) n
I /I
70111676-01-....,.,
~..7..02...07...s::~0..;.·:..!3-.....01.......
____ --_11!-1_"It::I2:J
STANDARD POWER II
70111630-01 SUPPLY ASSEMBLY P
• '-... (7020033-01 OR -02)
~ ~I TB1-3 ~0201117_01 G:::J lE:j

70111631-01", iIil 8 7 6 J 4 3 2/1 ~~ io~:~~:'~~~fL PANE K1-1
1
lofil~
1"-""L.L.I TB1-1 ~- ~ /702011111-01 AIRFLOWS~::::'RCABLy
(1701278-O1) ----y "'"~;tJ [f~o fD:~K';'-'_S/'-3
~'--
~ 1 /2113\4 5 6 7
/ \~ 12260112-01 V
TB1.2
h~!e~~\'~~~e3 r--B-A-CK-P-L-A-NE-(R-EA-R-Vl-E-W}--
(5414048-00) BLOWER
~ :~':.~W
1 2 3~ S 6 7 /~p;I~TB1 ~(1218828-00) (PART OF
. rf,/ ... /' ~ @",,_170'11684-O'/'II,·IJUNECORD

l;, !:'"' o' tl~':::~
"-'1r------- ~-~~~_!B:!;LA~C~K:....._JJ 1\ iii ~ TB3 / (1701276-02 OR -03)
~ ~ ~.
11-11--111---
J 70111685-01
"O-P"T-PO-W-E-R-S-UP-P-L-Y-A-S-SE-M-B-L-'V
(7020184-01OR,~2)
70201118-01
_ _
~ :
~
r-
i' Y
~
II
1113
lin 1
TB4
~
/
~ b. '1IJ2TBS
1 2 4 S 6 7 · ' - 3
TBOI==="""iJ
~aOODO~A
TBl .11UI '701I168S.02 C1 CABLES
(1700717..01
'''~V~ /'..- ~ '",.....,

TAO= ,...4X
RB OI--II-----4l iL""JI../
l11
~ :..... BLACK GRNIYEL RAO

1r-:r'21 nTn
UfU "" 1213756-28---...
=
/ . / 70111686-01 OR -02 _ (/
POW~E=R=S=~=PP=L=v==~~~F==t-=====t~~~=~=========;t~======r=====*===~~~~~~~
AC LINE CORD 1'\ t, I I 70111682-0{
(1701276-01) r-..I~~2'---..i4l~3'--"'41~1-
"'",,-
BP TO PS i'REAR SHIELDI
INTERCONNECT C1 CABLE
(701116711-01) ASSEMBLV
(723140-01)
em
CONSOLE
POWER CONTROLLER
ASSEMBLY
(3024374-01 OR 02)
FED C B A
[J [J [J [J [J 10 0
~ g ~ ~ ~ -:~
- ......:::Cl=---:[J=---=Cl=--=Cl=-~[J=--E_=----3 r----.. "-IK>TTON 110
~ 3 PHASElNEUTRALlGNO
AC POWER CORD
BULKHEAD ASSV
(7023135-01 )
RE~VTOPCAIFSENSOR~~----==============a===~--==~-===------~~==d
(1701231-01)
CXO-2076A
Sheet 1 of 6
Figure A-3 (Cont.) HSC50 (Modified) Internal Cabling


VIOLET J11-1 J41-01 +12V
VIOLET J11-2 J44-01 +12V
RED J11-7 J41-04 +5V
RED J11-8 J44-04 +5V
WHITE J11-13 J60-01 XMT-
WHITE J11-14 J60-02 XMT+
WHITE J11-15 J60-03 RCV+
,.
I
I
, J60-20 J60-06 DATA SET READY COMMONING STRIP SEE NOTE"
J11-16 J60-07 ACV- .. PINS AT J43-06 AND
WHITE
J60-06 ARE WIRELESS
PINS. THEY ARE TIED
WHITE J11-19 J43-01 XMT- TO J43-20 AND J60-20
WHITE J11-20 J43-02 XMT+ BY COMMONING STRIPS.
J11-21 J43-03 RCV+
I
WHITE
,. ,
I J43-20 J43-06 DATA SET READY COMMONING STRIP SEE NOTE ..
WHITE J11-22 J43-07 RCV-
WHITE J11-26 J42-01 TUO/1 PRESS L
WHITE J11-27 J42-02 TUO/1 XMT-
WHITE J11-28 J42-03 TUO/1 XMT+
WHITE J11-29 J42-04 TUO/1 REV+
WHITE J11-30 J42-05 TUO/1 RCV+
WHITE J11-34 J45-01 TU2I3 PRESS L
WHITE J11-35 J45-02 TU2I3 XMT-
WHITE J11-36 J45-03 TU2I3 XMT+
WHITE J11-37 J45-04 TU2I3 RCV+
WHITE J11-38 J45-05 TU2I3 RCV-
YELLOW J11-45 J40-01 STATE LAMP L
YELLOW J11-46 J40-02 POWER ON L
YELLOW J11-48 J40-04 TERM ENA L
YELLOW J11-57 J40-13 BDCOK H (tNT L)
RED J11-60 J40-15 +5V
CXO-2076A
Sheet 2 of 6


VIOLET P41-1 P1-1 +12V TU POWER
BLACK P41-2 P1-3 GNO(+12V) TU POWER
BLACK P41-3 P1-6 GND (+SV) TU POWER
READ P41-4 P1-S +SV TU POWER
WHITE P42-1 P2-F GND TU SIGNAL
WHITE P42-2 P2-D RCV- TU SIGNAL
WHITE P42-3 P2-C RCV+ TU SIGNAL
WHITE P42-4 P2-J XMT+ TU SIGNAL
WHITE P42-S P2-H XMT- TU SIGNAL
I t t P2-E 7
I I
7 KEYING PLUG (TU SIG)
YELLOW P40-1 P3-1 STATE LAMP L OCP
YELLOW P40-2 P3-2 POWER ON L OCP
YELLOW P40-3 P3-4 LAMP 0 L OCP
YELLOW P40-4 P3-3 TERM ENA L OCP
YELLOW P40-13 P3-1S BDCOKH {lNIT L) OCP
BLACK P40-14 P3-14 GND OCP
RED P40-1S P3-16 +SV OCP
P3-20 I
KEYING PLUG (OCP)

VIOLET J12-1 P31-1 +12V
VIOLET J12-2 P31-3 +12V
VIOLET J12-4 P31-7 +12V
BLACK J12-S P31-9 GND +12V
ORANGE J12-11 P31-6 -SV SENSE TWISTED
BLACK J12-12 P31-S GND -SV SENSE) PAIR
J12-16 NO CONNECTION KEYING PLUG
BLACK J12-17 J32-1 GND +SV SENSE) TWISTED
RED J12-1S J32-2 +SV SENSE PAIR
BLACK J12-19 J32-3 GND (+ 12V SENSE) TWISTED
VIOLET J12-20 J32-4 +12V SENSE PAIR
CXO-2076A
Sheet 3 of 6
Figure A-3 (Cont.) HSCSO (Modified) Internal Cabling

A-18 Intemal Cabling Diagrams

VIOLET J31-1 DOUBLE
TB1-3-5 +12V
VIOLET J31-3 CRIMP
VIOLET J31-5 DOUBLE
TB1-3-6 +12V
VIOLET J31-7 CRIMP
BLACK J31-9 DOUBLE
TB1-3-3 GND (+12V)
BLACK J31-2 CRIMP
ORANGE J31-6 TB1-2-2 +5V SENSE TWISTED
BLACK J31-8 TB1-2-1 GNO{-5V SENSE) PAIR
BROWN J31-10 TB1-1-4 POWER FAIL
WIRE TABLE FOR 7Q19681-01


,, ,,
, ,,
,,
7
7
, P4-02 , {. ,, NO CONNECTION KEYING PLUG
RED P4-03 D1-1 +5V
BLACK P4-04 D1-2 GND (+5V)
, l
T
,
P4-05 ,I- ,, NO CONNECTION SPARE
I I P4-08 , {. ,, NO CONNECTION SPARE

YELLOW S2-2 P33-4 ON/OFF (-5.3V)
ORANGE S2-1 P33-3 S2-
BLUE S2-4 P33-2 ON/OFF (+5V)
BLACK S2-5 P33-1 S1-

,t. ,, J13-2 ;
I ,
1 ,{. ,, KEYING PLUG
RED J13-3 P50-1 +5V SENSE TWISTED
BLACK J13-4 P50-2 GND (+5V SENSE) PAIR

RED J50-1 TB1-1 +5V SENSE TWISTED
BLACK J50-2 TB1-2 GND (+5V SENSE) PAIR
CXO-2076A
Sheet 4 of 6


BLACK +V2 TB2-1 GRN (-SV)
BLACK +V2 TB2-2 GND (-SV)
BLACK +V2 TBS-1 GND (-SV)
BLACK +V2 TBS-2 GND (-SV)
ORANGE -V2 TB2-3 -SV
ORANGE -V2 TBS-3 -sv
YELLOW J33-4 TB1-2-3 ON/OFF (-S.3V)
BLUE J33-2
TB1-1-3 ON/OFF (+SV)
BLUE J34-2 :
BLACK J33-1
TB1-1-2 S1-
BLACK J34-1

BLUE JS-2 TB1-3 ON/OFF (+SV)
BLACK JS-1 TB1-2 S-

BLACK P34-1 PS1-1 S-
BLUE P34-2 PS1-2 ON/OFF (+SV)
WIRE TABLE FOR 1228092-01 A/F SENSOR

RED A1-+ J70-1 T
I
T
BLACK A1-GND J70-2 T ,t

WHITE A1-LOAD J70-3 t r
WIRE TABLE FOR 1701231-01 RELAY TO PC A/F SENSOR

WHITE K1-3 P8-1 TRIP
WHITE K1-S P8-2 RETURN
WIRE TABLE FOR 1701275-01 A/F SENSOR CABLE

VIOLET P70-1 DOUBLE
K 1-1 +SV
VIOLET P3S CRIMP
ORANGE P70-2 K106 LOAD (-SV)
ORANGE P70-3 -S .2V BUSBAR 0 BACKPLAN E -S.2V
WIRE TABLE FOR 1701276 .. 01 STD POWER SUPPLY
GRN/YEL GND STUD 2" GND
BLUE TB1-1-7 2" ACC *POWER CONTROLLER ~ 2
BROWN TB1-1-6 2 " AC
CXO-2076A
Sheet S of 6

A-20 Exception Codes and Messages
WIRE TABLE FOR 1701276-01 OPT POWER SUPPLY

GRN/YEL GND STUD 3 * GND
BLUE TB1-1-7 3 * ACC *POWER CONTROLLER, 3
TB1-1-6 3* AC

GREEN P80-3 GND

GREEN P80-3 GND
CXO-2076A
Sheet 6 of 6
Figure A-3 HSC50 (Modified) Internal Cabling

B
Exception Codes and Messages
Certain software inconsistencies can cause an exception (crash) in the HSC. This appendix describes
all HSC exception codes caused by software inconsistencies. It provides a description of the
exception codes, the facility or program reporting, and the action you should take. For ease of
reference, these codes are arranged in numerical order (octal radix).
To determine which exception code caused a particular crash, refer to the crash dump printed on
the terminal. Note that the code number, but not the text, appears on hardcopy printouts.
B.1 Crash Dump Printout

In order to determine which exception code caused a particular crash, refer to the crash dump
printed on the terminal. The following HSC crash dump example breaks down and describes the
various fields. The HSCxx refers to the HSC model.
-* SUBSYSTEM EXCEPTION *- V100 HSCxx HSC001 0

at 17-Nov-1858 00:13:34.20 up o 00:13:34.20
User et Pc: 015066 caused by (20 lOT 0
PSW: 140001
DEMON C) active, PCB addr = 054214
RO-R5:
000005 000000 023004 147602 160020 154752
Kernel SP: 000774
Kernel Stack:
005045
047044
e
000004
000000
053336
047450
046004 001012 000000 046236 000000
000000 052074 000000 055334 000000
User SP: 154734
User Stack:
002013 ~ 104262 140310 102250 000034 035064 004305 000000
000000 000003 000001 000004 000000 002445 000000 000000
KPAR (0-7) :
Booting
INIPIO-I Booting .•.
Example B-1 Crash Dump Example

o This line calls out a crash and indicates the HSCxx is at software version number VlOO. The
last field is the assigned node name (set with SET NAME).
et Indicates the processor mode in which the crash occurred. This can be either Kernel or User.
B-1
B-2 Exception Codes and Messages
6) A 3-letter mnemonic indicating the type of crash. The example mnemonic lOT indicates that
this is a software inconsistency. Any other combination of letters, such as NXM: (Nonexistent
Memory) would designate a crash outside the scope of this appendix. Hardware exceptions are
defined in Appendix D.
e The initial name on this line identifies the process active at the time of the crash. It is valid
only during user-mode crashes. Use this name as a crosscheck when looking up the crash
description.
€) If the mode notation is Kernel, check the first word of the Kernel Stack for the crash code.
(D Because the mode notation in this example indicated User, check the User Stack for the crash
code number. This code is always the first word of the stack (in this case, 002013).
B.2 SINI-E Error Printout

The following SINI-E error example appears immediately upon reboot after a subsystem exception.
Information contained in this error message is a condensation of the crash dump.
SINI-E Seq 1. at 17-Nov-1858 00:00:02.00
Software inconsistency It
Process DEMON - .
PC 000002
PSW 140001
Stack dump: 002013 104262 140310
Example B-2 SINI-E Exception Code

o This line defines the cause of the crash.
• This line and the following three lines duplicate the applicable information in the crash dump.
In each of the exception descriptions in this appendix, Facility indicates the process(es) running at
the time the crash occurred. The first name listed is the major process. The second is the module of
the process that generated the exception. This may be a subprocess of the main process or simply a
different code module.
B.3 Submitting a Software Performance Report

Some of the exception messages listed in this appendix suggest submitting a Software Performance
Report (SPR) with a copy of the crash dump. Before submitting the SPR, contact the Customer
Support Center or the local field office to see if additional information will be needed. Submit an
SPR only after eliminating other possibilities, such as hardware-related problems.
If the customer requires immediate use of the HSC, you can reboot the HSC using the Init switch
on the OCP. But this reboot causes cause a loss of the information necessary for an SPR, so be sure
to make a hardcopy crash dump message and other required information.
Mter two or three similar exception messages occur an SPR should be submitted. Look up the
exception message in this appendix. If a data structure (for instance, lIMB or PCB) should be
included with the SPR, set the ODT parameter to cause the HSC to enter ODT after an exception.
If data structures are not requested in the applicable exception code, you do not need to enter ODT.
Exception Codes and Messages B-3
Data structures needed with the SPR must be formatted. These data structures are addressed by
a register or the contents of another structure's field. To format the necessary data structure(s),
substitute the x in Table B-1 with the pointer from the specified register or location. Substitute
only the x and type the rest of the line exactly as you see it in the table, except for the information
in parentheses. The number of = signs designates the data structure memory:
= indicates program memory
= == indicates Control memory
=== indicates data memory
Table B-1 Obtaining Data Structure Information

Data
Structure
Needed Type At * Prompt
CB x==CB$
Counter x=C (and) x= =C.
DCB x= =DC$DISK (or) X= =DC$TAPE (if Tape Path problem)
DDCB x=DD$
FRB x==F$
HCB x=HC$
HMB x= =HM$ (command packet)
x= =HM$CPY (BACKUP)
x= =HM$DATA (with BMl3s)
x= =HM$QUIET (diagnostic)
x= =HM$XFR (used while work is outstanding)
x==HM$VC (used to alter VC state)
K Control Area x==KG$
PCB x=z.
SLCB x=SL$
TDCB x=TD.
TFCB x=TF.
TI'CB x=TT$
XFRB x==x.
After the information is complete, the customer should fill out the SPR and submit it, together with
all hardcopy, as instructed on the SPR form.
NOTE
If you instruct the customer to call the Customer Supp.ort Center for assistance, inform
the Center of the problem. Also, let them know your customer will need help gathering
information related to the software error.
8.4 Exception Messages

In each of the exception messages, Facility indicates the process(es) running at the time the crash
occurred. The first name listed is the major process. The second name is the module of the process
that generated the exception. This module may be a subprocess of the main process or simply
a different code module. Include the crash dump message and any other applicable hardcopy
information with an Software Performance Report (SPR) submission.
001001 ($CKERSTK)
execution of Kernel Stack
Facility: EXEC, EXEC
Explanation: The HSC executive executed stack space.
Action: Submit an SPR with a crash dump. You may reboot the HSC immediately.
001002 ($CPUM1)
Previous mode not user
Explanation: During a context switch of user processes, the previous mode (as indicated by
the Program Status Word (PSW» was not user mode.
Action: Submit an SPR with a crash dump. R5 points to PCB (Process Control Block).
001003 ($CEXPCB)
EXEC PCB was scheduled
Explanation: During process scheduling, the EXEC process control block (PCB) was scheduled.
This dummy PCB is used only for loading the process and should never be scheduled.
Action: Submit an SPR with a crash dump. R2 points to PCB.
001004 ($CDEBCAC)
Cache setting In PDR Is In Incorrect state
Explanation: This software inconsistency should not appear under normal circumstances. A
Page Descriptor Register (PDR) directed to program memory does not have "disable cache" set.
A PDR directed to data memory does have "disable cache" set.
Action: Submit an SPR with a crash dump. RO points to PDR.
001005 ($CPUM2)
Previous mode not user
Explanation: During a context switch of user processes, the previous mode (as indicated by
the PSW) was not user mode.
Action: Submit an SPR with a crash dump.
001006 ($CCB4)
Spurious Interrupt from K at Control Bus Level 4
Explanation: One of the Ks interrupted the P.ioc at Level 4, but, upon queue examination, no
elements were shown (an element should be on the Level 4 Interrupt queue).
001007 ($CCBS)
Spurious Interrupt from K at Control Bus Level S
Explanation: This software inconsistency should not appear under normal circumstances.
One of the Ks interrupted the P.ioc at Level 5, yet, upon queue examination, no elements were
shown (an element should be on the Level 5 Interrupt queue.)
Action: Submit an SPR. If this crash continues to occur, escalate the problem to Customer
Service support.
001010 ($CDC1)
Downcount failed
During processing of the Level 5 Interrupt queue, a down-count operation on a counter (down
counted by 1) failed.
Action: Submit an SPR with a crash dump. R1 points to the counter.
001011 ($CDC2)
Downcount failed
During processing of the Level 5 Interrupt queue, a down-count operation on a counter (down
counted by 1) failed.
Action: Submit an SPR with a crash dump. R1 points to the counter.
001012 ($CACQ)
Acquire on Semaphore with address of 0
Explanation: This software inconsistency should not appear under normal circumstances. The
ACQ$P System Service was called with a Semaphore address of O.
Action: The process specified as active is the offender. Submit an SPR with a crash dump.
001013 (SCAML)
Acquire MUltiple on Semaphore with address of 0
AMLT$P System Service was called with a Semaphore address ofO.
001014 (SCRLP)
Release on Semaphore with address of 0
REL$P System Service was called with a Semaphore address of O.
001015 ($CRRTI)
RRTI$ on Semaphore with address of 0
RRTI$P System Service was called with a Semaphore address of O.
001016 ($CRTI1)
001017 ($CRTI2)
001020 ($CRCPP)
ReceivelDequeue from Queue with address of 0
One of the RCV$P FROM$P or DEQ$P FROM$P system services was called with a queue head
address of O.
001021 ($CRCCP)
RecelvelDequeue from Queue with address of 0
One of the RCV$C FROM$P or DEQ$C FROM$P System Services was called with a queue
head address ofO.
001022 ($CRCCV)
RecelvelDequeue from Queue with address of 0
·Facility: EXEC, EXEC
One of the RCV$C FROM$P, DEQ$C FROM$P, RCV$C FROM$W, or DEQ$C FROM$W System
Services was called with a queue head address ofO.
001023 ($CRMPP)
Receive/Dequeue Multiple from Queue with address of 0
Explanation: This software inconsistency should not appear under normal circumstances. One
of the RMLT$P FROM$P or DMLT$P FROM$P System Services was called with a queue head
address of O.
001024 ($CRMCP)
of the RMLT$C FROM$P or DMLT$C FROM$P System Services was called with a queue head
address of O.
001025 ($CRMCV)
One of the RMLT$C FROM$P, DMLT$C FROM$P, RMLT$C FROM$W, or DMLT$C FROM$W
System Services was called with a queue head address of O.
001026 ($CRAMCV)
Receive AII-Maybe from Queue with address of 0
of the RCAM$C FROM$P or RCAM$C FROM$W System Services was called with a queue head
address of O.
001027 ($CSPP)
SendlEnqueue to Queue with address of 0
of the SEND$P TO$P or ENQ$P TO$P System Services was called with a queue head address
ofO.
001030 ($CSCP)
of the SEND$C TO$P or ENQ$C TO$P System Services was called with a queue head address
ofO.
Action: The process specified as active is the offender. Submit an SPR with a crash dump_
8-8 Exception Codes and Messages
001031 ($CSCV)
of the SEND$C TO$P, ENQ$C TO$P, SEND$C TO$W, or ENQ$C TO$W System Services was
called with a queue head address ofO.
001032 ($CSHPP)
SendlEnqueue-to-Head to Queue with address of 0
of the SNDH$P TO$P or ENQH$P TO$P System Services was called with a queue head address
ofO.
001033 ($CSHCP)
SendlEnqueue-to-Head to Queue with address of 0
of the SNDH$C TO$P, ENQH$C TO$P, SNDH$C TO$P, or ENQH$C TO$P System Services was
called with a queue head address ofO.
001034 ($CIHPP)
Insert at Head to Queue with address of 0
INSH$P TO$P System Service was called with a queue head address of O.
001035 ($CIHCP)
Insert at Head to Queue with address of 0
INSH$C TO$P System Service was called with a queue head address of O.
001036 (SCUPCV)
Upcount to Counter with address of 0
UPC$ System. Service was called with a queue head address of O.
Exception Codes and Messages 8-9
001037 ($CDWCV)
Downcount to Counter with address of 0
DWNC$ System Service was called with a queue head address of o.
001040
Set Timer operation to Timer with address of 0
SETTM$ System Service was called with a queue head address of O.
001041 ($CSNZ1)
Release of Semaphore wHh address of 0
During some circumstances, a semaphore will require a downcount without subsequent
scheduling considerations. This typically happens when a process enters hibernation or exits.
During the implicit release operation, the semaphore had an address of o.
001042 ($CTOVR)
Time-of-day overflowed
Explanation: During an update of the current time of day, the executive detected an overflow.
This can happen if a node on the CI sets a false time to the HSC.
Action: Examine previous console printouts to verify accurate date and time fields. If accurate,
submit an SPR with the console crash report. If inaccurate, set the HSC outband error level to
INFO. Then verify console report of date and time set by a host node on the next HSC reboot.
If a host node problem is NOT indicated, escalate the problem to Customer Service support.
001043 (SCPWFL)
Power Failure
Explanation: The processor is still operating 5 seconds after a power failure indication.
Therefore, CRONIC concludes that the power failure indication was false.
Action: Verify the ac voltages are cOlTect. If so, and the problem persists, notify Customer
Service support.
001201 ($CNOHIBER)
Process on Recoverable List not hibernating
Facility: EXEC, EXECLOAD
Explanation: Before loading a utility or diagnostic, the loader examined the Recoverable
Memory List of cache programs to determine whether a program might be loaded from memory
instead of from the load device. When a program was found on the Recoverable Memory List,
its state was not Hibernate State. This software inconsistency should not be seen under normal
circumstances.
Action: Submit an SPR with a crash dump, noting previous activity with the program
requested.
Register R3 points to PCB (Process Control Block) for process to restart.
001202 ($CIMAGE)
Memory extent encroaches defined area
Explanation: The process to be loaded specified additional memory or buffer space, as specified
on the Loadable File Header (LFHEADER) directive. When the additional memory was
allocated and mapped to the process, it had encroached upon the loaded area.
Action: Submit an SPR with a crash dump. Register RO points to XFRB (extended function
request block) for loading the image. Register R4 points to CH$ (Canonical File Header).
001203 ($CNOPROC)
No code parent process loaded
Explanation: When a process was loaded, its PCB specified it should execute and share code
associated with another process. When attempting to locate the code parent, the loader found
that the parent was not loaded.
Action: Submit an SPR with a crash dump. Register R2 equals process number of code parent.
Register R3 points to code child's PCB.
001204 ($CALLOCATE)
Insufficient Kernel Pool
Explanation: When EXEC attempted to allocate either a PCB (PCB-Z.) or an address
Descriptor (A.) structure from Kernel Pool for a new process, Kernel Pool was inadequate
to support the additional structures.
001205 ($CLFAO)
FAO overrun

