SMPE

IBM SMP/E for z/OS
User’s Guide
SA22-7773-11
IBM SMP/E for z/OS
User’s Guide
SA22-7773-11
Note!
Before using this information and the product it supports, be sure to read the general information under “Notices” on page
225.
Twelfth Edition, April 2007

This book replaces the previous edition, SA22-7773-10, which is now obsolete. Changes or additions to text and
illustrations are indicated by a vertical line to the left of the change.
This edition applies to IBM SMP/E for z/OS, V3R4, program number 5655-G44, and to all subsequent releases and
modifications, unless otherwise indicated in new editions.
Changes or additions to text and illustrations are indicated by a vertical line to the left of the change.
Order IBM publications through your IBM representative or the IBM branch office serving your locality.
Publications are not stocked at the address given below.
IBM welcomes your comments. A form for readers’ comments may be provided at the back of this publication, or
you may address your comments to the following address:
International Business Machines Corporation
Department 55JA, Mail Station P384
2455 South Road
Poughkeepsie, NY 12601-5400
United States of America
FAX (United States & Canada): 1+845+432+9405

FAX (Other Countries):
Your International Access Code +1+845+432-9405
IBMLink (United States customers only): IBMUSM10(MHVRCFS)

Internet e-mail: mhvrcfs@us.ibm.com
World Wide Web: http://www.ibm.com/servers/eserver/zseries/zos/webqs.html
If you would like a reply, be sure to include your name, address, telephone number, or FAX number.
Make sure to include the following in your comment or note:
v Title and order number of this book
v Page number or topic related to your comment
When you send information to IBM, you grant IBM a nonexclusive right to use or distribute the information in any
way it believes appropriate without incurring any obligation to you.
© Copyright International Business Machines Corporation 1986, 2007. All rights reserved.
US Government Users Restricted Rights – Use, duplication or disclosure restricted by GSA ADP Schedule Contract
with IBM Corp.
Contents
Figures . . . . . . . . . . . . . . vii SMP/E CSI application programming interface 35
Summary . . . . . . . . . . . . . . 35
Tables . . . . . . . . . . . . . . . ix
Chapter 2. SMP/E concepts . . . . . . 37
About this book . . . . . . . . . . . xi What is SMP/E? . . . . . . . . . . . . . 37
What are SYSMODs? . . . . . . . . . . . 37
Who should read this publication . . . . . . . xi
Data sets used by SMP/E . . . . . . . . . 39
Bibliography . . . . . . . . . . . . . . xi
How SMP/E can help you install and maintain
products . . . . . . . . . . . . . . . 41
Summary of changes . . . . . . . . xiii Where to begin . . . . . . . . . . . . 41
Installing SYSMODs . . . . . . . . . . 42
Chapter 1. SMP/E primer . . . . . . . 1 Monitoring your system . . . . . . . . . 44
What is SMP/E, and why should I use it? . . . . 1 Managing the SMP/E database . . . . . . . 44
Understanding your system . . . . . . . . 1 Managing zones . . . . . . . . . . . . 46
Changing the elements of the system . . . . . 2 Linking and relinking modules . . . . . . . 47
Keeping track of the elements of the system . . . 7 General SMP/E processing . . . . . . . . 47
How does SMP/E work? . . . . . . . . . . 9
The distribution and target libraries . . . . . 10 Chapter 3. Preparing to use SMP/E . . 49
The consolidated software inventory (CSI) . . . 11 Allocating and initializing data sets in the SMP/E
What are the basic SMP/E commands I need to database . . . . . . . . . . . . . . . 49
know? . . . . . . . . . . . . . . . . 12 CSI data sets . . . . . . . . . . . . . 49
Setting the zone you want to work on . . . . 13 PTS data sets . . . . . . . . . . . . . 62
Receiving the SYSMOD into SMP/E’s data sets 13 SCDS data sets . . . . . . . . . . . . 62
Applying the SYSMOD to the target libraries . . 13 How to dynamically allocate data sets to be used
Restoring the target libraries to the previous level 13 during SMP/E processing . . . . . . . . . 62
Accepting the SYSMOD and updating the Sources of information for dynamic allocation . . 63
distribution libraries . . . . . . . . . . 13 How dynamic allocation works . . . . . . . 64
Displaying SMP/E data . . . . . . . . . 13 Defining utility programs and associated parameters
Flow of SMP/E SYSMOD processing . . . . . 14 to SMP/E . . . . . . . . . . . . . . . 64
Receiving the SYSMOD into SMP/E’s data sets . . 14 Using default values for utility programs . . . 65
What happens during RECEIVE processing . . . 15 Defining values for utility programs . . . . . 66
What happens during internet service retrieval 15 Example: How to request the desired utility
How SMP/E keeps track of RECEIVE processing 16 processing . . . . . . . . . . . . . . 68
Using the RECEIVE command . . . . . . . 17 Recovering after errors from utility processing . . . 69
Summary of the RECEIVE command . . . . . 19 Overview of your input to retry processing . . . 69
Applying the SYSMOD to the target libraries . . . 19 Example: How to request the desired retry
What happens during APPLY processing . . . 19 processing . . . . . . . . . . . . . . 70
How SMP/E keeps track of APPLY processing . 20 Connecting SMP/E dialogs to ISPF . . . . . . 71
Using the APPLY command . . . . . . . . 21 Check for required programs . . . . . . . 71
Summary . . . . . . . . . . . . . . 23 Add dialog modules to the PCF command table 71
Restoring the target libraries to the previous level 23 Concatenate the dialog libraries . . . . . . 72
What happens during RESTORE processing . . 24 Connect the dialogs to ISPF . . . . . . . . 74
How SMP/E keeps track of RESTORE processing 24 Customize the SMP/E dialogs . . . . . . . 75
Using the RESTORE command . . . . . . . 25 Setting up SMP/E for easier operation . . . . . 76
Summary . . . . . . . . . . . . . . 26 Recommended values for OPTIONS entry . . . 76
Accepting the SYSMOD into the distribution Recommended link edit utility output DDDEF
libraries . . . . . . . . . . . . . . . 27 entries . . . . . . . . . . . . . . . 77
What happens during ACCEPT processing . . . 27 Specifying automatic cross-zone requisite
How SMP/E keeps track of ACCEPT processing 28 checking . . . . . . . . . . . . . . 78
Using the ACCEPT command . . . . . . . 28 Defining the information needed to invoke SMP/E 81
Summary . . . . . . . . . . . . . . 30 Required JCL statements . . . . . . . . . 81
Displaying SMP/E data . . . . . . . . . . 31 Sample cataloged procedure for SMP/E . . . . 82
Using the query dialogs . . . . . . . . . 31 Defining exit routines . . . . . . . . . . . 86
Using the LIST command . . . . . . . . . 33
Using the REPORT commands . . . . . . . 34
© Copyright IBM Corp. 1986, 2007 iii

Chapter 4. Preparing to use Internet Chapter 7. Installing corrective service 123
service retrieval . . . . . . . . . . . 87 Introduction . . . . . . . . . . . . . . 123
Identity and authentication overview . . . . . . 87 Building or checking the fix . . . . . . . . 124
Obtaining a user certificate . . . . . . . . . 88 Preparing your system . . . . . . . . . . 125
Uploading the user certificate to z/OS . . . . 88 Staging the SYSMODs: the RECEIVE process . . . 125
Setting up z/OS security server RACF . . . . . 89 Generating a service request using the RECEIVE
Access to the RACDCERT command . . . . . 89 ORDER Command . . . . . . . . . . . 126
Creating keyrings . . . . . . . . . . . 90 Updating the target libraries: the APPLY process 126
Enabling certificate authority certificates . . . . 90 Checking the update (APPLY CHECK) . . . . 127
Adding the user certificate to your RACF data Updating the target library (APPLY) . . . . . 127
base . . . . . . . . . . . . . . . . 91 Testing the corrective service . . . . . . . . 128
Sharing a user certificate among multiple userids 91 Updating the distribution libraries: the ACCEPT
| Debugging keyring and certificate issues . . . 92 process . . . . . . . . . . . . . . . 128
Replacing a user certificate that has expired . . 92 Checking the update (ACCEPT CHECK) . . . 128
Refreshing RACF classes . . . . . . . . . 93 Updating the distribution library (ACCEPT) . . 129
Setting up alternate security products . . . . 93
Define the ORDERSERVER input for RECEIVE Chapter 8. Installing a user
ORDER . . . . . . . . . . . . . . . . 93 modification . . . . . . . . . . . . 131
Define the CLIENT Input for RECEIVE ORDER . . 94 Introduction . . . . . . . . . . . . . . 131
Options that affect Java . . . . . . . . . 94 Preparing your system . . . . . . . . . . 131
Options that affect HTTPS operations. . . . . 95 Staging the SYSMODs: The RECEIVE process . . 131
Options that affect FTP operations . . . . . . 96 Updating the target libraries: The APPLY process 132
Network configuration notes . . . . . . . . 97 Checking the update (APPLY CHECK) . . . . 132
Summary . . . . . . . . . . . . . . . 98 Updating the target library (APPLY) . . . . . 133
Example . . . . . . . . . . . . . . . 99 Testing the USERMOD . . . . . . . . . . 133
Updating the distribution libraries: The ACCEPT
Chapter 5. Installing a new function 101 process . . . . . . . . . . . . . . . 133
Introduction . . . . . . . . . . . . . . 101
RECEIVE-APPLY-ACCEPT method . . . . . . 101 Chapter 9. Managing exception
Using the standard RECEIVE-APPLY-ACCEPT SYSMODs . . . . . . . . . . . . . 135
method . . . . . . . . . . . . . . . 102
Introduction . . . . . . . . . . . . . . 135
Preparing your system . . . . . . . . . 102
What SMP/E does with the HOLDDATA . . . . 136
Staging the SYSMODs: The RECEIVE process 103
Initial entry into staging data sets: RECEIVE 136
Updating the target libraries: The APPLY
Updating target libraries: APPLY . . . . . . 137
process . . . . . . . . . . . . . . 104
Updating distribution libraries: ACCEPT . . . 138
Testing the new function . . . . . . . . 106
Removing HOLDDATA from SMP/E data sets 138
Updating the distribution libraries: The
Sources of HOLDDATA . . . . . . . . . . 139
ACCEPT process . . . . . . . . . . . 107
CBPDO tapes . . . . . . . . . . . . 139
Checking other zones for requisites: REPORT
ESO tapes . . . . . . . . . . . . . 139
CROSSZONE . . . . . . . . . . . . 108
PSP information . . . . . . . . . . . 140
Automated service delivery package . . . . 140
Chapter 6. Installing preventive How to process HOLDDATA . . . . . . . 140
service . . . . . . . . . . . . . . 109
Introduction . . . . . . . . . . . . . . 109 Chapter 10. Creating cross-product,
CBPDO tapes . . . . . . . . . . . . 109 cross-zone load modules: The LINK
ESO tapes . . . . . . . . . . . . . 110
A RECEIVE ORDER request . . . . . . . 111
MODULE command . . . . . . . . . 147
Preventive service process: Summary . . . . . 111 When to use LINK MODULE . . . . . . . . 147
Preparing your system . . . . . . . . . . 112 How to use LINK MODULE . . . . . . . . 148
Staging the SYSMODs: The RECEIVE process . . 112
Updating the target libraries: The APPLY process 113 Chapter 11. Displaying the data
Checking the update (APPLY CHECK) . . . . 114 managed by SMP/E: The LIST
Updating the target library (APPLY) . . . . . 117 command . . . . . . . . . . . . . 149
Installing PTFs that need special processing . . 118 Introduction . . . . . . . . . . . . . . 149
Testing the new service level . . . . . . . . 118 Listing all the SMP/E data . . . . . . . . . 149
Updating the distribution libraries: The ACCEPT Listing by specific entry type . . . . . . . . 150
process . . . . . . . . . . . . . . . 119 Listing specific entries . . . . . . . . . . 151
Checking the update (ACCEPT CHECK) . . . 119 Listing by FMID or FMIDSET . . . . . . . . 152
Updating the distribution library (ACCEPT) . . 120 Listing to compare two zones . . . . . . . . 153
Installing PTFs that need special processing . . 121 Summary . . . . . . . . . . . . . . . 154
iv SMP/E V3R4.0 User’s Guide

Chapter 12. Changing the data SMP/E Example 7: Adding new source code that uses
manages: The UCLIN command . . . 155 an IBM-supplied macro . . . . . . . . . 179
Introduction . . . . . . . . . . . . . . 155 Example 8: Adding a new module that uses an
When to use UCLIN . . . . . . . . . . . 155 IBM-Supplied macro . . . . . . . . . . 180
How to use UCLIN . . . . . . . . . . . 156
Chapter 18. Determining which
Chapter 13. Identifying cross-zone SYSMODs led others to fail: The
requisites: The REPORT CROSSZONE causer SYSMOD summary report . . . 183
command . . . . . . . . . . . . . 159 Introduction . . . . . . . . . . . . . . 183
Introduction . . . . . . . . . . . . . . 159 Using causer SYSMOD information . . . . . . 183
Defining a ZONESET . . . . . . . . . . . 159 Resolving errors for all SYSMODs that failed 183
Running the REPORT CROSSZONE command . . 160 Resolving errors for a single SYSMOD that
Installing the SYSMODs . . . . . . . . . . 161 failed . . . . . . . . . . . . . . . 184
Example . . . . . . . . . . . . . . 184
Chapter 14. Identifying installed
Chapter 19. Java archive update
SYSMODs affected by error holds:
exploiter’s guide . . . . . . . . . . 187
The REPORT ERRSYSMODS
JAR replacements in FMIDs . . . . . . . . 187
command . . . . . . . . . . . . . 163 JAR updates in PTFs . . . . . . . . . . . 187
Introduction . . . . . . . . . . . . . . 163 JAR replacements in PTFs . . . . . . . . . 188
Running the REPORT ERRSYSMODS command 163
Installing the SYSMODs . . . . . . . . . . 164
Appendix A. Migration . . . . . . . 191
Migration overview . . . . . . . . . . . 191
Chapter 15. Listing the source IDs in a Terms you need to know . . . . . . . . 191
zone: The REPORT SOURCEID SMP/E release levels . . . . . . . . . . 192
command . . . . . . . . . . . . . 165 Developing a migration strategy . . . . . . 192
Introduction . . . . . . . . . . . . . . 165 SMP/E V3R4 overview . . . . . . . . . . 193
Running the REPORT SOURCEID command . . . 165 Enhancement to the RECEIVE command . . . 194
Listing the SYSMODs . . . . . . . . . . 165 Impacts to SMP/E zone entries . . . . . . 194
ICSF not required for GIMZIP and RECEIVE
Chapter 16. Comparing the SYSMODs FROMNETWORK . . . . . . . . . . . 195
Improved load module build processing . . . 195
installed in two zones: The REPORT
SMP/E administration dialog . . . . . . . 195
SYSMODS command . . . . . . . . 167 SMP/E query dialog . . . . . . . . . . 195
Introduction . . . . . . . . . . . . . . 167 SMP/E V3R3 overview . . . . . . . . . . 196
Running the REPORT SYSMODS command . . . 167 GIMGTPKG service routine . . . . . . . 196
Installing the SYSMODs . . . . . . . . . . 167 Enhancements to GIMZIP and GIMUNZIP
service routines . . . . . . . . . . . . 196
Chapter 17. Building a user RECEIVE FROMNETWORK FTP interface
modification . . . . . . . . . . . . 169 enhancements . . . . . . . . . . . . 196
Choosing between a USERMOD and a function REJECT CHECK command . . . . . . . . 197
SYSMOD . . . . . . . . . . . . . . . 169 Extended RECEIVE SOURCEID processing . . 197
Creating the MCSs . . . . . . . . . . . 170 SPCLCMOD and CMWA . . . . . . . . 197
The ++USERMOD MCS . . . . . . . . . 170 SMP/E V3R2 overview . . . . . . . . . . 197
The ++VER MCS . . . . . . . . . . . 170 LINK LMODS command . . . . . . . . 197
The ++JCLIN MCS . . . . . . . . . . 172 REPORT CALLLIBS command removal . . . 198
++MOD and ++ZAP MCSs . . . . . . . . 172 UPGRADE command . . . . . . . . . 198
++MAC and ++MACUPD MCSs . . . . . . 172 GIMXSID service routine . . . . . . . . 198
++SRC and ++SRCUPD MCSs . . . . . . . 173 GIMZIP: Archive segmentation . . . . . . 198
The ++PROGRAM MCS . . . . . . . . . 173 GIMZIP: User defined subdirectories . . . . 199
Data element MCSs . . . . . . . . . . 173 Java archive files . . . . . . . . . . . 199
Hierarchical file system element MCSs . . . . 174 Smaller SMPLTS data set . . . . . . . . 199
Examples of USERMODs . . . . . . . . . 174 DUMMY data set for SYSDEFSD . . . . . . 200
Example 1: Updating a module . . . . . . 174 SMP/E dialog customization . . . . . . . 201
Example 2: Replacing a module . . . . . . 175 GIMUTTBL removal . . . . . . . . . . 201
Example 3: Adding new modules. . . . . . 175 SMP/E V3R1 overview . . . . . . . . . . 201
Example 4: Replacing a macro or source code 176 Defining exit routines using SMPPARM member
Example 5: Updating a macro or source code 177 GIMEXITS . . . . . . . . . . . . . 202
Example 6: Adding new source code . . . . 177 Dynamic allocation using SMPPARM member
GIMDDALC . . . . . . . . . . . . . 202
Contents v
Enhanced link name values. . . . . . . . 202 Enhanced internal HOLD SYS processing . . . 210
Removal of function to create backup Enhanced ZONEEDIT command . . . . . . 211
IEANUC01 load modules . . . . . . . . 203 Enhancements to the binder utility in
Conditional JCLIN processing . . . . . . . 203 DFSMS/MVS . . . . . . . . . . . . 211
Network delivery of SMP/E input . . . . . 203 System/390 service update facility . . . . . 212
AMODE=64 and COMPAT=PM4 link edit OS/390 version 1 release 2 SMP/E overview . . . 212
parameters . . . . . . . . . . . . . 204 BLOCKSIZE=8800 for SMP/E data sets . . . . 212
Selected SMP/E data sets may now reside in a BUILDMCS command . . . . . . . . . 212
UNIX file system . . . . . . . . . . . 204 Bypassing system holds for specific SYSMODs 212
HFS data set identification . . . . . . . . 204 FMIDSET selection . . . . . . . . . . 212
SMPPTS spill data sets . . . . . . . . . 204 Receiving relative file data sets created from
HOLDDATA summary reports . . . . . . 205 PDSEs . . . . . . . . . . . . . . . 213
SMP/E load modules and service routines SMP/E dialogs: FIND command . . . . . . 213
moved to SYS1.MIGLIB . . . . . . . . . 205 SMP/E GIMOPCDE member moved from
GIMXTRX service routine . . . . . . . . 205 PARMLIB . . . . . . . . . . . . . 213
OS/390 version 2 release 7 SMP/E overview . . . 205 Summary of interface changes . . . . . . . . 213
SMP/E planning and migration assistant . . . 205 Commands . . . . . . . . . . . . . 213
Data element reformatting . . . . . . . . 205 Data sets and files . . . . . . . . . . . 214
Description for a SYSMOD . . . . . . . . 206 Exits . . . . . . . . . . . . . . . 215
Improved protection for UNIX file system files 206 Macros . . . . . . . . . . . . . . 215
Pre-built load module support . . . . . . . 206 Messages . . . . . . . . . . . . . . 215
Product data . . . . . . . . . . . . 206 Panels . . . . . . . . . . . . . . . 215
Sequential data set support . . . . . . . . 206 Programming interfaces . . . . . . . . . 217
Shell script support . . . . . . . . . . 206 Service routines . . . . . . . . . . . 217
Symbolic link support . . . . . . . . . 207 SMPPARM members . . . . . . . . . . 218
OS/390 version 2 release 5 SMP/E overview . . . 207 SYS1.SAMPLIB members . . . . . . . . 218
CBIPO dialogs . . . . . . . . . . . . 207
Client code installation . . . . . . . . . 207 Appendix B. Recommended service
Global zone merge . . . . . . . . . . 207 upgrade (RSU) . . . . . . . . . . . 221
Library change interface . . . . . . . . . 207
Improved load module build processing . . . 208
Load module return code . . . . . . . . 208 Appendix C. Accessibility . . . . . . 223
Performance improvements. . . . . . . . 208 Using assistive technologies . . . . . . . . 223
PTF compaction in SMPPTS data set . . . . 208 Keyboard navigation of the user interface . . . . 223
Enhanced RECEIVE command processing . . . 208 z/OS information . . . . . . . . . . . . 223
Reduced SMP/E message output . . . . . . 209
GIMAPI: All entries and subentries support . . 209 Notices . . . . . . . . . . . . . . 225
GIMAPI: Version support . . . . . . . . 209 Trademarks . . . . . . . . . . . . . . 226
OS/390 version 1 release 3 SMP/E overview . . . 209
API for user access to the CSI . . . . . . . 209 Glossary . . . . . . . . . . . . . 229
Enhanced cross-zone requisite checking . . . 210
Enhanced exception SYSMOD report . . . . 210
Index . . . . . . . . . . . . . . . 243
Enhanced ++IF FMID processing . . . . . . 210
vi SMP/E V3R4.0 User’s Guide

Figures
1. Load module creation . . . . . . . . . 2 24. Sample multiple-CSI structure . . . . . . 51
2. Introducing an element . . . . . . . . . 4 25. Using a separate global zone for each
3. Preventing problems with an element . . . . 5 subsystem . . . . . . . . . . . . . 52
4. Fixing problems with an element . . . . . . 6 26. Using one CSI for the whole system . . . . 53
5. Customizing an element . . . . . . . . 7 27. Using a master CSI . . . . . . . . . . 54
6. PTF replacement . . . . . . . . . . . 8 28. Using a master CSI and a separate CSI for
7. PTF prerequisite . . . . . . . . . . . 8 each zone . . . . . . . . . . . . . 55
8. Load module constructions . . . . . . . . 9 29. Using a master CSI and one CSI per SREL 56
9. The Public Library . . . . . . . . . . 10 30. Relationships between zone definition entries 59
10. The distribution and target libraries . . . . 11 31. Relationships of OPTIONS, UTILITY, zone
11. z/OS system with SMP/E . . . . . . . 12 definition entries and the SET command . . . 67
12. Flow of SMP/E SYSMOD Processing . . . . 14 32. Sample logon procedure that concatenates
13. Results of RECEIVE processing . . . . . . 17 SMP/E and ISPF libraries . . . . . . . . 74
14. Results of APPLY Processing . . . . . . . 21 33. Sample SMP/E cataloged procedure . . . . 84
15. Results of RESTORE Processing . . . . . . 25 34. APPLY SYSLIB concatenation: APPLY different
16. Results of ACCEPT Processing . . . . . . 28 from ACCEPT. . . . . . . . . . . . 86
17. Query selection menu . . . . . . . . . 32 35. ACCEPT SYSLIB concatenation: APPLY
18. CSI query panel . . . . . . . . . . . 32 different from ACCEPT . . . . . . . . 86
19. CSI query - Select entry panel . . . . . . 33 36. SYSMOD Status Report: Sample Report for
20. CSI query - SYSMOD entry panel . . . . . 33 APPLY . . . . . . . . . . . . . . 184
21. Example of a SYSMOD hierarchy . . . . . 39 37. Causer SYSMOD summary report: sample
22. Summary of zone relationships . . . . . . 40 report for APPLY . . . . . . . . . . 185
23. Sample single-CSI structure . . . . . . . 50
© Copyright IBM Corp. 1986, 2007 vii

viii SMP/E V3R4.0 User’s Guide
Tables
1. Publications for IBM SMP/E for z/OS, V3R4 xi 13. Alternatives to UCLIN . . . . . . . . 156
2. Entries controlling SMP/E processing . . . . 59 14. Comparison of USERMODs and function
3. Entries describing the status and structure of SYSMODs . . . . . . . . . . . . 169
the target and distribution libraries . . . . 60 15. Information needed to add new source code 177
4. Default values for UTILITY entries . . . . . 65 16. Summary of changed commands . . . . . 214
5. How to request the desired utility processing 68 17. Summary of changed data sets . . . . . . 214
6. How to request the desired retry processing 70 18. Summary of Changed Macros . . . . . . 215
7. ISPF libraries and related SMP/E target 19. Summary of new and changed panels 215
libraries . . . . . . . . . . . . . . 72 20. Summary of SMP/E programming interfaces 217
8. SMPTABL data set allocations . . . . . . 73 21. Summary of SMP/E service routines 217
9. Sources for functions and their installation 22. Summary of SMP/E changes to
information . . . . . . . . . . . . 101 SYS1.SAMPLIB . . . . . . . . . . . 218
10. Format of a CBPDO tape . . . . . . . . 110 23. Summary of SMP/E changes to
11. Format of an ESO . . . . . . . . . . 110 SYS1.SAMPLIB . . . . . . . . . . . 218
12. CBPDO/service level/PSP HOLDDATA
example . . . . . . . . . . . . . 144
© Copyright IBM Corp. 1986, 2007 ix

x SMP/E V3R4.0 User’s Guide
About this book
This publication documents a new and enhanced version of SMP/E. New or
changed information is identified by revision bars (|) to the left of the addition or
change.
Who should read this publication

Anyone who uses SMP/E, or who wants to understand SMP/E processes, should
read this publication.
After reading this publication, you should be able to do most SMP/E processes.
You may have to refer to SMP/E Commands for details on commands.
Bibliography
This section tells you more about the SMP/E library.
v The IBM SMP/E for z/OS, V3R4 publications are available as printable PDF files
and BookManager-viewable softcopy at
http://www.ibm.com/servers/eserver/zseries/zos/bkserv/
v Table 1 lists the IBM SMP/E for z/OS, V3R4 publications and briefly describes
each one.
v For information on z/OS publications and more information on the IBM SMP/E
for z/OS, V3R4 books, see z/OS Information Roadmap.
Table 1. Publications for IBM SMP/E for z/OS, V3R4
Title Description
SMP/E Messages, Codes, and Diagnosis, GA22-7770 Explains SMP/E messages and return codes and the
actions to take for each; and how to handle suspected
SMP/E problems.
SMP/E Commands, SA22-7771 Explains SMP/E commands and processing in detail.
SMP/E Reference, SA22-7772 Explains SMP/E modification control statements, data
sets, exit routines, and programming interfaces in detail
and provides additional SMP/E reference material.
SMP/E User’s Guide, SA22-7773 Describes how to use SMP/E to install programs and
service.
© Copyright IBM Corp. 1986, 2007 xi

Bibliography
xii SMP/E V3R4.0 User’s Guide

Summary of changes
Summary of changes
for SA22-7773-11 April, 2007
SMP/E Version 3 Release 4
This revision reflects the deletion, addition, or modification of information to

support miscellaneous maintenance items. A vertical bar ( | ) in the left margin
indicates changes to the text and illustrations.
New information
In Chapter 4, “Preparing to use Internet service retrieval,” on page 87, a new topic
eTrust CA-Top Secret Security for z/OS users has been added.
Debugging keyring and certificate issues has been added.
Summary of changes
for SA22-7773-10 September, 2006

Changed information
Replacing a user certificate that has expired has been added.
Section ″ORDER RETENTION Subentry on the OPTIONS Entry″ in Appendix A,

“Migration,” on page 191 has been updated.
Summary of changes
for SA22-7773-09 May, 2006

Changed information
Chapter 4, ″Preparing to use internet service retrieval″, has been rewritten to

support APAR IO03469. New information has been added and some topics have
been moved within the chapter.
You might notice changes in the style and structure of some content in this
document—for example, headings that use uppercase for the first letter of initial
words only, and procedures that have a different look and format. The changes are
ongoing improvements to the consistency and retrievability of information in our
documents.
© Copyright IBM Corp. 1986, 2007 xiii

Summary of changes
for SA22-7773-08 April, 2006

Changed information
v The last portion of the topic “Options that affect Java” on page 94 has been
updated.
Summary of changes
for SA22-7773-07 September 15, 2005

New Information
v “Refreshing RACF classes” on page 93 has been added.
Changed information
v The notes of “Setting up z/OS security server RACF” on page 89 has been
updated.
Summary of changes
for SA22-7773-07

Changed information
v Added javahome attribute information to “Options that affect Java” on page 94.
v Updated the summary and example at the end of Chapter 4, “Preparing to use
Internet service retrieval,” on page 87.
Summary of changes
for SA22-7773-06

New information
v “What happens during internet service retrieval” on page 15 has been added.
v “Requesting a new PTF order with the RECEIVE ORDER command” on page 18
has been added.
v “Using internet service retrieval to request PTF or HOLDDATA: RECEIVE
ORDER” on page 42 has been added.
xiv SMP/E V3R4.0 User’s Guide

v Chapter 4, “Preparing to use Internet service retrieval,” on page 87 has been
added.
Changed information
v Chapter 6, “Installing preventive service,” on page 109 has been updated.
v Chapter 7, “Installing corrective service,” on page 123 has been updated.
v Chapter 9, “Managing exception SYSMODs,” on page 135 has been updated.
v Chapter 11, “Displaying the data managed by SMP/E: The LIST command,” on
page 149 has been updated.
v Chapter 12, “Changing the data SMP/E manages: The UCLIN command,” on
page 155 has been updated.
Moved information
v None.
Deleted information
v None
Summary of changes
for SA22-7773-04

New information
v Figure 18 on page 32 has been updated to reflect the new entry name wildcard
capability of the CSI Query dialog panel (GIMQU1PO).
v “SMP/E V3R3 overview” on page 196 has been added to ″Appendix A.
Migration″. It includes migration information on:
– “SPCLCMOD and CMWA” on page 197
– “Extended RECEIVE SOURCEID processing” on page 197
– “GIMGTPKG service routine” on page 196
– “Enhancements to GIMZIP and GIMUNZIP service routines” on page 196
– “RECEIVE FROMNETWORK FTP interface enhancements” on page 196
– “REJECT CHECK command” on page 197
Changed information
v None.
Moved information
v None.
Deleted information
v None.
Summary of changes
for SA22-7773-03
Summary of changes xv
New information
v “SMP/E V3R2 overview” on page 197 has been added to ″Appendix A.
Migration″. It includes migration information on:
– “LINK LMODS command” on page 197
– “REPORT CALLLIBS command removal” on page 198
– “UPGRADE command” on page 198
– “GIMXSID service routine” on page 198
– “GIMZIP: Archive segmentation” on page 198
– “Java archive files” on page 199
– “Smaller SMPLTS data set” on page 199
– “DUMMY data set for SYSDEFSD” on page 200
– “GIMZIP: User defined subdirectories” on page 199
– “SMP/E dialog customization” on page 201
– “GIMUTTBL removal” on page 201.
v Chapter 19, “Java archive update exploiter’s guide,” on page 187 has been
added.
v “Relinking load modules that use CALLLIBS: LINK LMODS” on page 47 has
been added.
Changed information
v Member GIMSAMPU in SYS1.SAMPLIB has been updated to provide sample job
steps to allocate SMPCSI data sets and SMP/E operational data sets (such as
SMPPTS and SMPLOG) and UCLIN statements to initialize the newly allocated
SMPCSI data sets, as shown in “Allocating a CSI data set” on page 56 and
Table 23 on page 218.
v “Handling cross-zone link-edits: LINK MODULE” on page 47 has been updated.
v “How dynamic allocation works” on page 64 has been updated.
v “Customize the SMP/E dialogs” on page 75 has been updated.
v “Updating target libraries: APPLY” on page 137 has been updated.
v “How to use LINK MODULE” on page 148 has been updated.
v “SMP/E load modules and service routines moved to SYS1.MIGLIB” on page
205 has been updated.
v “GIMXTRX service routine” on page 205 has been updated.
v Appendix B, “Recommended service upgrade (RSU),” on page 221 has been
updated.
Moved information
v None.
Deleted information
v The chapter on using the REPORT CALLLIBS command has been deleted,
because that command is no longer supported.
Summary of changes
for SA22-7773-02
as Updated, March 2002
xvi SMP/E V3R4.0 User’s Guide

New information
v An appendix with z/OS product accessibility information has been added.
Changed information
v Table 23 on page 218 has been updated to include GIMCRSAM, GIMPRSAM,
and GIMSAMPU.
Moved information
v None.
Deleted information
v None.
Summary of changes
for SA22-7773-01
SMP/E Version 3
The book contains information previously presented in SA22-7773-00, which

applied to z/OS Version 1 Release 1.
New information
v Appendix A, “Migration,” on page 191 has been added.
v SMPPARM member GIMDDALC in “How to dynamically allocate data sets to
be used during SMP/E processing” on page 62 has been added.
Changed information
v “Example 1: Updating a module” on page 174 has been updated.
Moved information
v None.
Deleted information
v Information related to backup IEANUC01 load modules has been removed.
This book contains terminology, maintenance, and editorial changes. Technical

changes or additions to the text and illustrations are indicated by a vertical line to
the left of the change. You may notice changes in the style and structure of some
content in this book — for example, headings that use uppercase for the first letter
of initial words only, and procedures that have a different look and format. The
changes are ongoing improvements to the consistency and retrievability of
information in our books.
Summary of changes xvii

xviii SMP/E V3R4.0 User’s Guide
Chapter 1. SMP/E primer
This chapter provides an introduction to SMP/E to new SMP/E users. If you are
already familiar with SMP/E, you can skip this chapter.
What is SMP/E, and why should I use it?

SMP/E is a tool designed to manage the installation of software products on your
z/OS system and to track the modifications you make to those products. Usually,
it is the system programmer’s responsibility to ensure that all software products
and their modifications are properly installed on the system. The system
programmer also has to ensure that all products are installed at the proper level so
all elements of the system can work together. At first, that might not sound too
difficult, but as the complexity of the software configuration increases, so does the
task of monitoring all the elements of the system. To better understand this, let’s
take a closer look at your z/OS system and see how SMP/E can help you maintain
it.
Understanding your system

Your z/OS system may appear to be one big block of code that drives your CPU.
Actually, z/OS is a complex system comprising many different smaller blocks of
code. Each of those smaller blocks of code perform a specific function within the
system.
For example, some of the functions that can appear in a Z/OS system include:
v Base Control Program (BCP)
v C/C++ IBM Open Class Library
v Communications Server (CS z/OS)
v Cryptographic Services
v DCE Application Support
v DCE Base Services
v DFSMSdfp
v DFSORT
v Distributed File Service
v Encina Toolkit Executive
v Hardware Configuration Definition (HCD)
v High Level Assembler (HLASM)
v IBM HTTP Server
v IBM License Manager (ILM)
v Infoprint Server
v ISPF
v JES2 or JES3
v Language Environment
v Managed System Infrastructure (msys) for Setup
v Network File System
v Open Systems Adapter/Support Facility (OSA/SF)
v Resource Measurement Facility (RMF)
© Copyright IBM Corp. 1986, 2007 1

Introduction
v System Display and Search Facility (SDSF)

v SMP/E
v Time Sharing Option/Extensions (TSO/E)
v z/OS UNIX® System Services (z/OS UNIX)
Each system function is composed of one or more load modules. In a z/OS

environment, a load module represents the basic unit of machine-readable,
executable code. Load modules are created by combining one or more object
modules and processing them with a link-edit utility. The link-editing of modules
is a process that resolves external references and addresses. The functions on your
system, therefore, are one or more object modules that have been combined and
link-edited.
To see where the object modules come from, let’s take a look at the example in
Figure 1.
Figure 1. Load module creation
Most of the time, object modules are sent to you as part of a product. In this
example, the object module MOD1 was sent as part of the product. Other times,
you may need to assemble source code sent to you by product packagers to create
the object module. You can modify the source code and then assemble it to
produce an object module. In the example, SRCMOD2 is source code that you
assemble to create object module MOD2. When assembled, you link-edit object
module MOD2 with object module MOD1 to form the load module LMOD1.
In addition to object modules and source code, most products distribute many
additional parts, such as macros, help panels, dialog elements, and other z/OS
library members. These modules, macros, and other types of data and code are the
basic building blocks of your system. All of these building blocks are called
elements.
Changing the elements of the system

Over time, you may need to change some of the elements of your system. These
changes may be necessary to improve the usability or reliability of a product. You
may want to add some new functions to your system, upgrade some of the
elements of your system, or modify some elements for a variety of reasons. In all
cases, you are making system modifications. In SMP/E, we refer to these system
modifications as SYSMODs.
2 SMP/E V3R4.0 User’s Guide

Introduction
A SYSMOD is the actual package containing information SMP/E needs to install

and track system modifications. SYSMODs are composed of two parts:
v Modification control statements (MCS), designated by ++ as the first two
characters, that tell SMP/E:
– What elements are being updated or replaced
– How the SYSMOD relates to product software and other SYSMODs
– Other specific installation information
v Modification text, which is the object modules, macros, and other elements
supplied by the SYSMOD
There are four different categories of SYSMODs, each supporting a task you might
want to perform:
Function SYSMODs Introduce the elements for a product.
PTF (program temporary fix) SYSMODs
Prevent or fix problems with an element, or
introduce new element s.
APAR (authorized program analysis reports) SYSMODs
Fix problems with an element.
USERMOD (user modifications) SYSMODs
Customize an element.
Introducing an element—the function SYSMOD

One way you can modify your system is to introduce new elements into that
system. To accomplish this with SMP/E, you can install a function SYSMOD. The
function SYSMOD introduces a new product, a new version or release of a
product, or updated functions for an existing product into the system. All other
types of SYSMODs are dependent upon the function SYSMOD, because they are all
modifications of the elements originally introduced by the function SYSMOD.
When we refer to installing a function SYSMOD, we are referring to the placing of

all the product’s elements in the system data sets, or libraries. Examples of these
libraries are SYS1.LPALIB, SYS1.MIGLIB, and SYS1.SVCLIB. Figure 2 depicts the
process of creating executable code in the production system libraries.
Chapter 1. SMP/E primer 3

Introduction
Figure 2. Introducing an element
In this figure, the installation of a function SYSMOD link-edits object modules

MOD1, MOD2, MOD3, and MOD4 to create load module LMOD2. The executable
code created in load module LMOD2 is installed in the system libraries through
the installation of the function SYSMOD.
There are two types of function SYSMODs:

v A base function SYSMOD adds or replaces an entire system function. Examples
of base functions are SMP/E and JES2.
v A dependent function SYSMOD provides an addition to an existing system
function. It is called dependent because its installation depends upon a base
function already being installed. Examples of dependent functions are the
language features for SMP/E.
Both base function SYSMODs and dependent function SYSMODs are used to
introduce new elements into the system.
Here’s an example of a simple function SYSMOD that introduces four elements:

++FUNCTION(FUN0001) /* SYSMOD type and identifier. */.
++VER(Z038) /* For MVS SREL */.
++MOD(MOD1) RELFILE(1) /* Introduce this module */
DISTLIB(AOSFB) /* in this distribution library. */.
Preventing or fixing problems with an element—the PTF

SYSMOD
When a problem with a software element is discovered, IBM supplies its customers
with a tested fix for that problem. This fix comes in the form of a program
temporary fix (PTF). Although you may not have experienced the problem the PTF

Introduction
is intended to prevent, it is wise to install the PTF on your system. The PTF
SYSMOD is used to install the PTF, thereby preventing the occurrence of that
problem on your system.
Usually, PTFs are designed to replace or update one or more complete elements of
a system function. Let’s look at Figure 3.
Figure 3. Preventing problems with an element
In Figure 3, we see a previously installed load module, LMOD2. If we want to

replace the element MOD1, we should install a PTF SYSMOD that contains the
module MOD1. That PTF SYSMOD replaces the element in error with the corrected
element. As part of the installation of the PTF SYSMOD, SMP/E relinks LMOD2 to
include the new and corrected version of MOD1.
Here is an example of a simple PTF SYSMOD:

++PTF(PTF0001) /* SYSMOD type and identifier. */.
++VER(Z038) FMID(FUN0001) /* Apply to this product. */.
++MOD(MOD1) /* Replace this module */
...
... object code for module
...
PTF SYSMODs are always dependent upon the installation of a function SYSMOD.
In some cases, some PTF SYSMODs may also be dependent upon the installation
of other PTF SYSMODs. These dependencies are called prerequisites. We will look at
a typical PTF prerequisite when we discuss the complexity of keeping track of the
elements of the system.
Fixing problems with an element—the APAR SYSMOD

You may sometimes find it is necessary to correct a serious problem that occurs on
your system before a PTF is ready for distribution. In this situation, IBM supplies
you with an authorized program analysis report (APAR). An APAR is a fix
designed to quickly correct a specific area of an element or replace an element in
error. You install an APAR SYSMOD to implement a fix, thereby updating the
incorrect element.

Introduction
In Figure 4, the shaded section shows an area of MOD2 containing an error.
Figure 4. Fixing problems with an element
The processing of the APAR SYSMOD provides a modification for object module
MOD2. During the installation of the APAR SYSMOD, MOD2 is updated (and
corrected) in load module LMOD2.
Here is an example of a simple APAR SYSMOD:

++APAR(APAR001) /* SYSMOD type and identifier. */.
++VER(Z038) FMID(FUN0001) /* Apply to this product */
PRE(UZ00004) /* at this service level. */.
++ZAP(MOD2) /* Update this module */
...
... zap control statements
...
The APAR SYSMOD always has the installation of a function SYSMOD as a

prerequisite, and can also be dependent upon the installation of other PTF or
APAR SYSMODs.
Customizing an element—the USERMOD SYSMOD

If you had a requirement for a product to perform differently from the way it was
designed, you might want to customize that element of your system. IBM provides
you with certain modules that allow you to tailor IBM code to meet your specific
needs. After making the desired changes, you add these modules to your system
by installing a USERMOD SYSMOD. This SYSMOD can be used to replace or
update an element, or to introduce a totally new user-written element into the
system. In either case, the USERMOD SYSMOD is built by you either to change
IBM code or to add your own code to the system.
In Figure 5, MOD3 has been updated through the installation of a USERMOD

SYSMOD.

Introduction
Figure 5. Customizing an element
Here is an example of a simple USERMOD SYSMOD:

++USERMOD(USRMOD1) /* SYSMOD type and identifier. */.
++VER(Z038) FMID(FUN0001) /* Apply to this product */
PRE(UZ00004) /* at this service level. */.
++SRCUPD(JESMOD3) /* Update this source module */
...
... update control statements
...
Prerequisites for USERMOD SYSMODs are the installation of a function SYSMOD,

and possibly the installation of other PTF, APAR, or USERMOD SYSMODs.
SYSMOD prerequisites
As you have learned, PTF, APAR, and USERMOD SYSMODs all have the function
SYSMOD as a prerequisite. In addition to their dependencies on the function
SYSMOD:
v PTF SYSMODs may be dependent upon other PTF SYSMODs.
v APAR SYSMODs may be dependent upon PTF SYSMODs and other APAR
SYSMODs.
v USERMOD SYSMODs may be dependent upon PTF SYSMODs, APAR
SYSMODs, and other USERMOD SYSMODs.
Consider the complexity of these dependencies. When you multiply that

complexity by hundreds of load modules in dozens of libraries, the need for a tool
like SMP/E becomes apparent.
Let’s examine the impact of these dependencies on the maintenance of software in

a z/OS environment.
Keeping track of the elements of the system

The importance of keeping track of system elements and their modifications
becomes readily apparent when we examine the z/OS maintenance process. Often,
a PTF contains multiple element replacements. In the example in Figure 6, PTF1
contains replacements for two modules, MOD1 and MOD2. Although load module
LMOD2 contains four modules, only two of those modules are being replaced.

Introduction
Figure 6. PTF replacement
But what happens if a second PTF replaces some of the code in a module that was
replaced by PTF1? Let’s look at Figure 7.
Figure 7. PTF prerequisite
In this example, PTF2 contains replacements for MOD2 and MOD3. For MOD1,
MOD2, and MOD3 to interface successfully, PTF1 must be installed before PTF2.
That’s because MOD3 supplied in PTF2 may depend on the PTF1 version of
MOD1 to be present. It is this dependency that constitutes a prerequisite. SYSMOD
prerequisites are identified in the modification control statements (MCS) part of the
SYSMOD package we discussed in “Changing the elements of the system” on page
2.
In addition to tracking prerequisites, there is another important reason to track

system elements. The same module is often part of many different load modules.
Let’s take a look at the example in Figure 8 on page 9.

Introduction
Figure 8. Load module constructions
In Figure 8, the same MOD2 module is present in LMOD1, LMOD2, and LMOD3.
When a PTF is introduced that replaces the element MOD2, that module must be
replaced in all the load modules in which it exists. Therefore, it is imperative that
we keep track of all load modules and the modules they contain.
You can now appreciate how complicated the tracking of system elements and
their modification levels can become. Let’s take a brief look at how we implement
the tracking capabilities of SMP/E.
Tracking and controlling requisites

To track and control elements successfully, all elements and their modifications and
updates must be clearly identified to SMP/E. SMP/E relies on modification
identifiers to accomplish this. There are three modification identifiers associated
with each element:
v Function modification identifiers (FMIDs) that identify the function SYSMOD
that introduced the element into the system.
v Replacement modification identifiers (RMIDs) that identify the last SYSMOD
(usually a PTF SYSMOD) to replace the element.
v Update modification identifiers (UMIDs) that identify the SYSMODs that have
updated an element since it was last replaced.
SMP/E uses these modification identifiers to track all SYSMODs installed on your
system. This ensures that they are installed in the proper sequence. Now that we
realize the need for element tracking and know the types of things SMP/E tracks,
let’s look at how SMP/E performs its tracking function.
How does SMP/E work?

Let’s review our discussion of how functions are installed into the system. We
begin with elements, such as modules, macros, and source code. These elements
are then processed by utilities, such as an assembler or link-editor, to create load
modules. The load modules contain the machine-readable, executable code.

How SMP/E works
Your production system in a z/OS environment consists of the z/OS operating

system and all the code needed to do your everyday work. That’s fine, but where
is all that stuff kept, and how is it organized? Let’s find out.
The distribution and target libraries

To properly perform its processing, SMP/E must maintain a great deal of
information about the structure, content, and modification status of the software it
manages. Think of all the information SMP/E has to maintain as if it were all the
information contained in the public library. To follow this analogy, let’s refer to
Figure 9.
Figure 9. The Public Library
If you look at this figure depicting the public library, you see bookshelves filled
with books and a card catalog with drawers containing a card for each book in the
library. These cards contain information, such as the title, author, publishing dates,
type of book, and a pointer to the actual book on the shelf.
In the SMP/E environment, there are two distinct types of “bookshelves.” They are
referred to as the distribution libraries and the target libraries. Figure 10 on page 11
depicts these two types of SMP/E libraries.

How SMP/E works
Figure 10. The distribution and target libraries
In much the same way the bookshelves in the public library hold the library books,
the distribution and target libraries hold the elements of the system.
Distribution libraries contain all the elements, such as modules and macros, that
are used as input for running your system. One very important use of the
distribution libraries is for backup. Should a serious error occur with an element
on the production system, the element can be replaced by a stable level found in
the distribution libraries.
Target libraries contain all the executable code needed to run the system.
The consolidated software inventory (CSI)

As you refer to the analogy of the public library, you can see that there is one
important piece of Figure 9 that we have not yet considered. In the public library,
there is a card catalog to help you find the book or piece of information you are
looking for. SMP/E provides the same type of tracking mechanism in the form of
the consolidated software inventory (CSI).
The CSI data sets contain all the information SMP/E needs to track the distribution
and target libraries. As the card catalog contains a card for each book in the library,
the CSI contains an entry for each element in its libraries. The CSI entries contain
the element name, type, history, how the element was introduced into the system,
and a pointer to the element in the distribution and target libraries. The CSI does
not contain the element itself, but rather a description of the element it represents.
Let’s see exactly how these entries are arranged in the CSI.
The SMP/E zones

The cards in the public library card catalog are arranged alphabetically by the
author’s last name, and by the topic and title of the book. In the CSI, entries for
the elements in the distribution and target libraries are grouped according to their
installation status. That is, entries representing elements found in the distribution
libraries are contained in the distribution zone. Entries representing elements found
in the target libraries are contained in the target zone. Both of these zones serve the
same purpose as the drawers of the public library card catalog.
In addition to the distribution and target zones, the SMP/E CSI also contains a
global zone. The global zone contains:
v Entries needed to identify and describe each target and distribution zone to
SMP/E
v Information about SMP/E processing options
v Status information for all SYSMODs SMP/E has begun to process

How SMP/E works
v Exception data for SYSMODs requiring special handling or that are in error
In SMP/E, when we speak of exception data, we are usually referring to

HOLDDATA. HOLDDATA is often supplied for a product to indicate a specified
SYSMOD should be held from installation. Reasons for holding a SYSMOD can be:
v A PTF is in error and should not be installed until the error is corrected (ERROR
HOLD).
v Certain system actions may be required before SYSMOD installation (SYSTEM
HOLD).
v The user may want to perform some actions before installing the SYSMOD
(USER HOLD).
All the information found in the global zone, combined with the information found
in the distribution and target zones, represents the data SMP/E needs to install
and track your system software.
Remember the picture of the public library in Figure 9 on page 10? Now look at
Figure 11.
Figure 11. z/OS system with SMP/E
Now you can see how all the elements of the system fit together, and how they
can be installed, modified, and tracked using SMP/E.
What are the basic SMP/E commands I need to know?

Now that you are familiar with SMP/E and what it can do, you are probably
wondering what you need to know to get started using SMP/E. Let’s take a look
at the basic processing commands you need to know to use SMP/E.

Basic SMP/E commands
Setting the zone you want to work on

Before processing SMP/E commands, you must first set the zone on which you
want SMP/E to work (global, target, or distribution). You do this by issuing the
SET command. The SET command identifies the zone and, therefore, the libraries,
upon which subsequent SMP/E commands are to act.
The SET command can also be used to request a particular set of predefined
processing options. For more information about the SET command, refer to SMP/E
Commands.
Receiving the SYSMOD into SMP/E’s data sets

For SMP/E to install a SYSMOD, the SYSMOD must be “received” into data sets
that can be used by SMP/E. The SMP/E RECEIVE command performs the task of
copying the SYSMOD from the distribution medium from which it was sent into
the data sets used by SMP/E.
For more information about the RECEIVE command, refer to “Receiving the
SYSMOD into SMP/E’s data sets” on page 14.
Applying the SYSMOD to the target libraries

Once a SYSMOD has been received, you want to “apply” the SYSMOD to the
appropriate target libraries. The SMP/E APPLY command invokes various system
utilities to install the SYSMOD’s elements into the target libraries.
For more information about the APPLY command, refer to “Applying the SYSMOD
to the target libraries” on page 19.
Restoring the target libraries to the previous level

Should you experience problems after applying a SYSMOD, you may want to
“restore” its elements in error to a previous and stable level. The SMP/E RESTORE
command replaces a failing element with a copy from the distribution libraries.
For more information about the RESTORE command, refer to “Restoring the target
libraries to the previous level” on page 23.
Accepting the SYSMOD and updating the distribution libraries

After you have performed a SYSMOD RECEIVE and APPLY, you want to “accept”
the elements into the distribution libraries for backup. However, this should be
done only after you are satisfied with the performance and stability of the
elements of the SYSMOD. Once you ACCEPT a SYSMOD, you cannot RESTORE
its element to a previous level. The SMP/E ACCEPT command updates the
distribution libraries so they are available for backup of any future SYSMODs.
For more information about the ACCEPT command, refer to “Accepting the
SYSMOD into the distribution libraries” on page 27.
Displaying SMP/E data

The SMP/E CSI and other primary data sets contain a great deal of information
you may find useful when installing new elements or functions, preparing user
modifications, or debugging problems. There are several ways SMP/E allows you
to display that information, as well as information about modules, macros, and
other elements:

Basic SMP/E commands
v Query dialogs display specific information you request through interactive

dialogs with SMP/E.
v The LIST command generates a hardcopy listing of information about your
system.
v REPORT commands check, compare, and generate hardcopy information about
the contents of zones on your system.
v The SMP/E CSI application programming interface can be used to write
application programs to query the contents of your system’s CSI data sets.
For more information about displaying SMP/E data, refer to “Displaying SMP/E
data” on page 31.
Flow of SMP/E SYSMOD processing

To see the flow of SMP/E SYSMOD processing for the RECEIVE, APPLY,
RESTORE, and ACCEPT commands, let’s look at Figure 12.
Figure 12. Flow of SMP/E SYSMOD Processing
Receiving the SYSMOD into SMP/E’s data sets

To initiate SMP/E processing, you must first install the software into SMP/E data
sets. You can use the RECEIVE command to load the SYSMOD information from
the distribution medium into the SMPPTS and SMPTLIB data sets for later
installation of the SYSMODs.
In this chapter, you will learn about those data sets and the following topics:
v “What happens during RECEIVE processing” on page 15
Receiving the SYSMOD
v “What happens during internet service retrieval”

v “How SMP/E keeps track of RECEIVE processing” on page 16
v “Using the RECEIVE command” on page 17
v “Summary of the RECEIVE command” on page 19
What happens during RECEIVE processing

SMP/E knows software in terms of SYSMODs. Each SYSMOD processed by
SMP/E contains two types of information:
v Instructions telling SMP/E what elements are in the SYSMOD and how to install
them
v The actual element replacements or updates contained in the SYSMOD
The instructions are made up of a series of control statements called modification

control statements (MCSs). The element replacements or updates can be packaged in
several ways:
v The RELFILE method packages the elements in relative files that are separate
from the MCSs. This method is used mostly for function SYSMODs. (The
examples in the remainder of this book assume that function SYSMODs are
packaged in RELFILE format.)
v The inline method packages the elements immediately following the associated
MCSs.
v The indirect library method packages elements in DASD data sets that are
separate from the MCSs.
For more details about packaging, see the z/OS Packaging Rules manual.
During RECEIVE processing, the MCS for each SYSMOD is copied to an SMP/E
temporary storage area called the SMPPTS data set. The MCS entry contains the
MCS and any inline element replacements or updates for the SYSMOD. Relative
files, however, are stored in another temporary storage area called the SMPTLIB
data sets.
We briefly mentioned HOLDDATA earlier in the book (see “The SMP/E zones” on
page 11). HOLDDATA is processed by the RECEIVE command and is stored for
use later on during installation of the affected SYSMODs.
What happens during internet service retrieval

Internet Service Retrieval uses the SMP/E RECEIVE ORDER command to order
software service directly from IBM. With RECEIVE ORDER processing, you run an
SMP/E job that places the service order directly with the IBM Automated Delivery
Request server, waits for the order to be fulfilled, and then downloads and
processes the service package contents.
You can use the RECEIVE ORDER command to request HOLDDATA or PTF
service orders based on criteria you specify. When you request HOLDDATA only,
you receive the last two years of Enhanced HOLDDATA for the entire z/OS
platform. You can request two types of PTF service orders:
Corrective
You can order PTFs that resolve specific APARS. The package resulting
from such an order is tailored to your SMP/E environment and contains
the PTFs you requested plus any requisite PTFs not already present in
your environment.

Preventive
You can order all currently available PTFs that meet your selection criteria.
The package resulting from such an order is tailored to your SMP/E
environment and contains the PTFs that match your selection criteria plus
any requisite PTFs not already present in your environment. There are
three selection criteria:
Critical
Includes all available PTFs that resolve high impact pervasive
(HIPER) problems or PTFs in error (PE).
Recommended
Includes all available PTFs identified with Recommended Service
Update SOURCEID (RSUyymm) and all available PTFs that resolve
a critical problem (HIPER or PE). Recommended service includes
PTFs through the most recent RSU level.
All Includes all available PTFs.
All PTF packages also contain the last two years of Enhanced HOLDDATA for the
entire z/OS software platform.
Using the RECEIVE ORDER command, you can automate the service update
process. For example, you can have an SMP/E job run every night at 1:00 AM to
order and download the latest HOLDDATA and critical service, so the information
is available locally and ready for use every morning.
How SMP/E keeps track of RECEIVE processing

SMP/E updates the global zone with information about the SYSMODs that have
been received:
v SYSMOD entries are created in the global zone for each SYSMOD that has been
received.
v HOLDDATA entries are created in the global zone for each ++HOLD statement
that has been received. HOLDDATA entries identify SYSMODs that should be
held back from being installed because they require special handling or are in
error.
Figure 13 shows what you have learned about RECEIVE processing.

Figure 13. Results of RECEIVE processing
Using the RECEIVE command

In this section, you will see some basic examples of how you might use the
RECEIVE command.
Examples
Let’s look at a few of these examples.
Receiving SYSMODs and HOLDDATA: In the course of maintaining your

system, you need to install service and process the related HOLDDATA. Assume
IBM has supplied you with a service tape (such as a CBPDO tape or an ESO tape),
and you want to install it on your system. The first step is to receive the SYSMODs
and HOLDDATA that are contained on the tape. You can accomplish this by
specifying the following commands:
SET BDY(GLOBAL).
RECEIVE.
When you issue these commands, SMP/E receives all the SYSMODs and
HOLDDATA on the service tape into the global zone.
Receiving only HOLDDATA: There may be times when you do not want to
receive the SYSMODs from a service tape, but you do want to receive the
HOLDDATA. Because the HOLDDATA provides information about SYSMODs
requiring special handling or that are in error, it is important for you to receive the
HOLDDATA into SMP/E’s storage repository as soon as possible. The following
commands process only the HOLDDATA:
SET BDY(GLOBAL).
RECEIVE HOLDDATA.
By issuing these commands, you direct SMP/E to receive only the HOLDDATA
from the service tape into the global zone.

Receiving only SYSMODs: Assume you have previously received only the
HOLDDATA from a service tape and are now ready to install the SYSMODs. To
install these SYSMODs (using the APPLY and ACCEPT commands), you must first
receive them. This can be done by specifying the following commands:
SET BDY(GLOBAL).
RECEIVE SYSMODS.
When you issue these commands, you direct SMP/E to receive only the SYSMODs
from the service tape into the global zone.
Receiving SYSMODs and HOLDDATA for a specific product: You may want to
receive SYSMODs and HOLDDATA for a particular product from the service tape.
You can accomplish this task by specifying the following commands:
SET BDY(GLOBAL).
RECEIVE FORFMID(HOP0001).
By issuing these commands, you direct SMP/E to receive SYSMODs and

HOLDDATA for the product whose FMID is HOP0001 from the service tape into
the global zone.
Requesting a new PTF order with the RECEIVE ORDER command: Assume
you want to order two specific PTFs (UQ12345 and UQ98765). You can accomplish
this task by specifying the following SMP/E job:
//jobname JOB ...
//RECEIVE EXEC PGM=GIMSMP
//SMPCSI DD DSN=SMPE.GLOBAL.CSI,DISP=SHR
//SMPNTS DD PATH=’/u/smpe/smpnts/’,PATHDISP=KEEP
//SMPOUT DD SYSOUT=*
//SMPRPT DD SYSOUT=*
//SYSPRINT DD SYSOUT=*
//SMPCNTL DD *
SET BOUNDRY(GLOBAL).
RECEIVE SYSMODS HOLDDATA
ORDER( /* Place an order for service */
ORDERSERVER(ORDRSRVR)
CONTENT(
PTFS(UQ12345,UQ98765) /* Get these PTFs, and any.. */
) /* ..requisites.. */
FORTGTZONES(ZOS14) /* ..for this target zone */
).
/*
//ORDRSRVR DD *
<ORDERSERVER
url="https://eccgw01.boulder.ibm.com/services/projects/ecc/ws/"
keyring="MRWKYRNG"
certificate="SMPE Client Certificate">
</ORDERSERVER>
/*
Note: An alternative url for the IBM Automated Delivery Request server is
https://eccgw02.rochester.ibm.com/services/projects/ecc/ws/.
In addition to receiving the specified PTFs, you receive any requisites for these
PTFs that are not already present in the ZOS14 target zone.
For a more complete description of all the RECEIVE command operands and other
examples, see The RECEIVE Command in SMP/E Commands.
Reporting output
When RECEIVE processing is complete, these reports will help you analyze the
results:

v The RECEIVE summary report provides you with an at-a-glance look at all the
SYSMODs that were processed during the RECEIVE command run. It shows
you which SYSMODs were received, which were not received, and why.
Note: The SYSMODs listed in this report depend on the operands you specify
on the RECEIVE command.
v The RECEIVE exception SYSMOD Data report provides you with a quick
summary of the HOLDDATA information processed during the RECEIVE
command run. It lists the SYSMODs requiring special handling or that are in
error, and those SYSMODs no longer requiring special handling or that have had
an error fixed.
v The File allocation report provides you with a list of the data sets used for
RECEIVE processing and supplies information about these data sets.
For more information about these reports (and samples of actual reports), see
SMP/E Reports in SMP/E Commands.
Summary of the RECEIVE command

Let’s summarize what you have learned about using the RECEIVE command to
load a SYSMOD into SMP/E’s storage area. The RECEIVE command:
v Copies the MCS for each SYSMOD to the SMPPTS data set
v Loads elements into SMPTLIB data sets for SYSMODs using the relative file
packaging method
v Records what is received in the global zone
– SYSMOD entries
– HOLDDATA entries
v Reports the results of processing
The RECEIVE ORDER command:

v Places the service order directly with the IBM Server
v Waits for the order to be fulfilled
v Downloads and processes the service package contents.
Applying the SYSMOD to the target libraries

After the SYSMODs have been received, you can use the APPLY command to
install them into the appropriate target system libraries. The APPLY command calls
system utilities, which are responsible for the actual updating of those libraries.
In this chapter, you will learn about the following topics:

v What happens during APPLY processing
v How SMP/E keeps track of APPLY processing
v Examples of using the APPLY command
v A summary of the APPLY command
What happens during APPLY processing

Throughout the APPLY process, SMP/E helps you manage the complexities of
your system when installing SYSMODs.
Selecting the SYSMODs

You can specify operands on the APPLY command that tell SMP/E which of the
received SYSMODs are to be selected for installation in the target libraries. SMP/E
checks to make sure all other required SYSMODs (prerequisites) have been

Applying the SYSMOD
installed or are being installed concurrently and in the proper sequence. For more
information about prerequisites, see “Keeping track of the elements of the system”
on page 7.
Selecting the elements

During APPLY processing, SMP/E uses the information provided in the selected
SYSMODs to determine which elements should be installed in the target libraries.
The selection of elements is monitored by SMP/E to make sure that the correct
functional level of each element is selected.
Checking the APPLY process

SMP/E provides you with an option to stop APPLY processing just before any
updating takes place so you can ensure all prerequisites are satisfied before the
installation of the SYSMODs. This helps you see what will happen (and helps you
detect problem SYSMODs) without actually updating the target libraries.
Updating the target libraries

After the proper SYSMODs have been selected and the proper functional and
service level of each element has been determined, the APPLY command directs
SMP/E to call the system utilities. It is the system utilities that actually place the
elements into the target libraries described in the target zone. The source of the
elements is the SMPTLIB data sets, the SMPPTS data set, or the indirect libraries,
depending on how the SYSMOD was packaged.
Note: Because the APPLY command updates the system libraries, you should
never use it on a live production system. When you process the APPLY
command, you should always use a copy of the target libraries and target
zone. By using a copy, you minimize the risk of new code causing an outage
of your system. This process of copying is called cloning and is explained in
detail in the OS/390 Software Management Cookbook, SG24-4775.
How SMP/E keeps track of APPLY processing

SMP/E updates the information about the SYSMODs that have been applied.
Remember, the target zone reflects the contents of the target libraries. Therefore,
after the utility work is complete, and the target libraries have been updated, the
target zone is updated to accurately reflect the status of those libraries.
v A SYSMOD entry is created in the target zone for each SYSMOD that has been
applied. Element entries (such as MOD and LMOD) are also created in the target
zone for those elements that have been installed in the target libraries.
v SYSMOD entries in the global zone are updated to reflect that the SYSMOD has
been applied to the target zone.
v BACKUP entries are created in the SMPSCDS data set so the SYSMOD can later
be restored, if necessary.
Figure 14 on page 21 shows what you have learned about APPLY processing.

Applying the SYSMOD
Figure 14. Results of APPLY Processing
Using the APPLY command

The APPLY command has many operands that allow you great flexibility in
choosing which SYSMODs you want installed in your target libraries. It also
provides you with a variety of output based on the operands you specify.
Examples
Let’s look at a few examples of how you might use the APPLY command.
Applying PTF SYSMODs: After you have received the SYSMODs into the global
zone, you can tell SMP/E that you want to install only the PTF SYSMODs. You can
do this by specifying the following commands:
SET BDY(ZOSTGT1).
APPLY PTFS.
By issuing these commands, you direct SMP/E to apply all eligible PTF SYSMODs
to target zone ZOSTGT1.
Suppose you do not want to install all the PTF SYSMODs, but only a select few.
You can do this by specifying the following commands:
SET BDY(ZOSTGT1).
APPLY SELECT(UZ00001,UZ00002).
Issuing these commands results in the selection of only PTFs UZ00001 and
UZ00002 for installation in target zone ZOSTGT1.
Applying APAR and USERMOD SYSMODs: You may want to install just
corrective fixes (APARs) or user modifications (USERMODs) into the target
libraries. You can accomplish this with the following commands:

Applying the SYSMOD
SET BDY(ZOSTGT1).
APPLY APARS
USERMODS.
When you issue these commands, SMP/E installs all eligible APARs and
USERMODs into target zone ZOSTGT1.
Applying SYSMODs for selected products: There may be times when you want
to update only certain products on your system with the SYSMODs contained on a
service tape. Assume you want to install all PTFs for a particular product to your
system. This can be accomplished by specifying the following commands:
SET BDY(ZOSTGT1).
APPLY PTFS
FORFMID(HOP0001).
or
SET BDY(ZOSTGT1).
APPLY FORFMID(HOP0001).
In both cases, SMP/E applies all applicable PTFs for the product with FMID
HOP0001 to target zone ZOSTGT1. Unless you specify otherwise, PTFS is the
default SYSMOD type.
Applying SYSMODs having prerequisites: When installing a SYSMOD, you

might not always know if it has prerequisites, or if the prerequisites are available.
(Sometimes a prerequisite SYSMOD might not be received, or it might be held
because it is in error.) In cases such as this, you can direct SMP/E to check
whether an equivalent (or superseding) SYSMOD is available, by specifying the
GROUPEXTEND operand.
Assume you want to update a product with all the eligible PTFs and APARs. You
can do this by specifying the following commands:
SET BDY(ZOSTGT1).
APPLY PTFS
APARS
FORFMID(HOP0001)
GROUPEXTEND.
By issuing these commands, you direct SMP/E to apply all PTFs and APARs,
along with any other required SYSMODs to the product whose FMID is HOP0001
and is located in the ZOSTGT1 target zone. If SMP/E cannot find a required
SYSMOD, it looks for and uses a SYSMOD that supersedes the required one.
Applying SYSMODs using the CHECK operand: In the previous example, you
directed SMP/E to automatically include all SYSMODs needed for the specified
product. There may be times when you want to see which SYSMODs are included
before you actually install them. You can do this with the CHECK operand by
issuing the following commands:
SET BDY(ZOSTGT1).
APPLY PTFS
APARS
FORFMID(HOP0001)
GROUPEXTEND
CHECK.
After these commands are processed, you can check the SYSMOD Status report to
see which SYSMODs would have been installed if you had not specified the

Applying the SYSMOD
CHECK operand. If you are satisfied with the results of this trial run, you can
issue the commands again, without the CHECK operand, to actually install the
SYSMODs.
For a more complete description of all the APPLY command operands, and for
additional examples, see APPLY Command in SMP/E Commands.
Reporting output
When APPLY processing is complete, these reports will help you analyze the
results:
v The SYSMOD status report provides you with a summary of the processing
that took place for each eligible SYSMOD, based on the operands you specified
on the APPLY command. It shows you which SYSMODs were applied, which
were not applied, and why.
v The Element summary report provides you with a detailed look at each element
affected by APPLY processing. It tells you in which libraries the elements were
installed.
v The Causer SYSMOD summary report provides you with a list of SYSMODs
that caused other SYSMODs to fail, and describes the errors that must be fixed
to successfully process the SYSMODs. This report can reduce the amount of
work involved in figuring out which errors caused SYSMODs to fail.
APPLY processing and supplies information about these data sets.
Additional reports may be produced depending on the work being done and the
content of the SYSMODs. For more information about all the reports produced by
the APPLY command (and samples of actual reports), see The APPLY Command in
SMP/E Commands.
Summary
Let’s summarize what you have learned about using the APPLY command to
install a SYSMOD in the target libraries. The APPLY command:
v Selects SYSMODs to install
v Checks that all other required SYSMODs have been (or are being) installed
v Based on SYSMODs, selects elements to install
v Directs SMP/E to call the system utilities to update the target libraries
v Records what is applied:
– Target zone: Creates SYSMOD entries and element entries
– Global zone: Updates SYSMOD entries
– SMPSCDS data set: Creates BACKUP entries
Remember, you should never perform APPLY processing on a live production

system!
Restoring the target libraries to the previous level

If you discover that a particular SYSMOD is causing a problem in your target
libraries, you can remove it and replace the elements affected by it with the
previous level of those elements, which is obtained from the backup (or
distribution) libraries. If you are wondering how a backup version came to exist in
the distribution libraries, this topic is covered in “Accepting the SYSMOD into the
distribution libraries” on page 27.

Restoring target libraries
You can use the RESTORE command to remove SYSMODs from the target libraries
and restore them to a previous level. The RESTORE command reverses APPLY
processing, but has no effect on ACCEPT processing.

v What happens during RESTORE processing
v How SMP/E keeps track of RESTORE processing
v Examples of using the RESTORE command
v A summary of the RESTORE command
What happens during RESTORE processing

SMP/E provides you with a method for removing an applied SYSMOD when its
installation results in unexpected problems.
Removing the SYSMODs

SMP/E ensures the eligibility of the selected SYSMODs and checks whether other
SYSMODs are affected before continuing with RESTORE processing. Because of the
various relationships and dependencies among the many SYSMODs, this checking
is very important to the integrity of your system. In fact, to ensure that the
requisites for a SYSMOD being restored are processed appropriately, SMP/E may
require the whole chain of prerequisites to be restored.

During RESTORE processing, SMP/E uses the information provided in the selected
SYSMODs to determine which elements in the target zone should be replaced by
elements in the related distribution libraries. The selection of elements is monitored
by SMP/E to make sure that the correct functional level of each element is
selected.
Checking the RESTORE process

SMP/E provides you with an option to stop RESTORE processing just before any
updating takes place so you can ensure all prerequisites are satisfied before
restoring any SYSMODs. This helps you see what will happen without actually
making any changes to the elements in the target libraries.
Replacing the elements in the target libraries

When SMP/E is satisfied that the proper SYSMODs have been selected, it uses
information from the target zone to determine which distribution zone describes
the elements necessary to replace the SYSMOD’s elements in the target libraries.
The RESTORE command directs SMP/E to call system utilities that replace the
elements in the target libraries with the previous level of the elements from the
related distribution libraries.
How SMP/E keeps track of RESTORE processing

SMP/E updates the information about the SYSMODs that have been restored.
Remember, the target zone reflects the contents of the target libraries. Therefore,
after the utility work is complete, and the target libraries have been updated, the
target zone is updated to accurately reflect the status of those libraries.
v All information in the target zone pertaining to the restored SYSMOD is
removed. The element entries in the target zone are restored to reflect the
distribution zone level of the elements.
v The global zone SYSMOD entries and MCS statements, which are stored in the
SMPPTS data set, are deleted for those SYSMODs that have been restored. Any

SMPTLIB data sets created during RECEIVE processing are also deleted for the
restored SYSMOD. SMP/E automatically performs this global zone clean-up,
unless you specify otherwise.
Figure 15 shows what you have learned about RESTORE processing.
Figure 15. Results of RESTORE Processing
Using the RESTORE command

The RESTORE command has operands that allow you to specify the criteria for
removing SYSMODs from the target libraries. It also produces output that reports
on its processing.
Examples
Let’s look at a few examples of how you might use the RESTORE command.
Restoring a single SYSMOD: Assume you have applied a SYSMOD and, after
some initial testing, you discover that a PTF SYSMOD is causing problems on your
system. You can remove this SYSMOD by specifying the following commands:
SET BDY(ZOZTGT1).
RESTORE SELECT(UZ00001).
By issuing these commands, you instruct SMP/E to remove PTF UZ00001 from
target zone ZOZTGT1 and replace its elements in the target libraries with the
previous level of elements from the distribution libraries.
Restoring SYSMODs using the GROUP operand: When you want to remove a
particular SYSMOD, it is not always easy to determine other SYSMODs that need
to be restored in order to remove the bad one. Assume a particular PTF SYSMOD

is causing a problem, and you want to know if it is dependent on any other

SYSMODs so you can also restore those SYSMODs. This can be accomplished by
SET BDY(ZOZTGT1).
RESTORE SELECT(UZ00003)
GROUP.
By issuing these commands, you instruct SMP/E to restore PTF UZ00003 and any
other related PTFs from target zone ZOZTGT1, and replace their elements with the
previous level from the distribution zone.
Restoring SYSMODs using the CHECK operand: In the previous example, you
directed SMP/E to restore any dependent SYSMODs in order to remove the bad
one. There may be times when you want to see which SYSMODs are restored
without actually restoring them. You can do this with the CHECK operand by
issuing the following commands:
SET BDY(ZOZTGT1).
RESTORE SELECT(UZ00003)
GROUP
CHECK.
see which SYSMODs would have been restored if you had not specified the
issue the commands again, without the CHECK operand, to actually restore the
SYSMODs.
For a more complete description of all the RESTORE command operands, and for
additional examples, see The RESTORE Command in SMP/E Commands.
Reporting output
When RESTORE processing is complete, these reports will help you analyze the
results:
on the RESTORE command. It shows you which SYSMODs were restored, which
were not restored, and why.
replaced or modified by RESTORE processing. It tells you in which libraries the
elements were restored.
RESTORE processing and supplies information about these data sets.
the RESTORE command (and samples of actual reports), see The RESTORE
Command in SMP/E Commands.
Summary
Let’s summarize what you have learned about using the RESTORE command to
remove a SYSMOD from the target libraries. The RESTORE command:

v Removes the SYSMOD from the indicated target zone

v Calls system utilities to replace the SYSMOD’s elements in the target libraries
with elements from the related distribution libraries
v Records what is restored:
– Target zone: Restores element entries to reflect their distribution zone level
and deletes all information about restored SYSMOD.
– Global zone: Deletes SYSMOD entries and MCS statements in SMPPTS for
restored SYSMOD. Any SMPTLIB data sets created during RECEIVE
processing are also deleted for the restored SYSMOD. (This global zone
processing is optional.)
– SMPSCDS data set: Deletes BACKUP entries for restored SYSMOD.
Note: Not all SYSMODs can be restored. For example, SMP/E cannot restore a
SYSMOD that deletes another SYSMOD or that deletes a load module
during APPLY processing.
Accepting the SYSMOD into the distribution libraries

You can use the ACCEPT command to install software in backup (or distribution)
libraries. ACCEPT processing is very similar to APPLY processing with one
important exception: ACCEPT processing is irreversible.

v What happens during ACCEPT processing
v How SMP/E keeps track of ACCEPT processing
v Examples of using the ACCEPT command
v A summary of the ACCEPT command
What happens during ACCEPT processing

After you are satisfied that an applied SYSMOD has performed reliably in your
target system, you can install it in your backup system (distribution) libraries.
Selecting the SYSMODs

You can specify operands on the ACCEPT command that tell SMP/E which of the
received SYSMODs are to be selected for installation in the distribution libraries.
SMP/E ensures that all other required SYSMODs have been installed or are being
installed concurrently and in the proper sequence.

During ACCEPT processing, SMP/E uses the information provided in the selected
SYSMODs to determine which elements should be installed in the distribution
libraries. The selection of elements is monitored by SMP/E to make sure that the
correct functional level of each element is selected.
Checking the ACCEPT process

SMP/E provides you with an option to stop ACCEPT processing before any
updating takes place so you can ensure all prerequisites are satisfied before the
installation of the SYSMODs. This helps you see what will happen (and helps you
detect problem SYSMODs) without actually updating the distribution libraries.
Updating the distribution libraries

After the proper SYSMODs have been selected and the proper functional and
service level of each element has been checked, SMP/E calls the system utilities (in

Accepting the SYSMOD
the same manner as APPLY and RESTORE) to place the elements into the
distribution libraries described in the distribution zone. The source of the elements
is the SMPTLIB data sets, the SMPPTS data set, or the indirect libraries, depending
on how the SYSMOD was packaged.
Note: When ACCEPT processing has been completed, there is no way it can be
undone.
How SMP/E keeps track of ACCEPT processing

SMP/E updates the information about the SYSMODs that have been accepted.
Remember, the distribution zone reflects the contents of the distribution libraries.
Therefore, after the utility work is complete, and the distribution libraries have
been updated, the distribution zone is updated to accurately reflect the status of
those libraries.
v A SYSMOD entry is created in the distribution zone for each SYSMOD that has
been accepted. Element entries (such as MOD and LMOD) are also created in
the distribution zone for the elements that have been installed in the distribution
libraries.
v Global zone SYSMOD entries and MCS statements in the SMPPTS data set are
deleted for those SYSMODs that have been accepted into the distribution zone.
Any SMPTLIB data sets created during RECEIVE processing are also deleted. If
you do not want SMP/E to do this global zone clean-up, you have the option to
indicate this to SMP/E, and the information is saved.
Figure 16 shows what you have learned about ACCEPT processing.
Figure 16. Results of ACCEPT Processing
Using the ACCEPT command

The ACCEPT command has many operands that allow you great flexibility for
further defining which SYSMODs you want installed in your distribution libraries.
It also provides you with a variety of output based on the operands you specify.