Explanation: The FAO string returned during formatting of a module version mismatch
message was too large for the buffer.
Action: Submit an SPR with a crash dump. If possible, send a copy of the RX33 diskette.
001401 ($CBUSy)
Performed receive when already busy with request
Explanation: The READ$/WRITE$ service, while in its exception routine, was already busy
with one request while a RCV$P operation was performed.
001402 ($CNOLOADED)
Requested driver not loaded
Facility: EXEC, EXECRDWR
Explanation: A process within the HSC specified a READ$ or WRlTE$ operation with a
device control block (DDCB) for a device not configured on that model. For example, a program
specified a transfer for a TU5S on an HSC70 model. Because the device is not configured on the
system, the driver is not loaded.
Action: Submit an SPR with a crash dump, describing activity on the HSC at the time of the
exception. The process listed as active may be the READ$lWRlTE$ Service, and not the process
that performed the offending request. R3 points to XFRB (extended function request block). R4
points to DDCB. R5 equals CSR for device.
001403 ($CDDCB)
Invalid DDCB specified
Facility: EXEC, EXECRDWR
Explanation: A request to the READ$/WRlTE$ Service specified a DDCB that was invalid, or
it specified an invalid device type in the DD$TYPE field.
Action: Submit an SPR with a crash dump, describing activity on the HSC at the time of the
exception. The process listed as active may be the READ$IWRITE$ Service and not the process
that performed the offending request. R3 points to XFRB (extended function request block). R4
points to DDCB. R5 equals CSR for device. RO equals Device Type.
001501
Software Inconsistency-Motor not Running
Facility: EXEC, EXECRX33
Explanation: The motor was not running when the Motor Shutdown Timer expired.
001502
Software Inconsistency-Non-RX33 command requested
Explanation: The RX33 driver received an XFRB (CRONIC transfer request), but the XFRB
specified a DDCB for a non-RX33 device. R4 points to DDCB, R5 points to XFRB (extended
function request block).
001503
Software Inconsistency-Invalid Unit Number
Explanation: The DDCB (device control block) specified an RX33 device, but the unit
requested was not 0 or 1. R5 points to XFRB (extended function request block).
001504
Software Inconsistency-Zero byte count transfer
Explanation: A transfer was requested with a zero byte count. R2 equals byte count, R5
points to XFRB (extended function request block).
001505
Software Inconsistency-Invalid byte count
Explanation: A transfer was requested with a byte count that was not a multiple of 512
(sector size). R2 equals byte count, R5 points to XFRB (extended function request block).
001506
Software Inconsistency-Invalid Internal byte count
Explanation: The remaining byte count of a partially completed transfer was not a multiple
of 512 (sector size). The original (requested) byte count was a multiple of 512. R2 equals byte
count, R5 points to XFRB (extended function request block).
001507
Software/Hardware Inconslstency-RX33 hardware registers are Incorrect
Explanation: RX33 hardware signaled successful completion of an 110 operation, but the
hardware registers (current sector, current track, or memory address register) did not contain
the expected values.
Explanation: Check for RX33-related hardware failures. If the problem persists, submit an
SPR with the crash dump.
001510
Software Inconsistency-Invalid Head Select
Explanation: The Software attempted to select a head other than 0 or 1. RO equals head
select.
001511
Software Inconsistency-Memory Management
Explanation: Relocation is not enabled in the memory management hardware. Bit 0 is not set
inMMRO.
001512
Software Inconsistency-Invalid Virtual Address
Explanation: The virtual address passed in the XFRB is not in page 4. R5 points to XFRB
(extended function request block).
001513
Software/Hardware Inconsistency - Unexpected Interrupt from RX33
Explanation: An unexpected interrupt was received from the RX33 controller. This condition
is not detected until a command is about to be issued; that is, the crash does not happen when
the intelTUpt is detected.
Action: If the problem persists, submit an SPR with the crash dump. Further testing of the
HSC subsystem load device area may be necessary.
001514
Software Inconsistency-Invalid Internal Unit Number
Explanation: The unit number index value is not 0 or 2. This unit number index value is
contained in R4.
001515
Software/Hardware Inconsistency - Nonexistent Memory
Facility: EXEC, EXECRX.33
Explanation: The RX33 controller returned an NXM error.
User Action: Further testing of the HSC subsystem Goad device area) may be necessary. If
the problem persists, submit an SPR with the crash dump.
001601 ($CPAG1)
TYPE$ crosses page boundaries
Facility: EXEC, EXECTT

Explanation: A process requested a TYPE$ system service (or an ACPT$ service with a
prompt) specifying a buffer that crosses a memory management page boundary. This is a
restriction of the driver. RO equals size of print string. Rl points to String Buffer. R4 points to
DDCB (device control block). R5 points to XFRB (extended function request block).
Action: Submit an SPR with a crash dump, describing activity at the time of the exception.
001602 (SCPAG2)
ACPT$ crosses page boundaries
Facility: EXEC, EXECTr

Explanation: A process requested an ACPT$ System Service specifying a buffer that crosses
a memory management page boundary. This is a restriction of the driver. R4 points to DDCB
(device control block). R5 points to XFRB (extended function request block).
001603 ($CNOPCB)
PCB not found on run queue
Facility: EXEC, EXECTT
Explanation: When a process attached to a terminal is excepted by a keyboard command, the
exception manager of the Terminal Service performs an EXCPT$ on the Terminal Service and
load device driver. To prevent the attached process from running while the drivers potentially
run down any activity, the PCB (process control block) for the active process is removed from
the run queue. When EXEC searched the run queue specified in the Z.RVNQ field of the PCB,
it could not find the PCB. This is a software inconsistency. R4 points to attached PCB.
001701 ($CPAGE)
READ$ or WRrrE$ crossed page boundary
Facility: EXEC, EXECTU58

Explanation: A request to the TU58 driver specified a buffer that crossed a memory
management page boundary. This is a restriction of the driver.
The process listed as active may be the READ$/WRITE$ Service and not the process initiating
the offending request.
002001
Exception routine Invoked for unknown reason
Facility: DEMON
Explanation: DEMONs exception routine was activated, but not for CTRUY, CTRUC, or a
diagnostic timeout.
Action: Submit an SPR with a crash dump. If a certain sequence of HSC operations induced
this crash, include a description of that sequence. A software problem is the most likely cause
of this crash.
002002
Insufficient free memory to allocate a program stack
Facility: DEMON
Explanation: When DEMON was initialized, it could not allocate enough free program
memory for use as a stack.
002003
DEMON was Initiated when there was no diagnostic to run
Facility: DEMON
Explanation: DEMON did a receive on its work queue and received a nondiagnostic request.
this crash, include a description of that sequence.
002004
Failure in periodic control or data memory test
Facility: DEMON, PRMEMY
Explanation: One of the periodic control or data memory interface tests detected a failure.
Failures in these tests are fatal, and the HSC must reboot after displaying a message describing
the failure.
Action: A failing P.ioc module is the most probable cause of this crash. Further testing of the
HSC memory and Pioj may be necessary.
002005
Failure in periodic K.sdi or K.sti test
Facility: DEMON, PRKSDI, PRKSTI
Explanation: The periodic KsdilK.si or KstilK.si tests detected a failure. Failures in these
tests are fatal, and the HSC must reboot after displaying a message describing the type of error
and requestor number of the failed module.
Action: A failing KsdilK.si or K.stiIK.si module is the most probable cause of this crash. The
requestor number of the probable failing module is displayed in the error message preceding
the crash. Further testing of HSC data channels and HSC internal buses may be necessary.
002006
ILDISK received Illegal queue address
Facility: DEMON, ILDISK
Explanation: ILDISK requested exclusive access to a drive's state area. The acquire operation
should return the control memory address of the Attention/Available Service Queue for the
specified drive. The address returned was zero, an illegal address for a queue.
Action: If a certain sequence of HSC operations induced this crash, include a description of
that sequence. Also note if the problem occurs only when a particular disk drive is tested.
002007
ILDISK received Illegal buffer descriptor
Explanation: ILDISK received a buffer descriptor from the free buffer queue. A consistency
check on the buffer descriptor failed because the descriptor indicated the buffer was not in the
HSC's buffer memory. A software problem is the most likely cause of this crash.
this crash, include a description of that sequence. Also note if the problem occurs only when a
particular disk drive is tested.
002010
ILDISK detected inconsistency in exception routine
Explanation: ILDISK.'s internal flags indicated exclusive ownership of a drive's state area,
but the address of the K.sdilK.si control area was not available. When ILDISK has exclusive
ownership of a drive state area, the address of the KsdilKsi control area should always be
available. A software problem is the most likely cause of this crash.
002011
An ILEXER disk 110 request failed to complete
Facility: DEMON, ILEXER
Explanation: ILEXER attempted to abort all outstanding disk 110 requests. After waiting 2
minutes, the program found that one or more 110 requests had not completed. The HSC crashes
and reboots because ILEXER cannot exit with a request outstanding.
A faulty disk drive is the most likely cause of this problem. Further testing of the suspect disk
and associated requestor(s) may be necessary.
002012
An ILEXER tape 110 request failed to complete
Facility: DEMON
Explanation: ILEXER attempted to abort all outstanding tape 110 requests. After waiting 2
minutes, the program found that one or more 110 requests had not completed. The HSC crashed
and rebooted because ILEXER cannot exit with a request outstanding.
particular tape drive or formatter is tested.
A faulty tape drive or formatter is the most likely cause of this problem. Further testing of the
suspect tape subsystem and associated requestor(s) may be necessary. This crash may also be
caused by the K.stilK.si clocks stopping due to a hardware error such as an Instruction Parity
error.
002013
ILTAPE was supplied an Illegal requestor number
Facility: DEMON,ILTAPE
Explanation: ILTAPE was automatically initiated to test a particular formatter. One of the
parameters supplied to ILTAPE is the requestor number of the K.stiJK.si connected to the
formatter. ILTAPE checked the specified requestor and found it was not a K.stiIK.si.
Action: Submit an SPR with a crash dump. Also include a summary of any tape error
messages immediately preceding the crash. If a certain sequence of HSC operations caused
particular tape drive or formatter is used.
002014
ILTAPE timed out waRing for Drive State Area
Facility: DEMON, ILTAPE
Explanation: ILTAPE requested exclusive access to a tape formatter for testing. ILTAPE
timed out because the request did not complete within 60 seconds.
002015
ILTAPE detected Inconsistency after a command failure
Explanation: ILTAPE issued a command to the HSC tape diagnostic interface, but the
command failed. In the process of preparing an error message, ILTAPE found that the
command Opcode was an illegal or unknown value).
002016
ILTAPE detected inconsistency while restoring a TACB
Explanation: ILTAPE maintains a table of available tape access control blocks (TACBs). When
a particular TACB is in use by the program, the associated table entry is zeroed. When finished
with a TACB, ILTAPE stores the address of that TACB into one of the table entries containing
a zero. While trying to return a TACB to the table, ILTAPE discovered that all table entries
were nonzero, implying that no TACBs were in use.
messages immediately preceding the crash. If a certain sequence of HSC operations induced
002017
ILTAPE detected Inconsistency In exception routine
Explanation: ILTAPE's internal flags indicated exclusive ownership of a drive state area,
but the address of the K.stilK.si control area was not available. When ILTAPE has exclusive
ownership of a drive state area, the address of the .K.stilK.si control area should always be
available. A software problem is the most likely cause of this crash.
Action: Submit an SPR with a crash dump. If a certain sequence of HSC operations caused
particular tape drive is tested.
003001 ($CFNnrrYP)
Illegal format type specified
Facility: CERF
Explanation: An illegal format type was specified in an error message to CERF. R4 equals
Format Type.
003002 ($CFA01)
Output length too long
Facility: CERF
Explanation: When CERF processed an MSCP error message, the FAO output of the text
string was too long for CERF's buffer. Rl equals number of bytes output.
003003 ($CFA02)
Output length too long
Facility: CERF
Explanation: When CERF processed an out-of-band message, the FAO output of the text
string was too long for CERF's buffer. Rl equals number of bytes output.
004002
BMB reserved but not found
Facility: DISK, many

Explanation: A Big Memory Buffer (BMB) was reserved through a system function but
was not found when the table of BMBs was searched. This indicates memory corruption,
mismanagement of the BMB pool, or lack of enough BMBs to handle the load on the disk.
Action: Submit an SPR with a crash dump. Specify which process was running and make note
of the activity on the system at the time of the erase.
004004
Invalid action byte In Connect Block
Facility: DISK, SDI

Explanation: The subprocess within the disk path that processes requests from the CI
Manager received a connect block with an invalid action byte. This indicates that an invalid
structure was passed to the process, the structure was passed at the improper time, or memory
was corrupted.
Action: Submit an SPR with a crash dump. Include the contents of user register 2 in the crash
dump.
004005
Datagram received from a connection
Facility: DISK, MSCP

Explanation: The main MSCP disk command server process received a non sequenced message
from some connection. This may indicate memory corruption or improper message reception. It
may also indicate that an improper structure was passed to the process, possibly by the host
software.
Action: Submit an SPR with a crash dump. Note all levels of host software running in the
cluster.
004006
MSCP message size exceeded maximum