Examples
Let’s look at a few examples of how you might use the ACCEPT command.
Accepting PTF SYSMODs: After you have applied the SYSMODs into the target
zone, you can define to SMP/E that you want to install only the PTF SYSMODs
into the distribution zone. You can do this by specifying the following commands:
SET BDY(ZOSDLB1).
ACCEPT PTFS.
By issuing these commands, you direct SMP/E to accept all eligible PTF SYSMODs
into distribution zone ZOSDLB1.
Suppose you do not want to install all the PTF SYSMODs, but only a select few.
You can do this by specifying the following commands:
SET BDY(ZOSDLB1).
ACCEPT SELECT(UZ00001,UZ00002).
When you issue these commands, only PTFs UZ00001 and UZ00002 are installed in
distribution zone ZOSDLB1.
Accepting SYSMODs for selected products: There may be times when you want
to update only certain products on your system with the SYSMODs contained on a
service tape. Assume you want to install all PTFs for a particular product. This can
be accomplished by specifying the following commands:
SET BDY(ZOSDLB1).
ACCEPT PTFS
FORFMID(HOP0001).
or
SET BDY(ZOSDLB1).
ACCEPT FORFMID(HOP0001).
In both cases, SMP/E accepts all applicable PTFs for the product whose FMID is
HOP0001 and that is located in distribution zone ZOSDLB1. Unless you specify
otherwise, PTFS is the default SYSMOD type.
Accepting SYSMODs having prerequisites: When installing a SYSMOD, you

might not always know if it has prerequisites, or if the prerequisites are available.
(Sometimes a prerequisite SYSMOD might not be received, or it might be held
because it is in error.) In cases such as this, you can direct SMP/E to check
whether an equivalent (or superseding) SYSMOD is available, by specifying the
GROUPEXTEND operand.
Assume you want to process all PTFs for a product on your system, and you want
to ensure that all other required SYSMODs are also processed. You can do this by
SET BDY(ZOSDLB1).
ACCEPT PTFS
FORFMID(HOP0001)
GROUPEXTEND.
By issuing these commands, you direct SMP/E to accept all PTFs, along with any
other required SYSMODs, to the product whose FMID is HOP0001 and is located
in the ZOSDLB1 target zone. If SMP/E cannot find a required SYSMOD, it looks
for and uses a SYSMOD that supersedes the required one.

Accepting SYSMODs using the CHECK operand: In the previous example,

SMP/E was directed to automatically include all SYSMODs needed for the
specified product. There may be times when you want to see which SYSMODs are
included before you actually install them. You can do this with the CHECK
operand by issuing the following commands:
SET BDY(ZOSTGT1).
ACCEPT PTFS
FORFMID(HOP0001)
GROUPEXTEND
CHECK.
see which SYSMODs would have been installed if you had not specified the
issue the commands again, without the CHECK operand, to actually install the
SYSMODs.
For a more complete description of all the ACCEPT command operands and other
examples, see The ACCEPT Command in SMP/E Commands.
Reporting output
When ACCEPT processing is complete, these reports will help you analyze the
results:
on the ACCEPT command. It shows you which SYSMODs were accepted, which
were not accepted, and why.
affected by ACCEPT processing. It tells you in which libraries the elements were
installed.
ACCEPT processing and supplies information about these data sets.
the ACCEPT command (and samples of actual reports), see The ACCEPT
Summary
Let’s summarize what we have learned about using the ACCEPT command to
install a SYSMOD in the distribution (or backup) libraries. The ACCEPT command:
v Selects SYSMODs to install
v Checks that all other required SYSMODs have been (or are being) installed
v Based on SYSMODs, selects elements to install
v Directs SMP/E to call the system utilities to update the distribution libraries
v Records what is accepted:
– Distribution zone: Creates SYSMOD entries and element entries.

– Global zone: Deletes SYSMOD entries and MCS statements in SMPPTS. Any
SMPTLIB data sets created during RECEIVE processing are also deleted. (This
global zone processing is optional.)
Remember, once you have accepted a SYSMOD, it cannot be restored!
Displaying SMP/E data

You can use SMP/E to provide helpful information for planning new installations,
debugging problems, and other instances when you want to know the function
and service level of your product software. There are several ways you can display
data in the SMP/E database.
In this chapter, you will learn about the kinds of information that help you manage
your system and the best method by which the information can be obtained.
v Query dialogs: The easiest and fastest way to obtain just the information you
want
v LIST command: When you need an all-inclusive hardcopy listing of information
about your system
v REPORT commands: To check and compare the zone contents and generate
command output that can be used to update your system
v SMP/E CSI application programming interface: To write an application
program to query the contents of your system’s CSI data sets.
Using the query dialogs

The SMP/E dialogs provide you with an online method of system management,
software inventory, data base inquiries, and guidance. For example, with the Query
dialogs, you can look up information in the CSI data set. The Query dialogs are
one of the easiest and most direct methods you can use to obtain the content and
status of any SYSMOD that has been processed by SMP/E. You can use the Query
dialogs to display an entry in either a specific zone (CSI query) or in all zones
(cross-zone query).
You can use the SMP/E dialogs to view a SYSMOD, even if it has been compacted.
Use the Query dialog (zone is GLOBAL, entry type MCS, entry name is the
SYSMOD name) and you will be shown the complete SYSMOD in an edit session.
You may save the SYSMOD in a different location from this session. If you are
using SMPPTS spill data sets, there is another benefit of viewing the SYSMOD
from the Query dialog, in that you do not have to know in which SMPPTS data set
the SYSMOD is stored; SMP/E will find it for you.
To get to the Query dialogs, you select SMP/E (option 1) on the initial SMP/E
dialog panel (CIDPGV2). Then, on the main menu for SMP/E options
(GIM@PRIM), select Query (option 3). This takes you to the initial Query panel,
shown in Figure 17 on page 32. If you need assistance with using the Query
dialogs, (or any of the SMP/E dialogs), help panels are available.
Let’s assume you want to find out which SYSMODs have been applied to a
particular target zone on your system. You can accomplish this task using the
QUERY SELECTION MENU and selecting the CSI QUERY option (1), as shown in
Figure 17 on page 32.

Displaying SMP/E Data
GIMQUPO QUERY SELECTION MENU

===> 1
1 CSI QUERY - Display SMPCSI entries

2 CROSS-ZONE QUERY - Display status of an entry in
all zones
3 SOURCEID QUERY - Display SOURCEIDs for specified zone
D DESCRIBE - Overview of using QUERY
T TUTORIAL - Information on using QUERY
To return to the SMP/E primary option menu, enter END .
5694-A01 5655-G44(C) COPYRIGHT IBM CORP 1982, 2005
Figure 17. Query selection menu
When the CSI QUERY panel is displayed (see Figure 18), you can indicate that you
want SMP/E to check target zone ZOSTGT1 for all SYSMOD entries.
GIMQU1PO CSI QUERY

===>
Specify the zone, entry type, and name to be queried:
ZONE NAME ===> ZOSTGT1 Name of the zone to be queried.

To display a list of all zones,
leave blank
ENTRY TYPE ===> SYSMOD Entry type to be queried.

To display a list of all valid
entry types, leave ENTRY TYPE
and ENTRY NAME blank
ENTRY NAME ===> Entry name to be queried.

Leave blank or use a wildcard
(entry name pattern) to display
a selection list.
To return to the Query selection menu, enter END .
Figure 18. CSI query panel
Because the ENTRY NAME was left blank on the CSI QUERY panel, SMP/E
displays another panel (see Figure 19 on page 33) that lists all the SYSMOD entries
in target zone ZOSTGT1.

GIMQUSEA CSI QUERY - SELECT ENTRY

===> SCROLL ===> PAGE
Select one entry to query from target zone ZOSTGT1 :
S NAME ACTION
AZ00005
s UZ00001
UZ00002
Figure 19. CSI query - Select entry panel
The CSI QUERY - SELECT ENTRY panel shows that SYSMODs AZ00005, UZ00001,
and UZ00002 have been applied to target zone ZOSTGT1. If you want more
information about the contents of SYSMOD UZ00001, you can select that entry by
entering an S next to it, and another panel is displayed (see Figure 20).
GIMQIT26 CSI QUERY - SYSMOD ENTRY

===> SCROLL ===> PAGE
To return to the previous panel, enter END .
Entry Type: SYSMOD Zone Name: ZOSTGT1

Entry Name: UZ00001 Zone Type: TARGET
Type: PTF Status: APPLIED

FMID: HOP0001
Date/Time: 03.341 13.57:31 APPLIED
-------- -------- -------- -------- -------- -------- --------

MODS MOD01 MOD02
Figure 20. CSI query - SYSMOD entry panel
The CSI QUERY - SYSMOD ENTRY panel displays all the relevant information
pertaining to SYSMOD UZ00001.
As you can see, the QUERY dialog panels provide a quick and easy way for you to
obtain information about your system.
Using the LIST command

In the course of managing your system, there may be times when you need a
hardcopy listing of some type of information. You can use the LIST command to
accomplish this task. For example, it might be necessary for you to have a record
of the following:
v All entries of a specific type
v Selected entries of a specific type
v All entries that meet certain criteria
The LIST command can provide you with a listing of this information.
Examples
Let’s look at a few basic examples of how you might use the LIST command.

Listing entries in a particular zone: In the course of managing your system, you
might need to know which SYSMOD entries exist in the global zone. You can find
this out by specifying the following commands:
SET BDY(GLOBAL).
LIST SYSMODS.
By issuing these commands, you direct SMP/E to list all the SYSMOD entries in
the global zone.
Listing Specific Entries: Suppose you discover a problem on your system and
need to determine whether a particular SYSMOD has been installed in the target
zone. You can accomplish this by specifying the following commands:
SET BDY(ZOSTGT1).
LIST SYSMOD(UZ00001).
By issuing these commands, you direct SMP/E to provide you with information
about SYSMOD UZ00001 in target zone ZOSTGT1.
Listing SYSMODs that are received but not installed: You might have received
service into the global zone and are in the process of installing the service on your
system. You want to see which of the SYSMODs you have received have not yet
been installed in a target zone. This can be accomplished by specifying the
following commands:
SET BDY(GLOBAL).
LIST SYSMODS
NOAPPLY(ZOSTGT1).
By issuing these commands, you direct SMP/E to list the SYSMODs that have been
received, but have not yet been applied to target zone ZOSTGT1.
Reporting output
When LIST processing is complete, these reports will provide you with the
information that was requested:
v The LIST summary report provides you with information about the type of
entry, name of entry, and status of entry for zones and data sets you have
specified.
v The File allocation report provides you with a list of the data sets used for LIST
processing, and supplies information about these data sets.
For a more complete description of the LIST command, additional examples, and
samples of actual reports, see The LIST Command in SMP/E Commands.
Using the REPORT commands

You can use the REPORT commands to check and compare the SYSMODs installed
in the different zones that exist on your system. In addition to this checking, you
can tell SMP/E to generate the necessary commands to synchronize the specified
zones. You can later modify these commands, if necessary, and use them to install
the indicated SYSMODs.
One of the REPORT commands (REPORT SYSMODS) is very useful if you want to
compare the SYSMODs installed in two zones. This command allows you to
compare the following:
v One distribution zone to another distribution zone
v One target zone to another target zone
v A distribution zone to a target zone

v A target zone to a distribution zone
Example
Let’s look at a basic example of how you might use the REPORT SYSMODS
command. Assume you have two systems using the same global zone, and you
want to check which SYSMODs are installed in a target zone on one system, but
are not installed in a target zone on the other system. You can accomplish this by
SET BDY(GLOBAL).
REPORT SYSMODS
INZONE(ZOSTGT1)
COMPARED(ZOSTGT2).
By issuing these commands, you direct SMP/E to compare the SYSMOD content of
zone ZOSTGT1 to that of zone ZOSTGT2. Any SYSMODs that are in zone
ZOSTGT1 and are not in zone ZOSTGT2 appear in the resulting report.
SMP/E also provides output you can use to install those SYSMODs you deem
appropriate.
Reporting output
When REPORT SYSMODs processing is complete, these reports will provide you
with the information that was requested:
v The SYSMOD comparison report provides you with a summary of the
SYSMODs found in the input zone, but not found in the comparison zone. It can
help you determine which SYSMODs might need to be installed in the
comparison zone so its content reflects that of the input zone.
REPORT processing, and supplies information about these data sets.
For a more complete description of the REPORT commands, additional examples,

and samples of actual reports, see SMP/E Reports in SMP/E Commands.
SMP/E CSI application programming interface

The SMP/E CSI application program interface (GIMAPI) allows you to write
application programs that have read-only access to data stored in SMP/E’s CSI
(Consolidated Software Inventory) data sets. GIMAPI is described in detail in
SMP/E Reference.
Summary
Let’s summarize what you have learned about using the Query dialogs, the LIST
command, the REPORT command, and the CSI API to check SMP/E’s records for
your system:
v Query dialogs: Easy and fast way to obtain information
v LIST command: Best for hardcopy listing
v REPORT commands: Best for checking and comparing zone contents
v SMP/E CSI application programming interface: Best for writing an application
program to query the contents of your system’s CSI data sets.


Chapter 2. SMP/E concepts
This chapter summarizes some basic concepts that you will need to understand
before you can use SMP/E. It briefly describes:
v What SMP/E is
v What system modifications are
v The data sets used by SMP/E
v How SMP/E can help you install and maintain products, and monitor changes
to products
What is SMP/E?
SMP/E is the basic tool for installing and maintaining software in OS/390 or z/OS
systems and subsystems. It controls these changes at the element level by:
v Selecting the proper levels of elements to be installed from a large number of
potential changes
v Calling system utility programs to install the changes
v Keeping records of the installed changes
SMP/E is an integral part of the installation, service, and maintenance processes

for CBPDOs, ProductPacs, RefreshPacs, and selective follow-on service for
CustomPacs. In addition, SMP/E can be used to install and service any software
that is packaged in SMP/E system modification (SYSMOD) format.
SMP/E can be run either using batch jobs or using dialogs under Interactive
System Productivity Facility/Program Development Facility (ISPF/PDF). SMP/E
dialogs help you interactively query the SMP/E database, as well as create and
submit jobs to process SMP/E commands.
These are some of the types of software that can be installed by SMP/E:
v Products and service provided in CBPDOs and CustomPac offerings
v Products and service from IBM Software Distribution Centers not provided in
CBPDOs or CustomPac offerings
v Service provided in Expanded Service Options (ESOs)
v Other products and service
SMP/E can install software from any of these sources, provided it is packaged as a
system modification, or SYSMOD.
What are SYSMODs?

Software, whether it is a product or service, consists of elements such as macros,
modules, source, and other types of data (such as CLISTs or sample procedures).
For software to be installed by SMP/E, it must include control information for the
elements. This information describes the elements and any relationships the
software has with other products or service that may also be installed on the same
system. The combination of elements and control information is called a system
modification, or SYSMOD.
There are four types of SYSMODs:

SMP/E concepts
v Function SYSMODs (or functions). These introduce a new product, a new

version or release of a product, or updated functions for an existing product into
the system.
There are two types of function SYSMODs:
– A base function either adds or replaces an entire functional area in the system.
Examples of base functions are SMP/E and MVS.
– A dependent function provides an addition to an existing functional area in the
system. It is called dependent because its installation depends on a base
function already being installed. Examples of dependent functions are the
language features for SMP/E.
v PTFs. These are IBM-supplied, tested fixes for reported problems. They are
meant to be installed in all environments. PTFs may be used as preventive
service to avoid certain known problems that may have not yet appeared on
your system, or they may be used as corrective service to fix problems you have
already encountered. The installation of a PTF must always be preceded by that
of a function SYSMOD, and often other PTFs as well.
v APAR fixes. Authorized program analysis reports (APARs) are temporary fixes
designed to fix or bypass a problem for the first reporter of the problem. These
fixes may not be applicable to your environment. The installation of an APAR
must always be preceded by that of a function SYSMOD, and sometimes of a
particular PTF. That is, an APAR is designed to be installed on a particular
preventive-service level of an element.
v User modifications (USERMODs). These are SYSMODs built by you, either to
change IBM code or to add independent functions to the system. The installation
of a USERMOD must always be preceded by that of a function SYSMOD,
sometimes certain PTFs, APAR fixes, or other USERMODs.
Note: If you want to package a user application program or new system

function in SMP/E format, the correct way is to build a base or
dependent function SYSMOD, not a USERMOD.
Figure 21 on page 39 shows the hierarchy of the various SYSMOD types. This
example shows two service chains: one for the base function HZY1101 and one for
the dependent function JZY1121.

SMP/E concepts
Figure 21. Example of a SYSMOD hierarchy
SMP/E keeps track of the functional and service levels of each element and uses
the SYSMOD hierarchy to determine such things as which functional and service
levels of an element should be installed and the correct order for installing updates
for elements. For more information about how SMP/E determines the processing
order of changes, as well as the functional and service levels of elements, see The
APPLY Command and The ACCEPT Command in SMP/E Commands.
Data sets used by SMP/E

When SMP/E processes SYSMODs, it installs the elements in the appropriate
libraries and updates its own records of the processing it has done. SMP/E installs
program elements into two types of libraries:
v Target libraries contain the executable code needed to run your system (for
example, the libraries from which you run your production system or your test
system).
v Distribution libraries (DLIBs) contain the master copy of each element for a
system. They are used as input to the SMP/E GENERATE command or the
system generation process to build target libraries for a new system. They are
also used by SMP/E for backup when elements in the target libraries have to be
replaced or updated.
To install elements in these libraries, SMP/E uses a database made up of several

types of data sets:
v SMPCSI (CSI) data sets are Virtual Sequential Access Method (VSAM) data sets
used to control the installation process and record the results of processing.
Chapter 2. SMP/E concepts 39

SMP/E concepts
The CSI data set is a VSAM data set in which SMP/E maintains information
about the system. A CSI can be divided into multiple partitions through the
VSAM key structure. Each partition is referred to as a zone.
There are three types of zones:
– A single global zone is used to record information about SYSMODs that have
been received into the SMPPTS data set. The global zone also contains
information enabling SMP/E to access the other two types of zones,
information about system utilities that SMP/E calls to install elements from
SYSMODs, and information allowing you to tailor SMP/E processing.
– One or more target zones are used to record information about the status and
structure of the operating system (or target) libraries. Each target zone also
points to the related distribution zone, which can be used during APPLY,
RESTORE, and LINK when SMP/E is processing a SYSMOD and needs to
check the level of the elements in the distribution libraries.
– One or more distribution zones are used to record information about the
status and structure of the distribution libraries (DLIBs). Each DLIB zone also
points to the related target zone, which is used when SMP/E is accepting a
SYSMOD and needs to check if the SYSMOD has already been applied.
Figure 22 shows the relationships between SMP/E zones and libraries.
Figure 22. Summary of zone relationships
There can be more than one zone in an SMPCSI data set (in fact, there can be up
to 32766 zones per data set). For example, an SMPCSI data set can contain a
global zone, several target zones, and several distribution zones. The zones can
also be in separate SMPCSI data sets. One SMPCSI data set can contain just the
global zone, a second SMPCSI data set the target zones, and a third SMPCSI
data set the distribution zones. For more information on ways to structure
SMPCSI data sets, see the SMP/E Reference manual.
v An SMPPTS (PTS) data set is a data set for temporary storage of SYSMODs
waiting to be installed.

SMP/E concepts
The PTS is used strictly as a storage data set for SYSMODs. The RECEIVE
command stores SYSMODs directly on the PTS without any modifications of
SMP/E information. The PTS is related to the global zone in that both data sets
contain information about the received SYSMODs. Only one PTS can be used for
a given global zone. Therefore, you can look at the global zone and the PTS as a
pair of data sets that must be processed (for example, deleted, saved, or
modified) concurrently.
v The SMPSCDS (SCDS) data set contains backup copies of target zone entries
modified during APPLY processing. Therefore, each SCDS is directly related to a
specific target zone, and each target zone must have its own SCDS.
SCDS data sets are used by SMP/E to store backup copies of target zone entries
modified during APPLY processing. Therefore, each SCDS is directly related to a
specific target zone, and each target zone must have its own SCDS.
SMP/E also uses the following data sets:

v The SMPMTS (MTS) data set is a library in which SMP/E stores copies of
macros during installation when no other target macro library is identified.
Therefore, the MTS is related to a specific target zone, and each target zone must
have its own MTS data set.
v The SMPSTS (STS) data set is a library in which SMP/E stores copies of source
during installation when no other target source library is identified. Therefore,
the STS is related to a specific target zone, and each target zone must have its
own STS data set.
v The SMPLTS (LTS) data set is a library that maintains the base version of a load
module. The load module in this library specifies a SYSLIB allocation in order to
implicitly include modules. Therefore, the LTS is related to a specific target zone,
and each target zone must have its own LTS data set.
v Other utility and work data sets.
SMP/E uses information in the CSI data sets to select proper element levels for
installation, to determine which libraries should contain which elements, and to
identify which system utilities should be called for the installation.
System programmers can also use the CSI data sets to obtain the latest information
on the structure, content, and status of the system. SMP/E provides this
information in reports, listings, and dialogs to help you:
v Investigate function and service levels
v Understand intersections and relationships of SYSMODs (either installed or
waiting to be installed)
v Build job streams for SMP/E processing
How SMP/E can help you install and maintain products

This section briefly describes the tasks involved in general SMP/E processing,
installing SYSMODs, and monitoring and maintaining your system, as well as the
commands used to accomplish these tasks. Following is a guide to the commands
that help you with these tasks. (You can use SMP/E dialogs for all the tasks in this
list.) For more information about these commands, see the SMP/E Commands
manual.
Where to begin
You must specify a SET command before SMP/E can process any other commands.

SMP/E concepts
Specifying the zone to be processed: SET

The main SMP/E database is a set of CSIs. When you establish the SMP/E
database, you use the UCLIN command or the administration dialogs to divide the
CSI into one or more partitions called zones. A zone contains control information
for the following:
v A set of target libraries. (These zones are called target zones.)
v A set of distribution libraries. (These zones are called distribution zones.)
v The data present in the PTS and for general SMP/E processing. (This zone is
called the global zone.)
You must tell SMP/E which zone you are working with before it can execute any
other commands. You direct SMP/E processing to a specific zone by coding the
zone name on the SET command.
Installing SYSMODs
The primary purpose of SMP/E is to install SYSMODs. This section describes the
tasks and commands you can use.
Loading SYSMODs into the SMPPTS: RECEIVE

The RECEIVE command performs the following functions:
v Reads in SYSMODs from a sequential input file, defined by the SMPPTFIN DD
statement
v Reads in HOLDDATA for exception SYSMODs from another sequential input
file, defined by the SMPHOLD DD statement
v Determines which SYSMODs and HOLDDATA are applicable to your system
v Stores the SYSMODs and HOLDDATA in a storage data set (the PTS data set)
and in the global zone
Note: The RECEIVE command can also be used to transfer software packages from
a TCP/IP network connected server directly into an SMP/E directory or
data sets.
Using internet service retrieval to request PTF or HOLDDATA:

RECEIVE ORDER
You can use the RECEIVE ORDER command to request a new PTF service order
from the IBM Automated Delivery Request server, or to download pending PTF
service orders that resulted from a previous request.
Requesting a new PTF service order: The RECEIVE ORDER command performs
the following functions:
1. Builds a software inventory based on the target zones specified. If no zones
were specified, then SMP/E uses all target zones.
2. Submits a HOLDDATA or PTF order request to the IBM Automated Delivery
Request server. The order request includes the content criteria specified by the
user and contains the software inventory.
3. The server responds to SMP/E to indicate that the order was accepted.
4. SMP/E records information about the order in the global zone by creating a
new ORDER entry. At this point the order is in a pending state, and remains so
until its package is downloaded.
5. When the server receives the order request, it initiates the manufacturing
processes to fulfill the HOLDDATA or PTF order. The IBM manufacturing

SMP/E concepts
processes use the supplied content criteria and the software inventory
information to build a package containing either PTFs or HOLDDATA, or both,
as specified in the request.
6. SMP/E waits a period of time for the order to be fulfilled. During this time
SMP/E periodically queries the status of the order on the server. If the order is
not fulfilled within the maximum allowed time, then the RECEIVE command
processing ends and the order remains in a pending state.
7. If the order is fulfilled within the allowed time, the server sends SMP/E the
information required to download the resultant package (FTP server name,
directory where the package resides on the FTP server, userid, password, and
the package SHA-1 hash value).
8. SMP/E downloads the package to the local SMPNTS directory using FTP and
the capabilities of the RECEIVE FROMNETWORK command. Once the package
has been completely downloaded, SMP/E changes the status value for the
order’s ORDER entry in the global zone from PENDING to DOWNLOADED.
9. SMP/E expands the contents of the downloaded package and traditional
RECEIVE processing stores PTFs and HOLDDATA into the global zone and
SMPPTS data set (SMP/E skips this step if the user specified TRANSFERONLY
on the RECEIVE command).
Downloading pending orders: When a requested PTF service order is not

fulfilled within the allowed time, RECEIVE command processing stops and the
order remains in the pending state. The IBM Automated Delivery Request server,
however, continues to fulfill the order. You can use the RECEIVE ORDER
command to process a pending order.
When the RECEIVE command is executed to process a pending order, SMP/E

performs the following steps:
1. Reads the global zone to find the entry for the specified ORDER and
determines if the order has a status of PENDING.
2. Queries the status for the pending order on the server. The server responds to
indicate either that the order has been fulfilled and is ready to be downloaded,
or that the order is not yet ready.
3. If the order is not yet ready for downloading, SMP/E again waits a period of
time for the order to be fulfilled and be ready for downloading. During this
time, SMP/E periodically queries the status of the order on the server. If the
order is not fulfilled within the maximum allowed time, then RECEIVE
processing ends and the order remains in a pending state.
4. If the order is fulfilled within the maximum allowed time, the server responds
to SMP/E with the information required to download the resultant package
(FTP server, directory where the package resides on the server, userid,
password, and the package SHA-1 hash value).
5. SMP/E downloads the package to the local SMPNTS directory using FTP and
the capabilities of the RECEIVE FROMNETWORK command. Once the package
has been completely downloaded, SMP/E changes the status value for the
order’s ORDER entry in the global zone from PENDING to DOWNLOADED.
6. SMP/E expands the contents of the downloaded package and traditional
RECEIVE processing stores PTFs and HOLDDATA into the global zone and
SMPPTS data set (SMP/E skips this step if the user specified TRANSFERONLY
on the RECEIVE command).
Installing SYSMODs in the target libraries: APPLY

After the SYSMODs have been received, you can use the APPLY command to
direct SMP/E to install them in the appropriate target libraries.

SMP/E concepts
Installing SYSMODs in the distribution libraries: ACCEPT

After installing a SYSMOD in the target libraries and testing it to your satisfaction,
you can use the ACCEPT command to have SMP/E install that SYSMOD in the
distribution libraries and delete it from the PTS and the global zone.
Note: Though the usual order of processing a SYSMOD is RECEIVE, APPLY, and
then ACCEPT, you skip the APPLY step and go directly from RECEIVE to
ACCEPT. For further information, see Chapter 5, “Installing a new
function,” on page 101.
Building a system from a set of distribution libraries: GENERATE

Use the GENERATE command to create a job stream that builds a system (a set of
target libraries) from a set of distribution libraries.
Monitoring your system

This section describes the tasks and commands you can use to monitor your
system.
Displaying information from the SMP/E database: LIST

Use the Query dialogs or the LIST command to display the data SMP/E retains.
There are several operands you can use to list subsets of the data.
Checking and comparing zone contents: REPORT

The REPORT command helps you obtain information about SYSMODs installed on
your system. There are several types of REPORT commands that you can use to do
the following:
v REPORT CROSSZONE to list conditional requisites that must be installed in
certain zones because of SYSMODs installed in other zones. This information can
help you synchronize service for related products that are in different zones.
v REPORT ERRSYSMODS to list SYSMODs that have already been installed but
are affected by error holds subsequently received.
v REPORT SOURCEID to list source IDs assigned to SYSMODs in a given zone
or ZONESET.
v REPORT SYSMODS to list SYSMODs installed in one zone and applicable to a
second zone, but not yet installed in it.
Managing the SMP/E database

This section describes the tasks and commands you can use to remove SYSMODs
and process the SMP/E database as a part of your system maintenance.
Notes:
1. There is no SMP/E command to remove a SYSMOD from the distribution
libraries once you have accepted it.
2. Because all SMP/E processing is controlled by information in the CSI data set,
you must provide the required information in the CSI before you process any
SYSMODs.
Removing SYSMODs from the target libraries: RESTORE

After you have installed a SYSMOD in the target libraries, you should test it to
make sure the change works correctly. If during testing you find an error in one of
the SYSMODs you applied, you can use the RESTORE command to remove that
SYSMOD from the system.

SMP/E concepts
The RESTORE command causes the elements in the distribution libraries to be

reinstalled on the target libraries. Therefore, when restoring a SYSMOD, you may
have to restore more than the one SYSMOD in error. All SYSMODs that have not
been accepted and that affect some of the same elements or need the same service
level requisites as the one you are restoring must also be restored.
Removing SYSMODs from the PTS: REJECT

After receiving a SYSMOD, you can use the REJECT command to remove that
SYSMOD from the PTS and the global zone. This can be done to rework a
SYSMOD.
You can also use REJECT after you use NOPURGE to accept a SYSMOD into the
distribution libraries. Using NOPURGE in the OPTIONS entry prevents SMP/E
from deleting the global zone SYSMOD entry and the PTS MCS entry during
ACCEPT processing. Later, you can delete the SYSMOD and MCS entries by using
REJECT.
Processing the SMP/E database to create, update, or delete a

single entry: UCLIN
You can use the Administration dialogs or the UCLIN command to create or
change entries in CSI data sets. UCLIN can be used to create entries required
during SMP/E processing, such as entries for:
v Identifying the utilities SMP/E is to use
v Providing information for dynamic allocation support
You can use UCLIN to delete orders from the SMPCSI data set. You can also use
UCLIN to correct errors in entries or to alter processing information (including the
ORDER RETENTION subentry in the OPTIONS entry).
Note: This command should be used with extreme caution; an incorrect change
can cause a SYSMOD to be installed incorrectly later.
UCLIN updates only entries in SMP/E data sets. It does nothing to any elements
or load modules in any product libraries. You must ensure that the appropriate
changes are made to the libraries.
Processing the SMP/E database to update several entries of the

same type: ZONEEDIT
Use the ZONEEDIT command to quickly change the values for a field in different
DDDEF or UTILITY entries in the same zone. You can also use ZONEEDIT to
change the cross-zone subentries of MOD, LMOD, and TARGETZONE entries.
Processing the SMP/E database to define the structure of the

target libraries: JCLIN
To install a SYSMOD successfully, SMP/E must have information about the
structure of your target libraries, such as:
v The library in which an element resides
v How modules are link-edited together to form load modules
v Where the load modules exist and their characteristics
The JCLIN command enables you to define the target library structure. In some
instances, such as defining the target library structure for data elements, it is not
necessary to use JCLIN, because the definition in the MCS statement is sufficient.
When processing the JCLIN command, you provide SMP/E with a job stream
containing all the job steps (such as copies, link-edits, and assemblies) needed to

SMP/E concepts
create a set of target libraries from a set of distribution libraries. SMP/E then scans
that input and builds all required entries to define the target system structure.
Processing the SMP/E database to write information to the

SMPLOG data set: LOG
Use the LOG command to write to the SMPLOG and SMPLOGA data sets. This is
useful for maintaining documentation about your system, such as who installed a
certain SYSMOD, and why.
Processing the SMP/E database to clean up the SMPLTS,

SMPMTS, SMPSTS, and SMPSCDS data sets: CLEANUP
Use the CLEANUP command to delete entries from the SMPMTS, SMPSTS, and
SMPSCDS data sets for a particular target zone, if any entries still exist after the
associated SYSMOD has been accepted into the related distribution zone.The
CLEANUP command can also be used to remove the base version of load modules
that are no longer needed from the SMPLTS data set.
Managing zones
This section describes the tasks and commands you can use to manage zones.
Copying a zone from one CSI to another: ZONECOPY

Use the ZONECOPY command to copy an entire target or distribution zone from
one CSI data set to another. When you use the ZONECOPY command, SMP/E
checks that the zone you copy into is empty except for a ZPOOL record.
Copying a zone to a sequential data set: ZONEEXPORT

The ZONEEXPORT command creates a copy of a specified distribution, target, or
global zone and places it into another data set (a variable-block sequential data
set). Having this copy of a specified zone gives you two advantages:
v You can use it as a backup copy to recreate a zone that has been accidentally
erased or destroyed.
v You can use it as a “transportable” copy to create the same zone on another
system or in another CSI data set on the same system.
Copying a zone from a sequential data set to a CSI data set:

ZONEIMPORT
Use the ZONEIMPORT command to reload an exported zone into a distribution,
target, or global zone. ZONEIMPORT can be used only with output from
ZONEEXPORT.
Deleting a zone: ZONEDELETE

Sometimes you need to delete the SMP/E data related to one of the systems you
are supporting. These are some examples of when you need to delete a zone:
v After a full system generation, you have to delete the information describing the
previous target system libraries. Then you rebuild that information to describe
the new set of target system libraries built from the distribution libraries.
v After installing a new level of a product that existed in its own target zone and
distribution zone, you want to delete the information about the old level of the
product and continue processing only the new level.
Merging zones: ZONEMERGE

Use the ZONEMERGE command to merge one zone into another zone after you
use the GENERATE command or do a full system generation. When you use the
ZONEMERGE command, SMP/E looks through the zones that ZONEMERGE
works on and merges the data from them.

SMP/E concepts
Renaming a zone: ZONERENAME

Use the ZONERENAME command to rename a zone. This command can help you
keep meaningful names for your zones. You can also use ZONERENAME after a
GENERATE command or a full system generation to help you change the name of
the distribution zone that GENERATE or the system generation uses to build a
new target zone.
Linking and relinking modules

This section describes the tasks and commands you can use to link or relink
modules into load modules.
Handling cross-zone link-edits: LINK MODULE

Use the LINK MODULE command to link modules in one zone with load modules
in another zone and to track this inclusion so that subsequent APPLY and
RESTORE processing can automatically maintain the affected load modules.
Relinking load modules that use CALLLIBS: LINK LMODS

Use the LINK LMODS command to relink load modules that have a CALLLIBS
subentry list in their LMOD entry (and therefore use the automatic library call
option to implicitly include modules from a specified library). The LINK LMODS
command may also be used to relink specific load modules that do not contain
CALLLIBS, rebuilding them from scratch if necessary.
General SMP/E processing

This section describes general SMP/E processing tasks and the commands you can
use to do them.
Requesting diagnostic processing: DEBUG

Use the DEBUG command to display the name of the issuing routine in each
SMP/E message, to dump SMP/E control blocks, to dump VSAM request
parameter list (RPL) control blocks, and to get a SNAP dump of SMP/E and its
work areas whenever a specified message is issued.
Resetting return codes: RESETRC

Normally, the execution of an SMP/E command depends on the successful
execution of all preceding commands. However, you can use the RESETRC
command to reset the return code to zero. This allows SMP/E to run the next
command, even if a preceding command failed.

Chapter 3. Preparing to use SMP/E
This chapter discusses how to prepare to use SMP/E after it has been installed. It
describes how to do the following:
1. Allocate and initialize data sets in the SMP/E database.
2. Define entries in the CSI data set to do the following:
v Create the zones associated with the PTS, distribution libraries, and target
libraries.
v Define the product and SMP/E libraries used during SMP/E processing.
v Define the utility programs and associated parameters used during SMP/E
processing.
v Define the libraries that are eligible for retry processing after x37 abends.
3. Connect the SMP/E dialogs to ISPF
4. Set up SMP/E for easier operation:
v Specify SMP/E OPTIONS entry
v Specify link edit utility output DDDEF entries
v Specify automatic cross-zone requisite checking
5. Define the information needed to invoke SMP/E.
6. Define exit routines, if desired, to customize SMP/E processing.
Allocating and initializing data sets in the SMP/E database

To install SYSMODs, SMP/E needs information about them and about the target
and distribution libraries where they are to be installed. This information is kept in
a database composed of the following data sets:
v SMPCSI (CSI)
v SMPPTS (PTS)
v SMPSCDS (SCDS)
CSI data sets

SMP/E uses CSIs to keep records of the system. To define the CSI data sets for
your system, you need to do the following:
1. Decide how to organize the CSI data sets.
2. Understand catalog considerations for CSI data sets.
3. Allocate the CSI data sets.
4. Define alias entries for user catalogs.
5. Initialize the CSI data sets.
6. Define the zones for your system.
7. Understand how to reorganize a CSI data set to reclaim space.
Deciding how to organize CSI data sets

Before you allocate any CSI data sets, you must decide how to organize those data
sets. Consider the following when you make your decision:
v The DASD configuration of your system libraries
v The organization of your system support structure
v How you want to use SMP/E
There are two basic structures for CSI data sets:

v Single-CSI
v Multiple-CSI

Allocating and initializing data sets
Descriptions of these structures are followed by examples.
Single-CSI structure: You can define the CSI structure to have one CSI that keeps
track of all your system activity. The single-CSI data set has one global zone and
one or more target and distribution zones. These are some reasons for having a
single-CSI data set:
v The single-CSI data set optimizes the use of direct access storage.
v The single-CSI data set puts your whole establishment in one VSAM data set.
This provides you with a single control point and one source of information for
your whole system.
Single-CSI systems do have a drawback. When SMP/E needs to process a zone, it

cannot request access to that specific zone; it must request access to the CSI data
set containing that zone. As a result, if you have a single-CSI system, you can run
only one background SMP/E job at a time.
Figure 23 shows a single-CSI data set structure.
Figure 23. Sample single-CSI structure
Multiple-CSI structure: Multiple CSI data sets can be:

v Separate from each other, each with its own global zone.
v Connected by ZONEINDEX entries to a single global zone. (The global zone
must be in one of the CSI data sets.)

Multiple CSIs enable you to use more than one VSAM data set for the global,
target, and distribution zones. These are some reasons for having multiple CSI data
sets:
v Your system may need multiple CSIs because of the characteristics of a
particular installation—its programming support, its backup and update needs,
and its need for added security and data integrity. For example, keeping libraries
and their associated zones synchronized when you dump them for backup is
easier if you keep them on the same physical DASD.
v Your system may need multiple CSIs if the support teams for different
subsystems—such as MVS, CICS, IMS, and NCP—are at different places.
v You may want to be able to run more than one background SMP/E job at a
time. When SMP/E needs to process a zone, it cannot request access to that
specific zone; it must request access to the CSI data set containing that zone. If
your zones are in separate CSI data sets, processing for one zone does not
prevent access to another zone.
Figure 24 shows a multiple-CSI data set structure.
Figure 24. Sample multiple-CSI structure
Chapter 3. Preparing to use SMP/E 51

Examples of CSI structures: The following examples show several ways to define
a CSI structure describing a sample z/OS system with Job Entry Subsystem 3
(JES3) and an Information Management System (IMS) subsystem. In some of the
examples, a single CSI contains multiple zones. Others show zones that are
separated into multiple CSI data sets. Zones in separate CSIs can be connected by
a single global zone. The CSI that contains the global zone is called the master CSI.
Example 1: Using a separate global zone for each subsystem: In Figure 25, the existing
DASD structure for the libraries is maintained, and a separate global zone is
defined for each of the subsystems (MVS, JES3, and IMS). This CSI structure keeps
control of the three subsystems separate. You might use this structure if the system
programming support for the three subsystems is organizationally separate. A
disadvantage is that there is no single control point (global zone) from which to
manage or query the entire system.
Figure 25. Using a separate global zone for each subsystem
Example 2: Using a single CSI for the whole system: In Figure 26 on page 53, all the
SMP/E control information for the system is contained in a single CSI. The system
structure is reflected by separate zones for MVS, JES3, and IMS. This CSI structure
provides a single control point from which to manage or query the entire system.