Explanation: The main MSCP command server process received a sequenced message, with
a length greater than the MSCP maximum, from some connection. This may indicate memory
corruption or improper message reception. It may also indicate that an improper structure was
passed to the process, possibly by the host software.
cluster.
004007
Invalid error signaled by K.ci
Explanation: The main MSCP comma~d server received an MSCP command packet, with
invalid error bits set, from the K.ci. This may indicate memory corruption or improper message
reception. It may also indicate that an improper structure was passed to the process, possibly
by the host software.
cluster and the revision level of the K.ci microcode.
004010
Server queue on work queue with no Items
Explanation: The main disk process received a subprocess work queue, with no items, from
the main work queue. This indicates either memory corruption or improper manipulation of
items on the subprocess work queue. An invalid structure may have been queued to the main
work queue.
Action: Submit an SPR with a crash dump. Note the current process running.
004011
Invalid module number
Facili ty: DISK, many
Explanation: The main disk process detected an invalid module number when it tried to
switch to a different internal process represented by the module number. This indicates that
memory is corrupted or that an invalid structure has queried to the main work queue.
Action: Submit an SPR with a crash dump. Note the current process running.
004013
State change to ONLINE requested through gatekeeper
Facility: DISK, SDI
Explanation: The state change processor within the sequential command gatekeeper received
a DUCB extension requesting a state change to on line. This crash indicates an improper use of
the state change mechanism.
004014
Inconsistent drive state detected
Facility: DISK, SDI
Explanation: The state change processor within the sequential command gatekeeper received
a DUCB extension containing a different state than the current state in the DUCB. This crash
indicates an improper use of the state change mechanism. _
004015
Improper state change for shadow member
Facility: DISK, SDI
Explanation: The sequential gatekeeper mechanism completes action for shadow units before
allowing a state change on any of the members of the shadow set. This crash indicates the
mechanism failed to operate properly.
004016
Disk Unit Table (OU) Inconsistency
Explanation: The disk server tried to add a unit to the DU when it was already there, or tried
to remove a unit from the DU that was not present. This crash indicates improper sequencing
of actions to add or remove a unit in the DU. This crash can also occur if the ordered list of
DUCBs is destroyed.
004017
Invalid diagnostic HMB
Explanation: The diagnostic interface within the disk path received a host message block
(HMB) with a nonzero length field in the HM$LOF word. This indicates an invalid request
from some diagnostic or improper routing of the HMB by the disk path.
Action: Submit an SPR with a crash dump. List any utilities or diagnostics running at the
time of the crash.
004021
Diagnostic release of disk unit while on line
Explanation: A diagnostic or utility attempted to release a disk unit while the disk unit was
still on line.
Action: Submit an SPR with a crash dump. Specify the utilities or diagnostics running at the
time of the crash.
004025
Error Identification table overwritten
Facility: DISK, ERROR
Explanation: The disk error identification table was overwritten or a wild branch was taken.
The most probable cause is a bad load.
Action: If this crash occurs immediately after a boot, try rebooting with a backup copy of the
HSC software. Otherwise, submit an SPR with a crash dump.
004026
Invalid error bH value found during error recovery
Explanation: The bit value describing a KsdilK.si error was not valid for a given stage of the
error recovery. It is also possible, though unlikely, that a K.sdilK.si is malfunctioning.
Action: If this error appears to recur from the same K.sdilK.si, replace it.
004027
Invalid disk characteristics for operation
Explanation: An arithmetic operation to compute some disk parameter caused an overflow or
produced a result outside the allowed range. It is also possible, though unlikely, that a disk is
supplying invalid characteristics to the HSC.
Action: If possible, get the number of the requestor involved from the last error log printed on
the console or from the system error log. Further testing of the disk and attached requestor(s)
may be necessary. If this error appears to recur from the same disk unit, repair it.
004030
S bit not set in FRS error state
Explanation: The S bit in the K control area port subarea for a drive in FRB error state was
not set as expected. This logical inconsistency indicates improper manipulation of the port
state.
Action: If possible, get the number of the requestor involved from the last error log printed
on the console or from the system error log. Further testing of suspected requestor may be
necessary. If this error appears to recur from the same KsdilK.si, replace it. If no hardware
problem exists, submit an SPR with a crash dump.
004031
DT$ERQ not zero In FRB error state
Explanation: The FRB error queue in the DRAT being processed by error recovery was not
zero as expected. This logical inconsistency indicates improper manipulation of the port state.
Action: This error could be caused by a malfunctioning KsdilK.si. Further testing of the
suspected requestor may be necessary. If this error appears to recur from the same K.sdilK..si,
replace it. If no hardware problem exists, submit an SPR with a crash dump.
004032
Unable to get to FRB error state
Explanation: Error recovery was unable to place a port in the FRB error state to perform
an error recovery operation. This crash can occur in an extremely unlikely compound error
situation.
Action: Reboot the HSC. If this error persists, submit an SPR with a crash dump.
004033
Non-ECC/EDC errors remaining after Ece correction
Explanation: ECC error correction should take place after all other errors, except EDC, have
been corrected. This crash occurs because other error bits are set after ECC correction.
004034
Level B retry In wrong state
Explanation: A Level B retry operation was attempted without the drive port being in FRB
error state.
004035
Level C retry in wrong state
Explanation: A Level C retry operation was attempted without the drive port being in FRB
error state.
004036
DeB state is busy with empty DCB queue
Explanation: The drive state indicator in the K control area indicates a KsdilK.si is
processing a DeB, but the DCB queue is empty.
the console or from the system error log.
Further testing of the suspect requestor may be necessary. If this error appears to recur from
the same K.sdilK.si, replace it. If no hardware problem exists, submit an SPR with a crash
dump.
004037
Invalid error queue address in route
Explanation: When the disk server attempted to route an FRB to an error queue, the error
queue address in a route descriptor was invalid.
004040
Undefined error bit In error word from K
Explanation: The error recovery routine IDENTIFY found an undefined bit in the error word
stored by either a K.sdi/K.si or Kci.
Further testing of the suspect requestor may be necessary. If this error appears to recur from
the same K.sdilK.si, replace it. If no hardware problem exists, submit an SPR with a crash
dump.
004041
No buffer found in FRS when expected
Explanation: The error recovery routine MAPBUF attempted to map a buffer but found the
buffer address to be zero.
004042
FRB not in error state for level D 110 operation
Explanation: A call to the error recovery subroutine LVLDIO was made without the port
being in FRB error state. The only cause of this logical inconsistency is a design error within
the error recovery code.
004043
Stack too deep to save in thread block
Explanation: A call to the error recovery subroutine LVLDIO was made with too many items
on the stack to save in a thread block.
004044 .
Buffer not found for specified error
Explanation: A call to the error recovery subroutine RCDH.MX. specified a buffer that was not
in the list of buffers for the specified FRB.
004046
DRAT not found for FRB retirement
Explanation: While attempting to retire an FRB by simulating route completion, the error
recovery subroutine RETIRE could not locate the DRAT for downcounting.
This crash is caused by either overwritten memory or a malfunctioning K.sdilK.si. Further
testing of requestors and HSC internal buses may be necessary. If this error appears to recur
from the same K.sdilK.si, replace it.
If no hardware problem exists, submit an SPR with a crash dump.
004050
DRAT queue not empty for shadow copy
Explanation: Mter obtaining exclusive use of a drive, the shadow copy code found that a
DRAT queue for the drive was not empty.
004051
Inconsistent result for repair operation
Explanation: An impossible combination of results was found at the end of a shadow repair
operation.
004052
Known drive not found In the Disk Unit Table
Explanation: When the disk server attempted to remove a known disk unit from the Disk
Unit Table, the unit was not found in that table.
Action: Submit an SPR with a crash dump. Note any utilities or diagnostics running at the
time of the crash.
004055
Attempt to enable drive interrupt already enabled
Explanation: The ARM subroutine was called to enable K.sdilK.si interrupts to the disk server
for drive state changes when interrupts were already enabled.
Action: Submit an SPR with a crash dump. Note the process running at the time of the crash.
004056
Attempt to enable drive interrupt with pending state change
Explanation: The ARM subroutine was called to enable KsdilK.si interrupts for drive state
changes while a drive state change was being processed.
004057
Invalid drive state change requested
Explanation: The SCHSQM subroutine was called to schedule a state change operation for a
drive that has been declared inoperative but whose state is still recorded as available.
004070
Nonzero status for SUCCESSful DCB
Facility: DISK, SDI
Explanation: Although a DCB (SDI command) completed with a status of SUCCESS, the elTor
word indicated elTors, or the SDI command opcode was invalid.
the console or from the system error log. If this elTor appears to recur from the same KsdilK.si,
replace it. If no hardware problem exists, submit an SPR with a crash dump.
004072
DCB state is busy with empty DCB queue
Explanation: The drive state indicator in the K control area indicates a DCB is being
processed by the KsdilK.si, but the DCB queue is empty.
Further testing of requestors, HSC internal buses, and the memory subsystem may be
necessary. If this error appears to recur from the same KsdilK.si, replace it. If no hardware
problem exists, submit an SPR with a crash dump.
004073
K.sdi is not responding
Facility: DISK, SOl
Explanation: A K.sdilK.si failed to process an immediate DCB within a preset time.
the console or from the system error log. If the error persists, replace the KsdilK.si.
004100
No thread block for operation
Facility: DISK, SOl
Explanation: A thread block was not available to the SDI interface in order to block the
CUlTent thread process
004101
Stack too deep to suspend process in thread block
Facility: DISK, SOl
Explanation: The DCBWAIT routine was called with too many words on the stack to suspend
the process in a thread block.
004106
DRAT allocation failure
Facili ty: DISK, many
Explanation: There was not enough free control memory to allocate a DRAT for a specific
drive type.
Action: Submit an SPR with the crash dump.
004107
A command did not complete after the drive was declared Inoperative
Facility: DISK, MSCX
Explanation: Since no processing was being done on an outstanding command, Get Command
Status processing declared the drive inoperative. The outstanding command, however, the
command still failed to complete in the timeout period.
Action: Submit an SPR with the crash dump. Note the drive type of the drive identified in the
error message and any errors reported by the disk server prior to the crash.
004110
Get Command Status overflow
Explanation: Get Command Status processing determined the calculated status will result in
an overflow.
004111
A timer's link field values are inconsistent with its current operational state
Explanation: A timer was in a state that prevented adding or removing it from an active list.
004112
Inconsistent shadow member state detected
Explanation: A unit is incorrectly marked as a member of a shadow set, or the shadow unit
links are inconsistent given the current state of the shadow unit.
004113
NO DRAT list is invalid
Explanation: The NO DRAT list was found to be invalid when declaring a drive inoperative.
004114
Connection closed after delay in ATTN process
Facility: DISK, AVLATT
Explanation: While the disk server was waiting to acquire resources to send an attention
message to the host, the connection closed.
004115
DeB address inconsistency
Facility: DISK, SDI
Explanation: While processing an error on a seek DCB, the current seek DCB address was
inconsistent with the DCB address stored in the DRAT.
004116
Bad error completion queue in DeB
Explanation: An invalid error completion queue was found in the DCB when it was being
setup for a seek operation. This indicates that after the previous seek operation, the DCB error
completion queue was not properly restored before the DCB was retired.
004117
No DRAT was found on the K.sdi DRAT list when expected
Explanation: The DRAT list was empty when the disk server expected to find a DRAT queued
to the KsdilK..si DRAT list. This most likely cause is a disk server design error.
004120
Too many DRATS in use during ESE transfer operations
Explanation: The number of DRATs in use has exceeded the maximum value allowed. The
possible causes include a design error in the disk transfer code or corruption of the count of
DRATs in use.
004121
RBN access during an ESE transfer
Explanation: The disk server is preparing to perform a transfer operation to an RBN, but the
ESE has no RBNs.
004122
Invalid DRAT bit set
Facili ty: DISK, MSCP
Explanation: A DRAT on the K.sdiJK.si DRAT list did not have the "Set D bit on completion"
flag set as expected. This indicates that the DRAT was probably not set up properly.
004123
DeB K.sdi list inconsistency
Explanation: More than one non-seek DCB was queued to the Ksdi/K.si during I/O rundown
on an ESE. Only one non-seek DCB is expected to be active at a time.
004124
Buffer count Inconsistency for ESE
Explanation: During transfer processing, the DRAT buffer count indicated that the DRAT was
full. However, the DRAT full flag was not set. The most likely cause is a transfer design error.
005001
EeC self-diagnostic string too big for FAO
Facility: ECC
Explanation: A self-diagnostic string generated for the ECC process was too big to print with
the allocated FAO buffer. This crash can only occur if the self-diagnostic code is present and
enabled. The self-diagnostic code is not enabled for distributed base levels.
005002
No ECC errors to correct
Facili ty: ECC
Explanation: An FRB without any errors was sent to the ECC process.
005003
Can't allocate XFRB to print self-diagnostic messages
Facility: ECC
Explanation: The ECC process failed to allocate an XFRB (extended function request block) or
printing messages during self-diagnostic.
005004
ECC found more than a 10-blt symbol error
Facility: ECC
Explanation: The ECC process received a buffer containing more than a lO-bit symbol error.
Error recovery processing should never pass on such a buffer.
006000
This class of crashes Is for tape path software inconsistency errors
Facility: TAPE, TFxxxx
Explanation: A software inconsistency error occUlTed.
Action: Submit an SPR with a crash dump. Specify the utilities or diagnostics active at the
time of the crash.
006001
An STI GET LINE STATUS failed
Facility: TAPE, TFATNAVL
Explanation: When issued to the tape data channel, the STI command GET LINE STATUS
returned with a failure. This command should not fail when issued to a working tape data
channel. General Register 5 points to the windowed K Control Area for the tape data channel
in question. Offset KG$SLT points to the tape requestor in question.
Action: Verify that the K.stilK..si tape data channel is working; if so, submit an SPll with a
crash dump.
006002
Received an interrupt from an unknown tape data channel
Facility: TAPE, TFATNAVL
Explanation: The tape server received an interrupt from an unknown tape data channel.
This is a software inconsistency. General Register 1 points to the windowed tape data channel
control area for the tape data channel in question. General Register 2 contains the tape data
channel slot number the interrupt was received from.
006003
Received an Illegal connection block (CB) from the CIMGR
Facility: TAPE, TFCI
Explanation: A connection block (CB) with an illegal Opcode was sent to the tape diagnostic
interface. General Register 1 points to the windowed address of the connection block (CB) in
question. General Register 2 contains the Opcode in question.
Action: Submit an SPR with a crash dump. Include the connection block (CB) structure.
006004
An Illegal diagnostic Opcode was received
Facility: TAPE, TFDIAG
Explanation: A diagnostic HMB with an illegal Opcode was sent to the tape diagnostic
interface. General Register :3 points to the windowed diagnostic host message block (81MB).
General Register 1 contains the Opcode in question.
time of the crash. Include the HM.B structure.
006005
Diagnostics trying to acquire assigned drive state area
Explanation: Diagnostics are trying to acquire the previously assigned Drive State Area.
General Register 3 points to the windowed Control Memory address of the host message block
(HM:B). General Register 2 points to the tape formatter control block (TFCB).
Action: Submit an SPR with a crash dump. Specify the diagnostics or utilities active at the
time of the crash. Include the HMB, TFCB, and tape drive control block (TDCP) structures.
006006
Inconsistencies during drive state area acquisition
Explanation: The software context word K.T$SFW is not equal to the tape formatter control
block (TFCB) address and/or the DIALOG list head is nonzero when diagnostics are trying
to acquire the Drive State Area. General Register 0 points to the windowed K control area.
General Register 2 points to the tape formatter control block (TFCB).
Action: Submit an SPR with the crash dump. Indicate the utilities or diagnostics active at the
time of the crash. Include the tape formatter control block (TFCB) structure.
006007
No Block Header supplied by BACKUP
Explanation: BACKUP did not supply the initial Block Header buffer descriptor. General
Register 3 points to the windowed host message block (HMB) address. General Register 5
should point to the buffer descriptor and, in this case, be o.
Action: Submit an SPR with the crash dump. Include details of the BACKUP operation.
Include the host message block (HMB) (command packet) structure.
006010
No buffers supplied In BACKUP operation
Explanation: No disk data block buffers were supplied in the host message block (H.M:B) for
the backup operation. General Register 3 points to the windowed Control Memory address
of the HMB in question. General Register 0 should point to the buffer descriptor list for the
backup operation.
Action: Submit an SPR with a crash dump. Include details of the BACKUP operation. Include
the host message block (HMB) (command packet) structure.
006011
Could not allocate a XFRB
Facility: TAPE, TFLIB
Explanation: The tape server could not allocate an XFRB (extended function request block)
through ALOCB, a CHRONIC system service.
006012
Required CIMGR functionality not yet implemented
Facility: TAPE, TFMSCP
Explanation: The host sent the tape server a command packet with an Opcode that was not
a sequenced message. General Register 5 is the Opcode received. General Register 3 is the·
windowed Control Memory address of the command packet received (host message block).
Action: Submit an SPR with a crash dump. Indicate the host software version. Include the
host message block (HMB) (command packet) structure.
006013
Explanation: The tape server received a host command packet longer than allowed (36 bytes).
General Register 4 is the size of the command packet received. General Register 3 is the
windowed Control Memory address of the command packet in question.
Action: Submit an SPR with a crash dump. Indicate the host software version. Include the
006014
Explanation: The tape server received a host command packet with a status that is currently
illegal. General Register 3 points to the windowed Control Memory address of the command
packet in question. Offset HM$ERR is the field in question.
Action: Further testing of HSC hardware, particularly the K.ci, may be necessary. If no
hardware problem exists, submit an SPR. Indicate the host software version and include the
006015
Could not find correct tape drive control block (TDeB) pointer
Facility: TAPE, TFSEQUEN
Explanation: A call to remove a host's access to a drive resulted in the tape server searching
the current chain of tape drive control blocks (TDCBs) in that host's HCB. Inability to find the
correct tape drive control block (TDCB) pointer resulted in this message.
General Register 4 points to the tape drive control block (TDCB) trying to have host access
removed. General Register 3 points to the windowed Control Memory address of the host
message block (lIMB). Offset HM$CTX in the host message block (HMB) points to the host disk
block (HDB). Offset HDB.TDCB in the HDB points to the tape drive control block (TDCB).
006016
Unable to allocate an HDB
Explanation: The tape server's attempt to add a host access, which requires allocation of a
host disk block (HDB), failed for lack of resources.
Action: Submit an SPR \vith a crash dump.
006017
Tape formatter does not support allowed densities
Explanation: The tape formatter does not support a density that the HSC supports. General
Register 4 points to the tape drive control block (TDeB) for the drive in question.
Action: Submit an SPR with a crash dump. Include the host software version and tape
formatter revision. Also include the tape drive control block (TDCB) structure, host software
version, and tape formatter revision.
006020
An invalid density is set in the tape drive control block (TDCB)
Explanation: An invalid density was set in the tape drive control block (TDCB). General
Register 4 points to the tape drive control block (TDCB) in question.
Action: Submit an SPR with a Crash dump. Include the host message block (HMB) structure.
006021
Read-reverse emulation not flagged
Explanation: The tape server entered the read-reverse emulation code without read-reverse
emulation being flagged in the tape drive control block (TDCB) at offset TD.FLAGS bit
TDF.RREVEM. General Register 3 points to the windowed Control Memory address of the
host message block (HMB). General Register 4 points to the tape drive control block (TDCB) for
the drive in question. General Register 2 points to the tape formatter control block (TFCB) for
the formatter in question.
Action: Submit an SPR with a crash dump. Include the following structures: host message
block CEiMB), tape drive control block (TDCB), and tape formatter control block (TFCB).
006022
Route pointer for read-reverse emulation zero
Explanation: The tape server entered the read-reverse emulation code without having the
route pointer set in the host message block (HMB). General Register 3 points to the windowed
Control Memory address of the host message block (HMB) in question.
Action: Submit an SPR with a crash dump. Include the host message block (HMB) structure.
006023
Requested transfer larger than 64 Kb
Explanation: The requested transfer size for a read reverse is larger than 64 Kh. General
Register 3 points to the windowed Control Memory address of the host message block (lIMB) in
question and offset HP.BC indicates the transfer size requested.
Action: Submit an SPR with a crash dump. Include the host message block (HMB) structure.
006024
Explanation: The tape server entered the read-reverse emulation short retry code without
read-reverse emulation being flagged in the tape drive control block (TDCB) at offset TD.FLAGS
bit TDF.RREVEM. General Register 3 points to the windowed Control Memory address of the
host message block (HMB). General Register 4 points to the tape drive control block (TDCB) for
drive in question. General register 2 points to the tape formatter control block (TFCB) for the
formatter in question.
block (I:IM:B), tape drive control block (TDCB), and tape formatter control block (TFCB).
006025
Explanation: The tape server entered the read-reverse emulation long retry code without read-
reverse emulation being flagged in the tape drive control block (TDCB) at offset TD.FLAGS bit
TDF.RREVEM. General Register 3 points to the windowed Control Memory address of the host
message block (HMB). General Register 4 points to the tape drive control block (TDCB) for the
drive in question. General Register 2 points to the tape formatter control block (TFCB) for the
formatter in question.
block (I:IM:B), tape drive control block (TDCB), and tape formatter control block (TFCB).
006026
KT$SEM Is equal to zero
Explanation: The K control area offset KT$SEM is zero. General Register 3 points to the K
control area in question.
Action: Submit an SPR with a crash dump. Include the K control area structure.
006031
No available stacks
Facility: TAPE, TFSERVER
Explanation: There are no available stacks for a process trying to suspend.
006033
Top of user stack for a resume Is not set to server return
Facility: TAPE, TFSERVER
Explanation: The top of the user stack on a process resume is not set to the server return
routine.
006040
No stack available to suspend with
Facility: TAPE, TFSTI
Explanation: There is no stack available to suspend a process. General Register 2 points
to the tape formatter control block (TFCB). General Register 5 points to the K control area.
General Register 4 points to the dialogue control block (DCB).
Action: Submit an SPR with the crash dump and include the following structures: TFCB,
DCB, and K control area.
006041
DeB operation timed out
Facility: TAPE, TFSTI
Explanation: A dialogue control block (DCB) operation timed out.
Action: This usually indicates a problem in the tape data channel. The tape requestor slot in
question is given as the second word on the stack. If no hardware problem exists, submit an
SPR.
006043
Buffer descriptor address missing
Facility: TAPE, TXREVERSE
Explanation: The next address is missing from the linked list of buffer descriptors. General
Register 5 points to the fragment request block (FRB) in question. Offset F$BFHD points to the
buffer descriptor list in question.
Action: Submit an SPR with a crash dump. Include the fragment request block (FRB)
structure.
006044
Unexpected fragment request block (FRB) error received
Facility: TAPE, TFERR
Explanation: The tape server received an error from a software station rather than a
hardware station. General Register 5 points to the fragment request block (FRB) in error.
Action: Submit an SPR with a crash dump. Include the FRB structure.
006045
Unknown fragment request block (FRB) error received
Explanation: An unidentifiable error is flagged in a fragment request block (FRB).
Action: Submit an SPR with a crash dump. Include the FRB structure.
006046
K.cl did not return a fragment request block (FRB)
Explanation: Transfer request blocks (TRBs) have associated fragment request blocks (FRBs)
that point to data buffers. When a TRB is received in error, the FRBs must be deallocated. If
an FRB is held by Kci and not returned within 20 seconds, this crash occurs.
Action: Check the K.ci. If no hardware problem exists, submit an SPR with a crash dump.
006047
Invalid downcount occurred on a host message block (HMB) chain
Explanation: Whenever transfer request blocks (TRBs) were purged from the K.stiIK.si input
queue, the associated host message block (HMB) were returned to the host as an end message.
This catching mechanism relies on a change of HMBs with associated counters. This is a
software consistency. Check to ensure Control memory is not corrupted by the end of the chain.
General Register 5 points to the HMB.
Action: Submit an SPR with a crash dump. Include the lIMB.
006050
Sequence number corruption occurred
Explanation: Error recovery ensures against a deadlock on K.stilK.si by preventing a transfer
request block (TRB) from waiting for a diagnostic control block (DCB) that will never execute.
This is a software inconsistency.
007000
This class of crashes includes CIMGR software consistency errors
Facility: CIMGR, many
Explanation: A software inconsistency error occurred.
time of the crash.
007001
Received a sequence message without a credit
Facility: CIMGR, CIDIRECT
Explanation: The SCS$DIRECT process received a sequence message in a host message block
(HMB) flagged by the K.ci as not having a credit for the connection. General Register 1 has the
address of the HMB in error.
Action: Submit an SPR with a crash dump. Include the lIMB.
007002
Failed to acquire 8 control block from K.cl
Facility: CIMGR
Explanation: The POLLER process could not obtain a control block from the K.ci to resend a
timed-out STACK datagram.
Action: Further testing of the HSC subsystem may be necessary, particularly the available
control memory. If no hardware problem exists, submit an $PR with a crash dump.
007003
K.ci is hung
Facility: CIMGR
Explanation: During the polling interval (60 seconds), the CIMGR ensures K.ci is still
running. This trap indicates it is not.
Action: Further testing of the HSC subsystem may be necessary, particularly the K.ci. If no
hardware problem exists, submit an SPR with a crash dump.
007004
K.cl detected an unrecoverable error and stopped
Facility: CIMGR
Explanation: K.ci sent its control area to the CIMGR exception process. K.ci does this
whenever it detects a nonrecoverable hardware error.
Action: Further testing of the HSC subsystem may be necessary, particularly the K.ci and data
memory. If no hardware problem exists, submit an SPR with a crash dump.
007005
K.ci patch status check failed
Facility: CIMGR
Explanation: K.ci did not respond to a path status check within 8 seconds.
Action: If no hardware problem exists, submit an SPR with a crash dump_
007006
System name Is corrupted
Facility: CIMGR
Explanation: During initialization, the CIMGR discovered the system name in the SCT was
corrupted.
Action: Release the Online button (out) on the HSC. Reboot the HSC by holding the Fault
button in until the State light blinks. This will bypass using the SCT on the boot device. Run
SETSHO to reset the system name and ID, then reboot the HSC again before pushing in the
Online button on the front panel.
007007
HMB received with wrong number of BMBs
Facility: CIMGR
Explanation: CIMGR received a host message block (HMB) with the wrong number of big
message blocks (BMBs), or CIMGR detected an inconsistent state. General Register 0 points to
theHMB.
hardware problem exists, submit an SPR with a crash dump_
007011
Connection Incarnation inconsistent
Facility: CIMGR
Explanation: While a connection is in the process of opening, the incarnation of that
connection is flagged as formative. The final step of opening the connection is to remove the
flag. This crash indicates the flag was prematurely removed, indicating a state inconsistency
for the connection. General Register 2 points to the connection block (CB).
Action: Submit an SPR with a crash dump. Include the CB.
007012
Connection incarnation mismatch
Facility: CIMGR
Explanation: The incarnation of an opening connection is kept in both the connection block
(CB) and the connection block vector table. As a connection opens, a check is made to ensure
these incarnations agree. A disagreement indicates a dangling reference to an old carnation of
the connection.
General Register 2 points to the connection block (CB).
007013
Inconsistent connection state due to a VC closure
Facility: CIMGR
Explanation: CIMGR attempted an illegal state transition on a connection. The state
transition was initiated by a virtual circuit closure. General Register 2 points to the connection
block (CB).
007014
Unable to retrieve resource from K.cl during a disconnect
Facility: CIMGR
Explanation: During a disconnect, the CIMGR was unable to retrieve the resources from the
K.ci associated with the credits on that connection.
007015
K.cl did not respond to notification of a VC closure
Facility: CIMGR
Explanation: The K.ci did not respond to notification of a virtual circuit closure with the
12-second time limit. This crash occurs if the response times out.
Action: Further testing of the HSC subsystem may be necessary, particularly the Kci. If no
007016
Illegal connector state
Facility: CIMGR
Explanation: CIMGR detected an illegal connector block (CB) state. General Register 2 points
to the CB.
007017
Attempt to deallocate a connection block without an incarnation
Facility: CIMGR
Explanation: A connection block (CB) did not have a valid incarnation at the time it was
deallocated.
007020
Failure to retrieve SCS resources from K.ci
Facility: CIMGR
Explanation: When CIMGR tried to allocate resources for use across a virtual circuit, the
count of data memory resources was incorrect. The host message block (HM:B) for serializing
VC traffic must have two big message blocks (BMBs). General Register 0 points to the HMB.
Action: Submit an SPR with a crash dump. Include the HMB.
007021
The count of walters for virtual circuit resources went negative
Facility: CIMGR
Explanation: While processing the list of waiters for virtual circuit transmission resources,
CIMGR detected a nonempty list to indicate a negative number of waiters. General Register 1
points to the system block (SB).
Action: Submit an SPR with a crash dump. Include the SB.
007022
Invalid BMB address
Facility: CIMGR
Explanation: An HMB arrives at the resource collector with an invalid BMB address attached
to it.
Action: Use the SETSHO SHOW REQUESTORS command to view the K.pli microcode
revision level. If it is less than revision 45, contact your Digital Customer Service
representative for the update. Submit an SPR with the crash dump and note the disk
configuration.
007023
SCS buffer retrieval failure
Facility: CIMGR
Explanation: When changing the status of the virtual circuit, CIMGR tries to retrieve the SCS
buffer from the K.ci.K.HSRR queue. This buffer should be on the queue because it is not in use
at the time of the crash. No elements were enqueued on the .KHSRR queue, therefore, CIMGR
forced a crash.
012001
Cantt Find Connection Block
Facility: DUP
Explanation: When DUP receives an HMB, DUP tries to find a reference to the connection
block (referred to by HM$CTX in the HMB) in the DG$ structures (DUP context control blocks).
DUP was unable to find a reference to the connection block, even though it searched every DG$
structure.
Action: Submit an SPR with an exception dump or startup message indicating the contents of
the stack.
012002
Illegal BMB Count
Facility: DUP
Explanation: The HMB (MSCP packet carrier) has an illegal number of Big Message Buffers
(BMBs) allocated. DUP allows only one BMB. Therefore, the HMB is invalid. The third word of
the stack contains the value in HM$CN - the count of the number ofBMBs.
the stack. The second word of the stack contains the windowed address of the lIMB.
012003
Illegal HMB Opcode
Facility: DUP
Explanation: The Opcode specified in the HM$LOF field of the lIMB was not equal to
Hl\1:L$RM. (Received sequence message over connection; HML$RM=OOOOOO.) lIMB Opcodes
must indicate the HMB is for a sequenced message.
the stack. The second word of the stack contains the illegal Opcode.
012004
Illegal HMB Error
Facility: DUP
Explanation: The error specified in the HM$ERR field of the lIMB was not equal to 0,
HME$EC, or HME$NC. The second word of the stack contains the value in the HM$ERR
field. (Extra credits received; HME$EC=10. No credits received; HME$NC=4.)
the stack.
012021
Invalid Connection Block
Facility: DUP
Explanation: The DUP process received a connection block with an invalid value in the
CB$ACT field. The CB$ACT field contains the action value (action to be performed by the DUP
Server).
Action: Submit an SPR with an exception dump or startup message indicating the conten~ of
the stack. The second word of the stack contains the contents of the CB$ACT field.
012024
Bad Down Count
Facility: DUP
Explanation: DUP initiates a return of the endpacket to the host by down counting the
reference counter in the related control block. The down-count action should return a one. If
the downcount did not decrement the reference counter to 1, DUP crashes the HSC. The second
word of the stack is the value of the counter following the downcount.
the stack.
012036
Connection Broken
Facility: DUP
Explanation: While DUP was preparing to send a message to the Kci, the connection to the
host was broken. The connection was broken after DUP did an extensive check to ensure the
connection existed.DUP detected the connection break the second time because the DG$CB
field was set to zero.
042001
FAO message buffer overflow
Facility: DIRECT
Explanation: The program DIRECT was attempting to output the formatted directory end
message, but the length of that message was longer than the allotted FAO output buffer.
043001
Wrong HMB received when trying to bring source on line
Facility: DKCOPY
Explanation: DKCOPY sent a host message block (lIMB) to the disk server requesting the
source unit be brought on line in a shadow set. When the completion queue of this HMB was
checked, it pointed to a different (incorrect) HMB. This is crash $CDKCOPY+SRC_ONL_HMB.
Action: Submit an SPR with a crash dump. Top of stack equals crash code. Second word
points to previous HMB.
043002
Bad downcount when trying to bring source on line
Facility: DKCOPY
Explanation: When an MSCP end message was to be sent over a connection to a host, a
counter keeping track of the transaction (decrementing by 1) failed to operate properly. This
occurred after DKCOPY asked the disk server to bring the source unit on line in a shadow set.
This is crash $CDKCOPY+SRC_ONL_CNT.
points to counter.
043003
Wrong HMB received when trying to issue GCS to target unit
Facility: DKCOPY
Explanation: DKCOPY sent a host message block (HM:B) to the disk server requesting it to
send a GET COMMAND STATUS (GCS) command to the target unit. When the completion
queue of this HM:B was checked, it pointed to a different (incorrect) HMB. This is crash
$CDKCOPY+TGT_GCS_HMB.
points to previous HM:B.
043004
Bad downcount when trying to issue GCS to target unit
Facility: DKCOPY
occurred after DKCOPY asked the disk server to send a GET COMMAND STATUS (GCS)
command to the target unit. This is crash $CDKCOPY+TGT_GCS_CNT.
points to counter.
043005
Bad downcount when trying to bring target unit on line
Facility: DKCOPY
occurred after DKCOPY asked the disk server to bring the target unit on line into the shadow
set. This is crash $CDKCOPY+TGT_ONL_CNT.
points to counter.
043006
Bad downcount when trying to Issue abort command to target unit
Facility: DKCOPY
occurred after DKCOPY asked the disk server to abort an ONLINE command to the target unit.
This is crash $CDKCOPY+TGT_ABO_CNT.
points to counter.
043007
Wrong HMB received after Issuing AVL command to shadow unit
Facility: DKCOPY
Explanation: DKCOPY sent a host message block (HMB) to the disk server requesting the
shadow unit used to facilitate the copy operation be made available. When the completion
queue of this HMB was checked, it pointed to a different (incorrect) lIMB. This is crash
$CDKCOPY+SHA_AVL_HMB.
points to previous HMB.
043010
Bad downcount when trying to issue AVL command to shadow unit
Facility: DKCOPY
occurred after DKCOFY asked the disk server to send the available shadow unit. This is crash
$CDKCOPY+SHA_AVL_CNT.
points to counter.
051001
An XFRB was not acquired to print messages
Facility: SETSHO, SSMAIN
Explanation: The SETSHO main routine did not acquire an XFRB (extended function request
block). A crash was initiated because the lack of an XFRB prevents communication between the
HSC and the console. This is crash $CSETSHO+NOXFRB.
051002
Failed to properly send HMB to K.cl
Explanation: SETSHO sent a host memory block (HMB) to the Kci (the hardware that
handles communication between the hosts and the HSC). A crash was initiated because
SETSHO did not receive confirmation of the HMB from the Kci within the required time.
This is crash $CSETSHO+CIHMB.
051003
Too many characters Intended for console printout
Explanation: In this case, when SETSHO called Formatted ASCII Output (FAO), it generated
more characters than the buffer size allocated would allow. The maximum buffer size is 510
characters. This is crash $SETSHO+PNTOVF. R1 points to string size.
051004
The SCT (System Control Table) crossed a page boundary
Explanation: The SCT must remain on one page in memory. The crash typically indicates
an incorrect amount of padding was placed at the end of the file SSDATA.MAC. This is crash
$SETSHO+SCTXPG.
051101
Failed in sending HMB to disk server for SET Dn [NO]HOST
Facility: SETSHO
Explanation: SETSHO sent a host memory block (HMB) to the disk server to set a disk
drive to HOST or NOHOST access. The crash was initiated because the confirmation of this
command was not received within the required time. This is crash $CSETSHO+SETDSK.
051102
Failed in sending HMB to tape server for SET Tn [NO]HOST
Facility: SETSHO
Explanation: SETSHO sent a host memory block (HMB) to the tape server to set a tape
drive to HOST or NOHOST access. The crash was initiated because the confirmation of this
command was not received within the required time. This is crash $CSETSHO+SETIAP.
051201
Failed in sending HMB to disk server for SHOW Dn
Facility: SETSHO
Explanation: SETSHO sent a host memory block (HMB) to the disk server to show a specified
disk drive. The crash was initiated because the confirmation of this command was not received
within the required time. This is crash $CSETSHO+SHODSK..
051202
Failed in sending HMB to tape server for SHOW Tn
Facility: SETSHO
Explanation: SETSHO sent a host memory block (HM:B) to the tape server to show a specified
tape drive. The crash was initiated because the confirmation of this command was not received
within the required time. This is crash $CSETSHO+SHOTAP.
051203
SeT crash context table contained too many characters
Facility: SETSHO
Explanation: The SCT crash context table contained too many characters. In this case, when
SETSHO called FAO, it generated more characters than the buffer size would allow. The
maximum buffer size is 510 characters. This is crash $SETSHO+CSHOVF. R1 points to string
size.
052001 ($CDWMATH)
Doubleword math not consistent
Facility: SINI
Explanation: During calculation and allocation of control blocks (allocated in quantities of a
doubleword), the count of words in control blocks was not a doubleword multiple. RO points to
memory descriptor (MD).
052002 ($CDIV10)
Divide operation set overflow
Facility: SIN!
Explanation: During allocation of control blocks (set as 80 percent of available Control
memory), a divide operation set the PSW Overflow bit.
052003 ($CMUL8)
Multiply operation set overflow
Facility: SINI
Explanation: During allocation of control blocks (set as 80 percent of available control
memory), a multiply operation set the PSW Overflow hit.
061001
XCALL stack overflow
Facility: DIAGINT
Explanation: The DDUSUB transfer routines use a stack allocated from common pool for
XCALLs (cross-address space calls) from the disk server. The low word of this stack is
initialized to a special value that should never change. This crash occurs when the routine
DDUTIO is called. The low word of the stack contains a value different than the initialization
value. The most probable cause of the crash is corruption by the process running.
Action: Submit an SPR with a crash dump. Note the diagnostics or utilities running at the
time of the crash.
062001 ($CNOWINDOW)
Process does not have windows declared
Facility: SUBLIB, ERTYP
Explanation: A process requesting an out-of-band error log be issued through the ERTYP$
service in SUBLIB does not have windows declared in its PCB (process control block)
declaration. A window set is required to use this service.
062002
Common Pool memory returned twice
Facility: Many
Explanation: A process attempted to return a memory segment that was already in the
common pool.
Generic Error Log Fields C-1
C
Generic Error Log Fields
C.1 Introduction
Some fields described on HSC console message printouts are generic, regardless of error type.
The following example is a typical printout of the error log fields. Table C-l describes the error
fields.
ERROR-S Bad Block Replacement (Success)
Command Ref t OA66000D
RAel unit t 77.
Err Seq t 166.
Format Type 09.
Error Flags 80
Event 002B
ERROR-I End of Error
Example C-1 Error Log Fields Example
Table C-1 Generic Error Log Fields

Field Description
ERROR-x The x represents the severity level of the error message. Severity levels are E for
error, S for success, W for warning, I for informational, and F for fatal. What follows
is the English version of the error message describing the event code, date, and time.
Command Ref # This number, in hexadecimal, is the MSCP command number that caused the
error reported, or is zero if the error does not correspond to a specific outstanding
command.
Err Seq # This number, in decimal, is the sequence number of this error log message since
the last time the MSCP server lost context, or is zero if the MSCP server does not
implement error log sequence numbers.
Error Flags This number, in hexadecimal, indicates bit Hags, collectively called error log message
Hags, used to report various attributes of the error. Refer to Table 0-2 for a
description of the error Hags.
Event This number, in hexadecimal, identifies the specific error or event being reported by
this error log message. This code consists of a five-bit major event code and an 11-bit
subcode. The event codes and what they mean are listed in Table 0-3.
C-1
C-2 Generic Error Log Fields
C.2 Error flags

Table C-2 is a list of error flags that can be set. The first column is the bit number that is set. The
second column is the bit mask hex number. The third column is the format description of the error
flag.
Table C-2 Error Flags

Bit Bit Mask
Number Hex Format Description
7 80 If set, the operation causing this error log message has successfully completed.
The error log message summarizes the retry sequence necessary to successfully
complete the operation.
6 40 If set, the retry sequence for this operation continues. This error log message
reports the unsuccessful completion of one or more retries.
5 20 This is MSCP-specific. If set, the identified logical block number (LBN) needs
replacement.
4 10 This is MSCP-specific. If set, the reported error occUlTed during a disk access
initiated by the controller bad block replacement process.
o 1 If set, the error log sequence number has been reset by the MSCP server since
the last error log message sent to the receiving class driver.
C.3 MSCPfTMSCP Status or Event Codes

Event codes are values reported to error logs and are equivalent to each status code.
The following table is a sequential list of all known MSCP and TMSCP event codes. Each event
code cross references to an error description. The first column is the event code number in
hexadecimal. The second column references the class of error. The third column is the expanded
description that matches the event code.
Table C-3 MSCPITMSCP Status or Event Codes

Event
Code Hex Class Description
0000 Success Normal.

0001 Invalid Command Invalid message length.
Other invalid command subcode values should be referenced
as follows (note that this is combined with the status code):
offset * 256. + code
offset * 256. is the command message and offset value in
decimal for the field in error.
+ code is the symbol for the invalid command status code.
0002 Command Aborted Command aborted.
0003 Unit Off Line Unit unknown or on line to another controller.
0004 Unit Available Unit available.
0007 Compare Error Used only as an event code when the error occurs during a
Read-Compare or Write-Compare operation.
Table C-3 (Cont.) MSCP/TMSCP Status or Event Codes

Event
0008 Data Error Disk-Sector was written with Force Error modifier.
Tape--Long gap encountered.
0009 Host Buffer Access Error Cause not available.
The controller was unable to access a host buffer to perform a
transfer and has no visibility into the cause of the error.
OOOA Controller Error Reserved for host-detected command timeout logging. This
error is never reported by a controller.
OOOC Shadow Set Status Has Disk-Shadow set status has changed. Tape--Formatter
Changed error.
OOOD BOT Encountered BOT encountered.
OOOE Tape Mark Encountered Tape mark encountered.
0010 Record Data Truncated Record data truncated, data transfer operation.
0013 LEOT Detected LEOT detected.
0014 Bad Block Replacement Bad block successfully replaced.
0020 Success Disk-Spindown ignored; status only subcode. Tape--Unload
ignored.
0023 Unit Off Line Disk-No volume mounted or drive disabled via RUN/STOP
switch. Unit is in known substate; status only subcode.
Tape--No media mounted, disabled via switch setting, or on
line to another controller.
002A Controller Error SERDES overrun or underrun error. Either the drive is too
fast for the controller, or more typically, a controller hardware
fault has prevented controller microcode from keeping up with
data transfer to or from the drive.
002B Disk Drive Error Drive command timeout For SDI drives, the controller timeout
expired for either a level 2 exchange or the assertion of
ReadlWrite Ready after an Initiate Seek.
0034 Bad Block Replacement Block verified good-not a bad block.
0035 Media Loader Loader command timeout. The key length is too short for the
specified key type.
0040 Success Still connected; status only subcode.
0043 Unit Off Line Unit is inoperative; status only subcode. For SDI drives,
the controller has marked the drive inoperative due to an
unrecoverable error in a previous level 2 exchange, the drive
C1 flag is set or the drive has a duplicate unit identifier.
0044 Unit Available Shadow set copy in progress; status only subcode.
c-4 Generic Error Log Fields
Table C-3 (Cont.) MSCPITMSCP Status or Event Codes

Event
CodeBex Class Description
0048 Disk Data Error Invalid header. The subsystem read an invalid or inconsistent
header for the requested sector. For recoverable errors, this
code implies a retry of the transfer read or a valid header.
For unrecoverable errors, this code implies the subsystem
attempted nonprimary revectoring and determined the
requested sector was not revectored. As an example, the
RCT indicates the sector is not revectored. Causes of an
invalid header include header mis-sync, header sync timeout,
and an unreadable header.
0049 Host Buffer Access Error Odd byte count.
004A Controller Error EDC error. The sector was read with correct or correctable
ECC and an invalid EDC. A fault probably exists in the ECC
logic of either this controller or the controller that last wrote
the sector This can also be caused by any K module (including
the K.ci) writing bad EDC into Data memory.
004B Disk Drive Error Controller-detected transmission error For SDI drives, the
controller detected an invalid framing code or a checksum
error in a Level 2 response from the drive.
0054 Bad Block Replacement Replacement failur~REPLA.CE command or its analog
failed.
0055 Media Loader Controller-detected transmission error. The controller does
not implement the specified key type.
0068 Disk Data Data sync not found (data sync timeout).
0069 Host Buffer Access Error Nonexistent Memory error.
006A Controller Error Inconsistent internal control structure. A high-level check
detected an inconsistent data structure. For example, a
reserved field contained a nonzero value, or the value in a
field was outside its valid range. This error almost always
implies the existence of a microcode or hardware problem.
006B Disk Drive Error Positioner error (mis-seek). The drive reported a seek
operation was successful, but the controller determined the
drive had positioned itself to an incorrect cylinder.
0074 Bad Block Replacement Replacement failur~inconsistent RCT.
0075 Media Loader Error Controller-detected protocol error.
0080 Success Duplicate unit number; status only subcode.
0083 Unit Off Line Duplicate unit number; status only subcode.
0084 Shadowing Unit Available No members in shadow set. An on-line command was
addressed to a virtual unit of an existing shadow set from
which all members have been removed.
0085 Media Format (Shadowing) Characteristics or protection mismatch for shadow member.
Error
Generic Error Log Fields c-s

Event
0088 Disk Data Error Correctable error in ECC field. A transfer encountered a
correctable error where only the ECC field was affected. All
data bits were correct, but a portion of the ECC field was
incorrect. The severity of the error (the number of symbols
in error) is unknown. If the number of symbols in error is
known, an n symbol ECC error subcode should be returned
instead.
0089 Host Buffer Access Error Host memory parity error.
OOBA Controller Error Internal EDC error. A low-level check detected an inconsistent
data structure. For example, a microcode-implemented
checksum or vertical parity (hardware parity is horizontal)
associated with internal sector data was inconsistent. This
error usually implies a fault in the memory addressing logic
of one or more controller processing elements. It can also
result from a double bit error or other error exceeding the
error detection capability of the controller hardware memory
checking circuitry.
008B Disk Drive Error Lost Read/Write Ready during or between transfers. For SDI
drives, Read/Write Ready drops when the controller attempts
to initiate a transfer or at the completion of a transfer with
ReadlWrite Ready previously asserted. This usually results
from a drive-detected transfer error, where additional error
log messages containing the drive-detected error subcode may
be generated.
0094 Bad Block Replacement Replacement failure-drive access failure. One or more
transfers specified by the replacement algorithm failed.
OOAS Media Format Error Disk-Not formatted with 512-byte sectors; status only
subcode. The disk FCT indicates it is formatted with 576-
byte sectors, although either the controller or the drive
support only 512-byte sectors For tape--Block mode device
not formatted for tape operations.
OOA9 Host Buffer Access Error Invalid page table entry.
OOAA Controller Error LESI adapter card parity error on input (adapter to
controller).
OOAB Disk Drive Error Drive clock dropout. For SDI drives, either data or state clock
was missing when it should have been present. This is usually
detected by means of a timeout.
OOB4 Bad Block Replacement Replacement failure, no replacement block available.
Replacement was attempted for a bad block, but a replacement
block could not be allocated. For example, the volume's RCT
is full.
OOC5 Disk Media Format Error Disk not formatted or FCT corrupted; status only subcode.
The disk FCT indicates the disk is not formatted in either
512- or 576-byte mode.

Event
00C9 Host Buffer Access Error Invalid buffer name. The key in the buffer name does not
match the key in the buffer descriptor, the V bit in the buffer
descriptor is clear, or the index into the buffer descriptor table
is too large.
OOCA Controller Error LESI adapter card parity error on output (controller to
adapter).
OOCB Disk Drive Error Lost Receiver Ready for transfer For SDI drives, Receiver
Ready was negated when the controller attempted to initiate
a transfer or did not assert at the completion of a transfer.
This includes all cases of the controller timeout expiring for a
transfer operation (Level 1 real-time command).
00D4 Bad Block Replacement Replacement failure, recursion failure. Two successive RBNs
were bad.
00E8 Data Error Disk-Uncorrectable ECC error. A transfer without the
Suppress Error Correction modifier encountered an ECC error
exceeding the correction capability of the subsystem error
correction algorithms, or a transfer with the Suppress Error
Correction modifier encountered an ECC error of any severity.
For tape--Unrecoverable read error.
00E9 Host Buffer Access Error Buffer length violation. The number of bytes requested in
the MSCP or TMSCP command exceeds the buffer length as
specified in the buffer descriptor.
OOEA Controller Error LESI adapter card cable in place not asserted.
OOEB Disk Drive Error Drive-detected error. For SDI drives, the controller received
a get status or unsuccessful response with EL set, or the
controller received a response with the DR Hag set and it does
not support automatic diagnosis for that drive type.
0100 Success Already on line; status only subcode.
0103 Unit Off Line Unit disabled by field service or diagnostic; status only
subcode. For SDI drives, the drive DD fiag is set.
0105 Disk Media Format Error RCT corrupted. The RCT search algorithm encountered an
invalid RCT entry. The subcode may be returned under
the following conditions: during replacement of a block,
revectoring a faulty block, and when a unit is brought on line.
0106 Write-Protected Unit is data safety write-protected; status only subcode.
0108 Disk Data Error One-symbol ECC error. A transfer encountered a correctable
ECC error with the specified number of ECC symbols in error.
The number of symbols in error roughly corresponds to the
severity of the error.
0109 Host Buffer Access Error Access control violation. The access mode specified in the
buffer descriptor is protected against the PROT field in the
PTE.
010A Controller Error Controller overrun or underrun. The controller attempted to
perform too many concurrent transfers, causing one or more of
them to fail due to a data overrun or underrun.