Figure 26. Using one CSI for the whole system
Example 3: Using a master CSI and multiple CSIs: In Figure 27 on page 54, one
global zone is defined for the entire system in a master CSI, and separate CSIs are
allocated for the JES3 and IMS subsystems on the packs where the subsystem
libraries reside. This CSI structure provides the advantages of a common control
point (as in Example 2) and keeps the SMP/E control information physically
associated with the libraries it describes. This is useful when you dump packs for
backup.

Figure 27. Using a master CSI
Example 4: Using a master CSI and a separate CSI for each zone: In Figure 28 on page
55, one global zone is defined for the entire system in a master CSI, and separate
CSIs are allocated for the JES3 and IMS subsystems on the packs where the
subsystem libraries reside. Unlike Example 3, each zone is in its own separate CSI.
This CSI structure provides the advantages of a common control point (as in
Example 2) and keeps the SMP/E control information physically associated with
the libraries it describes. This is useful when you dump packs for backup.

Figure 28. Using a master CSI and a separate CSI for each zone
Example 5: Using a master CSI and one CSI per SREL: In Figure 29 on page 56, one
global zone is defined for the entire system in a master CSI, and separate CSIs are
allocated for each SREL (MVS, CICS, NCP, and IMS/DB2). The target zones and
DLIB zones associated with a given SREL are in the same CSI. This CSI structure
provides the advantages of a common control point through one global zone (as in
Example 2) and keeps the SMP/E control information physically associated with
the libraries it describes. This is useful when you dump packs for backup.

Figure 29. Using a master CSI and one CSI per SREL
Catalog considerations
When you catalog the CSI data sets used by SMP/E, remember these
considerations:
v User catalogs: You should catalog each CSI in a user catalog, not in the master
catalog. However, the user catalog does not need to be on the same volume as
the CSI.
v Alias entries for user catalogs: Catalog information should be accessible through
your master catalog. You can do it by defining each user catalog as an alias in
the master catalog. For an example of defining an alias for a user catalog, see
“Defining an alias entry for a user catalog” on page 58. Defining alias entries for
user catalogs enables you to access all the CSI data sets and it eliminates the
need to restore both the DASD containing the master catalog and the DASD
containing the CSI after an I/O error.
Allocating a CSI data set

CSI data sets are keyed VSAM clusters and are allocated by use of access method
services. For additional information and a description of the parameters, see z/OS
DFSMS Access Method Services for Catalogs.
The GIMSAMPU member in SYS1.SAMPLIB is a sample job to allocate and prime

CSI and SMP/E operational data sets. The following sample job step, which is
taken from the sample job in GIMSAMPU, allocates a CSI data set with enough
space to have multiple target or distribution zones and then initializes the CSI with
the zpool record:
//DEFZONES EXEC PGM=IDCAMS
//GIMZPOOL DD DSN=SYS1.MACLIB(GIMZPOOL),DISP=SHR

//SYSIN DD *
DEFINE CLUSTER( +
NAME(SMPE.GLOBAL.CSI) +
VOLUMES(volid) +
CYLINDERS(100 10) +
FREESPACE(10 5) +
KEYS(24 0) +
RECORDSIZE(24 143) +
SHAREOPTIONS(2 3) +
) +
DATA ( +
NAME(SMPE.GLOBAL.CSI.DATA) +
CONTROLINTERVALSIZE(8192) +
) +
INDEX (NAME(SMPE.GLOBAL.CSI.INDEX) +
CONTROLINTERVALSIZE(4096) +
)
REPRO INFILE(GIMZPOOL) +
OUTDATASET(SMPE.GLOBAL.CSI)
/*
In your own job, be sure to include:

v NAME
These are the naming conventions for CSI data sets:
– The high-level qualifier must not be SYS1 if the CSI data set is cataloged in a
user catalog rather than in the master catalog. However, the user catalog
should be accessible through an alias entry in the master catalog.
– The low-level qualifier must be CSI.
These are examples of SMP/E data set names:
’SMPE.SMPCSI.CSI’
’PP.SMPCSI.CSI’
’IMS.SMPCSI.CSI’
’TEST.CSI’
v KEYS(24 0)
v RECORDSIZE(24 143)
v SHAREOPTIONS(2 3)
SMP/E requires 2 as the cross-region SHAREOPTIONS value. It uses the default
value of 3 as the cross-system SHAREOPTIONS value. Because SMP/E does not
support cross-system sharing of the CSI, you cannot specify 4 as the
cross-system value for SHAREOPTIONS. If you want to support cross-zone
sharing, you must either use Global Resource Serialization (GRS) or a similar
function, or ensure that the data set is not updated by multiple systems
simultaneously.
v CONTROLINTERVALSIZE (CISIZE)
If you allocate more than one CSI data set, ensure that they all have the same
data CI size and index CI size. Doing so will allow SMP/E to take advantage of
local shared resources (LSR) and VSAM resource pools. If the CSI data sets have
different CISIZE values, SMP/E may open the data sets without using LSR.
v CYLINDERS
The CYLINDERS value is only an estimated starting value. Your cylinder value
may vary according to the device type, the software arrangement, the amount of
service you install, and the number of CSIs.

Defining an alias entry for a user catalog

After allocating the CSI data sets, you should define alias entries for the high-level
qualifiers of your CSI data sets in your master catalog and relate them to your
SMP/E user catalog.
The following job creates an alias entry in the master catalog for a CSI data set
named SMPE.SMPCSI.CSI that is cataloged in a user catalog named SMPECAT:
//CREATE JOB ’accounting info’,MSGLEVEL=(1,1)
//ALIAS EXEC PGM=IDCAMS
//SYSPRINT DD SYSOUT=A
//SYSIN DD *
DEFINE ALIAS -
(NAME(SMPE) -
RELATE(SMPECAT)) -
CATALOG(AMASTCAT/password)
/*
If the CSI data sets are cataloged in different user catalogs, they must have
different high-level qualifiers.
Defining zones for your system

Once you have allocated and initialized the CSI data sets, you need to create
within them the entries SMP/E uses to maintain your system. The first entries you
need to define are the zone definition entries — GLOBALZONE, TARGETZONE, and
DLIBZONE entries —,which set up zones in CSI data sets.
v GLOBALZONE entry. A global zone is created by defining a GLOBALZONE
entry. The GLOBALZONE entry contains processing-related information for
SMP/E. It is also used by SMP/E as an index to target and distribution zones,
either in the same CSI or in different CSI data sets. The GLOBALZONE entry
must be defined before you can do any other processing for that global zone.
v TARGETZONE entry. A target zone is created by defining a TARGETZONE
entry. The TARGETZONE entry contains information SMP/E uses to process a
specific target zone and the associated target libraries. It must be defined before
you can do any other processing for that target zone.
v DLIBZONE entry. A distribution zone is created by defining a DLIBZONE entry.
The DLIBZONE entry contains the information SMP/E uses to process a specific
distribution zone and the associated distribution libraries. It must be defined
before you can do any other processing for that distribution zone.
Figure 30 on page 59 illustrates how zone definition entries define the relationships
between zones.

Figure 30. Relationships between zone definition entries
After you have defined the zones for your system, you can create other entries.
SMP/E zones contain two basic types of entries:
v Entries controlling SMP/E processing.
You generally define processing control entries through the SMP/E
Administration dialogs or with the UCLIN command. Table 2 summarizes the
information specified in these entries.
v Entries describing the structure and status of the target and distribution libraries.
Status and structure entries are generally created by SMP/E when you install
SYSMODs, run the JCLIN command, or copy entries from one zone to another.
Table 3 on page 60 summarizes the information specified in these entries.
SMP/E provides a member in SYS1.SAMPLIB (GIMSAMPU) containing sample

UCLIN statements to define entries for a basic z/OS system. You can access this
member by use of standard system utilities. The sample definitions are
syntactically correct and can be used as the basis for your CSI entries. This sample
is not complete for all systems, but it is an example of the types of information
various entries need. For examples of UCLIN to define entries, see The UCLIN
Command in SMP/E Commands, which shows the UCLIN syntax for each entry
type, and SMP/E Data Set Entries in SMP/E Reference, which contains a description
of the syntax plus examples and notes on its use.
Table 2. Entries controlling SMP/E processing
Zone where defined
Type of information Entry type Global Target DLIB
1
Data set definitions for dynamic allocation DDDEF X X X
DLIB zone processing information DLIBZONE X
2 2
Exception (held) SYSMODs HOLDDATA X
FMIDs to limit the SYSMODs processed by an FMIDSET X
SMP/E command

Table 2. Entries controlling SMP/E processing (continued)

Zone where defined
Global zone processing information GLOBALZONE X
3
Processing options OPTIONS X
Target zone processing information TARGETZONE X
4
Utility program parameters UTILITY X
Zone names to limit the SYSMODs processed by ZONESET X
an SMP/E command
Notes:
1. For more information about dynamically allocating data sets, see “How to dynamically allocate data sets to be
used during SMP/E processing” on page 62.
2. For more information about processing exception SYSMODs, see Chapter 9, “Managing exception SYSMODs.”
HOLDDATA entries cannot be updated by UCLIN or the Administration dialogs.
3. For more information about defining information to be used during SMP/E’s retry processing after x37 abends,
see “Recovering after errors from utility processing” on page 69.
4. For more information about defining utility programs and associated parameters, see “Defining utility programs
and associated parameters to SMP/E” on page 64.
Table 3. Entries describing the status and structure of the target and distribution libraries
Zone where defined
Assembler statements that can be assembled to ASSEM X X
create an object module
Data elements installed in the target or Data element entries X X
distribution libraries (data elements are
elements other than macros, modules, or
source)
Distribution libraries that were totally copied to DLIB X X
target libraries
Elements installed in a UNIX file system Hierarchical file X X
system element entries
Java Archive files JAR X X
Java Archive update files JARUPD X X
Load module information LMOD X X
Macros that have been installed in the target or MAC X X
distribution libraries
Module source that has been installed in the SRC X X
target or distribution libraries
Modules used to create load modules in the MOD X X
target libraries
Program objects PROGRAM X X
Software product information PRODUCT X
Software product feature information FEATURE X
SYSMODs that have been processed SYSMOD X X X

Reorganizing a CSI data set to reclaim space

During normal SMP/E processing, VSAM control interval splits and control area
splits can occur. These splits use up some of the free space in each control interval
or control area, and this can degrade SMP/E performance and DASD space
utilization. You should monitor your VSAM data sets regularly to determine how
many splits have occurred and how much free space remains. The following job
lists the catalog entry for data set SMPE.SMPCSI.CSI:
//LISTCAT JOB ’accounting info’,MSGLEVEL=(1,1)
//STEP01 EXEC PGM=IDCAMS
//SYSIN DD *
LISTCAT -
ENTRIES(SMPE.SMPCSI.CSI) -
ALL
/*
After examining the LISTCAT output, you may determine that the CSI should be
reorganized to eliminate splits in the control intervals or control areas and to reset
the amount of free space available. This can be done through the access method
services EXPORT and IMPORT commands. Once a CSI has been exported, a new
CSI can be allocated, and the exported CSI can be imported so that normal SMP/E
processing can continue.
Note: These examples are not the only way of compressing the CSI. You may
prefer to use another method, drawing on your experience and knowledge
of VSAM.
The following is a sample job for exporting the CSI:

//EXPORT JOB ’accounting info’,MSGLEVEL=(1,1)
//SMPCSI DD DSN=SMPE.SMPCSI.CSI,DISP=OLD
//TEMPCSI DD DSN=SMPCSI.DATA,DISP=OLD
//SYSIN DD *
EXPORT SMPE.SMPCSI.CSI -
INFILE(SMPCSI) -
OUTFILE(TEMPCSI)
/*
The following is an sample job for importing the CSI:

//IMPORT JOB ’accounting info’,MSGLEVEL=(1,1)
//SMPCSI DD DSN=SMPE.SMPCSI.CSI,DISP=OLD
//TEMPCSI DD DSN=SMPCSI.DATA,DISP=OLD
//SYSIN DD *
IMPORT INFILE(TEMPCSI) -
OUTFILE(SMPCSI) -
INTOEMPTY
/*
Notes:
1. If you want to delete the original CSI (SMPE.SMPCSI.CSI) when the exported
copy (SMPCSI.DATA) is created, do not use the IDCAMS TEMPORARY
keyword on the EXPORT command.
If you want to make a backup copy of the CSI, you can use the TEMPORARY
keyword on the EXPORT command to keep the original CSI intact.
2. Use a sequential data set to receive the exported CSI.

3. After allocating a new CSI to be imported into, do not prime it with the
GIMZPOOL record provided in SYS1.MACLIB; if you do, the import operation
will fail.
PTS data sets

The PTS data set is used as temporary storage for SYSMODs. It contains one
member for each SYSMOD received. Each member is called a modification control
statement (MCS) entry and is an exact copy of the SYSMOD as it was received
from the SMPPTFIN data set. The name of an MCS entry matches the ID of the
SYSMOD it contains. Generally, the MCS entries are kept on the PTS until the
SYSMOD is accepted; then, under normal processing, they are deleted.
Each PTS data set must be associated with only one global zone. The allocation of
space and directory blocks for the SMPPTS depends on your plans for installing
and maintaining the functions managed by the global zone. For more information
about allocating the SMPPTS data set, see SMP/E Reference.
SCDS data sets

The SCDS data set contains backup copies of target-zone entries that are modified
during APPLY processing. These backup copies are made before the entries are (1)
changed by inline JCLIN, a ++MOVE MCS, or a ++RENAME MCS or (2) deleted
by an element MCS with the DELETE operand. The backup copies are used during
RESTORE processing to return the entries to the way they were before APPLY
processing.
Each backup copy of an entry is associated with the SYSMOD that caused the
entry to be backed up. Together, the collection of entries associated with a
SYSMOD is called the BACKUP entry for that SYSMOD. When you process the
SCDS (for example, to list entries), you can specify only BACKUP entries; you
cannot process individual entries within a BACKUP entry.
Each target zone within a CSI must have its own SCDS. The correct SCDS to be
used during processing is determined either by the SMPSCDS DDDEF entry in
each specific target zone or by a DD statement in the JCL. An SCDS can be
allocated the same way as any normal partitioned data set. The allocation of space
and directory blocks for this data set depends on your plans for installing and
maintaining functions. For more information about allocating the SMPSCDS data
set, see SMP/E Reference.
How to dynamically allocate data sets to be used during SMP/E

processing
The processing of SMP/E commands requires a variety of data sets. You can either
provide the DD statements for these data sets (such as in a cataloged procedure) or
have SMP/E allocate the data sets dynamically. Dynamic allocation has the
advantage that data sets are allocated only as they are needed; DD statements
must successfully allocate all data sets, regardless of whether they are needed for
the command being processed.
There are some drawbacks to using DD statements. For example, all the data sets
defined by DD statements must be successfully allocated, regardless of whether
they are needed for the command being processed. In addition, if you are running
several SMP/E commands, you must be careful to use the correct DD statements

for each command. If you are processing zones that are in different CSI data sets,
you must make sure to provide a DD statement that points to each of those zones
and their associated CSIs.
With dynamic allocation, you do not have these problems. Subsequent sections
describe the sources from which SMP/E can get the information it needs to
allocate data sets dynamically and how it chooses which of these sources to use.
Sources of information for dynamic allocation

SMP/E can use several sources of information to allocate data sets dynamically:
v DDDEF entries
v SMPPARM member GIMDDALC
v Standard defaults
DDDEF entries
You can use DDDEF entries to provide SMP/E with information it needs to
allocate any of the following:
v Permanent data sets, such as target libraries, distribution libraries, and SMP/E
data sets
v Temporary data sets
v SYSOUT data sets
v Work data sets
v Pathnames for elements and load modules residing in a UNIX file system
Note: A member of a partitioned data set cannot be specified using a DDDEF

entry.
The name of the DDDEF entry must match the ddname of the data set it describes
and the entry must exist in the zone that uses the data set. DDDEF entries provide
more flexibility than DD statements; they enable different zones to use different
data sets for the same ddname and they use resources more efficiently because
they allow SMP/E to allocate only the data sets it needs.
DDDEF entries can include the following information:

v Data set name
v Unit type
v Volume serial number
v Initial data set status: NEW, OLD, MOD, or SHR
v Final data set status: KEEP, DELETE, or CATALOG
v How the data set is to be allocated: blocks, cylinders, or tracks
v Primary and secondary values for space allocation
v Whether the data set should be RACF-protected
v Whether the data set is SMS-managed
v Directory information used to allocate the pathname for an element or load
module residing in a UNIX file system.
For more information about DDDEF entries, see the SMP/E Reference manual.
SMPPARM member GIMDDALC

Another way to provide SMP/E with information about data sets is through
GIMDDALC control statements in SMPPARM member GIMDDALC. For more
information about GIMDDALC, see the chapter on SMPPARM members in the
SMP/E Reference manual.

Standard defaults: SMPOUT and SYSPRINT are critical for SMP/E to operate
properly. Therefore, in case they are not defined, SMP/E has built-in defaults for
them.
v SMPOUT is allocated either as SYSOUT (for background processing) or to the
terminal (for foreground processing).
v SYSPRINT is allocated as SYSOUT.
How dynamic allocation works

Once SMP/E has determined which data sets are needed for the command it is
processing, SMP/E checks whether DD statements have been provided for any of
those data sets. SMP/E uses information from those DD statements in allocating
the data sets to which they apply. If any data sets lack DD statements, SMP/E
must allocate them dynamically. To get the information it needs to do this, SMP/E
checks the following sources in the order shown.
1. DDDEF entries. If the zone specified on the SET command contains a DDDEF
entry for the required data set, SMP/E uses that entry to allocate the data set.
Otherwise, it checks the next source.
2. SMPPARM member GIMDDALC. If a GIMDDALC control statement has been
defined in SMPPARM member GIMDDALC, SMP/E uses that information to
allocate the data set. Otherwise, it checks the next source.
3. Standard defaults. If the data set is for SMPOUT or SYSPRINT, SMP/E uses
the standard default to allocate the data set. Otherwise, data set allocation fails.
Notes:
1. When a data set is part of a concatenation (such as the SYSLIB concatenation),
SMP/E does not do the previously described checking. All data sets in a
concatenation must be defined the same way. For example, if a DDDEF entry
specifies a concatenation, all the specified entries must also be defined by
DDDEF entries.
2. For the cross-zone phase of APPLY and RESTORE processing, a DD statement
cannot be used to allocate the SYSLIB for a cross-zone load module. This
library can be allocated only through a DDDEF entry in the cross-zone
containing the LMOD entry for the cross-zone load module.
3. The DD name SMPDUMMY is always allocated as a DUMMY data set. SMP/E
ignores any allocation information specified for SMPDUMMY by a DDDEF
entry or GIMDDALC member. If SMPDUMMY was previously allocated
outside of SMP/E, SMP/E frees the SMPDUMMY DD and then reallocates it as
a DUMMY data set.
Defining utility programs and associated parameters to SMP/E

SMP/E calls utility programs to install SYSMODs in the target and distribution
libraries. These utilities must reside in either the LNKLST concatenation or the link
pack area (LPA), as defined in SYS1.PARMLIB. SMP/E uses default utility
programs unless you define OPTIONS entries and UTILITY entries specifying
other utilities and parameters.
v OPTIONS entries point to the specific UTILITY entry to be used for each type of
utility. These are the types of utilities you can point to:
– Access methods services
– Assembler
– Compress
– Copy
– Link-edit
– Retry after x37 abends

Defining utility programs
– Update
– Superzap
v Each UTILITY entry defines the information to be used when invoking a specific
type of utility:
– The name of the utility program
– The maximum utility return code that SMP/E should consider to be
successful
– The ddname to be used for utility output
– Parameters to be passed to the utility
Using default values for utility programs

If you do not define UTILITY entries and OPTIONS entries to specify which utility
programs to use, SMP/E uses default utility programs, as well as its own default
values, for return codes, print values, and the parameters to be passed. Table 4 lists
the default values for the various types of utility programs.
Table 4. Default values for UTILITY entries
NAME (see note
Utility 1) RC PRINT PARM
Access method IDCAMS 0 SYSPRINT
services
Assembler ASMA90 4 SYSPRINT XREF,
NOOBJECT,
DECK
Compress IEBCOPY 0 SYSPRINT
Copy IEBCOPY 0 SYSPRINT SPCLCMOD and
CMWA=256K
(for program
elements and
copied load
modules)
Hierarchical file BPXCOPY 0 SYSPRINT (see
system copy note 3)
| Link-edit utility IEWBLINK 8 SYSPRINT LET, LIST,
NCAL, XREF
(see note 2)
Retry after x37 IEBCOPY 0 SYSPRINT
abends
Update IEBUPDTE 0 SYSPRINT Determined by
SMP/E during
processing
Superzap IMASPZAP 4 SYSPRINT

Table 4. Default values for UTILITY entries (continued)

NAME (see note
Utility 1) RC PRINT PARM
Notes:
1. If you replace a default utility program, the replacement utility program must be
compatible with the default utility it replaces, both in the way it processes any control
statements and execution parameters generated by SMP/E and in the return codes that
it returns to SMP/E.
2. When the load module being link-edited contains a CALLLIBS subentry list, SMP/E
does not always use NCAL by default. In this case, SMP/E uses CALL for the link to
the actual target library or NCAL for the link to the SMPLTS library. SMP/E always
uses NCAL for ACCEPT processing.
3. If SYSTSPRT is specified as the PRINT value, it is ignored and the default of SYSPRINT
is used instead.
Defining values for utility programs

If you want to use utility programs other than the defaults, or if you want to
specify different parameters for the default utility programs, you need to identify
the programs to SMP/E by defining UTILITY entries and OPTIONS entries. For
example, the installation information for a particular product you are about to
install may direct you to use a specific link-edit utility and may indicate that the
maximum successful return code from the utility is 4, not 8.
Figure 31 on page 67 shows how OPTIONS entries, UTILITY entries, zone

definition entries, and the SET command are related.

Figure 31. Relationships of OPTIONS, UTILITY, zone definition entries and the SET
command
These are the basic steps you follow:

1 Define the desired utility name and parameters in a UTILITY entry.
2 Define an OPTIONS entry that points to that UTILITY entry.
3 Put the OPTIONS entry into effect by doing one of the following:
A If the information should be the default for processing a particular
zone, update the associated zone definition entry to point to the desired
OPTIONS entry. The default OPTIONS ENTRY is always used for
processing that zone, unless you override the OPTIONS entry with the
SET command.
B Otherwise, use the SET command to indicate which OPTIONS entry to
use when processing the zone specified on the SET command. The
information in the specified OPTIONS entry overrides the default
OPTIONS entry defined for the zone.
For examples of these steps, see “Example: How to request the desired utility
processing” on page 68. For more detailed information, see OPTIONS Entry,
UTILITY Entry, GLOBALZONE Entry, DLIBZONE Entry, and TARGETZONE Entry
in SMP/E Reference, and the SET command in SMP/E Commands.
Note: You can specify which utility programs SMP/E can call by using the
PROGRAM class of the z/OS Security Server (RACF). Refer to z/OS Security
Server RACF Security Administrator’s Guide for more information about how
to use this function.

Example: How to request the desired utility processing

Table 5 describes the steps you need to follow in order to get the desired utility
processing. For details on syntax and coding considerations, see the UTILITY Entry
(Global Zone) section in SMP/E Reference.
Table 5. How to request the desired utility processing
Steps Sample scenario
You want SMP/E to call a user routine, USERRCVR, rather than IEBCOPY,
1 Define the desired utility name
to recover from x37 abends. This is the processing you need:
and parameters in a UTILITY
entry. v The program must receive parameter TYPE=FAST.
v The output should go to X37PRINT rather than SYSPRINT.
v A return code of 4 or less is acceptable.
v You want to suppress the listing of member names during retry
processing done by your program.
The following UCL defines the desired UTILITY entry for your program:
SET BDY(GLOBAL) /* Set to global zone. */.
UCLIN /* */.
ADD UTILITY(MYX37) /* Retry/recovery program. */
NAME(USERRCVR) /* Program name. */
PARM(TYPE=FAST) /* PARM value. */
PRINT(X37PRINT) /* SYSPRINT ddname. */
RC(4) /* Highest acceptable */
/* return code. */
LIST(NO) /* No list of member names. */
/* */.
ENDUCL /* */.
Once you have created the desired UTILITY entry, you need to point to it
2 Connect the UTILITY entry to
from an OPTIONS entry. The following UCL defines an OPTIONS entry
an OPTIONS entry.
(MYOPT1) that points to UTILITY entry MYX37:
UCLIN /* */.
ADD OPTIONS(MYOPT1) /* New OPTIONS entry. */
RETRY(MYX37) /* Connect to retry. */
/* */.
ENDUCL /* */.
You might want your OPTIONS entry to be the default for processing target
3A Use the zone definition entry to
zone TGT1. In this case, the TARGETZONE entry for TGT1 must point to
specify your OPTIONS entry as
your OPTIONS entry. The following UCL updates the existing
the default OPTIONS entry.
TARGETZONE entry for TGT1 so it points to OPTIONS entry MYOPT1:
SET BDY(TGT1) /* Set to target zone, TGT1.*/.
UCLIN /* */.
REP TARGETZONE /* Update zone definition. */
OPTIONS(MYOPT1) /* OPTIONS entry to be used.*/
/* */.
ENDUCL /* */.
Instead of changing the default OPTIONS entry, you might want to override
3B Use the SET command to have
whatever the default might be and use OPTIONS entry (MYOPT1) for
your OPTIONS entry override
processing particular commands or SYSMODs. In this case, the SET
the default OPTIONS entry for
command preceding the commands you want to process must point to your
the zone.
OPTIONS entry. The following UCL points the SET command to OPTIONS
entry MYOPT1:
SET BDY(TGT1) /* Set to target zone, TGT1.*/
. OPTIONS(MYOPT1) /* OPTIONS entry used. */.
.
.

Recovering after errors from utility processing

To complete as many updates as possible to your product libraries, SMP/E tries to
recover from errors that occur during processing by the utilities it invokes. The
type of recovery SMP/E attempts for such errors depends on the type of error that
occurred and the type of utility that was called.
v Batched updates, no out-of-space problems. Items to be processed by the
link-edit utility or the copy utility are often batched. If an error other than an
x37 abend occurs during such a utility call, SMP/E debatches the items and
reinvokes the utility to attempt updates for individual members. This recovery is
attempted automatically by SMP/E; no user intervention is required.
v Out-of-space errors (x37 abends). If a list of data sets is specified by the user,
SMP/E attempts “retry” processing for data sets that have run out of space.
During retry processing, the data set that ran out of space is compressed, and
the utility is called again to retry the updates.
If retry processing does not reclaim sufficient space and input to the utility was
batched (copy or link-edit utility only), SMP/E debatches the input and retries
the utility for each member separately. If this final attempt fails, the resulting x37
abend is treated as an unacceptable utility return code, and processing continues
for other eligible updates.
This section explains what you need to define in order to have SMP/E attempt
retry processing for x37 abends.
Overview of your input to retry processing

Your input to retry processing is through subentries in the OPTIONS entry, an
optional retry exit routine, and the RETRY command operand.
v OPTIONS entry. The RETRYDDN and EXRTYDD subentries in the OPTIONS
entry indicate which libraries are eligible for retry processing.
The RETRYDDN subentry specifies which libraries should be included in retry
processing. Without this list of libraries, SMP/E does not attempt retry
processing.
The EXRTYDD subentry specifies which libraries should be excluded from retry
processing. This makes it easier for you to include all but a few specific libraries
in retry processing.
v Exit routine. The retry exit routine enables you to control retry processing when
an x37 abend occurs, instead of having SMP/E compress the out-of-space data
set and reinvoke the failing utility.
If SMP/E determines that a retry can be attempted, it cancels the abend dump
and calls the retry exit routine. The routine can then either cancel retry
processing or perform some other method of recovery.
v RETRY operand. The RETRY operand tells SMP/E whether to attempt retry
processing for the specific SMP/E command that is being processed. RETRY can
be specified on the ACCEPT, APPLY, LINK LMODS, LINK MODULE, and
RESTORE commands.
You do not need to specify this operand in order to request retry processing,
because the default is RETRY(YES). However, you can explicitly specify
RETRY(YES) if you want to.
To prevent retry processing for a specific command, specify RETRY(NO) instead
of using RETRY(YES).

Example: How to request the desired retry processing

Table 6 describes the steps you need to follow in order to get the desired retry
processing. For details on syntax and coding considerations, see SMP/E Reference.
Table 6. How to request the desired retry processing
You want retry processing for all libraries except
1 Decide which data sets should be included in or
LINKLIB, MIGLIB, and NUCLEUS.
excluded from retry processing.
You already have been using OPTIONS entry OPT1 for
2 Define the necessary subentries in the OPTIONS
all your ACCEPT, APPLY, LINK LMODS, LINK
entries that you plan to use.
MODULE, and RESTORE commands. You need to add
RETRYDDN and EXRTYDD values to specify the
libraries that are eligible for retry processing.
SET BDY(GLOBAL).
UCLIN .
ADD OPTIONS(OPT1)
RETRYDDN(ALL)
EXRTYDD(LINKLIB,MIGLIB,
NUCLEUS)
.
ENDUCL .
You will use the defaults for the retry utility as shown in
3 Decide whether to use the default RETRY UTILITY
Table 4 on page 65.
entry, or to point to and define a RETRY UTILITY
entry that specifies the desired information.
You will use SMP/E’s retry processing instead of writing
4 If desired, define an exit routine for retry
a retry exit routine.
processing.
The Retry Exit Routine section in SMP/E Reference
provides all the information you need about the
interface for this routine and what SMP/E expects
the routine to do.
As indicated in step 2, you have been using OPTIONS
5 Make sure the desired OPTIONS entry is in effect
entry OPT1. Instead of specifying it on the SET
for the zone you are processing.
command, you have defined TZONE entry (TGT1) and
The methods you can use are shown in Figure 31 DZONE entry (DLIB1), which specify OPT1 as the
on page 67. OPTIONS entry to be used. (These zones are already
defined in the global zone by ZONEINDEX subentries.)
SET BDY(TGT1).
UCLIN .
ADD TARGETZONE(TGT1)
OPTIONS(OPT1)
SREL(Z038)
RELATED(DLIB1)
.
ENDUCL .
SET BDY(DLIB1).
UCLIN .
ADD DLIBZONE(DLIB1)
OPTIONS(OPT1)
SREL(Z038)
RELATED(TGT1)
.
ENDUCL .

Table 6. How to request the desired retry processing (continued)

You are installing product HYY2102 and the related
6 Use RETRY(YES) on the commands you want retry
service, and you want SMP/E to attempt retry
processing to be done for (RETRY can be specified
processing. You choose to use RETRY(YES) as the default
on the ACCEPT, APPLY, LINK LMODS, LINK
instead of specifying it explicitly.
MODULE, and RESTORE commands.).
Note: To prevent retry processing for a specific SET BDY(TGT1).
command, specify RETRY(NO) instead. APPLY FORFMID(HYY2102)
FUNCTIONS PTFS
GROUPEXTEND
.
Connecting SMP/E dialogs to ISPF

The SMP/E dialogs run under ISPF. Therefore, if you plan to use the dialogs, you
must connect them to ISPF. Follow these steps:
1. Make sure you have the required programs on your system.
2. If you have the TSO/VS2 programming control facility (PCF), add the
necessary dialog modules to the PCF command table.
3. Concatenate the dialog libraries in a logon procedure or a command list
(CLIST).
4. Connect the dialogs to ISPF.
5. Customize the dialogs, if desired.
These steps are described in the sections that follow.
Check for required programs

Make sure the following programs are on your system:
v ISPF (Version 4 Release 2, or later)
v ISPF/PDF (Version 2 Release 3, or later)
v TSO/E (Version 2 Release 5, or later)
The SMP/E dialogs use the TSO OUTPUT, SUBMIT, and STATUS commands.
Therefore, to support these commands, you must provide a TSO IKJEFF53 user
installation exit routine that does not restrict the TSO OUTPUT and STATUS
commands to job names beginning with the user’s user ID. For details, see z/OS
TSO/E Customization.
Add dialog modules to the PCF command table

If you have the TSO/VS2 programming control facility (PCF), add the following
modules to the PCF command table:
GIMPSBTP
GIMISID1
GIMISID2
GIMISID3
GIMISID4
GIMISID5

Concatenate the dialog libraries

Table 7 lists the SMP/E dialog libraries needed to concatenate with ISPF libraries
To use the dialogs, concatenate these libraries in a TSO logon procedure or in a
CLIST. Here are some recommendations for concatenating the libraries:
v There are two sets of dialog libraries for SMP/E: one for English and one for
Japanese. You can include the libraries for only one of these features in a given
logon procedure or CLIST. If you want to be able to use both features on the
same system, you need a logon procedure or CLIST for each feature. To switch
from one feature to the other, you have to exit ISPF, run the procedure or CLIST
for the other feature, and get back into ISPF.
v Do not use the ISPF LIBDEF service to concatenate the following libraries.
Instead, use either the TSO ALLOCATE command or DD statements and JCL.
– Dialog CLISTs (GIM.SGIMCLS0): If you use LIBDEF, the CLISTs and EXECs
in the data set cannot be executed.
– Dialog load module library (GIM.SGIMLMD0): If LIBDEF is used to
concatenate the load module libraries, the SMP/E dialogs cannot find
required load modules.
Table 7. ISPF libraries and related SMP/E target libraries
DDNAME SMP/E library Contents
SYSPROC GIM.SGIMCLS0 Dialog CLISTs
ISPLLIB GIM.SGIMLMD0 Dialog load modules
ISPMLIB GIM.SGIMMENU Dialog messages
ISPPLIB GIM.SGIMPENU Dialog panels
ISPSLIB GIM.SGIMSENU Skeleton JCL procedures
ISPTLIB GIM.SGIMTENU (see Note Table input library
1)
ISPCTL1 (see Note 2) Generated JCL
ISPCTL2 (see Note 2) Generated JCL
SMPTABL A user-defined name (see SMP/E table library
Note 3)
Notes:
1. Include GIM.SGIMTENU and the SMPTABL data set in the ISPTLIB
concatenation.
2. Use the ISPCTL1 and ISPCTL2 files to generate JCL for submitted SMP/E jobs.
The SMP/E job submit facility lets you browse and edit this JCL. You can omit
these files from your logon procedure and let ISPF automatically allocate them
as needed.
To save the input JCL generated by the dialogs, either:
v Use EDIT CREATE while in the generated JCL to save it in another
(permanent) data set, or
v Allocate a permanent sequential data set to ISPCTL1 (LRECL=80,
RECFM=FB) before you enter the SMP/E dialogs.
3. Allocate a single, installation-wide table data set to the ISPTLIB and SMPTABL
DD statements.
v SMP/E uses this table data set to save process status information for the
SYSMOD management dialogs.

v The data set must be a partitioned data set (LRECL=80, RECFM=FB).

Because the data set is also in the concatenation of ISPTLIB, make the block
size compatible with the block size of the corresponding ISPF data sets.
v For more information about the SMPTABL data set, see “SMPTABL space
allocation.”
SMPTABL space allocation

The amount of space required for SMPTABL depends on the number of concurrent
processes you want to support for the SMP/E SYSMOD management dialog and
on the amount of data that is saved for each process. Use Table 8 as a guide.
Table 8. SMPTABL data set allocations
Number of processes BLKSIZE Number of blocks Directory blocks
16 8800 60 4
32 8800 120 8
Sample logon procedure

Figure 32 is an example of a logon procedure that concatenates the libraries for the
SMP/E dialogs.
Note: Depending on the system you are connecting the dialogs to, you may need
to change your logon procedure, a CLIST, or both. It is common practice to
have a logon procedure call a CLIST every time a user logs on. This is
normally done so that minor changes to concatenations do not require
changes to the logon procedure, or so users with the same logon procedure
can have the CLIST perform different allocations from the ones performed
for other users.
If you make changes only to your logon procedure and the procedure calls a
CLIST that changes a concatenation, you may end up missing changes you
made in your logon procedure. In this case, instead of (or in addition to)
updating the concatenations in your logon procedure, you should update
the concatenations in the CLIST that is called.

//SMPE EXEC PGM=IKJEFT01

//SYSPROC DD DSN=GIM.SGIMCLS0,DISP=SHR /* SMP/E CLISTs. */
// DD DSN=ISP.SISPCLIB,DISP=SHR
//SYSHELP DD DSN=SYS1.HELP,DISP=SHR
//SYSPRINT DD TERM=TS
//SYSIN DD TERM=TS
//************************************************************
//*
//* ISPF temporary data sets (no ISPCTL1 or ISPCTL2)
//*
//************************************************************
//ISPLST1 DD DISP=NEW,UNIT=SYSALLDA,SPACE=(CYL,(1,1)),
// DCB=(LRECL=121,BLKSIZE=0,RECFM=FBA)
//ISPLST2 DD DISP=NEW,UNIT=SYSALLDA,SPACE=(CYL,(1,1)),
// DCB=(LRECL=121,BLKSIZE=0,RECFM=FBA)
//************************************************************
//*
//* ISPF/SMPE load modules, panels, skeletons, messages,
//* tables
//* NOTE: SMPTABL is a single installation-wide
//* ISPF table data set shared by all SMP/E
//* users. It is initially allocated as an empty PDS
//* LRECL=80, RECFM=FB, BLKSIZE= (compatible with
//* ISP.V2R3M0.ISRTLIB).
//*
//************************************************************
//ISPLLIB DD DSN=GIM.SGIMLMD0,DISP=SHR /* SMP/E LMODs. */
//ISPPLIB DD DSN=GIM.SGIMPENU,DISP=SHR /* SMP/E panels. */
// DD DSN=ISP.SISPPENU,DISP=SHR
//ISPSLIB DD DSN=GIM.SGIMSENU,DISP=SHR /* SMP/E skeletons.*/
// DD DSN=ISP.SISPSENU,DISP=SHR
//ISPMLIB DD DSN=GIM.SGIMMENU,DISP=SHR /* SMP/E messages. */
// DD DSN=ISP.SISPMENU,DISP=SHR
//ISPTLIB DD DSN=GIM.SGIMTENU,DISP=SHR /* SMP/E tables. */
// DD DSN=ISP.SISPTENU,DISP=SHR
// DD DSN=user-defined name,DISP=SHR /* SMP/E tables.*/
//SMPTABL DD DSN=user-defined name,DISP=SHR /* SMP/E tables.*/
Figure 32. Sample logon procedure that concatenates SMP/E and ISPF libraries
Connect the dialogs to ISPF

You can get to the SMP/E dialogs from the SMP/E primary option menu. To
provide access to that menu, you must connect the dialogs to ISPF. Sample ISPF
panels are provided with z/OS to enable panels for most z/OS elements, including
SMP/E. These panels are in the SISPPENU data set after APPLY processing. The
panels supplied are:
ISR@390 This is an ISPF Primary Options Menu. It is identical to the
ISR@PRIM Primary Options Menu, except that it includes the
additional options 12 and 13, which point to the next two panels.
ISR@390S This is a secondary panel, with options used by system
programmers and administrators, including SMP/E.
ISR@390U This is a secondary menu panel, including the options used by
most ISPF users.
Use one of the methods documented in the Program Directory for this level of
SMP/E to use ISR@390 instead of ISR@PRIM. The SMP/E dialogs will then be
accessible through panel ISR@390S.