Event
010B Disk Drive Error Controller-detected pulse or state parity error. For SDI drives,
the controller detected a pulse error on either the state or data
line, or the controller detected a parity error in a state frame.
0125 Disk Media Format Error No replacement block available. Replacement of a faulty block
was attempted, but a replacement block could not be allocated
(i.e., the RCT is full). This subcode may be returned during
actual replacement and when an interrupted replacement is
completed as part of bringing a unit on line.
0128 Disk Data Error Two-symbol ECC error A transfer encountered a correctable
012A Controller Error Controller memory error. The controller detected an error in
an internal memory, such as a parity error or nonresponding
address. This subcode applies only to errors not affecting
the ability of the HSC to properly generate end and error log
messages. Errors affecting end and error log messages are
not reported via MSCP. For most controllers, this subcode is
return.ed only for controller memory errors in data or buffer
memory and noncritical control structures. If the controller
has several such memories, the specific memory involved is
reported as part of the error address in the error log message.
012B Disk Drive Error Drive-requested error log (EL bit set).
0145 Disk Media Format Error No multicopy protection. All but one copy of a block in a
multicopy structure are bad. The disk should be reformatted
or replaced at the earliest convenient time.
0148 Disk Data Error Three-symbol ECC error. A transfer encountered a correctable
014A. Controller Error Insufficient resources. The controller is unable to honor a
(Shadowing) request to create a shadow set or to add an additional member
to an existing shadow set. This is due to the lack of internal
resources to support the new entity.
014B Disk Drive Error Controller-detected protocol error. For SDI drives, a level
2 response from the drive had correct framing codes and
checksum but was not a valid response within the constraints
of the SI protocol. The response had an invalid opcode, was an
improper length, or was not a possible response in the context
of the exchange.
0168 Disk Data Error Four-symbol ECC errer. A transfer encountered a correctable
OlGA Controller Error PLI transmission buffer parity error.
c-a Generic Error Log Fields

Event
016B Disk Drive Error Drive failed initialization. For SDI drives, the drive clock
did not resume following a controller attempt to initialize the
drive. This implies the drive encountered a fatal initialization
error.
0188 Disk Data Error Five-symbol ECC error. A transfer encountered a correctable
018B Disk Drive Error Drive ignored initialization. For SDI drives, the drive clock
did not cease following a controller attempt to initialize
the drive. This implies the drive did not recognize the
initialization attempt.
OIA8 Disk Data Error Six-symbol ECC error. A transfer encountered a correctable
olAB Disk Drive Error Receiver Ready collision. For SDI drives, the controller
attempted to assert its Receiver Ready when the Receiver
Ready of the drive was still asserted.
01C8 Disk Data Error Seven-symbol ECC error. A transfer encountered a correctable
OICB Disk Drive Error Response overflow. A drive sent back more frames than the
reception buffer could hold. This can be caused by a hung
drive microdiagnostic or a malfunctioning K.sdilK.si.
01E8 Disk Data Error Eight-symbol ECC error. A transfer encountered a correctable
0200 Success Still on line.
0203 Unit Off Line Exclusive use.
0208 Disk Data Error Nine-symbol ECC error. A transfer encountered a correctable
0220 Success Still on line, unload ignored.
0228 Disk Data Error Ten-symbol ECC error. A transfer encountered a correctable
0248 Disk Data Error Eleven-symbol ECC error. A transfer encountered a
correctable ECC error with the specified number of ECC
symbols in error. The number of symbols in error roughly
corresponds to the severity of the error.

Event
0268 Disk Data Error Twelve-symbol ECC error. A transfer encountered a

0288 Disk Data Error Thirteen-symbol ECC error. A transfer encountered a
02A8 Disk Data Error Fourteen-symbol ECC error. A transfer encountered a
02C8 Disk Data Error Fifteen-symbol ECC error. A transfer encountered a
0400 Success Disk-Incomplete replacement; status only subcode. For
tape-EOT encountered.
0404 Unit Available Already in use; status only subcode.
044B Tape Drive Drive error. Controller retry limit exhausted.
0800 Drive error Invalid RCT; status only subcode.
1000 Success Read only volume format; status only subcode.
1006 Write-Protected Unit is software write-protected; status only subcode.
2006 Write-Protected Unit is hardware write-protected; status only suhcode.
F3AA Controller Error Unknown KstilK..si error.
FCAA Controller Error Word rate clock timeout. The KstilK.si detected the loss of
clocks from a drive during a transfer.
FCEA Controller Error Receiver Ready not asserted at start of transfer. The HSC is
ready to start a transfer by sending the formatter a Level 1
command, and the formatter does not have Receiver Ready
asserted.
FD2A Controller Error Data ready timeout. This controller did not detect data ready
from the formatter within 5 ms after sending it a Levell
command.
FD6A Controller Error Acknowledge not asserted at start of transfer. The HSC is
ready to start a transfer by sending the formatter a Level
1 command, and the formatter does not have Acknowledge
asserted.
FDEC Tape Formatter Could not get extended drive status.
FEOC Tape Formatter Could not get formatter summary status while trying to
restore tape position.

Event
FE2A Controller Error Record EDC error. On a read from tape operation the EDC
calculated by the K.stilK.si did not match the EDC generated
by the tape formatter.
FE2B Tape Drive Could not set byte count.
FE4B Tape Drive Could not write tape mark.
FE6B Tape Drive Could not set unit characteristics.
FE8A Controller Error Lower Processor timeout. The Upper Processor in the
KstilK.si detected the Lower Processor had stopped and
restarted it.
FE8B Tape Drive Unable to position to before L_EOT.
FEAB Tape Drive Rewind failure.
FECB Tape Drive Could not complete on-line sequence.
FEEB Tape Drive Erase gap failed.
FFOB Tape Drive ERASE command failed.
FFOC Tape Formatter TOPOLOGY command failed.
FF31 Tape Drive Position Lost Retry limit exceeded while attempting to restore tape position.
FF68 Tape Data Formatter retry sequence exhausted.
FF6A Controller Error Lower Processor error. A bit was set in the Lower Processor
error register. Bits included in the Lower Processor error
register are Data bus NXM, data SERDES overrun, Data
bus ovelTUIl, Data bus par err, data pulse missing, and sync
real-time par err.
FF6B Tape Drive Tape drive requested error log.
FF6C Tape Formatter Formatter requested error log.
FF71 Tape Drive Position Lost Formatter-detected position lost.
FF88 Tape Data Controller transfer retry limit exceeded.
FF8A Controller Error Buffer EDC error. The K.stilK.si detected an EDC error on
the Data Buffer it read from memory on a Write operation.
FFAS Tape Data Host requested retry suppression on a K.sti/K..si-detected
error.
FFAA Controller Error Data overflow due to pipeline error. No Data Buffers in HSC
Data memory were available when the K.sti/K..si needed one
during a data transfer.
FFC8 Tape Data Reverse retry currently not supported.
FFCB Tape Drive Could not position for (formatter) retry.
FFCC Tape Formatter Cannot clear formatter errors.
FFDl Tape Drive Position Lost Formatter and HSC disagree on tape position.
Generic Error Log Fields 0-11

Event
FFE8 Tape Data Host requested retry suppression on a formatter-detected

error.
FFEB Tape Drive Cannot clear drive errors.
FFEC Tape Formatter Could not get formatter summary status during transfer error
recovery.
FFFI Tape Drive Position Lost Controller-detected position lost.
Interpretation of Status Code Bytes 0-1
D
Interpretation of Status Code Bytes
0.1 Introduction
This appendix lists all possible codes each K (e.g. K.ci or data channel) can generate after detecting
a fatal error. Only K-detected errors are listed here.
When a K detects· a fatal elTor, it: puts a code in its status register and performs a level 7 Control
bus interrupt to the P.io. ·This interrupt causes the HSC to trap through location 134 and crash.
The crash message contains the status codes from all Ks in the Status of requestors (1-9): field.
The following shows a printout example from a K-detected error. In this case, as in many others,
the crash was not caused by the K but was detected by the K which forced the crash. Section D.2
explains this crash is detail. For additional explanations of the fields in the crash message, refer to
AppendixB.
-* SUBSYSTEM EXCEPTION *- VI Y10B HSC70 HSC002

at 18-Jan-1986 01:15:14.50 up 0 00:08:46.20
User PC: 0027360 caused by (134 ) Kint
PSW: 140000
KBCTRL active, PCB addr = 102636
RO-R5:
024302 047632 000020 047626 0000000 141404
Kernel SP: 000774
Kernel Stack:
005046 000004 053354 046022 001012 050476 050476 000000
047062 047466 047466 000000 047264 000000 055352 000000
User SP: 023346
User Stack:
052525 052525 025252 025252 025252 025252 025252 025252
025252 025252 025252 025252 025252 025252 025252 025252
KPAR (0-7) :
000440 000640 001040 001440 002040 001240 000240 177600
KPDR(0-7) :
077506 077506 177506 077506 077406 077506 077506 077506
UPAR(0-7) :
000440 000640 001040 001440 002040 001240 000240 177600
UPDR (0-7) :
007406 007406 177406 007406 007406 007406 007406 100016
MMSR(0-2): 000017 000020 037654
Example 0-1 (Cont.) K-detected Error Example
0-1
0-2 Interpretation of Status Code Bytes
Window index reg: 000015

Window Bus Reg: 140105
WADR (0-7) :
160004 161004 162004 163004 164004 165004 166034 167034
001401 001401 001401 001401 001401 001401 001407 001607
Error regs: 170024 000077
Status of requestors (1-9):
000177 000002 000002 000377 000377 000377 000377 000377 000203
(PC-6) TO (PC):
104002 012600 000003 011505
Control area for slot #000001
Control area address: 022010
Register area contents:
000000
000000
100307
040003
104000
140143
100007
000552
000200
012002
000000
000533
104000
000401
022000
000000
000001
000003
004572
000003
017176
000003
000063
000150
000000
000000
000372
040003
002501
002431
000000
000000
000000
Example 0-1 K·detected Error Example
D.2 K-Detected Error Example Examination

Notice the third line of Example D-1 states the crash was caused by (134) Kint. The 134 indicates
a K detected a fatal problem and interrupted the P.ioj with a level 7 interrupt.
In this crash, requestor number 1 (the K.ci) status shows a 000177. The K.ci detected a fatal
condition. The two digits in the status code are 77 (from the 000177 failure code).
Table D-1 provides additional information regarding status code 77. The description of this error
indicates the HSC received a HOST CLEAR command from a host node. The description for the 77
status also shows that the node number of the host which sent the HOST CLEAR is found in R17.
To find R17, look at the Register area contents: field on the second page of the example. The first
entry in the register area contents is always· the Q register from the K. The Q register contains
important information for some crashes. The second entry is RO. In the example, count in octal
up to R17 (remember the first entry is the Q register). The contents of R17 are 000001. Many of
the error descriptions in the following tables indicate additional information exists in one of these
registers.
Notice other entries below R17 in the register area contents. In the K.sdi, Ksti, and Ksi register
areas, these other entries are RAMO through RAM 17, and they sometimes contain important
information. On the K.ci, these entries are not significant for troubleshooting crash messages.
D.3 K-Detected Failure Code Analysis

The following sections aid field service in analyzing the K-detected failure codes through use of the
status code tables. This appendix contains one status code table for each type of K.
• Table D-1 describes the K.ci status codes and applies only to requestor number 1.
• Table D-2 describes the K.sdi status codes.
• Table D-3 describes the K.sti status codes.
• Table D-4 describes the K.si disk status codes.
• Table D-5 describes the K.si tape status codes.
The use of the status code tables requires information about the type of requestor involved. In
order to determine which requestor detected the error, check the Status of requestors (1-9): field
in the crash message. This field shows the status register contents of all requestors present in the
subsystem.
NOTE
The registers referred to in this appendix are not general registers, but the internal
K registers. All status code~ followed by an asterisk (*) are hardware-detected errors.
More detailed information for these errors is found in the appropriate sequencer error
register.
The normal operational status codes for requestors are:
001 for a K.ci
002 for a K. sdiIK. si
203 for a K.stiIK.si
377 means no requestor is in the slot
Any value other than a 001, 002, 203, or 377 means the K detected an error. Because the Kci
is always requestor 1, a Kci-detected error always shows in the far left position in the Status
of requestors (1-9): field of the message. In any other position, the type of requestor must be
determined.
Count over the Status of requestors (1-9): field to the status contents showing an error (this is the
requestor number). When the HSC reboots, type SHOW REQUESTOR at the HSC> prompt to see
whether the requestor detecting the error is a K.sdi, K.sti, or Ksi. Find the number of the data
channel that found the error in the displayed response. This display shows whether that requestor
number is either a K.sdi, K.sti, or K.si.
D-4 Interpretation of Status Code Bytes
NOTE
If the HSC is not operational or the requestor in question fails initialization self-
tests, check the module utilization label above the card cage to determine whether
the involved requestor number is a K.sdi, K.sti, or K.si.
Tables in this appendix consider only the rightmost two octal characters in failure code. Use the
appropriate table (dependent upon requestor type) to find the meaning of the status code.
NOTE
"(See NOTE.)tt appears in several places in the following tables. In each table, this
information appears at the end of that table.
Table 0-1 K.el Status Code Bytes

Status Code
(Octal) Description
00 Two conditions cause failure of the 2911 sequencer test upon powerup or
reinitialization. In one case, the requestor sent status back to the P.io while Init
was asserted. In the other case, the sequencer had already released the !nit signal
but the sequencer failed to reach the point in its code where it could change the
status bits.
A common reason for this status code is from an HSC false power fail crash dump. In
this type of crash dump OOT through 20), all requestors present report a 00 status
code.
01 2901 ALU test failed upon powerup or reinitialization.
02 Data bus (DBUS) test failed upon powerup or reinitialization.
03 Control bus (CBUS) test failed upon powerup or reinitialization.
04 CROM test failed upon powerop or reinitialization.
06 K pli RAM test failed upon powerup or reinitialization.
07 PLI interface test failed upon powerup or reinitialization.
10 Packet buffer test failed upon powerup or reinitialization.
11 LINK. board test failed upon powerup or reinitialization.
12 Control bus/memory error occurred during a lock cycle while the Kci was attempting
to locate the K-Init packet in Control memory upon powerup or reinitialization.
13 Kci could not find a properly formatted K-Init packet in Control memory after
completing poweruplInit diagnostics.
14 An error was detected by the upper (control) sequencer. While attempting to update
the next buffer pointer in an FRB, the pointer was found to be zero (illegal). Rl1
contains the FRB address.
15 * An error was detected by the upper (control) sequencer. (See NOTE.)
16 An error was detected by the upper (control) sequencer. The control stream found
a structure on its own work queue which is not an HMB or FRB. Rll contains the
structure address.
17 An error was detected by the upper (control) sequencer. While constructing a slot
(SNDDAT, REQDAT) from an FRB, the FRB address was found to be zero (illegal).
R12 contains the slot address.
Table D-1 (Cont.) K.cl Status Code Bytes

Status Code
(Octal) Description
21 An error was detected by the upper (control) sequencer. A buffer allocate request
was initiated without sufficient buffers on the allocated queue in the control area to
satisfy the request. RII contains the FRB address.
22 An error was detected by the upper (control) sequencer. The queue head for an
allocated send buffer was zero.
24 An error was detected by the lower (control) sequencer. The lower sequencer
encountered an inconsistent internal data structure. R2 contains the message slot
address.
25 An error was detected by the lower (control) sequencer. During the RTNDAT routine,
the lower sequencer finds a zero (illegal) FRB address.
27 An error was detected by the lower (control) sequencer. This error occurs when the
lower sequencer polling loop calls a routine which adds or removes Big Message
Block (BMB) pointers to or from the BMB chain, if the queue that is supposed to
contain these pointers is empty.
30 An error was detected by the lower (control) sequencer. This error occurs when the
lower sequencer determines that BMBs need to be returned to the free BMB pool and
during a consistency check finds no BMBs to return. R2 contains the message slot
address.
32 An error was detected by the upper (control) sequencer. While attempting to transmit
over a connection, the upper sequencer found an incarnation number of zero (invalid)
in the Connection Block structure. Rll contains the HMB address and R14 contains
the CB address.
33 through 41 * An error was detected by the upper (control) sequencer. (See NOTE.)
42 An error was detected by the upper (control) sequencer. A hardware error was
detected following a block move to Control memory. RIO contains the Upper
Processor error register contents. RIG contains the last Control memory address
in the block that was moved.
43 * An error was detected by the upper (control) sequencer. A hardware error was
detected following a block move out of Control memory. RIO contains the Upper
Processor error register contents. RIG contains the last Control memory address in
the block that was moved.
44* An error was detected by the upper (control) sequencer. A hardware error was
detected following a Control memory Receive operation. RIO contains the Upper
Processor error register contents. RIG contains the Control memory address of the
item received. R17 contains the Control memory address of the queue head.
45 and4G * An error was detected by the upper (control) sequencer. (See NOTE.)
detected during a Downcount operation. RIO contains the Upper Processor error
register value. R17 contains the counter address.
detected while de-queueing a Control memory item from a scratchpad list. RIO
contains the Upper Processor error register contents. Rl1 contains the Control
memory address of the item.
Table D-1 (Cont.) K.ci Status Code Bytes

Status Code
(Octal) Description
51 * An error was detected by the upper (control) sequencer. A hardware error was
detected while internalizing an FRB. RIO contains the contents of the Upper
Processor error register, Rll contains the FRB address and R14 contains the CB
address. The Q register contains the work queue index.
52 An error was detected by the upper (control) sequencer.
Either a consistency problem was found with the scratchpad queue or an attempt
was made to send to a queue at address zero (illegal address).
53 through 55 * An error was detected by the upper (control) sequencer. (See NOTE.)
56 through 71 * An error was detected by the lower (control) sequencer. (See NOTE.)
72 * An error was detected by the lower (control) sequencer. This error occurs while the
Lower Processor is trying to link a BMB on the BMS free chain. RIO contains the
Lower Processor error register contents. R5 contains the BMB Data memory address.
73 * An error was detected by the lower (control) sequencer. A hardware error was
detected during a BMB list operation. RIO contains the Lower Processor error
register contents. R5 contains the BMB Data memory address.
74 * An error was detected by the lower (control) sequencer. A hardware error was
detected during a BMB list operation. RIO contains the Lower Processor error
register contents. R5 contains the BMB Data memory address.
75 * An error was detected by the lower (control) sequencer. (See NOTE.)
76 An error was detected by the upper (control) sequencer. While copying data. from an
HMB to a message slot, the upper sequencer found the byte count of the HMB was
larger than the slot capacity. R12 contains the slot address and R17 contains the text
length.
77 An error was detected by the upper (control) sequencer. A host clear sequence has
been received. R17 contains the address of the issuing node number.
NOTE
The sequencers access Control memory several times before checking for a hardware
error. Thus, to help determine the particular cause of the error, the sequencer saves the
contents of the error register present at the time of the error check in RIO (octal). The
contents of RIO are visible within the crash dump and can help in narrowing the error
possibili ties.
The following lists show the bits available from both the Upper and Lower Processor error registers.
Those bits marked with an asterisk (*) may cause a crash.
• Upper Processor error register:
Bit 0 = Even/odd bit Control memory address
Bits 3, 2, 1 = CCYCLE 2, 1, 0
* Bit 4 = Control bus error (illegal cycle)
* Bit 5 = Control bus NXl\1:
* Bit 6 = Control data parity error
* Bit 7 =Instruction (CROM) parity error
* Bit 8 = Scratchpad parity error
Bit 9 =PLI parity error

Bits 10 through 15 indicate K.ci hardware revision level
• Lower Processor error register:
Bit 0 = Data memory address hit 16
Bit 1 = Data memory address hit 17
Bit 2 =Data memory NMA
* Bit 5 = Data hus NXl\f
* Bit 6 = Data memory parity error
* Bit 7 = Data memory overrun
* Bit 8 = Scratchpad parity error
* Bit 9 = PLI parity error
Bits 10 through 15 indicate K.ci hardware revision level
Table 0-2 K.sdi Status Code Bytes

Status Code
(Octal) Description
reinitialization. In one case, the requestor Bent status back to the P.io while lnit
was asserted. In the other case, the P.io had already released the lnit signal but the
sequencer failed to reach the point in its code where it could change the status bits.
A common occurrence of this status code is from an HSC false power fail crash dump.
In this type of crash dump (lOT through 20), all requestors present report a 00 status
code.
04 PROM test failed upon powerup or reinitialization.
06 Scratchpad RAM test failed upon powerup or reinitialization.
07 R-S/Gen test failed upon powerup or reinitialization.
10 Partial SDI test failed upon powerup or reinitialization.
12 The K.sdi encountered a Control bus/memory problem while searching for the K-Init
packet in Control memory.
13 After completing poweruplInit diagnostics, the K..sdi could not find a properly
formatted K-Init packet in Control memory.
14 While trying to write the microcode version into the control area at address R7+44
(R7 is base address), the upper sequencer encountered a Control bus error. Rll
contains the contents of the upper error register. (See NOTE.)
15 The Upper Processor tried to advance the buffer descriptor pointer when the old
value of the pointer is zero (illegal).
Table 0-2 (Cont.) K.sdl Status Code Bytes

Status Code
(Octal) Description
16 While attempting to read the block number (LBN) from a buffer descriptor in Control
memory, the Upper Processor encountered a hardware error. R11 contains the
contents of the upper error register. (See NOTE.)
17 through 30 * The Upper Processor encountered an error while attempting to access Control
memory. R11 contains the Upper Processor error register contents. (See NOTE.)
31 This error occurs if, during transfer completion, a DRAT counter goes to zero and the
DRAT list head in the control area is not locked and not equal to the current DRAT
value.
memory. RI1 contains the Upper Processor error register contents. (See NOTE.)
43 This error occurs while processing an active DCB if the dialogue state indicator is
not locked (a value of 100000 is not in KS$DHD) and not valid (KS$IND does not
contain the values 0, 1, 2, 3, OR 4, or -1).
44 The Upper Processor encountered an error while attempting to access Control
45 This error occurs if, after completing state 0 processing, the upper sequencer cannot
find a valid DCB opcode. (No valid state is present to go to next.)
74 through 76 The Upper Processor attempted to downcount a counter that was already at zero.
NOTE
The upper sequencer accesses Control memory several times before checking for a
Control bus error. Thust to help determine the particular cause of the error, the upper
sequencer saves the contents of the error register present at the time of the error in
Rll (octal). The contents of Rll are visible within the crash dump and may help in
narrowing the error possibilities.
Interpretation of Status Code Bytes D-9
The following list defines all the bits contained within the Upper Processor error register (value
loaded in Rll). Those bits that may cause a crash are denoted with an asterisk (*).
0-1 0 Interpretation of Status Code Bytes

Bit 0 =Even/odd bit Control memory address
Bits 3, 2, 1 = CCYCLE 2, 1, 0
* Bit 5 = Control bus NXM
* Bit 6 = Control data parity error
* Bit 7 = Instruction (CROM) parity error
Bits 8 through 12 not used
* Bit 13 = Response pulse missing on SDI RDIRES Line (pulse error)
Bit 14 = Upper Processor RTCS clock present
Bit 15 = Parity error on RTDS Line
Table D-3 K.sti Status Code Bytes

Status Code
(Octal) Description
14 through 22 * Control bus error. (See NOTE.)