Customize the SMP/E dialogs

After you install the SMP/E dialogs, you can change the default values they use.
When you select option 0 (SETTINGS) on the SMP/E Primary Option Menu,
SMP/E displays panel GIM@PARM, which allows you to
v Modify the values for allocating temporary SMP/E utility (SYSUTn) and SMP/E
work (SMPWRKn) data sets . The options you specify are saved permanently in
the ISPF profile pool for use later by other SMP/E dialog processes.
v Specify a data set name to be used by SMP/E for the SMPPARM data set during
background execution. The SMPPARM data set is used to define exit routines
and specify allocation information used by SMP/E processing. If a data set name
is specified on panel GIM@PARM, SMP/E generates an SMPPARM DD
statement in all JCL jobs created by the SMP/E dialogs that invoke SMP/E.
GIM@PARM SMP/E DIALOG SETTINGS

===>
Reset to use SMP/E’s default settings? ===> NO (Yes or No)
Enter or verify the information below used to allocate temporary data sets.
These temporary data sets contain generated JCL jobs, output for jobs which
have completed execution, and MCS entries for viewing:
UNIT ===> SYSALLDA

VOLUME ===>
Enter or verify the information below used to create DD statements in

generated JCL jobs:
Space for SMP/E Utility Work data sets (SYSUTn):

Block Size Primary Secondary
SYSUT1 3120 380 760
SYSUT2 3120 380 760
SYSUT3 3120 380 760
SYSUT4 3120 38 100
Space for SMP/E Work data sets (SMPWRKn):

Block Size Primary Secondary Directory Blocks
SMPWRK1 3120 364 380 500
SMPWRK2 3120 364 380 500
SMPWRK3 3120 364 380 500
SMPWRK4 3120 364 380 500
SMPWRK6 3120 364 380 500
Unit for SYSUTn and SMPWRKn data sets:

UNIT ===> SYSDA
Data set name to use for SMPPARM. If a name is specified, an SMPPARM DD

statement will be generated.
DSN ===>
Enter or verify the information below used to allocate permanent SMPWRK3

data sets created and used by the SYSMOD Management dialogs:
UNIT ===>
VOLUME ===>
PREFIX ===> (TSO Prefix is used if no prefix is specified)
To save the changes, press ENTER .

To ignore the changes, enter END .

JOB statement customization

If you have never entered a JOB statement on panels GIMCGSUB, GIMRCSUB, or
GIMSB01, then SMP/E primes those panels with the following default JOB
statement:
//useridA JOB (ACCOUNT),’NAME’
//*
//*
//*
This initial default JOB statement is not customizable, but when you enter a JOB
statement on panels GIMCGSUB, GIMRCSUB, or GIMSB01, the statement you
enter is saved in your ISPF profile pool and will be used as the new default when
you use those panels again.
Setting up SMP/E for easier operation

SMP/E provides several optional facilities that you can use to make SMP/E
operations easier and more efficient. To take advantage of these facilities, you must
set up a few SMP/E options. Normally, these set up procedures need only be done
once.
The major tasks are:

v Specifying SMP/E OPTIONS entry
v Specifying link edit utility output DDDEF entries
v Specifying automatic cross-zone requisite checking
Recommended values for OPTIONS entry

IBM recommends the following OPTIONS entry values:
Recommended value Purpose
MSGFILTER(YES) MSGFILTER(YES) causes SMP/E to filter the
messages it writes to SMPOUT during APPLY,
ACCEPT, and RESTORE command processing.
When SMP/E filters messages, most non-critical
informational messages are not written to
SMPOUT. The result is less output to read through
when it is necessary to investigate an SMP/E
operation. MSGFILTER(NO) is the default.
MSGWIDTH(80) MSGWIDTH(80) causes SMP/E to format its
messages to an 80 character width. This makes
online viewing simpler by eliminating the need to
scroll right to view the entire message text.
MSGWIDTH(120) is the default.
RECZGRP Often the RECEIVE command will receive a PTF
that has already been accepted and purged from
the global zone and SMPPTS data set. There is no
need to receive such PTFs and they only add to the
space used by the SMPPTS. To prevent RECEIVE
from receiving such PTFs, you need to tell SMP/E
what dlib zones to check when determining if a
PTF has already been accepted. You can specify the
list of dlib zones using the RECEIVE Zone Group
(RECZGRP) subentry in an OPTIONS entry.

The RECZGRP subentry allows you to set a policy

and specify the list of zones once. This list is then
used for all future RECEIVE operations whenever
the OPTIONS entry is active. With the list of dlib
zones set, during RECEIVE processing, SMP/E will
check each of the zones specified first before
receiving a PTF. If that PTF is accepted in any of
the specified zones, the PTF will not be received
again.
RETRYDDN(ALL) RETRYDDN(ALL) causes SMP/E to compress
out-of-space libraries and retry processing after an
x37 abend. When you use this option, make sure
you are not updating production data sets.
Note: Do not specify a PEMAX value. Allow SMP/E to use its default value of
25,000.
Sample UCLIN job

Here is a sample UCLIN job to build an OPTIONS entry with the recommended
values:
//job JOB accounting info...
//step EXEC PGM=GIMSMP
//SMPCSI DD DSN=smp.global.csi,DISP=SHR
//SMPCNTL DD *
SET BDY(GLOBAL).
UCLIN.
ADD OPTIONS(OPTENT)
MSGFILTER(YES)
MSGWIDTH(80)
RETRYDDN(ALL)
RECZGRP( zosdlib
os390dlib
jes2dlib
jes3dlib
cicsdlib
db2dlib
imsdlib ).
ENDUCL.
/*
Activating the OPTIONS entry

After the OPTIONS entry has been defined, IBM recommends that you make it
active by defining it as the default OPTIONS entry for the global, target, and DLIB
zones. Otherwise, you must specify it on the SET command before using any other
SMP/E command.
Recommended link edit utility output DDDEF entries

To exploit utility multi-tasking in SMP/E, ensure the ddname that is to contain the
link edit utility output is defined with a DDDEF entry that identifies a SYSOUT
class. SMP/E’s default ddname for utility output is SYSPRINT, but can be changed
using the PRINT subentry of the LKED UTILITY entry. When multi-tasking,
SMP/E will invoke multiple instances of the link edit utility at the same time, thus
decreasing the total time required to complete an APPLY, ACCEPT, or RESTORE
command. If you do not define the print ddname using a DDDEF entry, or if the
DDDEF identifies something other than a SYSOUT class, SMP/E will not
multi-task link edit utility operations.

Specifying automatic cross-zone requisite checking

The installation of software service often requires the synchronization of service
levels across multiple SMP/E zones. For example, service for software in the MVS
zone may require related service for the JES2, CICS, DB2, and other zones to
permit all software within the system image to operate properly. To help ensure
proper synchronization across zones, you can tell SMP/E to automatically check
for cross-zone requisites during APPLY, ACCEPT, and RESTORE command
processing.
To enable automatic cross-zone requisite checking, you must tell SMP/E which
zones contain software to be checked for requisites. The set of zones identified for
cross-zone requisite checking is called the zone group. SMP/E provides two
methods to identify the zones within the zone group:
1. Define a default zone group
2. Specify the zones directly on the APPLY, ACCEPT, or RESTORE command.
Defining a default zone group

You can define a default zone group by creating a ZONESET entry that contains
the XZREQCHK(YES) subentry and the list of zones to be included in the default
zone group. SMP/E will use this default zone group to determine which zones to
check for requisites whenever the APPLY, ACCEPT, or RESTORE commands
process a zone named in this ZONESET. To create such a ZONESET, use the
SMP/E Administration Dialogs or use the UCLIN command, as in this example:
//job JOBaccounting info...
//SMPCNTL DD *
SET BDY(GLOBAL).
UCLIN.
ADD ZONESET(ZONEGRP)
XZREQCHK(YES)
/* use this ZONESET for cross-zone req checking */
ZONE(zostgt zosdlib
os390tgt os390dlib
jes2tgt jes2dlib
jes3tgt jes3dlib
cicstgt cicsdlib
db2tgt db2dlib
imstgt imsdlib).
ENDUCL.
/*
The ZONESET should contain the names of all the zones to be checked for
cross-zone requisites. Once the ZONESET is created and the XZREQCHK(YES)
subentry is set, the zones defined in the ZONESET are used as the default zone
group any time the APPLY, ACCEPT, or RESTORE commands process any zone
found in the ZONESET. For example, if an APPLY command is initiated for the
cicstgt zone, all zones found in the ZONESET entry named ZONEGRP are used for
the zone group.
Specifying the zone group on a command

If you don’t have a default zone group defined, or you want to use a different set
of zones for the zone group, you can specify the zones on the APPLY, ACCEPT, or
RESTORE command using the XZGROUP operand. This is simply a matter of
specifying the zones to be checked for cross-zone requisites, as shown in this
example:


//SMPCNTL DD *
SET BDY(zostgt).
APPLY SOURCEID(HIPER)
CHECK
XZGROUP(os390tgt,jes2tgt,jes3tgt,cicstgt,db2tgt,imstgt)
BYPASS(HOLDSYS).
/*
Define a ZONEINDEX for each zone

Each of the zones specified in a ZONESET or on the XZGROUP operand must be
defined by a ZONEINDEX in the current global zone, even if the zones are already
defined in another global zone (more than one global zone may contain a
ZONEINDEX for the same target or dlib zone). This allows the APPLY, ACCEPT,
and RESTORE commands initiated from the current global zone to access the
specified zones. To add ZONEINDEX subentries for each of the zones, use the
SMP/E Administration Dialogs or use the UCLIN command, as in this example:
//SMPCNTL DD *
SET BDY(GLOBAL).
UCLIN.
ADD GLOBALZONE ZONEINDEX(
(zostgt,zos.target.csi,TARGET)
(zosdlib,zos.dlib.csi,DLIB)
(os390tgt,os390.target.csi,TARGET)
(os390dlib,os390.dlib.csi,DLIB)
(jes2tgt,jes2.target.csi,TARGET)
(jes2dlib,jes2.dlib.csi,DLIB)
(jes3tgt,jes3.target.csi,TARGET)
(jes3dlib,jes3.dlib.csi,DLIB)
(cicstgt,cics.target.csi,TARGET)
(cicsdlib,cics.dlib.csi,DLIB)
(db2tgt,db2.target.csi,TARGET)
(db2dlib,db2.dlib.csi,DLIB)
(imstgt,ims.target.csi,TARGET)
(imsdlib,ims.dlib.csi,DLIB)
).
ENDUCL.
/*
Cross-zone requisite checking

Whether you define a default zone group or specify a zone group on the APPLY,
ACCEPT, and RESTORE command, SMP/E will determine during command
processing whether any cross-zone requisites are unsatisfied. Cross-zone requisites
are caused by ++IF statements, where a SYSMOD containing a ++IF statement
resides in one zone and the function identified on the ++IF resides in another zone.
If the requisite identified on the ++IF statement does not reside in the same zone
as the identified function, then the condition is not satisfied.
Unsatisfied cross-zone requisite conditions will cause APPLY, ACCEPT, and

RESTORE command processing to fail for the SYSMOD containing the ++IF
statement. Processing will continue to fail until the requisite is satisfied in the other
zone, unless the BYPASS(XZIFREQ) operand is specified on the command.

Bypassing unsatisfied cross-zone requisites

The BYPASS(XZIFREQ) operand on the APPLY, ACCEPT, and RESTORE
commands tells SMP/E to continue processing the command even if missing
cross-zone requisites are detected. SMP/E warning messages will be issued to
identify the missing cross-zone requisites.
//SMPCNTL DD *
SET BDY(zostgt).
APPLY SOURCEID(HIPER)
CHECK
BYPASS(HOLDSYS
XZIFREQ).
/*
Note: This example assumes a default zone group has been defined and will
therefore be used during APPLY command processing.
You can be broad or very granular in the specification of what cross-zone requisites
to bypass. You can indicate all cross-zone requisites are to be bypassed (as in the
previous example), you can indicate that specific cross-zone requisite SYSMODs
are to be bypassed, or you can indicate that only specific cross-zone requisite
SYSMODs from specific zones are to be bypassed. Details of the
BYPASS(XZIFREQ) operand and processing can be found in SMP/E Commands.
Resolving cross-zone requisites

If cross-zone requisites are bypassed and therefore cause unsatisfied cross-zone
requisites, you must resolve those unsatisfied requisites. To do this, you need to
APPLY or ACCEPT those requisites to the appropriate zones. To aid in this task,
SMP/E provides a method to identify missing cross-zone requisite SYSMODs and
make them candidates for APPLY and ACCEPT processing to resolve missing
cross-zone requisites.
In order to select cross-zone requisite SYSMODs to be installed in a particular

zone, the XZREQ operand can be used on the APPLY and ACCEPT commands.
The XZREQ operand causes SMP/E to search the zones in the zone group for
unsatisfied cross-zone requisites. If any are found which can be satisfied by
installing a requisite SYSMOD to the current zone, those SYSMODs are made
candidates for the APPLY and ACCEPT commands. Here is an example:
//job JOBaccounting info...
//SMPCNTL DD *
SET BDY(cicstgt).
APPLY CHECK
BYPASS(HOLDSYS)
XZREQ.
/*
Note: This example assumes a default zone group has been defined and will
therefore be used during APPLY command processing.
Using the XZREQ operand identifies and installs the needed cross-zone requisites.
You can also use the REPORT CROSSZONE command to identify the needed
cross-zone requisites. See The REPORT CROSSZONE Command in SMP/E
Commands for details.

Defining the information needed to invoke SMP/E

There are several ways to call SMP/E after it has been installed:
v Use the SMP/E dialogs.
v Submit a background job that calls GIMSMP, the program name for SMP/E. This
job can call SMP/E either directly or in a cataloged procedure.
This section describes the types of information you need to provide if you use a
cataloged procedure to invoke SMP/E. It discusses the following:
v Required JCL statements
v A sample cataloged procedure for SMP/E
Required JCL statements

Unless you are using the SMP/E dialogs, you must provide the following JCL
statements to invoke SMP/E:
v A JOB statement
v An EXEC statement
v DD (data definition) statements
JOB statement
The JOB statement describes your installation-dependent parameters. The JOB
statement (or the EXEC statement, or both) can also include the REGION
parameter to set the size of the region in which SMP/E runs. For details, see z/OS
MVS JCL User’s Guide, SA22-7598 or z/OS MVS JCL Reference, SA22-7597.
Note: To enable the SMP/E job step to get the maximum space above 16
megabytes. you can specify REGION=0M. Or, if you prefer, you can specify
a more specific region size.
EXEC statement
The EXEC statement must specify PGM=GIMSMP or the name of your cataloged
procedure for calling SMP/E. (For an example of a cataloged procedure, see
“Sample cataloged procedure for SMP/E” on page 82.) The following can be
specified in the EXEC statement PARM parameter:
CSI= dsname
where dsname is the name of the CSI data set containing the global zone. (This
data set is also known as the master CSI.) This parameter is used to enable
SMP/E to allocate the master CSI data set dynamically.
Note: If there is an SMPCSI DD statement, the CSI=dsname operand is not

allowed. If both are specified, SMP/E does not run.
DATE= date
where date can be one of the following:
U or IPL To use the IPL date of the system.
REPLY To request the date from the operator. As a result, SMP/E
issues message GIM399I.
yyddd To specify a specific date, where yy is the year and ddd is the
day of the year (the Julian date).
If DATE is not specified, the IPL date of the system is used.
PROCESS=WAIT or

Required JCL statements
PROCESS=END
The PROCESS parameter is used to control how long a job should wait if a CSI
or PTS data set is not immediately available because it is currently being used
either by another job or by a dialog.
v WAIT causes the job to wait until the data set is available. A message is
issued to the system operator every 30 minutes while the job is waiting.
v END causes the job to wait for 10 minutes. If the data set is still not
available after the 10-minute wait, the command requiring the data set is
stopped.
If PROCESS is not specified, the default is PROCESS=WAIT.
For more information on obtaining and sharing CSI data sets, see “Sharing
SMP/E Data Sets” in SMP/E Commands.
Processing of the PTS data set is also affected by the WAITFORDSN value
specified in its DDDEF entry. WAITFORDSN determines whether SMP/E
should wait to allocate a data set that is not immediately available. If the
DDDEF entry specifies WAITFORDSN=NO (or lets this value default to NO)
and the data set is not available, allocation of the data set fails, regardless of
the PROCESS value specified on the EXEC statement. If WAITFORDSN=NO,
SMP/E does not wait to retry allocation of the data set.
For example, suppose a PTS with a disposition of OLD is already being used
by a job, and a second job tries to access the same PTS data set by allocating it
through a DDDEF entry. The DDDEF entry used by the second job for the PTS
specifies WAITFORDSN=NO. As a result, allocation of the PTS fails for the
second job.
DD statements
DD statements define the data sets that can be used in SMP/E processing. For
information about the data sets required for each command, see the chapters on
individual SMP/E commands in SMP/E Commands.
Note: You can use DDDEF entries, rather than DD statements, to allocate many of
the necessary data sets. For more information, see “How to dynamically
allocate data sets to be used during SMP/E processing” on page 62.
Sample cataloged procedure for SMP/E

Figure 33 on page 84 is a sample cataloged procedure, SMPPROC, that can be used
to run SMP/E. The numbers to the left of the statements correspond to the notes
that follow the example. When you write a cataloged procedure for SMP/E,
remember the following:
v Tailor your own cataloged procedure to fit your system and processing
requirements.
v You may want to use a single procedure for all SMP/E processing, or you may
want to define multiple procedures for specific SMP/E commands and include
in each one just those DD statements required for that command. For example, a
special procedure for RECEIVE might include the SMPPTFIN DD statement but
no DD statements for the target and distribution libraries.
Note: The SYSLIB concatenations for APPLY and ACCEPT should point to
different libraries.
v Most of the data sets in the cataloged procedure can be allocated without DD
statements. If you use the methods described for the data sets listed below, you
may not need a cataloged procedure.

Sample cataloged procedure
– Master CSI data set. The master CSI data set can be specified on the CSI=
parameter of the EXEC statement for GIMSMP, rather than on the SMPCSI
DD statement. For more information about parameters you can specify on the
EXEC statement, see “Required JCL statements” on page 81.
– Target and distribution zones. CSI data sets for target and distribution zones
are normally dynamically allocated with zone indexes in the global zone. If
you want to use the batch local shared resources (BLSR) subsystem, you must
supply your own JCL statements. For examples of defining zone indexes and
of specifying JCL for BLSR, see the SMPCSI section in SMP/E Reference.
– Other data sets. Other data sets in the cataloged procedure can be defined
with DDDEF entries. When you use DDDEF entries, only the data sets
SMP/E needs for a particular run are allocated.
When you use DD statements, all the data sets defined must be online and
allocated. Therefore, you might want to use a combination of DDDEF entries
and a cataloged procedure shorter than the one in Figure 33 on page 84. For
more information about DDDEF entries, see SMP/E Reference.
v Although they are not shown in the sample cataloged procedure, the following
DD statements may also be required:
– An SMPCNTL DD statement, pointing to the commands that SMP/E
processes, is required by all commands.
– An SMPPTFIN DD statement, pointing to the source of the SYSMODs that are
processed, is required by the RECEIVE command.
– An SMPHOLD DD statement, pointing to the source of the HOLDDATA that
is processed, is required by the RECEIVE command.
– An SMPPTS DD statement should be coded with DISP=SHR. This allows
concurrent jobs to share the PTS as much as possible. For more information
on how SMP/E shares data sets, see Sharing SMP/E Data Sets in SMP/E
Commands.
– The SMPLTS data set is required when processing load modules with
CALLLIBS.
– An SMPMTS DD statement is required for changes to macros that do not
reside in a target library.
– An SMPSTS DD statement is required for changes to source code that does
not reside in a target library.
– If any of the SYSMODs being installed contain elements packaged with the
LKLIB operand, a DD statement for the ddname specified on that operand is
required by the APPLY and ACCEPT commands.
– If any of the SYSMODs being installed contain elements or JCLIN packaged
with the TXLIB operand, a DD statement for the ddname specified on that
operand is required by the APPLY and ACCEPT commands.
v If any of the required data sets (such as the SMPPTFIN) are not defined in the
cataloged procedure or by DDDEF entries, they must be specified in the JCL
used to call SMP/E.
v For more information about the data sets required by each SMP/E command,
see SMP/E Commands.

//SMPPROC PROC
//SMP EXEC PGM=GIMSMP
//* ----- SYSOUT data sets --------------------------------
//SMPOUT DD SYSOUT=A
//SMPRPT DD SYSOUT=A
//SMPLIST DD SYSOUT=A
//SMPSNAP DD SYSOUT=A
//SMPDEBUG DD SYSOUT=A
1 //SMPPUNCH DD SYSOUT=B
//*------ SMP/E data sets ---------------------------------
//SMPLOG DD DSN=SYS1.SMPLOG,DISP=MOD
//SMPLOGA DD DSN=SYS1.SMPLOGA,DISP=MOD
2 //* ----- Master CSI -------------------------------------
//SMPCSI DD DSN=SMPE.SMPCSI.CSI,DISP=SHR
//SMPDATA1 DD DSN=MVSTGT1.SMPDATA1,DISP=MOD
//SMPDATA2 DD DSN=MVSTGT1.SMPDATA2,DISP=MOD
3 //* ----- SMP/E temporary data sets------------------------
//SMPWRK1 DD UNIT=SYSDA,SPACE=(CYL,(2,1,5)),DISP=(,DELETE),
// DCB=BLKSIZE=6160
// DCB=BLKSIZE=6160
4 //SMPWRK3 DD DSN=data set name,
UNIT=SYSDA,SPACE=(CYL,(2,1,5)),DISP=(,CATALOG),
// DCB=BLKSIZE=3120
// DCB=BLKSIZE=3120
// DCB=BLKSIZE=6160
//* ----- Utility data sets -------------------------------
//SYSUT1 DD UNIT=SYSDA,SPACE=(CYL,(2,1)),DISP=(,DELETE)
//SYSUT4 DD UNIT=SYSDA,SPACE=(TRK,(2,1)),DISP=(,DELETE)
//* ----- Assembler SYSLIB data set -----------------------
5 //SYSLIB DD DSN=data set name,DISP=SHR
• • •
//* ----- Target libraries --------------------------------
//LINKLIB DD DSN=SYS1.LINKLIB,DISP=OLD
• • •
//* ----- Distribution libraries --------------------------
//AOSC5 DD DSN=SYS1.AOSC5,DISP=OLD
• • •
// PEND
Figure 33. Sample SMP/E cataloged procedure
1 SMPPUNCH is required for the GENERATE, REPORT, and UNLOAD

commands. Because it may have a high level of output, SMPPUNCH should
be directed to disk or tape.
2 SMPCSI DD statements should be coded with DISP=SHR. This allows
SMP/E to share the CSI data sets as much as possible. If DISP=OLD is
specified, no data-set sharing is attempted. For more information on how
SMP/E shares data sets, see Sharing SMP/E Data Sets in SMP/E Commands.
3 SMPWRK1–SMPWRK6 show only sample sizes for the data sets. The actual
size required depends on the number of SYSMODs being processed and the
number of elements within those SYSMODs.
4 SMPWRK3 can be permanently allocated in order to reuse assemblies. For
more information, see the description of the REUSE operand in The APPLY
5 SYSLIB concatenation depends on how you intend to use the distribution

libraries. For details on which data sets to include and in what order, see
“How to determine the appropriate SYSLIB concatenation.”
If you use a different SYSLIB concatenation for APPLY and ACCEPT and
prefer to use a SYSLIB DD statement, you should have at least two
procedures. If you use DDDEFs to point to the different library
concatenations, you can use one procedure. You can modify the examples to
use the appropriate procedure.
The following job uses the cataloged procedure in Figure 33 on page 84 to call
SMP/E.
//SMPJOB JOB ’accounting info’,MSGLEVEL=(1,1)
//SMPSTEP EXEC SMPPROC
//SMPPTFIN DD ... points to the file or data set that contains
//* the SYSMODs to be received
//SMPHOLD DD ... points to the file or data set that contains
//* the HOLDDATA to be received
//SMPTLIB DD UNIT=3380,VOL=SER=TLIB01
//SMPCNTL DD *
SET BDY(GLOBAL) /* Set to global zone */.
RECEIVE SYSMOD /* receive SYSMODS and */
HOLDDATA /* HOLDDATA */
SOURCEID(MYPTFS) /* Assign a source ID */
/* */.
LIST MCS /* List the cover letters */
SOURCEID(MYPTFS) /* for the SYSMODs */
/* */.
SET BDY(TARGET1) /* Set to target zone */.
APPLY SOURCEID(MYPTFS) /* Apply the SYSMODs */
/* */.
LIST LOG /* List the target zone log */.
/*
How to determine the appropriate SYSLIB concatenation

The recommended method for determining the appropriate SYSLIB concatenation
treats the distribution libraries as totally separate from the target libraries.
Note: The example shown is for processing a zone for SREL Z038 containing the
z/OS base control program (BCP). For other zones, follow the
recommendations for the products residing in those zones.
Treating the distribution libraries as totally separate from the target libraries
ensures that only the latest tested version of a macro is used during an assembly.
Thus, the SYSLIB concatenation at ACCEPT is different from that at APPLY.
The SMPMTS data set contains macros from SYSMODs that are applied. Therefore,
the proper SYSLIB concatenation for APPLY processing includes the SMPMTS data
set, as is shown in Figure 34 on page 86.

//* ------ Include SMPMTS first

//* ------ followed by all macro target libraries
//* ------ followed by all distribution
//* libraries
//SYSLIB DD DSN=SYS1.SMPMTS,DISP=OLD
// DD DSN=SYS1.MACLIB,DISP=OLD
// DD DSN=SYS1.MODGEN,DISP=OLD
//* ------ IF YOU NEED TO ASSEMBLE JES SOURCE MODULES:
//* ------ either HASPSRC (JES2) or JES3MAC (JES3)
// DD DSN=SYS1.HASPSRC,DISP=OLD
// DD DSN=SYS1.SISTMAC1,DISP=OLD
// DD DSN=SYS1.ATSOMAC,DISP=OLD
// DD •
// DD • other macro target libraries
// DD •
// DD •
// DD • any macro distribution libraries needed
// DD •
Figure 34. APPLY SYSLIB concatenation: APPLY different from ACCEPT
During ACCEPT processing, the macros in the SMPMTS and in the target macro
libraries are not considered to have been tested. The SMPMTS is, therefore, not
concatenated. Figure 35 shows the proper SYSLIB concatenation for ACCEPT.
//* ------ Include only macro distribution

//* libraries
// DD DSN=SYS1.AMACLIB,DISP=OLD
//SYSLIB DD DSN=SYS1.AMODGEN,DISP=OLD
//* ------ IF YOU NEED TO ASSEMBLE JES SOURCE MODULES:
//* ------ either HASPSRC (JES2) or JES3MAC (JES3)
// DD DSN=SYS1.HASPSRC,DISP=OLD
// DD DSN=SYS1.AISTMAC1,DISP=OLD
// DD DSN=SYS1.ATSOMAC,DISP=OLD
// DD •
// DD • other macro distribution libraries
// DD •
Figure 35. ACCEPT SYSLIB concatenation: APPLY different from ACCEPT
Defining exit routines

There are two types of exit routines you can define to tailor SMP/E processing:
v The RECEIVE exit routine enables you to scan statements in the SMPPTFIN
data set during RECEIVE processing.
v The retry exit routine enables you to control retry processing when a data set
runs out of space during RETRY can be specified on the ACCEPT, APPLY, LINK
LMODS, LINK MODULE, and RESTORE commands. processing. (In retry
processing, the data set is compressed and the utility that failed is called again.)
For more details, see the “SMP/E Exit Routines” chapter, in SMP/E Reference.

Chapter 4. Preparing to use Internet service retrieval
You can use Internet Service Retrieval to submit requests for PTFs and
HOLDDATA to a remote IBM® server and automatically download the packages
that result when those requests are fulfilled. Use the RECEIVE ORDER command
to submit an Internet Service Retrieval request to the IBM Automated Delivery
Request server. SMP/E uses the hypertext transfer protocol and Secure Sockets
Layer (HTTPS) to communicate with the server and FTP to download the
packages. To support the HTTPS communication infrastructure, SMP/E uses the
capabilities of Java™ and x.509 certificates to identify you to the server and perform
SSL authentication. To use both Java and x.509 certificates, you need to perform
some one-time configuration steps before you can actually use the SMP/E
RECEIVE ORDER command.
This chapter gives you an overview of using x.509 certificates to establish identity
and authenticity for client-server communications and also defines the steps you
need to take to:
v Obtain a user certificate
v Set up the z/OS® Security Server to work with certificates and install the user
certificate
v Define the ORDERSERVER input for the RECEIVE ORDER command
v Define the CLIENT input for the RECEIVE ORDER command, including
information necessary to allow SMP/E to use Java.
Identity and authentication overview

SMP/E communicates with the remote IBM Automated Delivery Request server
using the HTTP protocol, and all HTTP communications with the server are
performed using Secure Sockets Layer (SSL). Both the client (SMP/E) and the
server use x.509 certificates to secure communications when using SSL. When
initializing an SSL connection with a server, the client requests the server’s x.509
certificate to authenticate the server. The server’s certificate identifies the server to
the client and provides the server’s public key.
SSL server authentication allows a client application to confirm the identity of the
server application. The client application through SSL uses standard public-key
cryptography to verify that the server’s certificate and public key are valid and
that the certificate has been signed by a trusted Certificate Authority (CA) that is
known to the client application. The client and the server then use the negotiated
session keys and begin encrypted communications.
One of the most important pieces of the SSL server authentication scheme is the
trusted Certificate Authority (CA). Certificate Authorities are trusted organizations
that verify information about servers and then issue digital certificates that may be
accepted by applications as authentication of server identities when used in a
secure handshaking protocol such as SSL. Trusting a certificate issued by a
Certificate Authority is analogous to accepting a passport issued by a national
passport agency as proof of identity. We trust that the agency has taken proper
measures to verify the identity of the bearer of the passport. In a similar manner,
applications may accept certificates signed by a Certificate Authority.
Two types of certificates are of interest to SMP/E processing:

User certificate
A certificate that is associated with a z/OS user ID and is used to
authenticate the user’s identity. Such a certificate may also be known as a
Personal, or Client certificate.
Certificate-authority certificate
A certificate that is associated with a Certificate Authority and is used to
verify signatures in other certificates. Such a certificate may also be known
as a root certificate. GeoTrust is an example of a Certificate Authority that
provides a Certificate Authority certificate.
Obtaining a user certificate

| Before you can use the RECEIVE ORDER command, you must register to use
| IBM’s server and obtain a certificate that identifies you to the server. You must use
| ShopzSeries to register, and the registration process generates a user certificate for
| you. Certificates have an expiration date and you will need to obtain a certificate
| once per year. A single certificate that is generated for you can be used for many
| PTF and HOLDDATA orders, and SMP/E will notify you when the certificate is
| about to expire and you need to obtain a new one. ShopzSeries can be found here:
| www14.software.ibm.com/webapp/ShopzSeries/ShopzSeries.jsp.
1. If you are not yet a ShopzSeries user, register to become one.
2. If you are, or have become a registered ShopzSeries user, logon and create a
new order.
3. Select a Package category of ″Automated delivery certificates.″
4. On the next screen, enter an encryption pass phrase. ShopzSeries uses this pass
phrase to encrypt the PKCS12 package that contains your certificate and its
associated private key. Remember this pass phrase because you will need to
specify it again later when decrypting the package.
Note: A PKCS12 package contains both a certificate and the associated private
key of the certificate. Because a private key is sensitive and considered
secret, the package must be encrypted to protect it. To encrypt the
package, a one-time encryption key must be used. That encryption key is
also known as the pass phrase. You specify a pass phrase (any phrase
you will remember) in ShopzSeries when you request a certificate to be
generated for you.
5. Download to your workstation the generated PKCS12 certificate file as directed
by ShopzSeries.
After downloading the certificate file to your workstation, you need to upload it to
your z/OS system and add the certificate to your security product data base.
SMP/E obtains the certificate from your security product data base and uses it
during communications with IBM’s Automated Delivery Request server.
Uploading the user certificate to z/OS

After you have downloaded the certificate file to your workstation, you must
upload it to your z/OS system. There are many ways to transfer files from your
workstation to your z/OS system. For example, you can upload the certificate file
with Personal Communications 3270 or you can use TCP/IP FTP. The important
things to remember are the certificate file must be uploaded to z/OS as binary
data, the certificate file must be stored in a sequential data set, and the sequential
data set must have RECFM=VB and LRECL=256 or greater.

Setting up z/OS security server RACF
Before you can begin making changes to your security product data base, you
must ensure your userid is authorized to manipulate certificates and keyrings.
Consult your system’s RACF® administrator and refer to z/OS Security Server RACF
Security Administrator’s Guide before modifying the security characteristics for your
system.
Note: SMP/E requires either the z/OS Security Server (RACF) or an equivalent
security product on z/OS to store and manage x.509 certificates. The
remainder of this chapter assumes you are using RACF. If you are using an
equivalent security product, you should refer to that product’s
documentation to understand the equivalent actions. See Setting up alternate
security products for further information.
Access to the RACDCERT command

First, you need to define the necessary FACILITY class profiles to permit you
access to use the RACDCERT commands. RACF’s control levels in increasing
strength are NONE, READ, UPDATE, CONTROL, and ALTER. To use the
RACDCERT command, the command issuer requires appropriate permission to the
IRR.DIGTCERT.function profile under the FACILITY class. In general, READ access
is required to manipulate your own certificates and keyrings, UPDATE access is
required to manipulate them for other users, and CONTROL access is required to
manipulate CERTAUTH (Certificate Authority) certificates. Therefore, you can use
the following sample RACF RDEFINE and PERMIT commands to define necessary
FACILITY class profiles and to permit you access to use the RACDCERT
commands:
RDEFINE FACILITY IRR.DIGTCERT.ADD UACC(NONE)
RDEFINE FACILITY IRR.DIGTCERT.ADDRING UACC(NONE)
RDEFINE FACILITY IRR.DIGTCERT.ALTER UACC(NONE)
RDEFINE FACILITY IRR.DIGTCERT.CONNECT UACC(NONE)
RDEFINE FACILITY IRR.DIGTCERT.LIST UACC(NONE)
RDEFINE FACILITY IRR.DIGTCERT.LISTRING UACC(NONE)
PERMIT IRR.DIGTCERT.ADD CLASS(FACILITY) ID(userid) ACCESS(CONTROL)

PERMIT IRR.DIGTCERT.ADDRING CLASS(FACILITY) ID(userid) ACCESS(READ)
PERMIT IRR.DIGTCERT.ALTER CLASS(FACILITY) ID(userid) ACCESS(READ)
PERMIT IRR.DIGTCERT.CONNECT CLASS(FACILITY) ID(userid) ACCESS(UPDATE)
PERMIT IRR.DIGTCERT.LIST CLASS(FACILITY) ID(userid) ACCESS(READ)
PERMIT IRR.DIGTCERT.LISTRING CLASS(FACILITY) ID(userid) ACCESS(READ)
Notes:
1. CONTROL access is required to the IRR.DIGTCERT.ADD profile in the
FACILITY class only if you are using z/OS Release 4 or 5 and need to add a
Certificate Authority (CA) certificate. Otherwise READ access is sufficient. See
“Enabling CA certificates on z/OS release 4 or release 5” on page 90 for details
of CA certificates.
2. UPDATE access is required to the IRR.DIGTCERT.CONNECT profile in the
FACILITY class in order to connect a Certificate Authority (CA) certificate to
your keyring. See “Enabling certificate authority certificates” on page 90 for
details of CA certificates.
3. To use the SMP/E RECEIVE ORDER command, access is required only to the
IRR.DIGTCERT.LIST and IRR.DIGTCERT.LISTRING profiles. Access to the
other profiles is required to create and manipulate keyrings and certificates.
When your userid has the proper authorization, continue with “Creating keyrings”
on page 90.
Chapter 4. Preparing to use Internet service retrieval 89

Creating keyrings
A keyring is a named collection of certificates associated with a specific user. A
certificate is identified by its label as well as the keyring it is connected to. A
keyring must be created using a RACF command like the following:
RACDCERT ID(userid) ADDRING(keyringname)
Where keyringname is a name you choose for the keyring.
Enabling certificate authority certificates

A Certificate Authority (CA) certificate is used to verify signatures in other
certificates such as the server and user certificates. The IBM Automated Delivery
Request server uses a server certificate signed by the GeoTrust Certificate
Authority. Therefore, the GeoTrust CA certificate must be accessible in the RACF
data base during RECEIVE ORDER command processing so the server certificate
can be verified.
Enabling the CA Certificate varies depending on the release of z/OS you are using.
Enabling CA certificates on z/OS release 6 or higher

With z/OS Release 6 and higher, the GeoTrust CA certificates are supplied by
default in RACF. However, by default the supplied certificates are not trusted. Use
the following RACF command to trust the GeoTrust CA certificate:
RACDCERT CERTAUTH +
ALTER(LABEL(’Equifax Secure CA’)) TRUST
Note: Equifax was acquired by GeoTrust, and the server’s certificate was signed by
Equifax before the company’s acquisition. Hence the misleading certificate
name.
Connect the GeoTrust CA certificate to the user’s keyring using the following
RACF command:
RACDCERT ID(userid) CONNECT( CERTAUTH RING(keyringname) +
LABEL(’Equifax Secure CA’) USAGE(CERTAUTH) )
Where keyringname is the name for the keyring you choose in “Creating keyrings.”
Enabling CA certificates on z/OS release 4 or release 5

Before z/OS Release 6, the GeoTrust CA certificate must be obtained from the
GeoTrust Web site at http://www.geotrust.com/resources/root_certificates/
index.htm.
1. Download to your workstation the DER encoded form of the Equifax Secure
Certificate Authority root CA certificate from the GeoTrust Web site.
Note: There are several different CA certificates to choose from on the GeoTrust
Web site. Make sure you request the Equifax Secure Certificate
Authority root CA certificate.
2. Upload the CA certificate to your z/OS system. There are many ways to
transfer files from your workstation to your z/OS system. For example, you
can upload the certificate file with Personal Communications 3270 or use
TCP/IP FTP. The important things to remember are the certificate file must be
uploaded to z/OS as binary data, the certificate file must be stored in a
sequential data set, and the sequential data set must have RECFM=VB and
LRECL=256 or greater.