23 During transfer completion, the buffer descriptor link. word in the FRB was zero.
RAM7 contains the Lower Processor status.
34 The Lower Processor has timed out on a Transfer operation and the Upper Processor
cannot restart it.
35 and 36 * Control bus error. (See NOTE.)
37 A software inconsistency. The STI state zero processing code was entered when the
drive state indicator was not zero.
40 State zero processing is complete. However, the next state (such as Send Level 1
frame or Get Drive Status) is not specified. Thus, the state is undefined.
44 While setting up a transfer, the next buffer descriptor in the FRB was zero (no buffer
was there).
74 Attempted to downcount a counter that was already zero. R14 contains the FRB. R16
contains the counter minus one. R17 contains the address of the counter structure.
75 and 76 * Control bus error. (See NOTE.)
Table D-3 (Cont.) K.sti Status Code Bytes

Status Code
(Octal) Description
In the following status codes, bit 7 is the parity bit. Parity is always odd for
microdiagnostic failures and is always even for functional code failures. Bit 6
is the error bit and is set for microdiagnostic and functional code failures.
reinitialization. In one case, the requestor sent status back to the P.io while Init
was asserted. In the other case, the sequencer had already released the Init signal
but the sequencer failed to reach the point in its code where it could change the
status bits.
A common occurrence of this status code is from an HSC false power fail crash dump.
In this type of crash dump (lOT through 20), all requestors present report a 00 status
code.
106 Scratchpad RAM test failed upon powerup or reinitialization.
110 Partial STI test failed upon powerup or reinitialization.
112 The Ksti encountered a Control bus/memory problem while searching for the K-Init
packet in Control memory.
304 PROM test failed upon powerup or reinitialization.
307 SERDES test failed upon powerup or reinitialization.
313 After completing poweruplInit diagnostics, the Ksti could not find a properly
formatted K-Init packet in Control memory.
NOTE
The upper sequencer accesses Control memory several times before checking for a
Control bus error. Thus, to help determine the particular cause of the error, the upper
sequencer saves the contents of the error register present at the time of the error in
Rll (octal). The contents of Rll are visible within the crash dump and may help in
narrowing the error possibilities.
The following list defines all the bits contained within the Upper Processor error register (value
loaded in R11). Those bits that may cause a crash are denoted with an asterisk (*).
Bit 0 = Even/odd bit Control memory address
Bits 3, 2, 1 = CCYCLE 2, 1, 0
* Bit 5 = Control bus NXM
* Bit 6 =Control data parity error
* Bit 7 = Instruction (CROM) parity error
Bits 8 through 12 not used
* Bit 13 = Response pulse missing on SDI RDIRES Line (pulse-error)
Bit 14 = Upper Processor RTCS clock present
0-14 Interpretation of Status Code Bytes
Table D-5 (Cont.) K.si Tape Status Code Bytes

Status Code
(Octal) Description
03 Control Bus (CBUS) test failed at powerup or reinitialization.

04 The PROM/writable control store (WCS) parity test failed at powerup or
reinitialization.
06 Scratchpad RAM test failed at powerup or reinitialization.
07 RTS Gate Array test failed at powerup or reinitialization.
10 Calibration of the SIECL failed during powerup or reinitialization.
11 WCS moving inversions test failed during off-line diagnostics.
12 The K.si encountered a Control Bus/memory problem while searching for the K-init
packet in Control Memory.
13 Attempt to load module's WCS failed during powerup or off-line load attempt.
14 While trying to write the microcode version into the control area at KG$VRSN, the
upper sequencer encountered a Control Bus error. Rll contains the contents of the
upper error register.
1~22 Control Bus error. See note at the end of this table.
23 During transfer completion, the buffer descriptor link word in the FRB was zero.
RAM7 contains the lower processor status.
24-33 Control Bus error. See note at the end of this table.
34 The lower processor has timed out on a transfer operation and the upper processor
cannot restart it.
37 Software inconsistency. The STI state zero processing code was entered when the
drive state indicator was not zero.
40 State zero processing is complete. However, the next state (such as Send Level 1
frame, or Get Drive Status) is not specified. Thus, the state is undefined.
41-43 Control Bus error. See note at the end of this table.
44 While setting up a transfer, the next buffer descriptor in the FRB was zero because
the buffer descriptor named a nonexistent buffer.
45 Control Bus Error.
70-73 If SIECUSERDES path loop testing failed during powerup or reinitialization because
of one of the following conditions:
70 - Frame loopback from upper to lower
71- Frame loopback from lower to upper
72- Sector loopback from upper to lower
73- Sector loopback from lower to upper
74 The upper processor attempted to downcount a counter that was already zero. R14
contains the FRB. R16 contains the counter minus one. R17 contains the address of
the counter structure.
NOTE
When an error occurs, the upper processor transfers the contents of the upper processor
error register at the time of the error to register Rl1 (octal).
The contents of R11 are given in the crash dump to help you narrow the error
possibilities. The following list defines all the bits contained in Rl1 from the upper
processor error register. Those bits that can indicate the possible cause of a crash are
denoted with an asterisk (*).
Bit 0 Even/odd bit for control memory address
Bits 3,2,1 CCYCLE 2,1,0
Bit 4* Control bus error (illegal cycle)
Bit 5* Control bus NXM
Bit 6* Control data parity error
Bit 7* Instruction (CROM) parity error
Bits 8 through 12 Not used
Bit 13 Response pulse missing on SDI RDIRES line
Bit 14 Upper processor RTC clock pulse present
Bit 15 Parity error on RTDS line
:II m
co
c
HSC70-AA/CA IREV A1 B1 E5 .t,
; NUMBER DESCRIPTION REVISIONS
rr $I
E2-ETCH 01
-...
-...
III (CI LINK)
.... L0100-00 S.
en
g C-ETCH 01 . C)'
::J
::s 3:
.:..,.. L0107-YA K.PLI C2 C3
..
~ C4 e3.
:::!.
X
:t:
~
L0108-YA
(HSC5X-BA)
K.SDI C-ETCH C8 ~
-
.... ()
::r
D-ETCH C8 C9 C10 -.. ~
iI
~ E-ETCH C1 C2 C3 --.
0"
::s F-ETCH C22 C22 C23 C24 C25
iC
a:::!. L0108-YB
K.STI D-ETCH C10 .
~
(HSC5X-CA)
><
o E-ETCH C3 C4 ...
':3'
Q)
::I. F-ETCH C23 C23 C24 C25 C26
L0109-00 PILA E1 E2 ....
L0111-00 P.IOJ C-ETCH A1 -

D-ETCH A2 ~
L0117-AA M.STD2 A-ETCH A2

-...
L0118 III (CI LINK)
L0119 K.SI D-ETCH
5417764-01 BACKPLANE C-ETCH A1 C1 01
CXO-1271B
Sheet 1 of 4
."
cC HSC70-AA/CA IREV A1 81 E5
c
CiJ NUM8ER DESCRIPTION REVISIONS
7'-. 70-20033-03 STD PS ASSY - 120 VAC IN C1 ~
g 70-20184-01 OPT PS ASSY - 120 VAC IN 82 ---...

-
~
..:.,.. 30-24374-01 881A PWR CTRl ASSY B1 --..
% 70-23138-01 OCP ASSEMBLY A2 -.
~ 54-15286-01 ** OCP C -..
f
~ ---...
•0"
~
70-23129-01 FLOPPY DRIVE BKT ASSY A2
30-24962-01 RX33 DRIVE A1 ~

~
!.
:!. EK-HSCMN-IN INSTAllATION MANUAL 001 --..
)(
o:r aX926-H7 HSC70 SOFTWARE V100 V300 V370 ~370+

t»
:s. Bl-FH74X-DE HSC70 OFF-LINE DIAGS A ~
f.
CJ)
S·
::J
3:
~
)(
()
;:r
~
CXO-1271B m
"'THIS BREAKDOWN IS FOR FIELD SERVICE INFORMATION ONLY.
Sheet 2 of 4 ~
E-6 Revision Matrix Charts
E.3 HSC50 (Modified) Revision Matrix Chart

Figure E-2 shows the revision status of all applicable HSC50 (modified) FRUs. An HSC50
(modified) must have all the FRUs at a particular revision level in order to be supported.
l'! E5
CQ HSC50-AA IREV
C
CiJ NUMBER DESCRIPTION REVISIONS
7'
I')
L0100-00 III (CI LINK) E2-ETCH 01
'0
o
C-ETCH 01
-
:::J
r" L0105-00 P.IOC D-ETCH E1 E3 E5
:::t E-ETCH E2 E4 E6
~
g M.STD C1
-
L0106-AA
I: L0107-YA K.PLI C1 C2 C3 C4
o
Q.
:;; L0108-YA C-ETCH
K.SDI C6 C7 C8
-
& (HSC5X-B)
D-ETCH C5 C6 C7 C8 C9 C10
~
fI) E-ETCH C1 C2 C3
0"
:::J
K.SOI F-ETCH C23 C24 C25
I:
a:!.
><
o::r L0108-YB
K.STI O-ETCH C10
I» (HSC5X-C)
:J.
E-ETCH C3 C4
K.STI F-ETCH C24 C25 C26
L0109-00 PILA E1 E2
's,i?
CJ)
54-14048-00 BACKPLANE D-ETCH A1 A2
0'
::s
L0118 III (CI LINK) 3:
~
L0119 K.SI D-ETCH x'
C
:r
~
CXO-2078A m
Sheet 1 of 4 .!.,
11
co
c
NUMBER
HSC50-AA
DESCRIPTION
IREV E4 E5
REVISIONS
~
C1
rr
~
70-20033-01 STO PS ASSY - 120 V~C IN Ai B1 C1 C2 C3 Ai B1 C1 C2 C3 f.
en
0'
~ 70-20033-03 STO PS ASSY - 120 VAC IN C1 C2 C1 C2 :::J
3:
l HSC5X-EA OPT PS KIT - 120 VAC IN Ai A2 B2 a
::::I,
><
:c 70-20184-01 OPT PS ASSY - 120 VAC IN Ai B1 B2 A1 B1 B2 o'::J'
~ 70-19122-00 PWR CTRL ASSY - 120/208 V Ai A2 A3 B1 B2 C1 C2 ~
~
'i
o Z0300-CG HSC50 OKlTP SRVR FRMWR
a.
:;
70-20524-01 OCP ASSEMBL Y A2 A3 "'A2 A3
I
- 54-15286-01 OCP C C
i
(I) 70-20186-01 BEZEL ASSEMBLY (TU58) A1 A
o
~
TU58-XA ORV MECH (70-15510-00) F K F K
a:
a
~ TU58-XB S INTRFC (54-13489-00) F3 F4 K L M F3 F4 K L M
><
o
:r EK-HSCMN-IN INSTALLATION MANUAL 001 002 001 002
I»
:I.
AA-GMEAA-TK USER GUIDE
OX926-HG HSC50 SOFTWARE V350 V350 V370 V370+
BE-T 493X-XX HSC50 OFF-LINE DIAGS E-DE E-DE
CONTROLLER, PWR 120 V,

30-24374-01 3-PHASE, 9-0UTLET B1
CXO-2078A
Sheet 2 of 4
:II
ca HSC50-AB IREV E6
c
iI1 NUMBER DESCRIPTION REVISIONS
7'
I\)
L0100-00 III (CI LINK) E2-ETCH 01
~ C-ETCH 01
a
..:..,.. L0105-00 P.IOC D-ETCH E1 E3 E5
:E: E-ETCH E2 E4 E6
~
g M.STD C1
-f
L0106-AA
L0107-YA K.PU C1 C2 C3 C4
a.
it=
L0108-YA C-ETCH
K.SDI C6 C7 CS
-i (HSC5X-B)
O-ETCH C5 C6 C7 CS C9 C10
tn E-ETCH C1 C2 C3
0"
::J
K.SDI F-ETCH C23 C24 C25
~
a::1
><
o L010S-YB
K.STI D-ETCH C10
i
:1.
(HSC5X-C)
E-ETCH C3 C4
K.STI F-ETCH C24 C25 C26
L0109-00 PILA E1 E2
~
D-ETCH
s.
en
54-1404S-00 BACKPLANE A1 A2
c)"
::::J
L011S III (CI LINK) 3:
L0119 K.SI D-ETCH
~
x
9
i
CXO-207SA
Sheet 3 of 4
rr
(t)
::!!
cg
c
HSC50-AB lREV E4 E5 E6 rr-.
o
CiJ NUMBER DESCRIPTION REVISIONS
rr
N
70-20033-02 STD PS ASSY - 240 VAC IN A1 B1 C1 C2 C3 A1 B1 C1 C2 C3 A1 B1 C1 C2 C3
JJ
~,
rA
70-20033-04 STD PS ASSY - 240 VAC IN C1 C2 C1 C2 C1 C2 0'
::I: :::3
~ HSC5X-EB OPT PS KIT - 240 VAC IN A1 A2 A1 A2 B2 3:

fa
::s,
x
ia. 70-20184-02 OPT PS ASSY - 240 VAC IN A1 81 82 A1 81 82 A1 81 82 (')
=r
70-20613-01 PWR CTRL ASSY - 240/416 V A1 A2 B1 C1 A1 A2 B1 C1
~
-Ii ZD300-CG
70-20524-01
HSC50 DKlTP SRVR FRMWR
OCP ASSEM8L Y A2 A3 A2 A3 A2 A3
o
0"
~ 54-15286-01 OCP C C C
!:
a:1 70-20186-01 BEZEL ASSEMBLY (TU58) A1 A1 A1
><
o:r TU58-XA DRV MECH (70-15510-00) F K F K F K
ID
:1. TU58-XB S INTRFC (54-13489-00) F3 F4 K L M F3 F4 K L M F3 F4 K L M
EK-HSCMN-IN INSTALLATION MANUAL 001 002 001 002 001 002
AA-GMEAA-TK USER GUIDE
OX926-HG HSC50 SOFTWARE V300 V350 V350 V370 V370+
BE-T 493X-XX HSC50 OFF-LINE DIAGS E-DE E-DE ~-DE
CONTROLLER, PWR 240 V,

30-24374-02 3-PHASE, 9-0UTLET
CXO-2078A
Sheet 4 of 4
Revision Matrix Charts E-11
E.4 HSC50 Revision Matrix Chart

Figure E-3 shows the revision status of all applicable HSC50 FRUs. An HSC50 must have all the
FRUs at a particular revision level in order to be supported.
2 Index
Control bus error conditions (hardware- Disk functional out-of-band errors (cont'd.)
detected), 8-54 clock dropout from ONLINE disk unit
Controller byte field, 8--34 xx., 8-62
Controller errors deferred ATN. message for node xx,
compare error, ~ unit xx, 8-67
data bus overrun, 8-65 disk unit xx. (requestor xx., port xx.)
data memory error (NXM or parity), being INITialized, 8-67
8-66 disk unit xx. ready to transfer, 8-68
EDC error, 8-70 disk unit xxx. (requestor xx., port xx.)
internal consistency error, 8-77 declared inoperative, 8-68
MSCP, 8-29 DRAT/SEEK timeout, disk unit xxx.,
PLI receive buffer parity error, 8-86 8-68
PLI transmit buffer parity error, 8-86 DRIVE CLEAR attempt on disk unit
SERDES overrun, 8-93 xx. (requestor xx., port xx.)., 8-69
TMSCP, 8-29 duplicate disk unit xx, 8-70
Control program, 1-21 FRB error: Kci, 1st LBN xx., xx.
Cooling, 1-5 buffers, FE$SUM xx, 8-72
Crash dump, B-1 FRB error: Ksdi, unit xx., 1st LBN
CSR breakdown xxx., xx. buffers, FE$SUM xx,
RX33 disk drive, 8-51 8-73
illegal bit change in status from disk
unit xxx, 8--76
D KsdilK.si in slot xx. failed its lnit DIT
Data channel module (K.si), 1-17 status = xxx, 8-78 '
LBN restored with forced error in
see also K. si
dc power switch, 2-6 RESTOR operation!, 8-79
Description and :flags LBN xx. repaired for shadow member
unit xx., 8-79
MSCP error format, 8-26
TMSCP error format, 8-26 positioner error on disk unit xxx.
DRAT addr:xxx, 8-87
Device integrity tests
premature LP flag in RTNDAT
generic error message format, 5-2
sequence from host node xx, 8-87
generic prompt syntax, 5-1
ILRX33, 5-2 SDI exchange retry on disk unit xxx,
8-91
ILTU58 , 5-5
Diagnostic indications unexpected AVAILABLE signal from
ONLINE disk unit xx, 8--98
boot, 8-26
unit xx. declared inoperative because
Diagnostic manager, 1-22
no progress made on Command
Diagnostic subroutines, 1-22
Disk data channel module (K.sdi), 1-16 Reference xxx:xx., 8-98
unrecoverable error on disk unit xx.
see also K.sdi Drive appears inoperative, 8-99
Disk drive, 3-38 unsuccessful SEEK initiation, disk unit
see RX.33 disk drive xxx. DCB addr: xxx, 8-99
prompts, 5-40 VC closed due to timeout of
Disk drive integrity test RTNDAT/CNF from host node
ILDISK, 5-9 xx, 8-99
Disk error processor, 1-22 Disk 110 manager, 1-22
Disk functional errors, 8-52 Disk status code bytes
Disk functional out-of-band errors K.si, D-12
aborting error recovery due to excessive Disk transfer errors
recals, 8-58 data sync not found, 8-67
aborting error recovery due to excessive eight-symbol ECC error, 8-81
timeouts, 8-59 five-symbol ECC error, 8-81
attention condition serviced for forced error, 8-71
ONLINE disk unit xxx., 8-60 four-symbol ECC error, 8-81
ATl'N. message sent to node xx, for MSCP, 8-35
unit xx, 8-59 MSCP field description, 8-35
Index 3
Disk transfer errors (cont'd.) DKUTIL (cont'd.)

one-symbol ECC error, 8-81 DKUTIL-E n is an invalid par number;
RCT corrupted error, 8-88 maximum is n, 7-14
seven-symbol ECC error, 8--81 DKUTIL-E nonexistent unit number,
six-symbol ECC error, 8-81 7-13
three.-symbol ECC error, 8-81 DKUTIL-E revector for LBN n failed,
two-symbol ECC error, 8-81 MSCP status: (status), 7-14
uncorrectable ECC error, 8-81 DKUTIL-E SDI command was
DKUTIL, 7-1 to 7-14 unsuccessful, 7-14
command descriptions, 7-3 DKUTIL-E there is no buffer to dump,
command modifiers, 7-2 7-13
command prompt, 7-10 DKUTIL-E unable to read error log,
command summary, 7-3 7-14
command syntax, 7-2 DKUTIL-E unit is not available, 7-13
DEFAULT command, 7-4 DKUTIL-E xxx is an invalid xxx, 7-14
DEFAULT command modifiers, 7-4 DKUTIL-F 110 request was rejected,
DEFAULT command usage, 7-4 7-13
DISPLAY command, 7-5 DKUTIL-F insufficient resources to
DISPLAY command examples, 7-6 RUN, 7-13
DISPLAY command modifiers, 7-5 DKUTIL-I CTRLIY or CTRUC abort,
DISPLAY command parameters, 7-5 7-14
DISPLAY command syntax, 7-5 DUMP command, 7-6
DISPLAY command usage, 7-5 DUMP command examples, 7-7
DKUTIL-E all copies of xCT block n DUMP command modifiers, 7-7
are bad, 7-14 DUMP command parameters, 7-6
DKUTIL-E cannot bring unit ONLINE, DUMP command syntax, 7-6
7-13 error messages, 7-12
DKUTIL-E copy n of xCT block n (XBN error message severity levels, 7-13
n) is bad, 7-14 error message variables, 7-12
DKUTIL-E drive must be acquired to EXIT command, 7-7
execute this command, 7-14 EXIT command syntax, 7-7
DKUTIL-E drive must be on line to EXIT command usage, 7-8
execute this command, 7-14 fatal error messages, 7-13
DKUTIL-E drive went AVAILABLE, GET command, 7-8
7-13 GET command modifiers, 7-8
DKUTIL-E drive went OFFLINE, GET command parameters, 7-8
7-13 GET command syntax, 7-8
DKUTIL-E error log corrupted, can not GET command usage, 7-8
display entries, 7-14 information and error messages, 7-13
DKUTIL-E error log corrupted, can not POP command, 7-8
display header, 7-14 POP command syntax, 7-8
DKUTIL-E error log not implemented POP command usage, 7-8
in drive, 7-14 PUSH· command, 7-9
DKUTIL-E illegal response to start-up PUSH command syntax, 7-9
question, 7-13 PUSH command usage, 7-9
DKUTIL-E invalid block number for REVECTOR command, 7-9
XBN space, 7-14 REVECTOR command examples, 7-9
DKUTIL-E invalid decimal number, REVECTOR command parameters,
7-13 7-9
DKUTIL-E invalid octal number, REVECTOR command syntax, 7-9
7-13 REVECTOR command usage, 7-9
DKUTIL-E invalid sector size; only 512 sample session, 7-10
and 576 are legal, 7-14 SET command, 7-9
DKUTIL-E missing modifier only tit' SET command examples, 7-9
was specified, 7-13 SET command parameters, 7-9
DKUTIL-E missing parameter, 7-13 SET command syntax, 7-9
SET command usage, 7-9
4 'Index
DKUTIL (cont'd.) Error message listing (cont'd.)

starting, 7-1 bad block replacement (RCT
inconsistent), 8-60
E bad block replacement (recursive
failure), 8-60
Enable indicator bad block replacement (REPLACE
HSC, 2-5 failed), 8-61
HSC50, 2-6 bad block replacement (success), 8-61
Error byte field, 8-33 bad dispatch state in CB ... , 8-61
Error conditions (hardware-detected) booted from drive 1. Drive 0 error
control bus, 8-54 (text), 8-61
Error information buffer EDC error, 8-62
categories of software errors, 8-26 cables have gone from uncrossed to
DKUTIL, 7-12 crossed, 8-83
Hags, C-2 cache disabled due to failure, 8-62
FORMAT, 7-26 clock dropout from ONLINE disk unit
generic error log fields, C-1 xx., 8-62
ILEXER error messages, 5-50 compare error, 8-63
ILMEMY error messages, 5-8 controller-detected position lost, 8-63
ILTCOM error messages, 5-36 controller-detected transmission or time
initialization, 8-2 out error, 8-64
MSCP Hags, 8-28 controller transfer retry limit exceeded,
off-line bus interaction test, 6-27 8-63
off-line cache test, 6-20 could not complete on-line sequence,
off-line diagnostics bootstrap, 6--4 8-64
off-line KIP memory test, 6-45 could not get extended drive status,
off-line K test selector, 6-35 8-64
off-line memory test, 6-55 could not get formatter summary status
off-line OCP test, 6-77 during transfer error recovery,
off-line refresh test, 6-72 8-64
off-line RX33 exerciser, 6-67 could not get formatter summary
PATCH, 7-31 status while trying to restore tape
SDI, 8-30 position, 8-65
SINI-E printout, B-2 could not position for formatter retry,
TMSCP flags, 8-28 8-65
VERIFY, 7-18 could not set byte count, 8-65
Error message fields could not set unit characteristics, 8-65
MSCP, 8-27 data bus overrun, 8-65
RX33 message last line breakdown, data error Hagged in backup record,
8-51 8-66
TMSCP, 8-27 data memory error (NXM or parity),
Error message listing 8-66
aborting error recovery due to excessive data ready timeout, 8-66
recals, 8-58 data sync not found, 8-67
aborting error recovery due to excessive date/time set by node nn, 8-67
timeouts, 8-59 deferred ATN. message for node xx,
acknowledge not asserted at start of unit xx, 8-67
transfer, 8-59 disk unit XL (requestor xx., port xx.)
attention condition serviced for being INITialized, 8-67
ONLINE disk unit xxx., 8-60 disk unit xx. ready to transfer, 8-68
A1TN. message sent to node xx, for disk unit xxx. (requestor xx., port xx.)
unit xx, 8-59 declared inoperative, 8-68
bad block replacement (block OK), DRAT/SEEK timeout, disk unit xxx.,
8-60 8-68
bad block replacement (drive DRIVE CLEAR attempt on disk unit
inoperative), 8-60 xx. (requestor XL, port xx.)., 8-69
drive clock dropout, 8-69
Index 5
Error message listing (cont' d.) Error message listing (cont'd.)

drive-detected error, &-69 K..stilK.si in requestor xx has microcode
drive inoperative, 8-70 incompatible with this TMSCP
drive-requested error log (EL bit set), Server, 8-78
8-70 last soft Init resulted from unknown
duplicate disk unit xx, 8-70 cause, 8-78
EDC error, 8-70 LBN restored with forced error in
eight-symbol ECC error, 8-81 RESTOR operation!, 8-79
ERASE command failed, 8-71 LBN xx. repaired for shadow member
ERASE GAP command failed, 8-71 unit xx., 8-79
five-symbol ECC error, 8-81 less than 87.5 percent of xx memory is
forced error, 8-71 available, 8-79
formatter and HSC disagree on tape lost ReadlWrite Ready, 8-80
position, 8-72 lost Receiver Ready, 8-80
formatter-detected position lost, 8-71 Lower Processor error, 8-81
formatter-requested error log, 8-72 Lower Processor timeout, 8-81
formatter retry sequence exhausted, no control block available to satisfy
8-72 HMB request., 8--82
four-symbol ECC error, 8-81 node nn cables have gone from crossed
FRB error: Kci, 1st'LBN xx., xx. to uncrossed, 8-83
buffers, FE$SUM xx, 8-72 node nn path (A or B) has gone from
FRB error: Ksdi, unit xx., 1st LBN good to bad, 8-84
xxx., xx. buffers, FE$SUM xx, node nn path n has gone from bad to
8-73 good, 8-84
hard transfer error loading (file) xx, no tape drive structures available for
8-73 Requestor xx Port xx Unit xx,
hard transfer error writing SCT xx, 8-82
8-73 no tape formatter structures available
header error, 8-73 for Requestor xx Port xx, 8-83
HML$ER set--HM$ERR = nn, 8-74 no usable K..stiJK.si boards were found
host clear from CI node, 8-75 by the TMSCP Server, 8-83
host interface (Kci) failed INIT diags, one-symbol ECC error, 8-81
=
status xxx, 8-75 P.ioj/c running with memory bank or
host interface (Kci) is required but not board swap enabled, 8-86
present, 8-75 parity error Trap through 114, 8-85
host requested retry suppression on a PLI receive buffer parity error, 8-86
formatter-detected error, 8-76 PLI transmit buffer parity error, 8-86
host requested retry suppression on a positioner error on disk unit xxx.
KstilK.si-detected error, 8-76 DRAT addr:xxx, 8-87
illegal bit change in status from disk position or unintelligible header error,
unit xxx, 8-76 8-87
increase drive structures through SET premature LP flag in RTNDAT
MAX_TAPE command, 8-82 sequence from host node xx, 8-87
increase formatter structures pulse or parity error, 8-88
through SET MAX...FORMATI'ER ROT corrupted error, 8-88
command, 8-83 Receiver Ready not asserted at start of
insufficient Control memory for transfer, 8-88
K..stilK.si in requestor xx, 8-76 record EDC error, 8-89
insufficient private memory remaining requestor xx failed INIT diags, status =
for TMSCP Server, 8-77 :xxx, 8-89
internal consistency elTor, 8-77 requestor xx has failed initialization
=
Kci exception detected, code nnn, =
diagnostics with status xx, 8-89
8-77 reserved instruction Trap through 10,
Kci loopback microcode loaded, 8-78 8-89
KsdilK.si in slot xx. failed its Init DIT, resource lost to K..ci-xxx xxx HMBs,
=
status xxx, 8-78 8-90
6 Index
Error message listing (cont' d.) Error message listing (cont'd.)

retry limit exceeded while attempting TMSCP Server operation limited by
to restore tape position, ~90 insufficient private memory, 8-96
reverse retry currently not supported, topology command failed, ~97
~90 TTRASH fatal initialization error,
rewind failure, ~90 8-97
SCT read or verification error. Using two-symbol ECC error, 8-81
template SCT., 8-91 unable to position to before LEOT,
SDI clock persisted after Init, 8-91 8-97
SDl exchange retry on disk unit xxx, unclearable drive error, 8-97
8-91 unclearable formatter error, 8-98
SERDES overrun, 8-93 uncorrectable ECC error, 8-81
seven-symbol ECC error, 8-81 unexpected AVAILABLE signal from
SI clock resumption failed after lnit, ONLINE disk unit xx, ~98
8-91 unit xx. declared inoperative because
SI command timeout, ~92 no progress made on Command
SI Receiver Ready collision, 8-92 Reference xxxxx., ~98
SI response length or Opcode error, unknown K.tape error, 8-98
8-93 unrecoverable error on disk unit xx.
SI response overflow, ~93 Drive appears inoperative, 8--99
six-symbol ECC error, ·8-81 unsuccessful SEEK initiation, disk unit
software inconsistency Trap through xxx. DCB addr: xxx, 8-99
20, ~94 VC closed due to timeout of
subsystem exception, level 7 K RTNDAT/CNF from host node
interrupt Trap through 134, ~79 xx, 8-99
subsystem exception, MMU Trap VC closed with node nn due to
through 250, 8-81 disconnect timeout, ~99
subsystem exception, NX.M Trap VC closed with node nn due to request
through 4, 8-85 from K.ci, 8--100
subsystem exception, parameter VC closed with node nn due to START
change, process yyy, 8-85 received, ~100
subsystem exception, PC xxx, 8-85 VC closed with node nn due to
subsystem exception, PSW xxx, 8-85 unexpected disconnect, ~100
subsystem exception, Reason xxx, VC open with node nn, 8-101
8-85 ***WARNING*** K.sti microcode too
tape drive requested error log, ~94 low for large transfers., 8-101
tape formatter declared inoperative, word rate clock timeout, 8-101
8-94 Error messages
tape unit number xx connected to BBR, ~8
requestor xx port xx ceased to exist DKUTIL, 7-12
while on line, 8-94 FORMAT, 7-26
tape unit number xx connected to ILDISK, 5-14
requestor xx port xx dropped state ILEXER, 5-52
clock, 8-95 ILMEMY, 5-9
tape unit number xx connected ILRX33, 5--4
to requestor xx port xx is not ILTAPE, 5-30
asserting Available when it should ILTCOM, 5--36
be, ~95 ILTU58, 5-6
tape unit number xx connected to miscellaneous, 8-52
requestor xx port xx went available off-line bus interaction test, 6--28
without request, 8-95 off-line cache test, 6-21
tape unit number xx connected to off-line KIP memory test, 6--45
requestor xx port xx went off line off-line K test selector, 6-35
without request, 8-96 off-line memory test, 6-56
three-symbol ECC error, 8-81 off-line OCP test, 6--77
TMSCP fatal initialization error- off-line refresh test, 6--72
TMSCP functionality not available, off-line RX33 exerciser, 6-67
~96
Index 7
Error messages (cont'd.) FORMAT (cont'd.)

PATCH, 7-31 FORMAT-F cannot position to DBN
RX33 disk drive, S-50 area, 7-26
SIN!, S-52 FORMAT-F current maximum sector
TMSCP, S-39 size is 512, 7-26
VERIFY, 7-18 FORMAT-F DBN format error, 7-26
EXAMINE and DEPOSIT commands· FORMAT-F drive does not support 576
asterisk (*) symbolic address, 6-10 mode on this media, 7-26
at (@) symbolic address, 6-11 FORMAT-F drive is write-protected,
command repeats, 6-11 7-26
minus sign (-) symbolic address, 6-11 FORMAT-F FCT does not have enough
plus sign (+) symbolic address, 6-10 good copies of each block, 7-26
qualifiers (switches), 6-12 FORMAT-F FCT is improper, 7-26
qualifier switch /byte, 6-12 FORMAT-F FCT nonexistent, 7-26
qualifier switch !DECIMAL, 6-13 FORMAT-F FCT read error, 7-26
qualifier switch !HEX., 6-13 FORMAT-F FCT write error, 7-27
qualifier switch !INHIBIT, 6-13 FORMAT-F formatter initialization
qualifier switch /long, 6-12 error, 7-27
qualifier switch /next, 6-12 FORMAT-F GET STATUS failure,
qualifier switch /OCTAL, 6-13 7-27
qualifier switch /quad, 6-12 FORMAT-F LBN format error, 7-27
qualifier switch /word, 6-12 FORMAT-F nonexistent unit number,
relocation register, 6-11 7-27
set default command, 6-13 FORMAT-F RCT does not have enough
symbolic addresses, 6-10 good copies of each block, 7-27
Exception codes and messages, B-1 FORMAT-F RCT is full, 7-27
Exception messages listing, B-4 to B-44 FORMAT-F RCT read error, 7-27
External interfaces, 1-11 FORMAT-F RCT write error, 7-27
External loop test FORMAT-F SDI receive error, 7-27
Data channel module (Ksi), 3-29 FORMAT-F too many bad RBNs found
before RCT was formatted, 7-27
F FORMAT-F unsuccessful SDI
command, 7-27
Failover procedure, 3-2 FORMAT-I bad LBN n (x), a non-
Fault code displays primary revector, 7-27
interpretation, 4-9 FORMAT-I bad LBN n (x), a primary
OCP, 8-2 revector to RBN n., 7-27
Fault indicator and switch, 2-2 FORMAT-I bad LBN n (x), in the RCT
Field descriptions area, 7-27
BBR errors, 8-38 FORMAT-I bad RBN n (x), 7-27
original error flags, 8-36 FORMAT-I CTRIJY or CTRUC abort,
recovery flags, 8-37 7-28
SDI error, S-30 FORMAT-I cylinder n, group n, track
FORMAT, 7-22 to 7-28 n, position n, PBN n, 7-27
caution, 7-22 FORMAT-I FCT was not used, 7-28
CTRUC caution, 7-23 FORMAT-I FCTwas used successfully,
CTRUY caution, 7-23 7-28
error and information messages, 7-26 FORMAT-I n cylinders left in XBN
error messages, 7-28 space at hh:mm:ss.xx, 7-28
error message severity levels, 7-26 FORMAT-I only DBN area formatted
error message variables, 7-26 (n bad DBNs), 7-28
fatal error messages, 7-26 FORMAT-S format begun, 7-28
FORMAT-E illegal response to start-up FORMAT-S format completed, 7-28
question, 7-28 FORMAT-W possible head addressing
FORMAT-E nondefaultable parameter, problem, 7-27
7-28 information messages, 7-27
running, 7-23
10 Index
ILDISK ILEXER
error messages (cont'd.) disk errors (cont'd.)
error 52, 576-byte format failed, error 103, this drive removed from
5-20 test, 5-54
error 53, 512-byte format failed, error 104, couldn't put drive in
5-20 DBN space, 5-54
error 54, insufficient resources to error 105, no DACB available,
perform test, 5-20 5-54
error 55, drive transfer queue not error 106, some disk I/O failed to
empty before format, 5-20 complete, 5-54
error 56, K.sdilK.si detected error error 107, command failed - invalid
during format, 5-20 header code, 5-54
error 57, wrong structure on error 108, command failed-no
completion queue, 5-21 control structures available,
error 58, Read operation timed out, 5-54
5-21 error 109, command failed-no
error 59, K.sdilK.si detected error buffer available, 5-54
in read preceding format, error 111, write requested on
5-21 write-protected drive, 5-54
error 60, read DRAT not returned error 112, data compare error,
to completion queue, 5-21 5-54
error 61, Format operation timed error 113, pattern number error,
out, 5-21 5-54
error 62, format DRAT was not error 114, EDC error, 5-54
returned to completion queue, error 116, unknown unit number
5-21 not allowed in ILEXER, 5-54
error 63, can't acquire specific unit, error 117, disk unit numbers
5-21 must be between 0 and 4095
error 64, duplicate unit detected, decimal, 5-54
5-21 error 118, hard failure on disk,
error 65, format tests skipped due 5-54
to previous error, 5-21 error 119, hard failure on Compare
error 66, testing aborted, 5-22 operation, 5-54
error 67, not good enough DBN s error 120, hard failure on Write
for format, 5-22 operation, 5-54
hardware requirements, 5-10 error 121, hard failure on Read
MSCP status codes, 5-22 operation, 5-54
operating instructions, 5-10 error 123, hard failure on initial
progress reports, 5-12 Write operation, 5-54
software requirements, 5-10 error 124, drive no longer on line,
specifying requestor and port, 5-12 5-55
system requirements, 5-10 error message format, 5-50
test parameters, 5-11 error messages, 5-52
tests performed, 5-9 generic errors, 5-52
test summaries, 5-12 error 01, no disk or tape
test termination, 5-11 functionality, 5-52
ILEXER error 02, could not get control block
communications error format, 5-51 for timer, 5-52
communications error report, 5-48 error 03, couldn't get timer for
data compare error format, 5-50 MDE, 5-52
data patterns, 5-44 error 04, disk functionality
data transfer error report, 5-46 unavailable, 5-52
disk drive prompts, 5-40 error 05, tape functionality
disk errors, 5-54 unavailable, 5-53
error 102, drive error not up to error 06, couldn't get drive status,
speed, 5-54 5-53
error 07, drive is unknown, 5-53
Index 11
ILEXER ILEXER
generic errors (cont'd.) tape errors (cont'd.)
error 08, drive is unavailable, error 204, comm error: TDUSUB
5-53 call failed, 5-55
error 09, drive cannot be brought error 205, read data error, 5-55
on line, 5-53 error 206, tape mark error, 5-55
error 12, couldn't return drive to error 207, tape position lost, 5-55
available state, 5-53 error 209, data pattern word error,
error 13, user requested write on 5-55
write-protected unit, 5-53 error 210, data read EDC error,
error 14, no tape mounted on unit, 5-55
5-53 error 211, couldn't set unit char,
error 15, record length larger that 5-55
12K or 0, 5-53 error 213, truncated record data
error 16, this unit already error, 5-55
acquired, 5-53 error 214, drive error... hard error,
error 18, invalid time entered, 5-55
5-53 error 215, unexpected error
error 20, couldn't get buffers for condition, 5-55
transfers, 5-53 error 216, unexpected BOT
error 21, tape rewind commands encountered, 5-55
were lost, 5-53 error 217, unrecoverable write
global prompts, 5-43 error, 5-55
informational message error 218, unrecoverable read
at most, 16 words may be entered error, 5-55
in a data pattern, 5-52 error 219, controller error... hard
disk interface not available, 5-52 error, 5-55
number must be between 0 and 15, error 220, formatter error...hard
5-52 error, 5-55
pattern number must be within error 221, retry required on tape
specified bounds, 5-52 drive, 5-56
please mount a scratch tape, 5-52 error 222, hard error limit
please wait-clearing outstanding exceeded, 5-56
110, 5-52 error 224, drive went off line,
starting LBN is either larger than 5-56
ending LBN or larger than error 225, drive went available,
total LBN on disk,5-52 5-56
tape interface not available, 5-52 error 226, short transfer error,
informational messages, 5-52 5-56
multi drive exerciser, 5-37 error 227, tape position
operating instructions, 5-38 discrepancy, 5-56
pattern word error format, 5-51 test parameter, 5-39
performance summary, 5-46 test summaries, 5-48
progress reports, 5-46 test termination, 5-39
prompt error format, 5-50 ILMEMY
setting/clearing :flags, 5-46 error message example, 5-8
system requirements, 5--37 error messages, 5-9
tape drive exercise commands, 5-49 error 000, tested twice with no
tape drive prompts, 5-42 error, 5-9
tape errors, 5-55 error 001, returned buffer to free
error 201, coulddt get formatter buffer queue, 5-9
characteristics, 5-55 error 002, memory parity error,
error 202, couldn't get unit 5-9
characteristics, 5-55 error 003, memory data error, 5-9
error 203, some tape I/O failed to error 004, NXM Trap (Buffer
complete, 5-55 Retired), 5-9
14 Index
K pli (cont'd.) LINK (cont'd.)

indicators, 2-10 indicators, 2-10
interfaces, 1-16 interfaces, 1-16
removal, 3-22 jumpers, 3-16
replacement, 3-23 packet reception, 1-16
switches, 3-22 packet transmission, 1-15
switch settings, 2-14 removal, 3-14
testing, 3-23 replacement, 3-20
Ksdi SERDES, ENDEC, 1-15
LEDs, 8--13 switches, 3-15
removal, 3-23 switch settings, 2-13
replacement, 3-24 testing, 3-20
status code bytes, D-7 Loader
testing, 3-24 HSC50 off-line diagnostics, 6-2
Ksi HSC off-line diagnostics, 6-1
configuration problems, 3-31 off-line diagnostics, 6-7
disk status code bytes, D-12
external loop test, 3-29 M
indicators, 2-11
initialization, 3-30 M.std
LEDs, 8--13 indicators, 2-11
mismatch conditions, 3-30 LEDs, 8--12
new boot microcode, 3-32 removal, 3-37
removal, 3-26 replacement, 3--37
replacement, 3-28 testing, 3--38
requestor configuration, 3-27 M.std2
switches, 3-26 control memory (M.ctI), 1-18
switch settings, 2-16 Data memory (M.dat), 1-18
tape status code bytes, D-13 indicators, 2-11
testing, 3-33 LEDs, 8--12
Ksti Program memory (M.prog), 1-18
indicators, 2-11 removal, 3-35
LEDs, 8--13 replacement, 3--36
removal, 3-25 RX33 diskette controller (Krx), 1-18
replacement, 3-25 testing, 3-36
status code bytes, D-I0 Main power supply
testing, 3-25 HSC, 3-69
K-detected errors, D-l HSC50, 3-73
K-detected failure code analysis, D-3 Maintenance terminal connection
HSC50, 4-2
L Memory integrity tests
ILMEMY, 5-7
LA12 Parameters, 4-4 Memory module (M.stci), 1-19
Lamp bit check see also M. std
off-line OCP test, 6-79 Memory module (M.std2), 1-18
Lamp test, 2-3 see also M.std2
LEDs Memory test configuration
Kci modules, 8--13 Off-line bus interaction test, 6-28
Ksdi module, 8--13 Miscellaneous errors, 8-52
Ksi module, 8--13 Mismatch conditions
Ksti module, 8--13 Data channel module (Ksi), 3-30
M.std2 module, 8--12 Mode byte field, 8-33
M.std module, 8--12 Module indicators, 2-8
P.ioj/c module, 8--12 Module names, 1-14
LINK Module removal and replacement, 3-13
ACKINACK, 1-15 Module switches, 2-11
CRC, 1-15 MSCP
Index 15
MSCP (cont'd.) OCP fault code interpretation (cont'd.)