3. After you have stored the certificate in a sequential data set, add it to your
RACF database using the following RACF command:
RACDCERT CERTAUTH ADD(’ca-cert.dataset.name’) +
WITHLABEL(’Equifax Secure CA’) TRUST
Where ca-cert.dataset.name is the data set name used to store the certificate
received from the GeoTrust Web site.
4. Connect the GeoTrust CA certificate to the user’s keyring using the following
RACF command:
RACDCERT ID(userid) CONNECT( CERTAUTH RING(keyringname) +
LABEL(’Equifax Secure CA’) USAGE(CERTAUTH) )
Where keyringname is the name for the keyring you chose in “Creating
keyrings” on page 90.
Adding the user certificate to your RACF data base

A user certificate will be used by the SMP/E RECEIVE ORDER command to
uniquely identify you to the IBM Automated Delivery Request server. As described
above, the user certificate was generated for you by ShopzSeries, downloaded to
your workstation, transferred to your z/OS system as binary data, and stored as a
sequential data set. From the sequential data set, the certificate can be stored in the
RACF data base using the following RACF command:
RACDCERT ID(userid) ADD(’user.certificate.dataset.name’) +
WITHLABEL(’SMPE Client Certificate’) PASSWORD(’pass phrase’) TRUST
Where user.certificate.dataset.name is the data set name used to store the PKCS12
certificate package obtained from ShopzSeries, SMPE Client Certificate is the label
you choose to identify this certificate (32 characters or less), and pass phrase is the
encryption pass phrase you specify when generating the PKCS12 certificate
package on ShopzSeries.
Note: After you issue the above RACDCERT command, RACF should return this
message: "Certificate Authority not defined to RACF. Certificate added
with TRUST status." This is the expected response and is acceptable.
After you add the certificate to the RACF data base, you must connect it to the
keyring:
RACDCERT ID(userid) CONNECT(LABEL(’SMPE Client Certificate’) +
RING(keyringname) USAGE(CERTAUTH))
Where SMPE Client Certificate is the label you choose in the previous step to
identify this certificate, and keyringname is the name of the keyring you choose in
“Creating keyrings” on page 90.
Note: To enable the user certificate to be easily shared by other userids without
requiring unnecessarily high levels of access for those other userids, the user
certificate must be connected to the keyring as a Certificate Authority (CA)
certificate (USAGE of CERTAUTH). This allows the user certificate to be
shared without requiring other userids to access the certificate’s associated
private key.
Sharing a user certificate among multiple userids

It is possible for multiple users to share a single user certificate obtained from
ShopzSeries. To do so, you must first create a keyring, enable the CA certificate,
and add the user certificate to your RACF data base as explained in the preceding

topics. Assume userid USER1 is associated with the keyring and is the owner of
the user certificate. In order to allow userid USER2 to share the user certificate,
you must give USER2 permission to read other users’ keyrings and certificates.
You can use the following RACF commands:
PERMIT IRR.DIGTCERT.LIST CLASS(FACILITY) ID(USER2) ACCESS(READ)
PERMIT IRR.DIGTCERT.LISTRING CLASS(FACILITY) ID(USER2) ACCESS(UPDATE)
Permitting USER2 UPDATE access to the IRR.DIGTCERT.LISTRING FACILITY

class is not a security exposure. It is true that USER2 will have the ability to read
anyone’s keyring. However, that only allows the ability to extract and use the
certificates from the keyring. It does not allow use of the private keys associated
with those certificates. Therefore, USER2 cannot masquerade as another userid.
After USER2 has the appropriate permission, in order for USER2 to use the
certificate for the SMP/E RECEIVE ORDER command, you must ensure SMP/E
finds the certificate in the correct keyring when running the command. To do this,
USER2 must specify not only the keyring name, but also the userid associated with
the keyring, USER1, on the keyring attribute in the ORDERSERVER data set for the
RECEIVE ORDER command as follows:
keyring="USER1/keyringname"
See “Define the ORDERSERVER input for RECEIVE ORDER” on page 93 for
further information about the keyring attribute and the ORDERSERVER data set.
| Debugging keyring and certificate issues

| The following RACDCERT commands may be useful if the SMP/E RECEIVE
| ORDER command detects errors or failures related to your keyring or
| certificates.You can use the following RACDCERT commands to list the keying and
| certificates to verify their existence and proper attributes.
| v To list the keyring:
| RACDCERT ID(userid) LISTRING(keyringname)
| v To list the certificate authority (CA) certificate:
| RACDCERT CERTAUTH LIST(LABEL(’Equifax Secure CA’))
| v To list the user certificate:
| RACDCERT ID(userid) LIST(LABEL(’SMPE Client Certificate’))
Replacing a user certificate that has expired

A user certificate obtained from ShopzSeries has a finite life span; it will expire
after a specific time period and will no longer be valid. The SMP/E RECEIVE
ORDER command will warn you if the user certificate will expire within 30 days,
and the command will fail if the certificate has finally expired. When this occurs,
you should replace your existing certificate with a new one. The steps to replace an
existing certificate with a new certificate are very much like those you performed
when obtaining and adding the first user certificate.
1. Obtain a new user certificate from ShopzSeries as described in “Obtaining a
user certificate” on page 88.
2. Upload the new user certificate to z/OS as described in “Uploading the user
certificate to z/OS” on page 88.
3. Before adding the new certificate to your security product database, you must
delete the existing certificate so that you can use the same label for the new
user certificate. Use the following RACF command:
RACDCERT ID(userid) DELETE(LABEL(’SMPE Client Certificate’))

4. Follow the steps described in “Adding the user certificate to your RACF data
base” on page 91 to add the new user certificate and to connect it to your
keyring.
Because the new certificate uses the same label as the existing certificate, no other
changes are necessary to ensure that your RECEIVE ORDER command jobs
continue to run as expected.
Refreshing RACF classes

After you have performed the RACF updates to add certificates and keyrings, you
need to refresh the in-storage RACF profiles. That is, if you have RACLISTed the
DIGTCERT or DIGTRING classes, then you need to refresh the in-storage profiles
before the updates can take affect. You can use the following RACF command:
SETROPTS RACLIST(DIGTCERT DIGTRING) REFRESH
Setting up alternate security products

SMP/E requires either the z/OS Security Server (RACF) or an equivalent security
product on z/OS to store and manage x.509 certificates. This chapter assumes you
are using RACF. If you are using an equivalent security product, you should refer
to that product’s documentation to understand the equivalent actions.
eTrust® CA-ACF2 security for z/OS users

Computer Associates™ has created a Hyper Notification, QI73845, to address the
setup requirements associated with SMP/E Internet Service Retrieval for its eTrust
CA-ACF2 Security for z/OS customers. Its purpose is to provide detailed
instructions equivalent to the RACF instructions found herein. Refer to this Hyper
Notification for details.
| eTrust CA-Top Secret security for z/OS users

| Computer Associates has created a Technical Document, ID# TEC394405, to
| address the setup requirements associated with SMP/E Internet service retrieval
| for its eTrust CA-Top Secret Security for z/OS customers. Its purpose is to provide
| detailed instructions equivalent to the RACF instructions found herein. Refer to
| this Technical Document for details.
Define the ORDERSERVER input for RECEIVE ORDER

The ORDERSERVER data set is used by the SMP/E RECEIVE ORDER command
to provide necessary information about the IBM Automated Delivery Request
server. The information is described using the <ORDERSERVER> tag and attributes
that are defined in detail in SMP/E Commands. Here is an example and brief
explanations:
<ORDERSERVER
url="https://eccgw01.boulder.ibm.com/services/projects/ecc/ws"
keyring=”SMPEKeyring”
certificate=”SMPE Client Certificate”>
</ORDERSERVER>
url specifies the Uniform Resource Locator (URL) for the order server. The
actual urls for the IBM Automated Delivery Request server are
https://eccgw01.boulder.ibm.com/services/projects/ecc/ws and
https://eccgw02.rochester.ibm.com/services/projects/ecc/ws
keyring
identifies the RACF keyring that contains the user certificate required for
access to the order server. If the keyring is owned by a userid other than

that used to run SMP/E, then the keyring name must be prefixed with the
associated userid. The userid and keyring name must be separated by a
forward slash. For example, keyring="USER1/SMPEKeyring".
certificate
specifies the label to identify the user certificate used for access to the
order server. This is the certificate obtained from ShopzSeries and added to
your security product data base.
Define the CLIENT Input for RECEIVE ORDER

The CLIENT data set is used by the SMP/E RECEIVE ORDER command to
provide information about the local z/OS system and network. The information is
described using the <CLIENT> tag and attributes that are defined in detail in
SMP/E Commands. Here is an example:
<CLIENT
javahome="/usr/lpp/java/J1.4"
classpath="/usr/lpp/smp/classes"
javadebugoptions="-Dcom.ibm.smp.debug=severe -showversion">
<HTTPPROXY host=”local.httpproxy.com”>
</HTTPPROXY>
</FIREWALL>
<SERVER host=”local.ftpproxy.com”> </SERVER>
<FIRECMD>&REMOTE_USER;@&REMOTE_HOST;</FIRECMD>
<FIRECMD>&REMOTE_PW;</FIRECMD>
</FIREWALL>
</CLIENT>
Not all tags and attributes shown in the example above will be required for every
user, but are explained below. The options of the CLIENT data set can be grouped
into three categories: options that affect Java, options that affect HTTPS, and
options that affect FTP.
Options that affect Java

To support the HTTPS communications with the IBM Automated Delivery Request
server, SMP/E uses the capabilities of Java. Specifically, SMP/E requires IBM
Software Development Kit (SDK) for z/OS, Java 2 Technology Edition, Version 1.4
(5655-I56) with PTF UK04987 (level 1.4.2) or its logical successor. SMP/E supplies
several Java application classes, which after SMP/E is installed, reside in the
/usr/lpp/smp/classes directory in the UNIX file system on the z/OS system. In order
for SMP/E to use these application classes, they and the Java runtime must be
available in the execution environment for SMP/E. Because neither can be accessed
using a STEPLIB DD statement, you specify the locations for the Java runtime and
the SMP/E application classes using the javahome and classpath attributes in the
CLIENT data set. The javahome attribute is used to specify the directory where the
Java runtime resides, and the classpath attribute is used to specify the search path
for Java application classes. For example:
Note: Support for the javahome attribute was added to SMP/E V3.4 in APAR
IO01770.
If the SMP/E Java application classes are not installed on the driving z/OS
system’s UNIX file system, you can use appropriate mountpoints and directory
structure to point to a target system’s directories where SMP/E is installed. This is
useful if you use STEPLIB to access a target system’s base SMP/E programs in
order to ensure the base SMP/E programs and the application classes are at the
same service level. For example:

classpath="/TGTSYS/usr/lpp/smp/classes"
As an alternative to the javahome and classpath attributes in the CLIENT data set,
you can also use the SMPJHOME and SMPCPATH DD statements or DDDEF
entries to specify the location of the Java runtime and the SMP/E application
classes. If specified, the SMPJHOME and SMPCPATH DD statements override any
values specified on the javahome and classpath attributes. For example:
//SMPJHOME DD PATH=’/usr/lpp/java/J1.4’
//SMPCPATH DD PATH=’/usr/lpp/smp/classes’
Note: Support for the SMPJHOME and SMPCPATH ddnames was added to
SMP/E V3.4 in APAR IO03469.
The javadebugoptions attribute is used to specify any Java command line
parameters, including debug and trace options. Such options determine what
debug and trace information should be produced when SMP/E invokes its Java
application classes. The debug and trace information is written to the PRINT file
for the HFSCOPY utility (SYSPRINT is the default). Although it is not necessary all
of the time, it is a good idea when first using the RECEIVE ORDER command, to
specify the following value to ensure basic debug and trace information is
produced for reference:
javadebugoptions="-Dcom.ibm.smp.debug=severe -showversion"
Although it is not necessary, it is also possible to specify other Java command line
parameters that affect the operation of the Java Virtual Machine (JVM). For
example, if you encounter a Java error that indicates there is insufficient space in
the Javaheap, you can specify an option to override the default maximum Java
heap size:
javadebugoptions="-Xmx128m"
Options that affect HTTPS operations

SMP/E communicates with the IBM Automated Delivery Request server using the
HTTP 1.1 protocol using Secure Sockets Layer (SSL), also known as HTTPS. All
communications with the server are performed using the well known HTTPS port
of 443.
Optional tags are available in the CLIENT data set for the RECEIVE ORDER
command to describe local HTTP or SOCKS proxy servers. A proxy server redirects
HTTP requests to the IBM Automated Delivery Request server on the Internet. The
<HTTPPROXY> tag is used to identify a local HTTP proxy server, and the
<HTTPSOCKSPROXY> tag is used to identify a local SOCKS proxy server. For
example:
<HTTPPROXY host=”local.httpproxy.com”> </HTTPPROXY>
The <HTTPPROXY> and <HTTPSOCKSPROXY> tags are optional, and should be

specified only if the HTTP requests to the Internet from your z/OS system are
required to pass through a specific HTTP or SOCKS proxy server. For example, if
you must specify a proxy server in your Internet browser configuration to allow
you access to websites on the Internet, then you may need to specify the
<HTTPPROXY> or <HTTPSOCKSPROXY> tag in the CLIENT data set. If your
HTTP or SOCKS proxy server requires authentication, the user and pw attributes
can be used to specify the proxy server userid and password. Also, if your HTTP
or SOCKS proxy server listens on a port other than the well known ports 80 and
1080 respectively, the port attribute can be used to specify an alternate port value.
For example:

<HTTPPROXY host=”local.httpproxy.com”
user=”userid” pw=”password” port=”8080”> </HTTPPROXY>
See the SMP/E Commandsfor complete details of the <HTTPPROXY> and

<HTTPSOCKSPROXY> tags and attributes, and consult your network
administrator for help determining what, if anything, you must specify for an
HTTP or SOCKS proxy server.
Options that affect FTP operations

SMP/E uses HTTP to communicate with the IBM Automated Delivery Request
server, and it uses FTP to download package files containing PTFs and
HOLDDATA from an IBM FTP server to your local z/OS system. After the package
files are staged to the IBM FTP server, the IBM Automated Delivery Request server
provides SMP/E with the information it needs to login to the FTP server and
download the package files. Specifically, it provides SMP/E with the FTP server
host name, and a userid and password for that FTP server which are unique for
the specific package to be downloaded.
In order for SMP/E to login to the IBM FTP server, it may be necessary to navigate
a local FTP firewall server. If so, optional tags are available in the CLIENT data set
to describe information necessary to navigate a local FTP firewall server.
Note: SMP/E’s use of FTP is the same for the RECEIVE ORDER and RECEIVE
FROMNETWORK commands. Therefore, if you already use the RECEIVE
FROMNETWORK command successfully, you should use the same FTP
options in your CLIENT data set for the RECEIVE ORDER command.
The <SERVER> tag is used to identify the FTP firewall server. You can also specify
a userid and password if your firewall server requires authentication. The
<FIRECMD> tags are used to identify the commands necessary to navigate your
firewall server. The commands you specify in the <FIRECMD> tags should be the
same as those you use with the z/OS Communications Server FTP client
(PGM=FTP). The behavior of firewall servers differ, therefore, the best method to
determine what you should specify in the <FIRECMD> tags is to login to an
external FTP server using the z/OS Communications Server FTP client in a JCL
job, and then specify the same commands in the <FIRECMD> tags. For example:
//jobname JOB ...
//FTP EXEC PGM=FTP,PARM=’testcase.boulder.ibm.com’
//OUTPUT DD SYSOUT=*
//INPUT DD *
anonymous ; userid for remote FTP server
email_address ; password for remote FTP server
cd mvs/toibm ; simple change directory command
quit ; done, so log off
/*
This sample job logs into the IBM Testcase FTP server, and assumes there are no
special commands or login procedures required to navigate a local firewall server.
If such a job works for you, then no firewall server information needs to be
specified in your CLIENT data set for the RECEIVE ORDER command. However,
if you have a local firewall server that requires such commands or login
procedures, you will need to account for them. For example, if your working FTP
job is like this:
//jobname JOB ...
//FTP EXEC PGM=FTP,PARM=’local.ftpproxy.com’
//OUTPUT DD SYSOUT=*
//INPUT DD *

anonymous@testcase.boulder.ibm.com
email_address ; password for remote FTP server
cd mvs/toibm ; simple change directory command
quit ; done, so log off
/*
Then your CLIENT data set for the RECEIVE ORDER command should include
the following:
<FIREWALL>
<FIREWALL
The substitution variables &REMOTE_HOST;, &REMOTE_USER;, and

&REMOTE_PW; will be replaced automatically by SMP/E during RECEIVE
ORDER command processing with the specific values for the FTP server host
name, userid, and password that are correct for your order. The FTP server
information is returned to SMP/E automatically during the RECEIVE ORDER
command from the IBM Automated Delivery Request server. SMP/E then uses the
specified commands to log into the FTP server.
Refer to the SMP/E Commands for complete details of the <FIREWALL>,

<SERVER>, and <FIRECMD> tags, as well as a list of the substitution variables
that can be used within the firewall commands, and consult your network
administrator for help determining what, if anything, you must specify to navigate
your local FTP firewall server.
Network configuration notes

SMP/E assumes you have network connectivity from your z/OS system to the
IBM servers through the Internet. Consult your network administrator and the
z/OS Communications Server: IP Configuration Guide for information about how to
setup your z/OS system’s network configuration properly.
The HTTP and FTP operations performed by SMP/E require host name to IP
address resolution. This is usually accomplished using a Domain Name System
(DNS) name server. A name server is defined using the NSINTERADDR or
NAMESERVER statement within a resolver configuration file (TCPIP.DATA
information). There are several different locations where a resolver configuration
file can be found when using an application such as SMP/E. Because SMP/E uses
a UNIX process for its HTTP and FTP operations, the z/OS UNIX search order is
used to find the resolver configuration file. See the sections titled “Resolver
configuration files” and ″Search orders used in the z/OS UNIX environment″ in
thez/OS Communications Server: IP Configuration Guide for details on specifying the
location of the resolver configuration file. However, because of the asynchronous
UNIX process used by SMP/E for its HTTP and FTP operations, there are two
exceptions to the documented search order: neither the SYSTCPD DD statement,
nor the RESOLVER_CONFIG environment variable can be used to define the
location of the resolver configuration file.
You can verify your name server setup by using the following sample job to
invoke the NSLOOKUP command:
//jobname JOB ...
//NSLOOKUP EXEC PGM=BPXBATCH,
// PARM=’PGM /bin/nslookup eccgw01.boulder.ibm.com’
//STDOUT DD PATH=’/tmp/&SYSUID..bpxbatch.stdout’,
// PATHOPTS=(OWRONLY,OCREAT,OTRUNC),PATHMODE=SIRWXU

//OUTSTEP EXEC PGM=IKJEFT01
//SYSTSPRT DD SYSOUT=*
//INPUT DD PATH=’/tmp/&SYSUID..bpxbatch.stdout’,
// PATHOPTS=(ORDONLY),PATHDISP=DELETE
//OUTPUT DD SYSOUT=*,DCB=(RECFM=V,LRECL=256)
//SYSTSIN DD *
OCOPY INDD(INPUT) OUTDD(OUTPUT)
/*
Note: Although the NSLOOKUP command may be run simply from the OMVS
shell, this sample job runs the command in an environment similar to that in
which SMP/E runs.
If your name server is setup properly, the nslookup command will return the IP
address for the server. The output of the command will be similar to the following:
Defaulting to nslookup version 4
Starting nslookup version 4
Server: local.dns.com
Address: 9.0.2.1
Non-authoritative answer:
Name: eccgw01.boulder.ibm.com
Address: 207.25.252.197
If an IP address for the eccgw01.boulder.ibm.com server is not returned, then

verify your name server setup and resolver configuration file are proper. See the
section titled ″Diagnosing resolver problems″ in z/OS Communications Server: IP
Diagnosis Guide for details about using the Trace Resolver debug facility.
Summary
The steps described in this chapter help you to configure your system to enable the
SMP/E RECEIVE ORDER command. After you perform the described tasks, you
can run the RECEIVE ORDER command to order and download PTFs and
HOLDDATA from the IBM Automated Service Delivery server. Here is a summary
of the steps:
1. Obtain a user certificate:
a. Go to ShopzSeries to generate a user certificate and download the certificate
file to your workstation.
b. Transfer the certificate file to your z/OS system as binary data and store as
a sequential data set with RECFM=VB and LRECL=256 or greater.
2. Set up z/OS security server (RACF):
a. Ensure your userid has appropriate RACF access to create keyrings and add
certificates.
b. Create a RACF keyring.
c. Alter the GeoTrust CA certificate to mark it trusted.
d. Connect the GeoTrust CA certificate to your keyring.
e. Add the user certificate to the RACF data base.
f. Connect the user certificate to your keyring.
3. Setup for using Java:
a. Ensure IBM SDK for z/OS, Java 2 Technology Edition, Version 1.4 with PTF
UK04987 is installed on your z/OS system (level 1.4.2).
b. Specify the Java runtime directory on the javahome attribute in the CLIENT
data set for the RECEIVE ORDER command.

c. Specify the SMP/E Java application classes directory on the classpath
attribute in the CLIENT data set for the RECEIVE ORDER command.
4. Setup for network configuration:
a. If necessary, identify your HTTP proxy server in the CLIENT data set for
the RECEIVE ORDER command.
b. If necessary, identify your FTP firewall server and navigation commands in
the CLIENT data set for the RECEIVE ORDER command.
Example
After you complete the necessary steps to configure your system to enable the
SMP/E RECEIVE ORDER command, you can run an SMP/E RECEIVE ORDER job
like the following:
//jobname JOB ...,REGION=0M
//SMPCSI DD DSN=smpe.global.csi,DISP=SHR
//SMPCNTL DD *
SET BDY(GLOBAL).
ORDER(ORDERSERVER(ORDSRVR)
CLIENT(MYCLIENT)
CONTENT(RECOMMENDED))
DELETEPKG.
/*
//ORDSRVR DD *
<ORDERSERVER
keyring="keyringname"
</ORDERSERVER>
/*
//MYCLIENT DD *
<CLIENT
javadebugoptions="-Dcom.ibm.smp.debug=severe -showversion">
<HTTPPROXY host=”local.httpproxy.com”>
</HTTPPROXY>
<FIREWALL>
<FIREWALL>
</CLIENT>
/*
This sample job instructs SMP/E to order and then download all recommended
(RSU) PTFs that are applicable to the target zones defined in your global zone. See
the RECEIVE chapter of the SMP/E Commands, SA22-7771 for details of the
RECEIVE ORDER command.

Chapter 5. Installing a new function
This chapter discusses the method you use to install a new function. It describes:
v Sources of new functions and sources of installation information
v An example of installing a function
Introduction
The primary purpose of SMP/E is to help you install SYSMODs in your target and
distribution libraries. To do this, SMP/E provides three basic commands:
RECEIVE, APPLY, and ACCEPT.
This chapter summarizes the general steps you follow to install a function. You
should look at the installation materials that arrive with a function to find out
about special requirements and procedures for installing the function. Table 9 lists
sources of new functions and places where you can find information for installing
the new functions.
Table 9. Sources for functions and their installation information
Product delivery vehicle Products and service you get Installation materials you can use
CBPDO Products and service (not integrated) Sample jobs to receive products and
on a single logical tape service
Program directories for the products

you ordered
Installation manuals for the products

you ordered
Independent product Product tape Program directory for the product
Installation manual for the product (if

one is provided)
RECEIVE-APPLY-ACCEPT method
RECEIVE-APPLY-ACCEPT is the standard installation method. It uses SMP/E
RECEIVE, APPLY, and ACCEPT commands to install functions onto a subsystem.
Usually, you do not have to do any special processing outside SMP/E to install
your functions. The JCLIN needed to set up the load modules for the function is
sent along with the function.
In this method of installation, you:

1. Use the RECEIVE command to get the SYSMODs from the input data set and
put them in the SMP/E data sets (the PTS and global zone within the CSI).
2. Use the APPLY command to install the SYSMODs in the target system libraries;
then test them as required.
3. Use the ACCEPT command to install the SYSMODs into the distribution
libraries.

Installing functions
Using the standard RECEIVE-APPLY-ACCEPT method

This section describes the standard process for using the RECEIVE, APPLY, and
ACCEPT commands to install a function.
Note: You can use either the SMP/E dialogs or JCL jobs to receive, apply, and
accept functions. The basic steps to follow are the same. If you have access
to the SMP/E dialogs, you should use them. Otherwise, you can use the
steps described in this chapter as examples.
Preparing your system

Before you start doing any operations on the product function or service tape you
have received, there is work you should do to your system to make sure it is ready
and to be certain you can recover in case of a serious failure during installation.
Following are some of these steps:

1. Read the documentation for your new product. This includes the program
directory and, if provided, an installation guide. Also check the IBM Preventive
Service Planning (PSP) files for the latest information about the product. This is
important because there may be a PTF for the product that is not included in
an ESO, or one of the PTFs may contain an error you should know about.
2. When you order a product, you should update the FMID list in the global
zone with the FMIDs of the products you will be receiving. (Check the program
directory for this information.) This enables you to receive any preventive
service shipped between the time you order the product and the time you
install it.
3. Read the program directory. It tells you which libraries are affected, whether
any existing libraries must be expanded and by how much, and whether any
new libraries are required.
4. Prepare the target and distribution libraries. If these libraries are properly
prepared before you apply or accept a SYSMOD, very little time is lost if an
error occurs.
List the VTOC of the target and distribution packs. This shows you which data
sets are into their secondary extents, or are too full to contain additional
elements that may be applied or accepted. If you are unsure how large a data
set will grow, you may want to check full data sets against the SYSMODs you
will be processing.
Partitioned data sets with a high percentage of their space used can be
compressed by use of IEBCOPY. If it looks as if more space may be needed
even after the compression, allocate a larger data set and copy the nearly full
data set into it; then delete the old data set. Rename the new one properly and,
if it has had to be allocated on a different pack, update any procedure
necessary with the new VOLUME data.
This preparation is time-consuming but takes less time and work than
recovering from an out-of-space abend (E37, B37, and so on).
SMP/E command operands can also help you handle out-of-space abends.
v The COMPRESS operand tells SMP/E to compress the data sets before they
are updated; this can help you avoid an x37 ABEND. For more information
about the COMPRESS operand, see the SMP/E Commands manual.
v The RETRY(YES) operand tells SMP/E to attempt recovery after an x37
ABEND occurs by compressing the affected data sets and retrying the failing
utility. If you still need space after SMP/E’s initial retry attempt and input to
the utility was batched (copy or link-edit utility only), SMP/E debatches the

input and retries the utility separately for each member. For information
about this retry processing, see “Recovering after errors from utility
processing” on page 69.
5. Allocate any new libraries required. Determine where they are to be allocated
and then allocate them. Remember that the program directory ordinarily shows
how much space will be used—not how much space to allocate for the
libraries. Libraries should be allocated with more space than required to allow
for later modifications. Usually, twice the required space is recommended to
allow for the replacement of every element in the library without running out
of space.
Remember to add the appropriate DDDEF entries to the target zones and DLIB
zones into which you will be installing this function.
6. Check the SMP/E data sets to make sure they have enough space. If
necessary, compress or expand the partitioned data sets. A data set that is easily
overlooked in this process is the SMPSTS, which fills very rapidly when you
are receiving source updates (JES2 and JES3, for example). Reorganize or
expand (if necessary) the CSI data set (using access method services EXPORT
and IMPORT).
7. Create a backup for the volumes affected. This is a very important step that
should not be overlooked. Without a current backup copy, a serious system
failure during installation means not only redoing the installation in process
but also means going to the last backup level and redoing all the work done
since then.
8. Estimate the time required for APPLY and ACCEPT processing. Make sure
enough time is available to allow these jobs to run to completion.
The program directory or installation guide may contain information to help
you estimate the time required.
Staging the SYSMODs: The RECEIVE process

When you get a new function as part of a CBPDO, you get one logical tape that
contains the function and the unintegrated service. If there is no preventive service
for the function, it is not included in your order.
The first step in installing the new function is the RECEIVE process, which reads
in the SYSMODs so they can be installed later:
v If you have access to the SMP/E dialogs, you can use the RECEIVE dialog to
receive the function and any related service and HOLDDATA.
v You can use the RIMLIB job included on the CBPDO tape to receive the
function, service, and HOLDDATA shipped on the CBPDO. For more
information, see z/OS and z/OS.e Planning for Installation and the documentation
that was included with the CBPDO.
Receiving the function SYSMOD

Function SYSMODs obtained from IBM are packaged in RELFILE format. Before
any actual processing takes place, SMP/E must first determine if the SYSMODs
packaged in RELFILE format are to be received from tape or DASD. If the
SMPPTFIN data set is located on tape, SMP/E assumes that the RELFILEs are on
tape. If the SMPPTFIN data set is located on DASD, SMP/E assumes the RELFILEs
are on DASD and are cataloged.
During RECEIVE processing, the contents of the RELFILEs are placed into the
SMPTLIBs, which are used as temporary storage. SMPTLIBs that are uncataloged
Chapter 5. Installing a new function 103

are automatically cataloged by SMP/E. When the RELFILEs are on DASD, SMP/E
checks to ensure that the RELFILE and SMPTLIB names are not the same. If they
are, RECEIVE processing stops.
Note: Do not delete the SMPTLIBs after the RECEIVE step; they must be retained
until after the function is applied and accepted.
Updating the target libraries: The APPLY process

After preparing your target and distribution libraries and receiving the function
and any related service and HOLDDATA, the next step is to update your target
libraries. Review the program directory for the products you are installing.
When installing a new function, you are concerned with three groups of
SYSMODs:
v The function itself
v All PTFs applicable to the function
v All PTFs applicable to other functions that are specified as requisites of the
function or service applicable to the function
You may be able to apply all the required SYSMODs with one APPLY command.
This method has several advantages:
v It eliminates the need to run the APPLY command several times in order to
install the complete set of SYSMODs required.
v You replace elements in the target libraries less often; therefore, there is less risk
of running out of space.
v Because the SMP/E overhead and the number of invocations of the system
utilities are reduced, overall processing time is decreased.
Therefore, although SMP/E supports the separate installation of a new function

and its service, the common installation method is preferred unless the product
program directory for other unique installation requirements directs otherwise.
This is the method illustrated in subsequent examples. For more information about
the APPLY command operands, see The APPLY Command in SMP/E Commands.
When you are updating the target libraries, there are actually three distinct SMP/E
jobs to be run:
v Receive additional HOLDDATA. Before starting the APPLY, you should contact
the IBM Support Center to get additional HOLDDATA from the product PSP file
or the CORPE PSP file. Obtaining and receiving additional HOLDDATA is
covered under “Processing HOLDDATA from PSP files” on page 143. The other
two steps are covered in more detail in the following sections.
v Run the APPLY CHECK job. This is a nonupdating mode of APPLY. Its purpose
is to help resolve any problems that may prevent the APPLY from completing
processing successfully.
v Apply the SYSMOD updates. This installs the new function and service into the
target libraries.
Checking the update: The APPLY CHECK process

The purpose of this step is to determine:
v Whether any errors will occur while the new function is being applied (except
for errors that occur as a direct result of an update, such as a target library
running out of space). This includes missing DDDEF entries.
v Whether any requisite SYSMODs are missing.

v The target libraries that will be updated.

v The SYSMODs, if any, that will be regressed.
Use the SMP/E dialogs or the following sample job to do an APPLY CHECK for
the function and related SYSMODs:
//APPLY JOB ’accounting info’,MSGLEVEL=(1,1)
//APPLYCHK EXEC SMPPROC
//SMPCNTL DD *
SET BDY(ZOSTGT1) /* Set to target zone. */.
APPLY FORFMID(HXX1200) /* Apply for this FMID */
FUNCTIONS /* the function itself */
PTFS /* and all its PTFs. */
GROUPEXTEND /* Also all requisite PTFS. */
CHECK /* But do not update libs. */
BYPASS(HOLDSYSTEM /* Bypass options */
HOLDCLASS(UCLREL,ERREL)
)
.
/*
Researching the APPLY CHECK reports: As a result of running the APPLY

CHECK job, SMP/E produces various messages and reports that you should now
use to do further research. Here are some of the errors that may have been
detected:
v Some DD statements may be missing. Check the program directory or the
SMP/E Reference manual to determine why they are required and how they
should be specified.
v Some APAR fixes or USERMODs may be regressed. If so, you must determine
why. For APAR fixes, you have to get the version of the APAR fix applicable to
the new product. For USERMODs, you have to rework the modification to make
it applicable to the new function, or eliminate the modification if the product
being installed provides the same function. When doing the actual APPLY
operation, you may need to specify the BYPASS operand to inform SMP/E that
you have resolved these problems.
v Some prerequisite or requisite PTFs may be missing. If so, you should determine
whether they can be obtained. Some may already be on an ESO tape you have
in-house but have not received; others may not have been shipped, in which
case you have to get an early copy of them by contacting the IBM Support
Center. Although you can also avoid these conditions by using the BYPASS
operand, you are advised not to do this because the regressions have not been
resolved.
Note: Obtaining a product in a CBPDO greatly reduces the amount of work

needed to find requisite PTFs, because CBPDOs include all the service for
products applicable to the selected SREL.
v Some elements may not have been selected for installation. For each such
element, if the current functional owner (that is, FMID) is an IBM product, there
may not be a problem; this condition is common and occurs because there are
multiple functions with common elements. Check the program directory or
installation guide for the product you are installing to determine whether this
condition is normal or if it indicates a problem.
If the FMID is not one for an IBM product, further research is necessary. Contact
the current owner of the element to determine how that product is related to the
one you are installing.
v Some of the PTFs may not have been selected for installation because of
exception SYSMOD conditions identified by the ++HOLD MCSs. When

installing a new function, you may want to research these PTFs further. You can
use the reason ID and the comments specified in the ++HOLD MCS to
determine which of the following actions is most appropriate:
– Bypass the condition using the BYPASS(HOLDERR) operand
– Do not install the PTF
– Obtain a fix for the APAR
Getting additional SYSMODs: After doing the research step, you may decide
that additional SYSMODs are needed. If so:
1. Obtain the additional SYSMODs by using CBPDO, ESO, IBMLINK, or the IBM
Support Center.
2. Receive the additional SYSMODs, using the same source ID value as used
when processing the CBPDO tape.
3. Rerun the APPLY CHECK job.
Repeat this process until no new errors are reported.
Updating the target library: The APPLY process

After you have done the APPLY CHECK and the associated research and the other
necessary preparation, the APPLY job itself should be fairly simple. The APPLY job
does the same checking as the APPLY CHECK and then continues by calling the
appropriate system utilities to get all the elements installed.
Note: If a product deletes another product, you cannot use the RESTORE
command to back off the applied product and bring back the deleted one.
Use the SMP/E dialogs or the following sample job to apply the function and any
related SYSMODs:
//APPLY EXEC SMPPROC
//SMPCNTL DD *
SET BDY(ZOSTGT1) /* Set to target zone. */.
APPLY FORFMID(HXX1200) /* Apply for this FMID */
GROUPEXTEND /* Also all requisite PTFs. */
/* No check this time. */
BYPASS(HOLDCLASS(UCLREL,ERREL)/*Bypass options.*/
HOLDSYS(reason-id,...) .
/*
If you have obtained additional APAR fixes or USERMODs, you should either
specify each of these SYSMODs in the SELECT operand, or if all applicable APARs
and USERMODs are to be applied, specify the APARS and USERMODS operands.
Note: For most products, you do not have to do any additional processing to get
the elements into executable format. However, some products may require
you to run additional utilities or to perform extra steps after applying the
SYSMODs. See your product documentation for more information.
Testing the new function

After installing the new function, you should perform two operations:
1. Create a backup of the updated data sets, including any SMP/E data sets
affected, in case something happens to the data sets during the next phase.
Note: If you are doing the installation on a clone of your original system, you
already have a backup—your original system.

2. Do some testing before putting the new function into production. This testing
should include some of the following:
v If the product has supplied installation verification procedures (IVPs), they
should be run.
v If your installation has a test job stream, the tests should be run.
v If the new function could at all affect your ability to IPL the system, try an
IPL at this time.
Only after verifying the new function on a noncritical test system should you put
it into production. You should not consider the test phase completed until you
have run the new function in production mode for some period (as determined by
your installation requirements). Only then, if no errors are found, are you ready to
go to the next step, updating your distribution libraries.
Updating the distribution libraries: The ACCEPT process

The last major phase of installing a new function is to update the distribution
libraries, using the SMP/E ACCEPT command. This is a very critical step: Once
the function and its service have been accepted, there is no SMP/E method for
removing it from either the target or distribution libraries.
When you are ready to update your distribution libraries, you have the same set of
considerations and SMP/E support as described under “Updating the target
libraries: The APPLY process” on page 104, and the same three-phase operation:
1. Receive additional data. Before starting the ACCEPT process, you should
obtain any additional HOLDDATA for the function you are installing by using
CBPDO, ESO, IBMLINK, or the IBM Support Center. See “Processing
HOLDDATA from PSP files” on page 143 for additional information.
Note: If there is a significant time between the APPLY process and ACCEPT
process, additional problems may have been reported for which
++HOLD statements have been created. If this new data is not obtained,
SMP/E may install PE PTFs into the distribution libraries.
2. Run the ACCEPT CHECK job. This is a nonupdating mode of ACCEPT. Its
purpose is to help resolve any problems that may prevent the ACCEPT from
completing processing successfully.
3. Accept the SYSMOD update. This installs the new function and service into
the distribution library.
Checking the update: The ACCEPT CHECK process

The ACCEPT CHECK job provides the same function for the distribution libraries
that the APPLY CHECK job provides for the target libraries. See “Checking the
update: The APPLY CHECK process” on page 104.
Use the SMP/E dialogs or the following sample job to do an ACCEPT CHECK for
the function and related SYSMODs:
//ACCEPT JOB ’accounting info’,MSGLEVEL=(1,1)
//ACCEPTCK EXEC SMPPROC
//SMPCNTL DD *
SET BDY(DLIB1) /* Set to DLIB zone. */.
ACCEPT FORFMID(HXX1200) /* Accept for this FMID */
BYPASS(HOLDSYSTEM /* Bypass options */

HOLDCLASS(UCLREL,ERREL)
)
.
/*
Researching the ACCEPT CHECK reports: The ACCEPT CHECK reports should
be researched in the same manner as the APPLY CHECK reports (see “Researching
the APPLY CHECK reports” on page 105).
Getting additional SYSMODs: The process of getting additional SYSMODs or

APAR fixes for those PTFs being accepted is the same as during the APPLY process
(see “Getting additional SYSMODs” on page 106).
Updating the distribution library: The ACCEPT process

The ACCEPT process updates the distribution libraries. Use the SMP/E dialogs or
the following sample job to accept the function and any related SYSMODs:
//ACCEPT EXEC SMPPROC
//SMPCNTL DD *
ACCEPT FORFMID(HXX1200) /* Accept for this FMID */
BYPASS(HOLDCLASS(UCLREL,ERREL)/*Bypass options */
HOLDSYS(reason-id,...) .
/*
Note: If you have obtained additional APAR fixes or USERMODs, you should
either specify each of these SYSMODs in the SELECT operand or, if all
applicable APARs and USERMODs are to be installed, specify the APARS
and USERMODS operands.
Checking other zones for requisites: REPORT CROSSZONE

After installing a function, you may need to check other zones for conditional
requisites. A conditional requisite is software (such as service) that must be
installed for a given function if another function is also installed. To help automate
the research for conditional requisites, the installation logic (SMP/E modification
control statements) for a function uses ++IF statements to identify the requisites.
Conditional requisites may be for functions that are installed in different zones. If
you have set up automatic cross-zone requisite checking, as described in
“Specifying automatic cross-zone requisite checking” on page 78, SMP/E will
enforce cross-zone requisites during APPLY or ACCEPT processing. Otherwise, you
must use the SMP/E REPORT CROSSZONE command after a function and the
related service has been installed to manually identify cross-zone requisites defined
in the installation logic. To help you install the identified requisites, REPORT
CROSSZONE can also write APPLY and ACCEPT commands to the SMPPUNCH
data set. So, if you have not specified automatic cross-zone requisite checking, and
the function you have installed (or any related service) specifies conditional
requisites, you should run the REPORT CROSSZONE command against the target
and DLIB zones containing that product, as well as against the zones containing
the functions identified on the ++IF statements. For more information about the
REPORT CROSSZONE command, see Chapter 13, “Identifying cross-zone
requisites: The REPORT CROSSZONE command,” on page 159.

Chapter 6. Installing preventive service
This chapter describes the steps for installing preventive service. After an
introduction to preventive service and a summary of the preventive service
process, it discusses the following topics:
v Preparing your system
v Staging the SYSMODs with the RECEIVE command
v Requesting preventive service with the RECEIVE ORDER command
v Updating the target libraries with the APPLY command
v Testing the new service level
v Updating the distribution libraries with the ACCEPT command
Introduction
You install preventive service through the use of the SMP/E preventive service
process. The process uses:
v The preventive service SYSMODS received as either a CBPDO tape, an expanded
service options (ESO) tape, or a package downloaded from the IBM Server as a
result of a RECEIVE ORDER request.
v A well-defined set of steps that you should follow to install each service level in
order to bring your system up to the current service level.
ESOs are used as input to the software manufacturing database for the service to
be included in CBPDOs. As a result, there are many similarities between these two
offerings.
If you receive service and HOLDDATA from both CBPDO and ESO tapes, make
sure you install them in service level order (for example, service level 0301,
followed by service level 0302, and so on). Each PTF that is currently available on
a service level has a corresponding source ID. After you have received PTFs from
one of these service offerings (ESO or CBPDO), do not try to receive them again
from the other.
CBPDO tapes
CBPDO tapes can be ordered periodically, as you decide to update your system. A
CBPDO tape contains the PTFs, HOLDDATA, and PSP upgrade files to bring your
system up to a service level that you select. A CBPDO is ordered for a particular
feature (CICS, database system, MVS, or NCP). These features correspond to the
SRELs to which products are applicable.
You have two options when ordering a CBPDO: you can get products plus service,
or service only. (If you are interested just in installing preventive service, order a
service-only CBPDO.) With both of these options, you automatically receive service
for products for which you are already licensed under a single customer number
for a single feature.
The amount of service you receive in your CBPDO depends on your selection of a
service level and whether this is your first CBPDO order.
v If you select a service level, you get all service from that level to the current
level.

Preventive service
v If you do not select a service level and this is your first CBPDO order, you get
all the service shown on the order checklist.
v If you do not select a service level and you have ordered a CBPDO before, you
get service following the service level that was shipped in your previous
CBPDO.
The CBPDO tape is a standard label (SL) tape, with files arranged as shown in
Table 10.
Table 10. Format of a CBPDO tape
File number Processed by SMP/E Contents
1 No Documents
2 No Installation RIMs
3 Yes HOLDDATA for exception SYSMODs
4 No Program directories and PSP information
5 Yes SMPMCS file (MCS statements for
SYSMODs on the tape), PTFs, and cover
letters
6–n Yes RELFILEs for products on the tape
ESO tapes
As an IBM customer, you may periodically receive a customized ESO tape based
on your IBM software distribution profile. This tape has the PTFs, HOLDDATA,
++ASSIGN statements, and UCLIN data needed to bring your system up to the
current service level.
Service levels are identified according to when they were available correctively. For
example:
Service level When available correctively

0308 August 2003
0401 January 2004
0411 November 2004
The ESO tape is a nonlabeled (NL) tape, with files arranged as shown in Table 11
on page 110.
Table 11. Format of an ESO
Processed by
File number SMP/E Contents
1 Yes ++ASSIGN statements
PTFs within the requested service level range
PTFs that resolve PE PTFs contained in ESO

2 No Installation and usage instructions
3 No Softcopy of packing list for all PTFs
4 Yes HOLDDATA for exception SYSMODs
5 Optional UCLIN file

Preventive service
A RECEIVE ORDER request

You may order all currently available PTFs that meet your selection criteria. The
PTF package that results from such an order is tailored to your SMP/E
environment and contains all currently available PTFs that match your selection
criteria and are not already present in your environment. There are three selection
options:
Critical
Critical service includes all available PTFs that resolve high impact
pervasive (HIPER) problems or PTFs in error (PEs).
Recommended
Recommended service includes all available PTFs identified with a
Recommended Service Update SOURCEID (RSUyymm) and all available
PTFs that resolve a critical problem (HIPER or PE). Recommended service
includes PTFs through the most current RSU level.
All All service includes all available PTFs.
The package also contains the last 2-years of Enhanced HOLDDATA for the entire
z/OS software platform.
Preventive service process: Summary

The preventive service process has five phases:
1. Prepare your system.
2. Stage the preventive service.
3. Update the target libraries.
4. Test the system.
5. Update the distribution libraries.
The preventive service phases are the same as those defined in Chapter 7,
“Installing corrective service,” on page 123, although the steps within each phase
differ. These phases are the basic SMP/E operations to install any SYSMOD.
Each of these phases consists of a series of steps (SMP/E jobs, research, or

invocations of system utilities) that must be done to make sure the preventive
service is installed correctly and to ensure the integrity of your system libraries.
The steps defined are the normal steps for the installation of most PTFs. Any PTF
that requires special processing will contain instructions for installing it.
Generally, you should first attempt to install all the normal PTFs that you have
received and then to install those having special requirements. The intention is to
install the maximum number of preventive fixes on your system as soon as
possible.
Note: You should let SMP/E manage PE PTFs (PTFs discovered to be in error),
rather than researching and resolving them yourself.
The following sections describe, in detail, the steps necessary for each of the
preventive service phases.
accept preventive service. The basic steps are the same. If you have access to
the SMP/E dialogs, you should use them. Otherwise, you can use the steps
described in this chapter as examples.
Chapter 6. Installing preventive service 111

Preventive service

Before you start installing preventive service, you should do the following to make
sure that your system is ready and that you can recover in case of a serious failure
during installation:
v Get the latest HOLDDATA from IBMLINK or the IBM Support Center. You
process this HOLDDATA during the next phase.
As described in Chapter 9, “Managing exception SYSMODs,” on page 135, an
ESO may contain PTFs that were discovered to be in error (called PE PTFs) after
they were shipped to an IBM software distribution center. You can find out
about PE PTFs from IBMLINK or the IBM Support Center and build the
corresponding ++HOLD/++RELEASE statements necessary to prevent
introducing problems by processing PTFs with known errors.
If you used the RECEIVE ORDER command to request a preventive service
package, it contains the last 2-year’s HOLDDATA for the z/OS software
platform, so you can skip this step.
v Make sure you have the publications you need.
v Estimate the time you need for APPLY and ACCEPT processing. Make sure there
is time for these jobs to run to completion.
v Back up your disk packs.

After you prepare your target and distribution libraries, receive the preventive
service SYSMODs (PTFs) and the HOLDDATA (including data on the CBPDO or
ESO and any obtained from the IBM Support Center) into the SMP/E database
(the global zone and the SMPPTS).
v If the service was obtained from a CBPDO, you can use SMP/E dialogs or the
RIMLIB job included on the CBPDO tape to receive the service and HOLDDATA
shipped on the CBPDO. For more information, see the documentation that was
included with your tape.
v If the service was obtained from an ESO, you can use the SMP/E dialogs or the
following sample job to receive the service and HOLDDATA.
v If you used the RECEIVE ORDER command to request a preventive service
package and did not specify TRANSFERONLY, the process downloaded the
SYSMODS directly into the SMP/E database (the global zone and SMPPTS), so
you can skip this step.
//RECESO EXEC SMPPROC
//SMPHOLD DD DSN=HOLDDATA,
// UNIT=TAPE,
// VOL=(PRIVATE,RETAIN,SER=tape1),
// LABEL=(4,NL),
// DCB=(LRECL=80,BLKSIZE=7200,RECFM=FB),
// DISP=(SHR,PASS)
//SMPPTFIN DD DSN=SMPPTFIN,
// UNIT=AFF=SMPHOLD,
// VOL=(PRIVATE,RETAIN,SER=tape1),
// LABEL=(1,NL),
// DCB=(LRECL=80,BLKSIZE=7200,RECFM=FB),
// DISP=(OLD,PASS)
//SMPCNTL DD *
RECEIVE SYSMODS /* Receive all SYSMODs. */
HOLDDATA /* Receive HOLDDATA. */
SOURCEID( /* Optional: Assign SOURCEID */
xxxxxxxx) /* (SOURCEIDs are included */.
/* on all ESO tapes) */

Preventive service
Note: If multiple ESO service tapes are being RECEIVED, the additional tape
volumes must be concatenated under the SMPPTFIN DD card. See File 2 of
an ESO tape for further information.
In CBPDO and ESO tapes, PTFs are assigned SOURCEID values by ++ASSIGN
statements. You can assign an additional value by specifying it on the RECEIVE
command. The SOURCEID operand is used with the MCS operand on the LIST
command to list the PTF cover letters. This is used rather than the LIST operand
on the RECEIVE command because the output is in a more usable format.
v For ESO tapes, you should substitute a meaningful value for xxxxxxxx in each
command shown above. The value should be unique and easily tied to an ESO.
v For CBPDO tapes, the recommended format is PDOyyww, where yy is the year
and ww the week of the CBPDO tape.
The DCB values shown for SMPHOLD and SMPPTFIN are those used for
preventive service when this publication was written. If these values change, use
the ones defined for the ESO.
For both CBPDO and ESO tapes, you should call the IBM Support Center to obtain
additional HOLDDATA (unless you just received your tape). For additional
information, see “Processing HOLDDATA from PSP files” on page 143. You can use
the SMP/E dialogs or the following sample job to process the data set with the
HOLDDATA obtained from the IBM Support Center.
//RECEIVE JOB ’accounting info’,MSGLEVEL=(1,1)
//RECESM EXEC SMPPROC
//SMPHOLD DD ...data describing your data set
//* ...must be RECFM=FB, LRECL=80
//SMPCNTL DD *
RECEIVE HOLDDATA /* Receive HOLDDATA. */.
/*
The HOLDDATA you obtain from the IBM Support Center should be in SMP/E
format. If not, or if you are creating your own HOLDDATA, see the SMP/E
Reference manual to review the syntax for the ++HOLD statements.

After you prepare your target and distribution libraries and receive all the
necessary PTFs and HOLDDATA, update the target libraries. Though most PTFs
can be installed directly into the target libraries, some require special processing,
such as a fix that must be concurrently installed on all processors in a network.
These PTFs contain a ++HOLD statement that automatically places them into
HOLD for SYSTEM action status; that is, SMP/E does not allow them to be
installed unless you take some direct action, such as specifying
BYPASS(HOLDSYS) on the APPLY command. These PTFs should not be processed
immediately; you should attempt to install all PTFs not requiring such actions and
then return to process these. For additional information about these PTFs, see
“Installing PTFs that need special processing” on page 118.
When installing preventive service, you are concerned with two groups of PTFs:
v All PTFs from the CBPDO, ESO, or requested service package you are installing
v Any other PTFs that are required to install these PTFs

Preventive service
SMP/E provides operands (SOURCEID and GROUP or GROUPEXTEND) on the

APPLY command that facilitate the installation of all required PTFs by use of one
APPLY command. Installing all PTFs with one APPLY command provides several
advantages:
v It eliminates the need to run the APPLY command several times in order to
install the complete set of PTFs required.
v It reduces the risk of running out of space, because you are replacing elements
in the target libraries less often.
v It decreases overall processing time, because there is less SMP/E overhead and
the system utilities are invoked less often.
When you update the target libraries, there are three distinct SMP/E jobs to be
run:
1. Receive additional HOLDDATA. Before starting the APPLY, you should
contact the IBM Support Center to obtain any additional HOLDDATA for the
CBPDO or ESO you are installing. This step is required if:
a. You did not obtain the additional HOLDDATA from the IBM Support
Center during the staging phase.
b. There has been a delay between the RECEIVE and APPLY staging phase
and the target update phase.
We will not discuss this first step further here. If you need to perform this step,
see “Staging the SYSMODs: The RECEIVE process” on page 112.
2. Run the APPLY CHECK job. This second step is a nonupdating mode of
APPLY, referred to as the APPLY CHECK run. Its purpose is to assist in
resolving any problems that prevent the APPLY itself from completing
processing successfully.
3. Run the APPLY job. This third step is the updating mode of APPLY, in which
the preventive service is installed into the target libraries.
The following sections describe the last two steps as well as the processing of PTFs
that require special processing.
Checking the update (APPLY CHECK)

The purpose of this step is to determine:
v Whether any errors will occur when you apply a SYSMOD (except for those
error conditions that occur as a direct result of an update, such as a target
library running out of space)
v Whether any requisite PTFs are missing
v The target system libraries that will be updated during APPLY
v The PTFs or APARs, if any, that will be regressed during APPLY
The GROUP and GROUPEXTEND operands allow SMP/E to include any PTF that
may be required to install PTFs on the current service level (PUT0303 in the
example that follows). Some of the PTFs on previous tapes may not have been
installed, because they were in hold status (PE PTFs) at the time the ESO
containing the service level was installed. The current service level may contain
fixes for the APARs that caused the original PTFs to be held. These PTFs, because
they have module intersections with the PE PTF, must either be prerequisite to the
old PTFs or must supersede them so SMP/E can automatically include the old
PTFs when the fixing PTF is installed.

Preventive service
The following sample job shows how to do an APPLY CHECK for preventive
service:
//SMPCNTL DD *
SET BDY(TGT1) /* Set to target zone. */.
APPLY SOURCEID(PUT0303) /* Apply this service level */
GROUPEXTEND /* and all requisite PTFs, */
CHECK /* but do not update libs. */
SELECT(sysmod-id,...) /* Select additional */
/* service if required. */
BYPASS(HOLDCLASS(ERREL,UCLREL)
HOLDSYSTEM) .
/*
You may be able to improve SMP/E performance by including the source IDs for
previous service levels within the SOURCEID operand. The following job provides
an example of an APPLY CHECK job for PTFs in service level 0303:
//SMPCNTL DD *
APPLY SOURCEID(PUT0303 /* Apply this service level */
PUT0302 /* and back-level tapes */
PUT0301) /* back to some reasonable */
/* level. */
GROUPEXTEND /* And all requisite PTFs. */
HOLDSYSTEM) .
/*
Note: This form of the SOURCEID operand can also be used to group service
levels initially in one APPLY command.
If you want to install preventive service only on selected functional areas of the
system, you can also specify the FORFMID operand on the APPLY command,
specifying either specific function identifiers (FMIDs) or the name of one or more
FMIDSETs.
Researching the APPLY CHECK reports

As a result of running the APPLY CHECK job, SMP/E produces various messages
and reports you should now use to perform further research. Here are some of the
errors that may be detected:
v Some DD statements may be missing. Check the SMP/E Reference manual to
determine why they are required and how they should be specified.
v Some APARs or USERMODs may be regressed. If so, you must determine why.
For APARs, obtain the version of the APAR fix applicable to the service level.
For USERMODs, rework the modification to be applicable to the new service
level. When performing the actual APPLY operation, you most likely need to
specify the BYPASS operand in order to inform SMP/E that you have resolved
these problems.
v Some requisite PTFs may be missing. If so, you should determine how they can
be obtained. Some may be on service levels you have not received; others may
not have been shipped, in which case you have to obtain an early copy of them

Preventive service
by contacting the IBM Support Center. Although you can get around these
conditions by using the BYPASS operand, you are advised not to do this because
the regressions have not been resolved.
v Some of the PTFs are not installed because of exception SYSMOD conditions
identified by the ++HOLD statements. You should ignore these PTFs until a
fixing PTF is delivered in a subsequent service level.
Note: Depending on your requirement to install such PTFs, you can use the
reason ID and the comments specified in the ++HOLD statement to
determine which of the following actions is most appropriate:
– Bypass the condition using the BYPASS(HOLDERR) operand
– Do not install the PTF
– Obtain a fix for the APAR
A common cause of regression is user modification. When USERMODs are applied

to your system, the service level information (RMID or UMID) is altered to reflect
these additions. The APPLY CHECK may have flagged a SYSMOD as one that
would cause regression.
This regression-handling procedure works under the assumption that you have
applied, but not yet accepted, a USERMOD. This means that the USERMOD has
applied service to the target libraries, but the service in your distribution library is
that which the SYSMOD should be applied against.
You can follow these steps for handling regression:

1. Restore the module from the distribution library back into the target system to
back off the USERMOD.
2. Apply the SYSMOD in question to the target system in order to keep SMP/E’s
information about the target system up to date.
3. Accept the SYSMOD into the distribution libraries.
When USERMODs are applied on a system, it is up to you to ensure that they are
at the proper level.
If you reapply your USERMOD at this point, remember to exclude it when

accepting the preventive service if you want to be able to restore your system to
the level assumed by the next preventive service update.
The following steps describe regression handling.

1. Restore APARs or USERMODs, if necessary. Use the RESTORE command to
remove the APAR or USERMOD from the target libraries. This places into the
target library the versions of the elements that currently exist in the distribution
library.
Use the SMP/E dialogs or the following sample job to restore the SYSMODs:
//SMPRESTR JOB ’accounting info’,MSGLEVEL=(1,1)
//RESTORE EXEC SMPPROC
//SMPCNTL DD *
RESTORE /* Put DLIB data into */
/* target libraries. */
SELECT(AZ00001, /* Must be SELECT or GROUP, */
AZ00002) /* NOT by source ID. */.
/*
2. Repeat the APPLY CHECK. This gives you an updated status report to
determine that all regression conditions have been addressed.

Preventive service
3. If a USERMOD or APAR is necessary, compare the PTF just flagged by

APPLY CHECK with the APAR or USERMOD that caused the regression. You
may need microfiche or a dump of the module. If any changes are needed,
follow the steps below. Otherwise, continue with the APPLY step.
v Do one of the following:
– For a USERMOD, add the REWORK operand to the ++USERMOD MCS.
The REWORK operand allows the updated SYSMOD to be automatically
rereceived, as long as it is more recent than the version that has already
been received. This takes the place of rejecting the SYSMOD and receiving
it.
– For an APAR or USERMOD, reject the prior copy from the SMPPTS.
The SMP/E REJECT job removes the USERMOD or APAR from the
SMPPTS. This prevents the prior copy from being applied again.
A sample REJECT job follows:
//SMPREJ JOB ’accounting info’,MSGLEVEL=(1,1)
//REJECT EXEC SMPPROC
//SMPCNTL DD *
REJECT /* Remove these two */
/* SYSMODS from the */
S(AZ00001, /* PTS and global zone. */
AZ00002) /* */.
/*
v Receive the USERMODs or APARs. This loads the SYSMODs into the
SMPPTS.
Getting additional SYSMODs

After doing the research step, you may decide that additional SYSMODs are
needed. These should be obtained from the IBM Support Center and then received
into the SMPPTS.
At this time, you should modify the APPLY command to add a SELECT operand
specifying each of the PTFs obtained from the IBM Support Center. An alternative
is to assign all such PTFs the same source ID value as the service level, or to assign
them a unique value and then add that value to the SOURCEID operand.
This process should continue until no new errors are reported.
Updating the target library (APPLY)

If the suggested preparation and all phases of the APPLY CHECK are completed,
the APPLY job itself should be fairly simple. The APPLY job performs the same
checking as the APPLY CHECK and then calls the appropriate system utilities to
install all the elements.
Use the SMP/E dialogs or the following sample job to apply the preventive
service:
//SMPCNTL DD *
APPLY SOURCEID(PUT0303) /* Apply this service level */
GROUPEXTEND /* and all requisite PTFs, */
SELECT(UZ00001 /* Plus two other PTFs. */
UZ00002 /* */
AZ12345 /* Plus two APAR fixes. */
AZ12346) /* */

Preventive service

HOLDSYSTEM) .
/*
v If you have obtained additional APAR fixes or USERMODs, you should either
specify each of these SYSMODs in the SELECT operand or, if all applicable
APARs and USERMODs are to be installed, specify the APARS and USERMODS
operands.
v If any of the SYSMODs specified in the SELECT list have already been applied
and you want to reinstall them, you must also specify the REDO operand on the
APPLY command.
v If you want to install preventive service only on selected functional areas of the
system, you can also specify the FORFMID operand on the APPLY command,
specifying either specific function identifiers (FMIDs) or the name of one or
more FMIDSETs.
Installing PTFs that need special processing

There are so many reasons a PTF may require special processing that it is
impossible to document how you should handle each case. Any PTF requiring
special processing should contain a ++HOLD statement (after all the ++VER
statements and before the first element MCS). That ++HOLD statement should be
as follows:
++HOLD(sysmod-id) /* Originating SYSMOD ID. */

SYSTEM /* Special processing info. */
FMID(sysmod-id) /* Functional owner. */
REASON(reason-id)
COMMENT(....
....any amount of comment text
....
)
.
See SMP/E Reference for a detailed description of the ++HOLD statement syntax.
The comment text within the ++HOLD statement, or in the PTF cover letter,
contains a description of all the special processing necessary to install this PTF.
Testing the new service level

After having installed the new service, you should perform two operations:
1. Create a backup of the updated data sets, including the SMP/E data sets
affected. This ensures that if something happens to the data sets during the
next phase, you do not have to repeat all the processing done in previous steps.
2. Perform some testing before putting the service into production. This testing
should include some of the following:
v Run selected product IVP jobs.
v Run test job streams, if your installation has them.
v Attempt an IPL.
Only after verifying the service on a noncritical test system should you put that
service into production. The test phase should not be considered complete until
you have run the service in production mode for some period (as determined by
the requirements for your installation). If no errors are found, you are ready to
proceed to the next step: updating your distribution libraries.

Preventive service

The last major phase of installing preventive service is updating the distribution
libraries with the SMP/E ACCEPT command. This is a very critical step. Once the
service is accepted, there is no SMP/E method to remove it from either the target
or distribution libraries.
When you are ready to update your distribution libraries, you have the same set of
considerations and SMP/E support as described under “Updating the target
libraries: The APPLY process” on page 113, and the same three-phase operation:
1. Receive additional HOLDDATA. Before starting the ACCEPT, you should
obtain any additional HOLDDATA for the service level you are installing. This
step is required if:
v You did not obtain the additional HOLDDATA from the IBM Support Center
during the staging phase.
v There has been a delay between the RECEIVE staging phase and the
ACCEPT DLIB update phase.
2.
Notes:
a. If there is a significant time between the APPLY and ACCEPT, additional
problems may have been reported for which ++HOLD statements have
been created. If this new data is not obtained, SMP/E may install PE PTFs
into the distribution libraries.
b. You may want to run the REPORT ERRSYSMODS command to see whether
any SYSMODs that are applied are now in error. For more information, see
Chapter 14, “Identifying installed SYSMODs affected by error holds: The
REPORT ERRSYSMODS command,” on page 163.
3. Run the ACCEPT CHECK job. The second job is a nonupdating mode of
ACCEPT, referred to as the ACCEPT CHECK run. Its purpose is to help resolve
any problems that prevent the ACCEPT itself from successfully completing
processing.
4. Run the ACCEPT update. The third job is the updating mode of ACCEPT, in
which the preventive service is installed into the distribution libraries.
Note: Special processing may be required during the ACCEPT process. PTFs
requiring this processing should be handled in the same manner as during
the APPLY process.
Checking the update (ACCEPT CHECK)

that the APPLY CHECK job provided for the target libraries. See “Checking the
update (APPLY CHECK)” on page 114.
Use the SMP/E dialogs or the following sample job to do an ACCEPT CHECK for
preventive service. This example is an ACCEPT CHECK job for PTFs in service
level 0303:
//SMPCNTL DD *
ACCEPT SOURCEID(PUT0303) /* Accept this service level*/
GROUPEXTEND( /* Include requisite PTFs */
NOAPARS /* Don’t include APARs or */
NOUSERMODS) /* USERMODs */

Preventive service
CHECK /* but do not update libs. */

HOLDSYSTEM) .
/*
Note: This example can be used for PTFs from either a CBPDO or an ESO.
If you want to install preventive service only on selected functional areas of the
system, you can also specify the FORFMID operand on the ACCEPT command,
specifying either specific function identifiers (FMIDs) or the name of one or more
FMIDSETs.
Researching the ACCEPT CHECK reports

The ACCEPT CHECK reports should be researched in the same manner as the
APPLY CHECK reports (see “Researching the APPLY CHECK reports” on page
115).

The procedure for getting additional SYSMODs or APAR fixes from those PTFs
being accepted is the same as that followed during the APPLY process (see
“Getting additional SYSMODs” on page 117).
If you obtain additional SYSMODs during the ACCEPT phase, you should process
these through the APPLY phase before accepting them.
Updating the distribution library (ACCEPT)

The ACCEPT command causes SMP/E to update the distribution libraries. You can
use the SMP/E dialogs or the following sample job to accept the preventive
service:
//SMPCNTL DD *
ACCEPT SOURCEID(PUT0303) /* Accept this service level*/
GROUPEXTEND( /* Include requisite PTFs */
NOAPARS /* Don’t include APARs or */
NOUSERMODS) /* USERMODs */
HOLDSYSTEM) .
/*
Notes:
1. If you have obtained additional APAR fixes or USERMODs, you should either
specify each of these SYSMODs in the SELECT operand or, if all applicable
APARs and USERMODs are to be installed, specify the APARS and
USERMODS operands.
2. If you want to install preventive service only on selected functional areas of the
system, you can also specify the FORFMID operand on the ACCEPT command,
specifying either specific function identifiers (FMIDs) or the name of one or
more FMIDSETs.

Preventive service
Installing PTFs that need special processing

During the ACCEPT process, the considerations for special processing are the same
as for the APPLY process (see “Installing PTFs that need special processing” on
page 118).

Chapter 7. Installing corrective service
This chapter describes the steps for installing corrective service. It discusses these
topics:
v An introduction to corrective service
v Building and checking a corrective service fix
v Staging the SYSMODs with the RECEIVE command
v Requesting corrective service with the RECEIVE ORDER command
v Testing the corrective service
Introduction
Corrective service is the process of installing a SYSMOD to resolve a specific
problem in your system. The problem has usually been brought to your attention
because the system has not functioned as expected (for example, an abend has
occurred, or jobs are not running as expected).
The first task is to investigate the problem, so that the failing component and
module can be identified. SMP/E Messages, Codes, and Diagnosis provides helpful
information on diagnosing and handling SMP/E problems. This can be done in
conjunction with the IBM Support Center. SMP/E can help you work with the IBM
Support Center to isolate and obtain a fix for the problem. Useful information
includes:
v The function and service level of the module involved
v The service level of your system—that is, the specific SYSMODs that have been
installed
v Any USERMODs involved
v The affected load modules
After determining the cause of the error, you want a fix for the problem. There are
several possibilities:
1. The problem has already been reported, and a PTF has been created to fix the
module. That PTF may not have been shipped on an ESO. If it has been
shipped, you should have it in your shop (already received). If not, the IBM
Support Center can help you get a copy of it, or you can use ShopzSeries and
the RECEIVE ORDER command to order the PTF from the Automated Service
Delivery server.
2. The PTF identified by the Support Center may have been received but not yet
applied. Use the LIST command or the SMP/E Query dialog to check the status
of the PTF.
3. The problem has been previously reported. No PTF has been created, but an
APAR fix is available either to fix or to bypass the problem.
4. The problem is a new one, and no fix is available. In this case, you have to
work with the IBM Support Center to construct a fix for your system.
No matter where you obtain the fix, the installation of that fix is said to be in
corrective mode.

Corrective service
Note: You can use either the SMP/E dialogs or JCL jobs to build, receive, apply,
and accept corrective service. The basic steps to follow are the same. If you
have access to the SMP/E dialogs, you should use them. Otherwise, you can
use the steps described in this chapter as examples.
Building or checking the fix

If the fix is a PTF, either from a CBPDO, an ESO, the IBM Support Center, or the
Automated Service Delivery server, you can assume that the construction of that
fix is accurate and the material in this section is not applicable.
If the fix obtained is not a PTF, you should make sure it was constructed
accurately. This is true even if the fix obtained from the IBM Support Center is
already in SMP/E format (that is, you received a ++APAR type SYSMOD).
If you have to build the fix yourself, see z/OS Packaging Rules for rules for
constructing APAR SYSMODs and SMP/E Reference for details on MCS syntax. (To
get started, you might find Chapter 17, “Building a user modification,” on page
169 helpful.) The general format of all ++APAR fixes is:
++APAR(xxxxxxx) /* APAR identifier */.
++VER (srel) /* System identifier */
FMID(aaaaaaa) /* Functional area */
PRE ( /* PRE some SYSMODs. */
bbbbbbb /* PRE RMID of element */
ccccccc ccccccc /* and any UMIDs present. */
ccccccc ccccccc /* */
) /* */
SUP ( /* SUP some SYSMODs. */
ddddddd ddddddd /* Fixes incorporated into */
ddddddd ddddddd /* this fix */
) /* */.
++ZAP (modname) /* */
DISTLIB(eeeeeeee) /* DLIB name */.
... Some superzap cards here
or
++MACUPD (macname) /* */
... Some IEBUPDTE cards here
or
++SRCUPD (srcname) /* */
... Some IEBUPDTE cards here
You should perform the following checks to ensure the accuracy of the fix:
1. Make sure the value xxxxxxx in the ++APAR statement is equal to the APAR
number associated with the problem you are fixing.
2. Make sure the system release value (SREL) in the ++VER statement matches
one of those defined as an SREL subentry in the TARGETZONE entry for your
target zone.
3. Make sure the FMID value aaaaaaa in the ++VER statement matches the FMID
value in the appropriate element entry in your target zone. You can determine
this value by listing the appropriate entry.
4. If the element entry in your target zone has an RMID value different from its
FMID value, make sure it is a prerequisite of the APAR fix (that is, make sure
the bbbbbbb value is accurate). If the RMID and FMID values are equal, the
bbbbbbb value need not be specified.

Corrective service
5. If the element entry in your target zone has any UMID values, you should first
make sure the fix itself was constructed so that it works correctly in that
environment.
You should then make sure each of the UMID values is specified in the PRE
operand in place of the ccccccc values shown in the example. This is not
absolutely required, but if it is not done, SMP/E issues warning messages
during installation indicating that these SYSMODs may have an intersection
with the one you are installing, and therefore may be regressed. Putting the
UMID values in the PRE list suppresses these messages.
6. If this SYSMOD is to fix multiple problems, each of the additional APARs that
are being fixed should be specified in the SUP operand (ddddddd values in the
example).
7. Make sure the name in the ++ZAP, ++MACUPD, or ++SRCUPD statement is
correct.
8. Make sure the value eeeeeeee specified in the DISTLIB operand matches the
DISTLIB value in the target zone entry.
Note: The DISTLIB value is optional, but it is a good idea to specify it to make
sure there is no mistake about which element you are dealing with.
Once you have made sure the SYSMOD is accurate, you are ready to start the
actual installation process.

Corrective service is very different from the installation of a new function or
preventive service.
v It usually affects a very limited area of the system.
v It is usually done because a severe problem is affecting the system.
v There is a need for an immediate fix.
v The fix usually takes little time to install.
Thus, there is usually no need or time to prepare the system by backing up packs,
compressing libraries, and so on. If possible, it is a good idea to have a backup
system available in case a problem does occur.
Staging the SYSMODs: the RECEIVE process

After verifying that the corrective SYSMOD is syntactically correct and specifies
the proper set of functions and PTFs, receive that SYSMOD (APAR or PTF) into the
SMP/E database so you can install it into the target libraries.
Because corrective service requires a very small number of SYSMODs—often only

one—the job of receiving it is very simple. You can use the SMP/E dialogs or the
following sample job:
//RECEIVE EXEC SMPPROC
//SMPPTFIN DD ...points to input with SYSMOD
//* If you put the SYSMOD in a data set
//* refer to that data set.
//* If the SYSMOD is in card format
//* use "DD *" followed by the cards.
//SMPCNTL DD *
RECEIVE SELECT( /* Receive selected SYSMODs.*/
xxxxxxx /* Specify SYSMOD number. */
Chapter 7. Installing corrective service 125

Corrective service
) /* */
SYSMODS /* Only process SMPPTFIN -
do not look at SMPHOLD. */.
/*
Note: No source ID was assigned. This is because the SYSMOD is installed

selectively in the APPLY step. If you want to assign a common value or tag
the SYSMOD with some sort of identifier (such as programmer initials), you
can use the SOURCEID operand.
If the input data set contains only the SYSMODs you are installing for this
corrective service problem, you can omit the SELECT operand. SMP/E then
attempts to process all the SYSMODs in the SMPPTFIN input data set.
Generating a service request using the RECEIVE ORDER Command

You may order specific PTFs by name and you may order PTFs that resolve
specific APARs. The package that results from such an order is tailored to your
SMP/E environment and contains the PTFs you requested, plus any requisite PTFs
if those requisites are not already present in your environment.
You can use the SMP/E dialogs or the following sample job:
//jobname JOB ...
//SMPCSI DD DSN=SMPE.GLOBAL.CSI,DISP=SHR
//SMPCNTL DD *
SET BOUNDARY(GLOBAL).
ORDER( /* Place an order for service */
ORDERSERVER(ORDRSRVR)
CONTENT(
PTFS(UQ12345,UQ98765) /* Get these PTFs, and any.. */
) /* ..requisites.. */
FORTGTZONES(ZOS14) /* ..for this target zone */
).
/*
//ORDRSRVR DD *
<ORDERSERVER
keyring="MRWKYRNG"
</ORDERSERVER>
/*
Updating the target libraries: the APPLY process

After receiving the corrective service, you are ready to install it into the target
libraries. You should not attempt to install the SYSMODs without first verifying
them. If you have done all the proper checking before this time, the SYSMODs
should be installed correctly. However, if you have overlooked something, the
direct installation may cause unexpected results.

Corrective service

The purpose of this job is to verify that the SYSMOD(s) can be installed correctly
and that you understand what libraries and load modules in the system are
affected. You can use the SMP/E dialogs or the following sample job to do an
APPLY CHECK for corrective service:
//SMPCNTL DD *
APPLY SELECT( /* Install selected SYSMOD. */
xxxxxxx /* Specify SYSMOD name here.*/
) /* */
CHECK /* In check mode only. */.

Review the reports from the check, looking for the following types of information:
v Were any error messages produced? If so, determine the cause and fix the
problem.
v Will any SYSMODs be regressed? If so, determine how to resolve the problems.
v Are any other areas of the system affected?
Using HOLDDATA to assist in identifying fixes: If the SYSMOD you are

installing is a PTF (obtained from a CBPDO, an ESO, or directly from the IBM
Support Center), SMP/E may have some HOLDDATA stored applying to that PTF.
If so, the reports will indicate all the reason IDs preventing PTF installation. You
should use these reason IDs to determine what the errors are. This can be done by:
1. Listing the SYSMOD and MCS entries for the PTF.
2. Looking at the ++HOLD statements that are listed.
3. Using the COMMENT field in the ++HOLD statement to determine the cause
of the error. If the COMMENT field is not present or does not describe the
problem adequately, ask the IBM Support Center for further information.
4. Determining whether the error in the PTF is critical enough to stop it from
being installed. Remember that you are trying to fix an existing problem; you
may decide to install the PTF to fix that problem because the exposure is
minimal.
5. If necessary, contact the IBM Support Center to obtain a corrective fix for the
PTF. If, in the preceding step, you decided that the PTF should be installed
immediately to fix your problem, you should perform this step at some later
date.

If the research of the APPLY CHECK reports discloses that additional SYSMODs
are required, these should be obtained in the same manner as the original
corrective SYSMOD.

Once the APPLY CHECK has run to your satisfaction, you are ready to install the
fix using the SMP/E dialogs or the following sample job to apply the corrective
service:
//SMPCNTL DD *

Corrective service

) /* */
/* Note no check operand. */.
Testing the corrective service

The testing needed after you install a corrective fix depends on the type of
problem you encountered. It may range from no testing to running the job in
which the error was detected.
Updating the distribution libraries: the ACCEPT process

Once you have installed the corrective service into the target libraries, you must
decide whether you want to update the distribution libraries. You should base this
decision on the products involved and on your processing requirements.
The following is a consideration for not accepting corrective service:

Corrective service has not been tested and, therefore, may be found to be in
error at some later date. Once you have accepted the SYSMODs, there is no
RESTORE capability.
The following are some of the considerations for accepting corrective service:
v If you do not accept the SYSMOD and you perform a system generation, all that
service is lost and must be reinstalled.
v If you must restore a SYSMOD, the work required increases with the number of
SYSMODs that have been applied but not accepted. All intersecting SYSMODs
must be restored, and then all but the desired SYSMOD must be reapplied. This
is especially true for source-modified products.
The following sections describe the steps you should use, assuming you have
decided to accept some corrective service.
Checking the update (ACCEPT CHECK)

that the APPLY CHECK job provided for the target libraries. It is important
because the function and service level of the modules in the distribution libraries
may be different from that in the target libraries. You can use the SMP/E dialogs
or the following sample job to do an ACCEPT CHECK for corrective service:
//SMPCNTL DD *
ACCEPT SELECT( /* Install selected SYSMOD. */
) /* */
Researching the ACCEPT CHECK reports

You should research the ACCEPT CHECK reports in the same way as for the
APPLY process (see “Researching the APPLY CHECK reports” on page 127).
Using HOLDDATA to assist in identifying fixes: If SMP/E reported any

exception SYSMOD data during the APPLY CHECK process, you should expect to
see the same information during the ACCEPT CHECK process. If you have

Corrective service
processed any HOLDDATA between the APPLY and ACCEPT, additional

information may be reported. This information should be handled in the same
manner as the APPLY information.