class server, 1-22 fault code 31, initialization failure,
controller errors, 8-29 8-9
Disk transfer error field description, illegal inst, 8-10
8-35 Kci host reset, 8-11
disk transfer errors, 8-35 level 7 interrupt, 8-10
error flags, 8-28 Memory Management Unit (MMU)
error format description and flags, trap, 8-10
8-26 NXM trap, 8-10
error message fields, 8-27 parity trap, 8-10
format type codes, 8-27 software crash, 8-10
generic error format, 8-27 fault code 32, software inconsistency,
SDI errors, 8-30 8-11
software errors, 8-26 fault code 33, illegal configuration,
status codes in ILDISK, 5-22 8-11
status or event codes, C-2 Off-line bus interaction test (OBIT), 6-24
Multidrive exerciser error OOO--memory test error, 6-28
ILEXER, 5-37 error 001-K timed-out during lnit,
Multiple HSCs in cluster failover, 3-2 6-28
error 002-K timed-out during test,
N 6-29
error 003-parity trap, 6-29
New boot microcode error 004-NXM trap, 6-29
Data channel module (Ksi), 3-32 error OOS-memory test error (P.ioj/c)
Nonfailing requestor detected, 6-29
status, 8-14 error 01l-RX.33 drive not ready, 6-30
error 012-RX33 CRC error during
o seek, 6-30
error 013-RX33 track 0 not set on
OCP recalibrate, 6-30
Blank indicators, 2-3 error 014-RX33 seek timeout, 6-30
controls and indicators, 2-2 error 01S-RX.33 seek error, 6-30
fault code displays, 8-2 error 016-RX33 read timeout, 6-30
fault codes, 2-2 error 017-RX33 CRCIRNF error on
lamp test, 2-3 read command, 6-30
off-line OCP test (OOCP), 6-73 error 10 (12 octal Hache parity error,
OCP fault code interpretation =
VPC XXXXXX, 6-30
fault code 1, Kpli error, 8-4 error information, 6-27
fault code 10, P.ioj cache failure, 8-5 memory test configuration, 6-28
fault code 11, Kci failure, 8-5 operating instructions, 6-24
fault code 12, data channel module parameter entry, 6--25
failure, 8-6 prerequisites, 6-24
fault code 2, KsdilK.si incorrect version progress reports, 6--26
of microcode, 8-4 requestor error summary, 6-27
fault code 21, P.ioj/c module failure, system requirements, 6--24
8-6 test summaries, 6--26
fault code 22, M.std2 module failure, test termination, 6--25
8-6 Off-line cache test (OFLCXT), 6-16
fault code 23, boot device failure, 8-7 error OO--memory parity error, 6--21
fault code 25, port link node address error 01-NXM trap, 6-21
switches out of range, 8-7 error 02-cache parity error, 6-21
fault code 26, missing files required, error 03-bit stuck in cache control
8-8 register, 6-21
fault code 3, KstilK.si incorrect version error 04--forced miss operation failed,
of microcode, 8-5 6-21
fault code 30, no working Kci, Ksdi, error OS-forced miss with abort failed,
K..sti, or Ksi in subsystem, 8-8 6-21
18 Index
Off-line RX33 exerciser (OFLRXE) Off-line RX33 exerciser (OFLRXE)

(cont'd.) (cont'd.)
error 06--track 0 did not set after test summaries, 6--65
recalibrate command, 6-68 test termination, 6--64
error 07-RX33 did not interrupt as Online indicator, 2--3
expected, 6-68 Online switch, 2--3
error 10--seek error detected during Operating instructions
positioning operation, 6--68 ILDISK, 5-10
error 11--current track resister ILEXER, 5-38
incorrect, 6--68 ILMEMY, 5-7
error 12-CRC error in header detected ILRX33, 5--3
during position verify, 6--68 ILTAPE, 5-23
error 13-processor type is not J11, ILTCOM, 5-34
6-68 ILTV58 , 5-6
error 14--drive under test is not ready, off-line bus interaction test, 6--24
6-68 off-line KIP memory test, 6-41
error 15-last command did not off-line K test selector, 6--31
complete, 6-68 off-line memory test, 6--52
error 16--RX33 header does not off-line OCP test, 6-73
compare, 6-68 off-line refresh test, 6-70
error 17-record not found during read off-line RX33 exerciser, 6--63
(could also say write), 6-68 Operational status codes
error 20-CRC error in date during requestors, D--3
read (could also say write), 6--69 Operator control panel (OCP), 2--2
error 21-lost data detected during Ordering related documentation, xxi
read (could also say write), 6--69 Original error flags
error 23-invalid pattern code in buffer, field description, 8-36
6-69 Out-of-band errors, 8-49
error 24--drive is write-protected,
6-69 p
error 25-CRC error in header during
read (could also say write), 6--69 P.ioj/c
error 26--data incorrect after DMA test indicators, 2--11
modeconnnand, 6-69 INIPIO diagnostic (P.ioj), 4--7
error 27-data compare error, 6--69 lnit P.ioc diagnostic, 4-8
error 30--RX33 detected parity error jumpers, 3-34
during read (could also say write), module LEDs, 8-12
6--69 module LEDs power-up sequence,
error 31-RX33 detected NXM during 8--12
read (could also say write), 6--69 P.ioj switch settings, 2-17
error 32-RX33 MAR value incorrect removal, 3--33
after DMA transfer, 6-69 replacement, 3-34
error 33-parity error was not forced in ROM bootstrap, 6--2
main memory, 6--69 testing, 3-35
error 34--parity error did not set in Parameter entry
CSR, 6--69 off-line bus interaction test, 6--25
error 35-NXM did not set in CSR, off-line cache test, 6--17
6--70 off-line KIP memory test, 6-42
error 36--parity error set along with off-line K test selector, 6--31
NXM in CSR, 6--70 off-line memory test, 6--52
error 37--cache parity error, VPC - off-line OCP test, 6--73
XXXXXX, 6--70 off-line refresh test, 6--71
error information, 6--67 off-line RX33 exerciser, 6--64
operating instructions, 6--63 Parity errors
parameter entry, 6-64 off-line KIP memory test, 6--43
progress reports, 6--65 off-line memory test, 6--54
system requirements, 6--63 PATCH, 7-28 to 7-33
Index 19
PATCH (cont'd.) PILA (cont'd.)

commands, 7-28 testing, 3-21
error and information messages, 7-31 Port buffer module (PlLA), 1-16
error messages, 7-32 see also PlLA
fatal error messages, 7-32 Port link module (LINK), 1-15
information messages, 7-33 see also LINK
PATCH-E incorrect checksum, 7-32 Port processor module (Kpli), 1-16
PATCH-E invalid command, 7-32
see also K. pli
PATCH-E invalid device name or
881 power controller, 2-17
switch, 7-32
BUS/ON/OFF switch, 2-18
PATCH-F cannot access PATCH data
circuit breaker, 2-18
file, 7-32
fuse, 2-18
PATCH-F cannot access version on
operating instructions, 2-17
off·line diagnostic medium, 7-32
Power Control bus connections, 2-18
removal, 3-64
system medium, 7-32
replacement, 3-64
TOTAL OFF connector, 2-19
utility medium, 7-32
Power controller
PATCH-F file not found, 7-32
HSC, 3-64
PATCH-F insufficient resources to run,
HSC50, 3-67
7-32
Power indicator, 2-2
PATCH-F read failure: block-number,
Power removal
7-32
HSC, 3-3
PATCH-F unites) write·protected:
HSC50, 3-5
update was not done, 7-32
Precautions, 3-2
PATCH-F version on system medium
Progress reports
does not match utility medium,
ILDISK., 5-12
7-32
ILEXER, 5-46
PATCH-F write failure: block-number,
ILMEMY, 5-8
7-32
ILRX33, 5-3
PATCH-F write failure during write
ILTAPE, 5-28
check, status: block number, 7-32
off-line bus interaction test, 6-26
PATCH-F you cannot PATCH this file,
off-line cache test, 6-18
7-32
off-line KIP memory test, 6-43
PATCH-I checksum =octal-checksum
off-line K test selector, 6-33
(0), 7-33
off-line memory tests, 6-53
PATCH-I CTRUY or CTRUC abort,
off-line refresh test, 6--71
7-33
off-line RX33 exerciser, 6-65
PATCH-I no patches recorded, 7-33
PATCH-I patches made:, 7-33
PATCH-S patch-count changes Made, Q
7-33
Qualifiers (switches)
PATCH-S wait... , 7-33
EXAMINE and DEPOSIT, 6--12
PATCH-W buffer space exhausted,
Quick verify algorithm
7-33
off-line memory test, 6-52
PATCH-W nonfile structured mode
assumed, 7-33
running, 7-29 R
sample session, 7-31
Recovery Hags
success messages, 7-33
field description, 8-37
warning messages, 7-33
Related documentation, xxi
PILA
Relocation register
indicators, 2-10
EXAMINE and DEPOSIT, 6--11
removal, 3-20
Removal
replacement, 3-21
Data channel module (Ksi), 3-26
switches, 3-21
Disk data channel module (Ksdi),
switch settings, 2-15
3-23
20 Index
Removal (cont'd.) Requestor error summary

HSC50 airflow sensor, 3-58 off-line bus interaction test, 6-27
HSC50 auxiliary power supply, 3-79 Requestor status
HSC50 blower, 3-62 nonfailing requestor, 8-14
HSC50 main power supply, 3-73 Revision matrix chart
HSC50 OCP, 3-53 HSC, E-1
HSC50 power controller, 3-67 HSC50, E-11
HSC airflow sensor, 3-56 HSC50 (modified), E-6
HSC auxiliary power supply, 3-76 ROM bootstrap, 6-2
HSC blower, 3-60 RX33 disk drive, 3-38
HSC main power supply, 3-69 CSR breakdown, 8-51
HSC OCP, 3-51 error code table, 6-5
I/O control processor module (P.ioj/c), error message last line breakdown,
3-33 8-51
Memory Module (M.std), 3-37 errors, 8-50
Memory module (M.std2), 3-35 indicators, 2-5
Port buffer module (PILA), 3-20 jumpers, 3-41
Port processor module (Kpli), 3-22 off-line RX33 exerciser (OFLRXE),
881 power controller, 3-64 6-63
RX33 disk drive, 3-38 removal, 3-38
Tape data channel module (Ksti), replacement, 3-45
3-25 status register summary, 8-51
TU58 tape drive, 3--45 testing, 3-45
Removal and replacement
FRUs, 3-3
subunits, 3-38 s
Replace flags bit descriptions, 8-39 Safety, 3-2
Replacement SDl bus, 1-6, 1-12
Data channel module (K.si), 3-28 SDI errors
Disk data channel module (Ksdi), controller-detected transmission or time
3-24 out error, 8-64
HSC50 airflow sensor, 3-58 drive clock dropout, 8-69
HSC50 auxiliary power supply, 3-79 drive-detected error, 8-69
HSC50 blower, 3-62 drive inoperative, 8-70
HSC50 main power supply, 3-73 drive-requested error log (EL bit set),
HSC50 OCP, 3-55 8-70
HSC50 power controller, 3-67 error example, 8-30
HSC airflow sensor, 3-56 field descriptions, 8-30
HSC auxiliary power supply, 3-76 header error, 8-73
HSC blower, 3-60 lost ReadlWrite Ready, 8-80
HSC main power supply, 3-69 lost Receiver Ready, 8--80
HSC OCP, 3-52 MSCP, 8-30
I/O control processor module (P.iojlc), position or unintelligible header error,
3-34 8-87
Memory module (M.std), 3-37 pulse or parity error, 8-88
Memory module (M.std2), 3-36 SDl clock persisted after Init, 8-91
Port buffer module (PILA), 3-21 SI clock resumption failed after Init,
port link module (LINK), 3-20 8-91
Port processor module (Kpli), 3-23 SI command timeout, 8-92
881 power controller, 3-64 SI Receiver Ready collision, 8-92
RX33 disk drive, 3--45 SI response length or Opcode error,
Tape data channel module (Ksti), 8-93
3-25 SI response overflow, 8-93
TU58 tape drive, 3-50 SDl manager, 1-22
Request byte field, 8-32 SecurelEnable switch
Requestor configuration HSC, 2-5
Data channel module (Ksi), 3-27 HSC50, 2-6
Index 21
SecurelEnable switch (cont'd.) Status register summary (cont'd.)

off-line OCP test (OOCP), 6-80 RX33 disk drive, 8-51
SINI errors, 8-52 STI bus, 1-6, 1-12
booted from drive 1. Drive 0 error STI errors, 8-40
(text), 8-61 drive error log, 8-42
cache disabled due to failure, 8-62 drive error log (TA78 drive specific),
SINI-E error printout, B-2 8-43
SINI out-of-band errors drive error log example, 8-42
hard transfer error loading (file) xx, drive error log field description, 8-43
8-73 example, 8-40
hard transfer error writing SCT xx, field description, 8-40
8-73 formatter E log, 8-42
host clear from CI node, 8-75 formatter error log, 8-40
host interface (Kci) failed INIT diags, formatter error log field description,
status = xxx, 8-75 8-42
host interface (Kci) is required but not GEDS text, 8-43
present, 8-75 STI manager, 1-22
last soft lnit resulted from unknown Submitting an SPR, B-2
cause, 8-78 Subsystem block diagram, 1-13
less than 87.5 percent of xx memory is Subunit removal and replacement, 3-38
available, 8-79 Switches
P.ioj/c running with memory bank or Data channel module (Ksi), 3-26
board swap enabled, 8-86 off-line OCP test, 6-78
parity error Trap through 114, 8-85 Port buffer module (PlLA), 3-21
requestor xx failed INIT diags, status = port link module (LINK), 3-15
xxx, 8-89 Port processor module (K. pli), 3-22
reserved instruction Trap through 10, System requirements
8-89 ILDISK, 5-10
SCT read or verification error. Using ILEXER, 5-37
template SCT., 8-91 ILMEMY, 5-7
software inconsistency Trap through ILRX33, 5-3
20, 8-94 ILTAPE, 5-23
subsystem exception, level 7 K ILTCOM, 5-33
interrupt Trap through 134, 8-79 ILTU58 , 5-5
subsystem exception, MMU Trap off-line bus interaction test, 6-24
through 250, 8-81 off-line cache test, 6-17
subsystem exception, NXM: Trap off-line diagnostics loader, 6-8
through 4, 8-85 off-line KIP memory test, 6-41
subsystem exception, parameter off-line K test selector, 6-30
change, process yyy, 8-85 off-line memory test, 6-52
subsystem exception, PC xxx, 8-85 off-line OCP test, 6-73
subsystem exception, PSW xxx, 8-85 off-line refresh test, 6-70
subsystem exception, Reason xxx, off-line RX33 exerciser, 6-63
8-85
Software errors T
error message categories, 8-26
MSCP, 8-26 Tape compatibility test
Software inconsistencies, B-1 ILTCOM, 5-32
Software overview, 1-20 to 1-22 Tape Data Channel Module (K.sti), 1-17
State and lnit indicators, 2-2 see also Ksti
State LED check Tape device integrity test
off-line OCP test, 6-81 ILTAPE, 5-23
Status code bytes Tape drive, 3-45
K.ci modules, D-4 see TU58 tape drive
K.sdi, D-7 prompts, 5-42
K.sti, D-10 Tape errors
Status register summary
24 Index
Utilities (cont'd.) VERIFY (cont'd.)

VERIFY, 7-15 VERIFY-I n blocks with solid (non-
Utility processes, 1-22 ECC) errors, 7-22
VERIFY-I n blocks with transient
v errors, 7-22
VERIFY-I n blocks with uncorrectable
VERIFY, 7-15 to 7-22 ECC errors, 7-22
after running FORMAT, 7-25 VERIFY-I n good RBNs marked bad in
error and information messages, 7-18 the RCT, 7-22
error classes, 7-15 VERIFY-I n LBN s with corrupted data,
error message severity levels, 7-18 7-22
fatal error messages, 7-19 VERIFY-I n revectors verified, 7-22
information messages, 7-21 VERIFY-I n total blocks with any
running, 7-16 error, 7-22
sample session, 7-17 VERIFY-I n total ECC symbols
steps to verify disk, 7-15 corrected, 7-22
variable output error fields, 7-18 VERIFY-I n unused RBNs with good
VERIFY-F all copies of the xCT block n EDC, 7-22
are bad, 7-19 VERIFY-I RBN block is good but not
VERIFY-F current system sector size is used for a revector, 7-21
512, 7-19 VERIFY-I RBN block_no marked bad
VERIFY-F drive went off line, 7-19 in the RCT was not bad, 7-21
VERIFY-F I/O request was rejected, VERIFY-I there were inconsistencies
7-19 found for this drive, 7-21
VERIFY-F insufficient resources to VERIFY-I xBN block has an
run, 7-19 uncorrectable ECC error, 7-21
VERIFY-F mode is bad or format is in VERIFY-I XBN n. has a n symbol
progress on this unit, 7-19 correctable ECC error, 7-21
VERIFY-F too many bad blocks, 7-19 VERIFY-I XBN n. has a transient (n
VERIFY-I copy of n of xCT block n out of 6) x error, 7-21
(XBN n.) is bad, 7-21 VERIFY-I XBN n. has solid errors: x.,
VERIFY-I CTRl1Y or CTRUC abort, 7-21
7-21 VERIFY-W cannot on-line unit, 7-19
VERIFY-I DBN area should probably VERIFY-W cannot read track with
be reformatted, 7-21 starting XBN n, 7-19
VERIFY-I drive is OK, 7-21 VERIFY-W copy n of xCT block n
VERIFY-I exiting, 7-22 (xBNn.) does not compare, 7-19
VERIFY-I initial write should be VERIFY-W illegal response to start-up
specified for ILEXER, 7-21 question, 7-19
VERIFY-I LBN block has corrupted VERIFY-W LBN block has corrupted
data (forced error), 7-21 data (forced error), 7-20
VERIFY-I LBN n., a primary, has a VERIFY-W LBN n., a non-primary
bad header (is non-primary), 7-21 revector, is improper, 7-19
VERIFY-I n bad DBNs, 7-22 VERIFY-W LBN n., a primary revector,
VERIFY-I n bad RBNs verified, 7-22 is improper, 7-19
VERIFY-I n blocks with EDC errors, VERIFY-W LBN n. revectors to RBN
7-22 n. which is bad, 7-19
VERIFY-I n blocks with hard EDC VERIFY-W LBN n marked primary in
errors, 7-21 RCT, not revectored to its primary,
VERIFY-I n blocks with header 7-20
compare errors, 7-22 VERIFY-W n bad PBNs not in the
VERIFY-I n blocks with non-header, RCT, 7-19
non-EDC errors, 7-22 VERIFY-W nonexistent unit number,
VERIFY-I n blocks with n symbol ECC 7-20
errors, 7-22 VERIFY-W RBN block is good but not
VERIFY-I n blocks with positioner used for a revector, 7-20
errors, 7-22
Index 25
VERIFY (cont'd.) VERIFY-WXBN n. lIO error in access

VERIFY-W RBN block_TW marked bad (MSCP code: 0), 7-20
in the RCT was not bad, 7-20 warning messages, 7-19
VERIFY-W unit is not available, 7-20 Voltage test points
VERIFY-W xBN block has an HSC50 auxiliary power supply, 3-81
uncorrectable ECC error, 7-20 HSC50 main power supply, 3-74
VERIFY-W XBN n. has a hard EDC HSC auxiliary power supply, 3-78
error, 7-20 HSC main power supply, 3-72

Ek-Hscma-Sv-002 HSC Service Manual Dec89

Uploaded by

Copyright:

Available Formats

Ek-Hscma-Sv-002 HSC Service Manual Dec89

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ek-Hscma-Sv-002 HSC Service Manual Dec89

Uploaded by

Copyright:

Available Formats

HSC Service Manual

Order Number EK-HSCMA-SV-002

Digital Equipment Corporation

Copyright ©1989 by Digital Equipment Corporation

All Rights Reserved.

The following are trademarks of Digital Equipment Corporation:

This document was prepared using VAX DOCUMENT, Version 1.1

About This Manual xx:

2 Controls and Indicators

2.6 Module Indicators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-8

3 Removal and Replacement Procedures

5 Device Integrity Tests

5.2.8 Error Messages ............................................... . 5-4

5.7.5 Test Summaries .............................................. . 5-36

6.3.3 Loader Operating Instructions ................................... . 6-8

6.6.6 K.ci Path Status Information .................................... . 6-33

6.10.7 Error Information ............................................. . ~72

6.11 OOCP-Off-line Operator Control Panel (OCP) Test .................... . ~73

6.11.1 System Requirements .................... -...................... . 6-73

7.3.3 Error and information Messages. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-18

8.3.7 Boot Diagnostic Indications ...................................... . 8-26

A Internal Cabling Diagrams

B Exception Codes and Messages

C Generic Error Log Fields

D Interpretation of Status Code Bytes

E Revision Matrix Charts

D-1 K-detected Error Example. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D-1

3-10 HSC50 Card Cage Cover Removal ................................... 3-14

3-4 RX33 Jumper Description. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-42

Table 1-1 Differences Between HSC Models

I/O control processor LOllI L0105 LOI05 L0111-YA

Kit number HSC7X-AA1AB is available to upgrade the HSC40 to an HSC70.

Figure 1-1 Redundant Cluster Configuration

1.2 HSC Cabinet Layout

Figure 1-2 HSC Cabinet Front View

Figure 1-3 HSC Cabinet Inside Front View

Figure 1-4 HSC Module Utilization Label Example

Figure 1-5 HSC Cabinet Inside Rear View

1.3 HSC50 Cabinet Layout

Figure 1-6 HSC50 Cabinet Front View

Figure 1-7 HSC50 Cabinet Inside Front View

Figure 1-8 HSC50 Module Utilization Label Example

BACK DOOR LATCH

Figure 1-9 HSC50 Cabinet Rear View

Figure 1-10 HSC50 Cabinet Inside Rear View

CI BUS----ONE OR MORE HOST COMPUTERS

SOl B U S - - - - DISK DRIVES

STI BUS - - - - TAPE FORMATTER

HSC70 A S C I I - - - - - CONSOLE TERMINAL

ASCII - - - - - (NOT USED)

RX33 DISK DRIVE }

Figure 1-11 HSC External Interfaces

CI BUS----ONE OR MORE HOST COMPUTERS

SOl B U S - - - - DISK DRIVES

STI BUS - - - - TAPE FORMATTER

Figure 1-12 HSC50 External Interfaces

The external hardware interface lines perform the following flUlctions:

1.5 HSC Hardware Overview

ASCII PORT SERIAL