If additional SYSMODs are required to ACCEPT the corrective service, you should
obtain them in the same manner as the original corrective service SYSMOD.
Note: If you obtain additional SYSMODs, make sure you process them through the
APPLY and test phases before accepting them.
Updating the distribution library (ACCEPT)

Once the ACCEPT CHECK runs to your satisfaction, you are ready to accept the
fix. Use the SMP/E dialogs or the following sample job to accept the corrective
service:
//SMPCNTL DD *
ACCEPT SELECT( /* Install selected SYSMOD. */
) /* */

Corrective service

Chapter 8. Installing a user modification
This chapter describes the steps for installing a user modification (USERMOD).
After an introduction to USERMODs, it describes the following processes:
v Staging the USERMOD with the RECEIVE command
v Testing the USERMOD
Introduction
A USERMOD is a SYSMOD used to make a modification to some IBM-supplied
software element (module, macro, source, or data element) to implement a new
function or to provide a hook into a user program that provides that function.
A USERMOD should not be confused with an APAR SYSMOD (corrective fix),

even if you built the initial version of that fix to fix a problem immediately. For a
description of how to construct USERMODs, see Chapter 17, “Building a user
modification,” on page 169. For details on the syntax of the MCS statements used
in constructing USERMODs, see the SMP/E Reference manual. The z/OS Packaging
Rules contains additional information on when a USERMOD should be used. The
rest of this chapter assumes that you have properly constructed the USERMOD
and are now ready to install it.
accept USERMODs. The basic steps to follow are the same. If you have
access to the SMP/E dialogs, you should use them. Otherwise, you can use
the steps described in this chapter as examples.

You must determine the amount of system preparation necessary for your
USERMOD. If it is extensive and affects critical components of the system, you
should perform the same tasks as defined under “Preparing your system” on page
102 or “Preparing your system” on page 112. If it is a minor change, affecting very
few modules and not critical to the operation of the system, no preparation is
needed.

Because a USERMOD is generally processed as a single SYSMOD, processing is
very similar to that for corrective service; that is, it is received by use of the
SELECT option. Use the SMP/E dialogs or the following sample job to receive the
USERMOD:
//SMPPTFIN DD ...points to input with your USERMOD
//* If you put the USERMOD in a data set
//* refer to that data set.
//* If the USERMOD is in card format
//* use "DD *" followed by the cards.
//* Create your data set in LRECL=80,
//* FB format.

Installing USERMODs
//SMPCNTL DD *
RECEIVE SELECT( /* Receive selected SYSMODs.*/
xxxxxxx /* Specify USERMOD number. */
) /* */
SYSMODS /* Only process SMPPTFIN -
do not look at SMPHOLD. */.
/*
Note: No source ID was assigned, because the SYSMOD is installed selectively in

the APPLY step. If you want to assign a common value to all the
USERMODs or tag each of them with some sort of identifier (such as
programmer initials), you can use the SOURCEID operand.
If the input data set contains only USERMODs that you want to receive now, you
can omit the SELECT operand. SMP/E then attempts to process all SYSMODs in
the SMPPTFIN input data set.

After receiving the USERMOD, you are ready to install it into the target libraries.
You may be tempted to install the SYSMODs without first performing the
verification pass. If you have constructed your USERMOD correctly, it should
install correctly. However, if you have overlooked something, the direct installation
may cause unexpected results. Thus, it is advisable to perform the verification
pass.

The purpose of this job is to verify that the SYSMODs are installed correctly and
that you understand which libraries and load modules in the system are affected.
Use the SMP/E dialogs or the following sample job to do an APPLY CHECK for
the USERMOD:
//SMPCNTL DD *
) /* */
Note: At times, it may be necessary to reinstall a USERMOD—for example, after

the installation of a PTF. If you are reinstalling it, the APPLY REDO operand
is necessary. You may also have to specify one of the BYPASS operands; that
depends on the relationship between your USERMOD and the PTF that was
installed.

Review the reports from the APPLY CHECK process, looking at the following
types of information:
v Were any error messages produced? If so, determine the cause and fix the
problem.
A common error here is that the FMID specified on the ++VER modification
control statement did not match the FMID value in the element entry; therefore,
SMP/E does not select the element to be installed. This condition does not stop
the USERMOD from being installed. However, messages are issued to say which
elements were not selected.

Installing USERMODs
v Will any SYSMODs be regressed? If so, determine how to resolve the problems.

Once the APPLY CHECK runs to your satisfaction, you are ready to install the
USERMOD. Use the SMP/E dialogs or the following sample job to apply it:
//SMPCNTL DD *
) /* */
Testing the USERMOD

The amount of testing needed after the installation of a USERMOD depends on the
changes you are making. You may want to review the recommendations found
under “Testing the new function” on page 106 and “Testing the new service level”
on page 118.
When originally constructing your USERMOD, you may want to provide a

document similar to a program directory, containing some of the following
information:
v The elements affected.
v The areas within each element.
v Externals of the change.
v An IVP job that can be used to ensure that the USERMOD is working correctly.
This can be used after subsequent preventive service is applied or if the
USERMOD must be reinstalled because a new release of the IBM product is
installed.
This information may be helpful to the next system programmer responsible for
installing and maintaining your USERMOD.

Once you have installed the USERMOD into the target libraries, you must decide
whether you want to update the distribution libraries. While this decision is up to
you, IBM generally does not recommend the accepting USERMODs, because if a
problem is encountered in the modified modules, you may be asked to re-create
the problem using an unmodified version. If you have accepted the USERMOD,
you cannot create an unmodified version of the module unless you are also
maintaining a separate set of distribution libraries without the USERMODs.
Note: You can use ++HOLD statements to prevent USERMODs from being
accepted. For each USERMOD that you want to keep from being accepted,
follow these steps after applying the USERMOD:
1. Create a ++HOLD statement with a user reason ID that you plan to use
only for USERMODs that are not supposed to be accepted. Here is an
example:
++HOLD(usermod) USER
REASON(NOUSERM)
COMMENT(do not accept this usermod).
Chapter 8. Installing a user modification 133

Installing USERMODs
2. Run the SMP/E RECEIVE command to read in the ++HOLD statement.

Use the SMPHOLD DD statement to point to the file or data set
containing the ++HOLD statement.
Because of the user hold, this USERMOD can be accepted only if you bypass the
specific user reason ID. The SYSMOD will not be automatically accepted if you
specify USERMOD or the specific SYSMOD ID on the ACCEPT command without
bypassing the user reason ID.
Be aware that if you receive the ++HOLD statement before applying the
USERMOD, you must bypass the user hold reason ID in order to apply the
USERMOD.

Chapter 9. Managing exception SYSMODs
This chapter explains how SMP/E manages SYSMODs that require special
processing. It discusses these topics:
v An introduction to exception SYSMODs
v What SMP/E does with HOLDDATA
v Sources of HOLDDATA
v Steps for processing the data
Introduction
Most SYSMODs you receive from IBM can be installed without additional
considerations; you can simply receive, apply, and then accept them. For some
SYSMODs, however, this is not possible. Examples of such SYSMODs are:
v SYSMODs that were sent out to correct a problem but that either have not fixed
the problem or have introduced a new problem. These are called PTFs in error,
or PE PTFs.
v SYSMODs that require special installation processing, such as a fix that must be
concurrently installed on all processors in a network.
v SYSMODs that introduce changes into the system that you should be made
aware of, such as changes to operator messages or critical documentation
changes.
In SMP/E terms, these SYSMODs are called exception SYSMODs. SMP/E supplies a
function to automate the management of exception SYSMODs. SMP/E supports
three categories of exception SYSMODs:
v Error. PTFs in error (PE PTFs).
v System. SYSMODs identified by IBM as requiring special processing or
notification.
v User. Any SYSMODs that you identify as requiring special processing.
Two MCSs are used to manage exception SYSMODs:

v ++HOLD puts a SYSMOD into exception status.
v ++RELEASE removes a PE PTF from exception status when it has been
determined that the PTF was held erroneously.
++HOLD statements for system holds are usually built as part of the held PTF.
++RELEASE statements and ++HOLD statements for error or user holds must be
in SMPHOLD. ++HOLD and ++RELEASE statements provided by SMPHOLD
(external holds) identify the following:
v The SYSMOD ID of the exception SYSMOD (that is, the SYSMOD being held).
v The exception SYSMOD category.
v The FMID to which that ++HOLD applies.
v The reason the SYSMOD is being put into or was in exception status. This is a 1-
to 7-character alphanumeric string called the reason ID.
– For error-category exception SYSMODs, SMP/E expects the reason ID to be
the SYSMOD ID of the APAR reporting the problem.
– For system-category exception SYSMODs, SMP/E expects the reason ID to be
a short description of the action required.

Exception SYSMODs
– For user-category exception SYSMODs, SMP/E makes no assumption about

what the reason ID represents.
For more information about reason IDs, see the SMP/E Reference manual.
v Text describing why the SYSMOD is being put into exception status. This field is
only for ++HOLD statements.
v An alternative way to release the exception SYSMOD. This field is only for
++HOLD statements.
Every ++HOLD statement specifies a HOLD category of ERROR, SYSTEM, or
USER. In addition to one of these categories, a ++HOLD statement may include
a HOLD CLASS, which is an alternative way to release a held SYSMOD. For
example, an exception SYSMOD may fix a problem more severe than the
problem it introduces. The ++HOLD statement for that SYSMOD would have an
ERROR reason ID that matches an APAR ID and a CLASS of ERREL.
++HOLD statements provided within a SYSMOD identify the same information.

However, even though these internal holds are effective against the containing
SYSMOD, the SYSMOD ID specified on the hold may be different from that of the
containing SYSMOD, as long as the SYSMOD ID specified on the hold is
superseded by the containing SYSMOD.
SMP/E then manages exception SYSMODs by actually managing the resolution of

the problems described by the reason ID specified on the ++HOLD statement.
Subsequent sections of this chapter describe how SMP/E uses HOLDDATA during
the installation of a SYSMOD, where the exception SYSMOD statements come
from, and how to process them. The chapters on the RECEIVE command, the
APPLY command, and the ACCEPT command in SMP/E Commands contain a much
more detailed explanation of the material covered here.
What SMP/E does with the HOLDDATA

This section describes what SMP/E does with the HOLDDATA when processing
the various commands associated with installing and removing SYSMODs.
Note: You must provide SMP/E with the most current HOLDDATA possible to get
the most benefit from this support.
Initial entry into staging data sets: RECEIVE

The RECEIVE command tells SMP/E to take the HOLDDATA from the input data
set on which it was delivered and store it in the SMP/E database.
The two operands that control input processes are:

v SYSMOD, which tells SMP/E to process the SYSMODs from the data set
specified by the SMPPTFIN DD statement
v HOLDDATA, which tells SMP/E to process the HOLDDATA (++HOLD and
++RELEASE statements) from the data set or file in a UNIX file system specified
by the SMPHOLD DD statement
You can specify one or both operands on the RECEIVE command. If neither
operand is specified, SMP/E attempts to receive the SYSMODs from SMPPTFIN,
and HOLDDATA from SMPHOLD.
When receiving a SYSMOD, SMP/E creates two entries:

Exception SYSMODs
1. An MCS entry is created on the SMPPTS. This entry is a copy (possibly

compacted) of the SYSMOD as it appeared in the SMPPTFIN data set.
2. A SYSMOD entry is created in the global zone. This entry contains information
that describes the installation requirements and element content of the
SYSMOD.
When receiving the HOLDDATA, SMP/E also creates (or modifies) two entries:
1. A HOLDDATA entry is created (or modified) in the global zone. This entry is
an exact copy of the ++HOLD statements as they appeared in SMPHOLD. The
name of the entry is the ID of the SYSMOD affected by this ++HOLD
statement. The HOLDDATA entry for a single SYSMOD can contain multiple
++HOLD statements.
Note: When a ++RELEASE statement is processed, SMP/E removes the

corresponding ++HOLD statement from the HOLDDATA entry. When all
++HOLD statements are removed, the HOLDDATA entry is
automatically deleted.
2. A SYSMOD entry is created (or modified) in the global zone. This entry
contains information that describes the exception SYSMOD conditions.
For each ++HOLD statement processed, SMP/E updates the global zone
SYSMOD entry to add a HOLD reason ID subentry. There are three types of
HOLD reason ID subentries, HOLDERROR, HOLDSYSTEM, and HOLDUSER,
corresponding to the three categories of exception SYSMODs.
Note: When a ++RELEASE statement is processed, SMP/E removes the

corresponding reason ID from the global zone SYSMOD entry. Do not
use the ++RELEASE statement to install a SYSMOD with an unresolved
reason ID. Use the appropriate BYPASS operand instead.
Updating target libraries: APPLY

When SMP/E applies a SYSMOD, SMP/E checks to see if that SYSMOD is
currently in exception SYSMOD status by seeing if there are any HOLD reason ID
subentries in the global zone SYSMOD entry. If so, SMP/E makes sure each reason
ID is resolved before allowing the SYSMOD to be installed.
For an error reason ID to be resolved, at least one of the following conditions must
be met:
v The reason ID must be superseded by another SYSMOD being installed.
v The reason ID must already exist as a SYSMOD entry in the target zone.
v You must specify BYPASS(HOLDERROR) on the APPLY command to show that
you are aware that an unresolved exception SYSMOD is being installed.
v If there is a HOLD CLASS associated with the reason ID, you can specify
BYPASS(HOLDCLASS) on APPLY to indicate that you are using an alternative
way to resolve the reason ID.
See the BYPASS operand in The APPLY Command in SMP/E Commands for
additional information.
Internal holds are considered resolved if any of the following conditions are met:
v The SYSMOD ID specified on the ++HOLD defining the exception is found as a
SYSMOD entry in the target zone
v The SYSMOD ID specified on the ++HOLD defining the exception is being
superseded by a SYSMOD being applied concurrently
Chapter 9. Managing exception SYSMODs 137

Exception SYSMODs
v The applicable BYPASS operand is specified.
You can resolve external system and user reason IDs by specifying
BYPASS(HOLDSYSTEM) or BYPASS(HOLDUSER) on the APPLY command. If
there is a HOLD CLASS associated with the reason ID, you can specify
BYPASS(HOLDCLASS) on APPLY to indicate that you are using an alternative way
to resolve the reason ID.
If you choose to resolve a reason ID by using the BYPASS operand, you must do
any required processing at the appropriate time. Otherwise, errors related to the
undone processing may occur, even though the reason ID was considered resolved.
Note: You may use the ++RELEASE statement for user category reason IDs if you
want to unconditionally release the SYSMOD for all systems. Remember
that, unlike BYPASS, ++RELEASE actually deletes the ++HOLD statement. If
you plan to use the user category ++HOLD statement, see the SMP/E
Reference manual for more information on the naming conventions for reason
IDs.
If all reason IDs are resolved, SMP/E allows the SYSMOD to be applied. If any
remain unresolved, SMP/E prevents the SYSMOD, and any other SYSMODs
dependent on this one, from being installed.
SMP/E leaves the reason IDs in the global zone SYSMOD entry when the
SYSMOD is applied, so if the SYSMOD is applied on another system later, the
same checking is done on that system. If the information had been deleted during
the first APPLY, SMP/E would not recognize the problem when the SYSMOD is
applied to subsequent systems. Therefore, the ++RELEASE statement should not
be used to install an exception SYSMOD with an unresolved reason ID. The
appropriate BYPASS operand should be used instead.
In summary, SMP/E ensures that no known problems are introduced into your
system by managing those problems at the level of the individual problem, rather
than the SYSMOD level. It is, therefore, very important that SMP/E have the most
current information on exception SYSMODs. For more information on the
importance of having current HOLDDATA and what you must do to provide that
information to SMP/E, see “How to process HOLDDATA” on page 140.
Updating distribution libraries: ACCEPT

Exception SYSMOD processing is the same when accepting a SYSMOD as when
applying one, except that the appropriate distribution zone is used to determine
whether the fixes for the reason IDs have been installed.
Removing HOLDDATA from SMP/E data sets

There are various ways to remove HOLDDATA from SMP/E data sets.
After a successful ACCEPT

When accepting a SYSMOD, if SMP/E determines that the global zone SYSMOD
entry and the SMPPTS MCS entry are to be deleted, SMP/E deletes any
HOLDDATA associated with those SYSMODs. Once the SYSMOD and MCS entries
for a SYSMOD have been deleted, you will probably not install that SYSMOD on
any other systems, so you would not need the HOLDDATA again.

Exception SYSMODs
During RESTORE processing

HOLDDATA is never deleted during RESTORE processing. The assumption is that
you may later want to reapply the SYSMODs you are restoring.
With the REJECT command

If you are using the SMPPTS as a history file of all SYSMODs, you may eventually
want to purge some of those SYSMODs. To do this, use the REJECT command. You
can use the HOLDDATA operand to have SMP/E delete not only the global zone
SYSMOD and SMPPTS MCS entries, but also any associated HOLDDATA.
Sources of HOLDDATA
Besides the data you create, these are the main sources of HOLDDATA provided
by IBM:
v CBPDO tapes
v Expanded service options (ESO) tapes
v Preventive service planning (PSP) information
v IBM service web pages
v Automated Service Delivery server as a result of a RECEIVE ORDER command
This section describes these sources.
CBPDO tapes
One of the primary means of obtaining HOLDDATA is CBPDO tapes. IBM
custom-builds these tapes to provide the products and service you request, taking
into consideration whether this is your first CBPDO order.
v If you select a particular service level, you get HOLDDATA for all service from
that level to the current level.
v If you do not select a service level and this is your first CBPDO order, you get
HOLDDATA for all the service shown on the order checklist.
v If you do not select a service level and you have ordered a CBPDO before, you
get HOLDDATA for service following the last service level that was shipped in
your previous CBPDO.
The HOLDDATA on a CBPDO tape has been customized to your product set. That
is, it contains only data applicable to PTFs for those products within a given
feature for which you are licensed under a single customer number. However, it
does not reflect the contents of any specific system within the establishment
defined by that customer number.
The HOLDDATA on the CBPDO tape should be processed immediately on receipt

of the tape. You can use either the SMP/E dialogs or the RIMLIB jobs provided
with the CBPDO tape to receive the HOLDDATA. For more information, see the
documentation that came with the CBPDO tape.
ESO tapes
Another way to obtain HOLDDATA is through ESO tapes. IBM regularly creates
the service levels shipped on these tapes, then custom-builds ESOs for users and
makes the tapes available through either subscription orders or special request
orders.
The HOLDDATA on an ESO tape has been customized to your product set. That is,
it contains only data applicable to PTFs for those products within a given feature

Exception SYSMODs
for which you are licensed under a single customer number. However, it does not
reflect the contents of any specific system within the establishment defined by that
customer number.
The HOLDDATA on the ESO should be processed immediately on receipt of the

tape. For more information on processing the ESO, see “Processing HOLDDATA
from an ESO” on page 142 and “Example of processing HOLDDATA” on page 144.
PSP information
Once a service level has been created, there is no further opportunity to change the
HOLDDATA on that tape, even though new errors are reported. It is, however,
important that you have this error information when you are about to install a
CBPDO or an ESO. Otherwise, you may direct SMP/E to install a PE PTF, thus
introducing a problem while fixing one. To solve this problem, PSP files have been
set up to hold this additional HOLDDATA. Within each SREL, there is one PSP file
for each service level.
When a service level is created, a PSP file is also created. As problems (APARs) are
reported, appropriate ++HOLD statements are added to the applicable PSP files.
IBM determines the applicable PSP files as follows:
1. Determine the PTF that introduced the APAR.
2. If the PTF is not yet assigned a monthly service level, add a ++HOLD
statement to the CORPE PSP file as well as the most current monthly bucket.
Note: The CORPE PSP file is for PE PTFs available correctively, but not yet
assigned to a monthly service level. The current monthly bucket will
contain ++HOLD statements for corrective service PTFs as well as those
assigned to a monthly service level.
3. If the PTF is available on a service level, determine the service level on which
that PTF was first shipped.
4. Add a ++HOLD statement to the PSP file corresponding to that service level.
5. Add a ++HOLD statement to each PSP file corresponding to any service levels
shipped after the one determined in the previous step, up to the current service
level.
6. Add a ++HOLD statement for the next service level.
7. No further PSP files are updated.
Before installing a CBPDO tape, an ESO tape, or PTFs in corrective mode, use
IBMLINK or contact the IBM Support Center to get the information in the
applicable PSP file. For more information on how to use the PSP information, see
“Processing HOLDDATA from PSP files” on page 143 and “Example of processing
HOLDDATA” on page 144.
Automated service delivery package

Another method of obtaining HOLDDATA is from an Automated Service Delivery
server by issuing a RECEIVE ORDER command. Whether you use the RECEIVE
ORDER command to obtain preventive service or corrective service, you
automatically receive the last 2-years of Enhanced HOLDDATA for the entire z/OS
software platform.
How to process HOLDDATA

The management of exception SYSMODs is a very important part of SMP/E.
SMP/E’s ability to manage exception SYSMODs, however, is limited by the quality

Exception SYSMODs
and timeliness of the HOLDDATA made available to it. To gain the full advantage
of this function, you must understand how SMP/E expects the three HOLDDATA
input sources to be used and the times during which SMP/E expects them to be
used.
The following steps summarize the process for managing exception SYSMOD data:
1. Receive all new products as you get them, or use UCLIN to add the FMIDs of
the new products to the global zone. This allows you to process exception
SYSMODs for them. Then receive any associated HOLDDATA shipped with the
product.
2. Receive HOLDDATA from subsequent CBPDO or ESO tapes in service level
order. Remember to do the following:
v Receive all new products as you get them so you can process exception
SYSMODs for them.
v Receive HOLDDATA as soon as you get your CBPDO or ESO tapes.
3. Before doing preventive service, do the following:
a. Get the PSP file associated with the last service level for which you
processed HOLDDATA, and receive this additional HOLDDATA for your
products.
b. Get the CORPE PSP file. This contains PE PTFs that are available
correctively, but are not yet available in a service level.
c. List and review HOLDDATA for SYSTEM HOLDs and, if possible, handle
the required special conditions. Then apply and accept these processed
SYSMODs by specifying BYPASS(HOLDSYS) and listing the individual
SYSMOD IDs on the SELECT operand. This helps makes sure all available
service is installed when you do preventive service.
Be sure to review “Example of processing HOLDDATA” on page 144. This

example shows why you should follow the procedures defined.
Processing HOLDDATA from a CBPDO tape

When you receive a CBPDO tape, the HOLDDATA it contains is based on the
service level you selected and on whether this is your first CBPDO. This
HOLDDATA pertains both to the PTFs actually on that tape and to PTFs shipped
on previous tapes. Follow these steps to process the HOLDDATA:
1. Receive the HOLDDATA from the CBPDO tape as soon as you get the tape.
Use the SMP/E dialogs or the RIMLIB job provided with the CBPDO tape.
By receiving the HOLDDATA as soon as possible, you make sure SMP/E has
the most current information available. Therefore, if you try to install any PTF
in response to a problem on your system and that PTF is in error, SMP/E
knows this and warns you so you can weigh the effect of installing the known
problem against the effect of fixing the problem you have encountered.
2. Receive the SYSMODs from the CBPDO tape as soon as you get the tape. Use
the SMP/E dialogs or the RIMLIB job provided with the CBPDO tape.
This makes sure that all available PTFs are ready to be installed. If you find a
problem in your system and determine that a PTF must be installed in
corrective mode, you have a better chance of having that PTF and all its
requisites readily available on the SMPPTS.
Note: You can receive the SYSMODs and HOLDDATA separately or in the same
job.

Exception SYSMODs
By following these procedures, you are essentially making a trade-off: system

resources as increased DASD space for the SMPPTS against the time the system
programmer would spend on searching for the service level with the required PTF
and on fixing problems caused by installing PE PTFs.
One important part of this procedure is that the HOLDDATA on each CBPDO and
ESO must be received in chronological order. SMP/E processes the ++HOLD and
++RELEASE statements in the order in which they are encountered. Therefore,
there can be an exposure if you receive the data out of sequence. For instance, the
tapes may be set up so that one contains a ++HOLD for a PTF and a subsequent
one contains a ++RELEASE for the same PTF. If the tapes are processed in the
wrong order, the RELEASE statement is processed first, and then the HOLD
statement. As a result, the PTF remains held.
Processing HOLDDATA from an ESO

When you receive your expanded service options (ESO) tape, the HOLDDATA it
contains pertains both to the PTFs actually on that tape and to PTFs shipped in
previous service levels. SMP/E and exception SYSMOD support have been
designed to work most efficiently and effectively if you adhere to the following
processing guides:
1. Receive the HOLDDATA file of the ESO as soon as you get the tape using the
SMP/E dialogs or the following sample job:
//RECHOLD JOB ’accounting info’,MSGLEVEL=(1,1)
//SMPHOLD DD ...data describing exception file
//SMPCNTL DD *
RECEIVE HOLDDATA /* Receive only exception
SYSMOD data. */.
/*
By receiving the HOLDDATA as soon as possible, you make sure SMP/E has
the most current information available. Therefore, if you try to install any PTF
in response to a problem on your system and that PTF is in error, SMP/E
knows this and warns you so you can balance the effect of installing the known
problem against the effect of fixing the problem you have encountered.
2. Receive the SYSMOD file of the ESO as soon as you get the tape using the
SMP/E dialogs or the following sample job:
//RECPTFS JOB ’accounting info’,MSGLEVEL=(1,1)
//SMPPTFIN DD ...data describing sysmods file
//SMPCNTL DD *
RECEIVE SYSMODS /* Receive only SYSMODs. */
SOURCEID(MYESO) /* Specify user-defined
source ID value. */.
LIST SYSMODS /* Now list the SYSMODs */
MCS /* including SMP MCS */
SOURCEID(MYESO) /* for those SYSMODs just
received. */.
/*
This makes sure that all available PTFs are ready to be installed. If you find a
problem in your system and determine that a PTF must be installed in
corrective mode, you have a better chance of having that PTF and all its
requisites readily available on the SMPPTS.
Note: The SOURCEID operand is optional. All PTFs in an ESO are assigned
SOURCEID values by ++ASSIGN statements.

Exception SYSMODs
Although the procedures for receiving the SYSMOD and the HOLDDATA file are
shown as separate jobs, they can be done in one RECEIVE command by specifying
both the SYSMODS and HOLDDATA operands. In fact, this is the preferred
method and is the default if neither operand is specified.
By following these procedures, you are essentially making a trade-off: system

resources as increased DASD space for the SMPPTS against the time the system
programmer would spend on searching for the service level with the required PTF
and on fixing problems caused by installing PE PTFs.
One important part of this procedure is that the HOLDDATA on each ESO and
CBPDO must be received in chronological order. SMP/E processes the ++HOLD
and ++RELEASE statements in the order in which they are encountered. Therefore,
there can be an exposure if you receive the data out of sequence. For instance, the
tapes may be set up so that one contains a ++HOLD for a PTF and a subsequent
one contains a ++RELEASE for the same PTF. If the tapes are processed in the
wrong order, the RELEASE statement is processed first, and then the HOLD
statement. As a result, the PTF remains held.
Processing HOLDDATA from PSP files

Another source of exception SYSMOD data is the PSP file, available through
IBMLINK or from the IBM Support Center. For each service level that is applicable
to a specific environment, there is a PSP file containing additional HOLDDATA.
This file contains all the ++HOLD and ++RELEASE statements applicable to PTFs
on either that service level or earlier ones. You should process this PSP file before
you install an ESO or before you install a CBPDO tape that includes PTFs from
that service level.
When you are ready to install a CBPDO or ESO, you must do the following:
1. Make sure you have received the HOLDDATA from, at minimum, all the
CBPDOs and ESOs up to the service level of the tape you are installing. You
should receive the HOLDDATA from all the available tapes to reduce the
amount of data that you have to get from PSP.
2. Use IBMLINK or contact the IBM Support Center to get the latest CORPE PSP
file, as well as the PSP file associated with the latest service level for which
you have received HOLDDATA (not the PSP file for the service level that you
are installing).
3. Create a data set containing the ++HOLD statements obtained from IBMLINK
or the IBM Support Center.
4. Receive that data set using the SMP/E dialogs or the following sample job.
//RECHOLD JOB ’accounting info’,MSGLEVEL=(1,1)
//SMPHOLD DD ...data describing your data set
//SMPCNTL DD *
RECEIVE HOLDDATA /* Receive only exception
SYSMOD data. */.
/*
You should also use IBMLINK or contact the IBM Support Center to get PSP
information whenever you are installing corrective service. Before installing a PTF
in corrective mode, determine the service level on which it was initially shipped.
Then contact the IBM Support Center to get the PSP data associated with that
service level to determine whether there are any ++HOLD statements for that PTF.
If so, process them and install the PTF.

Exception SYSMODs
Example of processing HOLDDATA

Assume you are ready to actually install a CBPDO or ESO. The following example
may help you understand the reasons behind the recommendations made in this
chapter. In this example:
1. Table 12 on page 144 shows information on exception SYSMODs. The PTFs and
PSP files are as follows:
a. Column 1 lists the three service levels involved in this example.
b. Column 2 lists the five SYSMODs in each service level.
c. Column 3 lists the ++HOLD statements contained on the CBPDO or ESO.
For simplicity, there are no PE PTFs before the first service level in the
example (PUT0301). The exact syntax and APAR numbers for the ++HOLD
are not significant for this example.
d. Column 4 lists the ++HOLD statements contained in the PSP file associated
with each of the service levels. The exact syntax and APAR numbers for the
++HOLD are not significant for this example.
2. The SYSMODs have been marked PE as follows:
v As of the 0301 service level, there were no PTFs in error.
v Between the 0301 and the 0302 service levels, PTFs UR00002 and UR00003
were marked as PE.
v Between the 0302 and the 0303 service levels, PTF UR00005 was marked PE.
v At the 0303 service level, PTF UR00001 was marked as PE.
3. Table 12 shows the contents of each of the files at some time after the creation
of the 0303 service level.
Table 12. CBPDO/service level/PSP HOLDDATA example
HOLDDATA in PSP
PTFs per PTFs with File for the source ID
Service level Service level HOLDDATA Service level
PUT0301 UR00001 UR00001
UR00002 UR00002
UR00003 UR00003
UR00004 UR00005
UR00005
PUT0302 UR00006 UR00002 UR00001
UR00007 UR00003 UR00005
UR00008
UR00009
UR00010
PUT0303 UR00011 UR00005 UR00001
UR00012
UR00013
UR00014
UR00015
4. You are now trying to install PTFs at service level 0301. The amount of
processing you have to do before installing 0301 service level PTFs depends on
what you did with PTFs in 0302 and 0303 service levels.
v If you have received the HOLDDATA for service levels 0301, 0302, and 0303,
you have to process only the one ++HOLD statement for PTF UR00001 from
the PSP file for service level 0303.

Exception SYSMODs
v If you have received only the HOLDDATA for service level 0302, you have to
process the two ++HOLD statements for PTFs UR00001 and UR00005 from
the PSP file for service level 0302.
v If you decided not to process any data for particular service levels until you
are actually ready to install them (that is, if you did nothing with service
level 0302 or 0303), you have to process the four ++HOLD statements for
PTFs UR00001, UR00002, UR00003, and UR00005 from the PSP file for service
level 0301.
Note: In each case, you used the PSP file associated with the last service level
for which you received the HOLDDATA, but if you had kept current in
processing the exception SYSMOD files from the service levels, you
would have had less information to obtain from the IBM Support Center.
In this example, the number of PTFs and HOLDDATA was small and, thus, the
data seems manageable. However, with a real service level with hundreds of
PTFs, the amount of manual work involved in getting the ++HOLD statements
from IBMLINK or the IBM Support Center, and then keying them into a data
set and receiving them could be very time-consuming. So, the cost of the
increased DASD space necessary to store the HOLDDATA each month is
commonly paid back in increased programmer productivity when the service
level is to be installed.

Exception SYSMODs

Chapter 10. Creating cross-product, cross-zone load modules:
The LINK MODULE command
This chapter discusses the LINK MODULE command, with an emphasis on when
and how to use it.
When to use LINK MODULE

Products sometimes contain modules from other products. For example, a product
may need to:
v Include another product’s modules in its load modules. In this case, as long as
the two products are in the same zone, SMP/E can automatically include the
required modules in the load modules that need them (if the modules reside in
the target library as single-CSECT load modules). SMP/E also tracks the
inclusion of these cross-product modules in the load modules.
v Update another product’s load module with one of its modules. In this case, as
long as the two products are in the same zone, SMP/E can automatically relink
the load module and include the supplied module. SMP/E also tracks the
inclusion of the modules in the cross-product load module.
However, when such products reside in different zones, SMP/E cannot

automatically perform the cross-zone link-edits. The LINK MODULE command
can be used to perform these cross-zone link-edits as postinstallation steps within
SMP/E control. The LINK MODULE command causes the required load modules
in one zone to be linked with modules residing in another zone, and tracks this
inclusion so that subsequent APPLY and RESTORE processing can automatically
maintain the affected load modules.
Notes:
1. When SMP/E processes the LINK MODULE command, it assumes that adding
the desired modules to the load modules does not require any changes to the
load module definition (that is, the linkage editor control statements or linkage
editor attributes). If any such changes are needed, make them through JCLIN
before using the LINK MODULE command.
2. For the LINK MODULE command, the SET BOUNDARY command must
specify the target zone that contains the LMOD entries for the load modules to
be link-edited.
3. There are times when the LINK MODULE command is not appropriate to
use—generally, for products that are written in a high-level language and, as a
result, include modules from libraries (such as compiler libraries) owned by a
different product. Your options for installing such a product depend on how
the product was packaged.
v SYSLIB DD statements are used in link-edit steps in order to implicitly
include the necessary modules.
In this case, when you install the product, the implicitly-included modules
are automatically linked into the load modules. If the libraries containing
those modules are updated, you can use the LINK LMODS command to
rebuild the affected load modules. For more information, see the LINK
LMODS chapter in SMP/E Commands.

LINK MODULE Command
v No SYSLIB DD statements are used in link-edit steps in order to implicitly

include the necessary modules. In this case, you must use postinstallation
link-edit steps outside of SMP/E.
How to use LINK MODULE

Assume you have installed GDDM and CICS, and some of the GDDM modules
must be linked into CICS load modules. GDDM resides in zone GDDTZN, and the
zone controlling CICS is CICTZN. Because GDDM and CICS are controlled by
different zones, SMP/E does not automatically link the GDDM modules into the
CICS load modules when GDDM is installed. The LINK MODULE command can
be used instead.
In this example, GDDM module ADMABCD needs to be linked into CICS load
module DFHWXYZ. Module ADMABCD is installed in a target library as a
single-CSECT load module when GDDM is installed. Therefore, SMP/E can use
the target library version of ADMABCD to update CICS load module DFHWXYZ.
(If a module does not reside in a target library as a single-CSECT load module,
SMP/E uses the related distribution zone copy of the module to update the load
module.)
The following commands can be used to have SMP/E install and track the
installation of GDDM module ADMABCD in the CICS load module:
SET BDY(CICTZN) /* Target zone for CICS. */.
LINK MODULE(ADMABCD) /* Link module ADMABCD */
FROMZONE(GDDTZN) /* residing in zone GDDTZN */
INTOLMOD(DFHWXYZ) /* into load module DFHWXYZ. */.
These commands cause GDDM module ADMABCD to be linked into CICS load
module DFHWXYZ. SMP/E also adds cross-zone subentries to the affected entries:
v An XZLMOD subentry is added to the ADMABCD MOD entry in target zone
GDDTZN so that if ADMABCD is updated, it can be automatically replaced in
the CICS load module.
Note: The CICS load module is automatically updated only if the XZLINK
subentry was previously set to AUTOMATIC in the TZONE entry for
zone CICTZN. Here is an example of the commands that can be used to
do this:
SET BDY(CICTZN) /* Target zone for CICS. */.
UCLIN.
ADD TZONE(CICTZN) /* Update TZONE entry */
XZLINK(AUTOMATIC). /* to do automatic links. */.
ENDUCL.
v An XZMOD subentry is added to the CICS DFHWXYZ LMOD entry in target
zone CICTZN to indicate that:
– DFHWXYZ now contains ADMABCD.
– Any updates for ADMABCD should be accepted only from zone GDDTZN.
v TIEDTO subentries are added to the TZONE entries for CICTZN and GDDTZN
to indicate that there is a relationship between modules and load modules in
these zones.
For more information on the LINK MODULE command and cross-zone updating
during APPLY and RESTORE processing, see the LINK MODULE chapter in
SMP/E Commands.

Chapter 11. Displaying the data managed by SMP/E: The LIST
command
This chapter discusses the LIST command. After an introduction, it discusses these
topics:
v Listing all the SMP/E data
v Listing by specific entry type
v Listing specific entries
v Listing by FMID or FMIDSET
v Listing to compare two zones
Introduction
The SMP/E database contains much information that is useful to you at certain
times. For instance, when a problem is encountered in your system:
v You need to know the functional and service level of the module with the error.
v You may also want to know when that module was last changed: a recent
change may have caused the problem.
v After reporting the problem, you can start working with the IBM Support Center
to debug the problem. They may want to know the status of other specific PTFs:
have you installed them; are they available on your system; is anything stopping
them from being installed?
v After identifying the problem in the module, you may want to know whether
any other parts of the system are affected by that module.
All of this information is available in the SMP/E database. You can obtain the
information by using the SMP/E dialogs as you debug the problem. You can also
use the SMP/E LIST command to create hardcopy listings of the information.
With the LIST command, you can display all entries of a specified type or specific
entries. The following sections demonstrate the flexibility of the LIST command.
For a complete description of all the LIST command operands, see The LIST
Listing all the SMP/E data

If you encounter a problem with your system, the SMP/E data describing that
system can be very important in diagnosing the problem. This information can be
obtained by using the SMP/E dialogs during the debugging process. However, if
the system is not running, the information is not available unless you have
periodically listed the SMP/E data for the system.
Therefore, it is advisable to list all the data for each of your systems and save the
hardcopy listing. The data can be listed by individual zone. The following is an
example of a job for listing all the entries in zone TGT01.
//LIST JOB ’accounting info’,MSGLEVEL=(1,1)
//LIST EXEC SMPPROC
//SMPCNTL DD *
SET BDY(TGT01) /* Set to desired zone. */.
LIST /* List all entries */

LIST command
XREF /* with XREF option to show

additional relationships
between entries. */.
/*
Because the global zone contains data used to process each of the target zones and
distribution zones, you may want to list that data more often. The following job
lists all the data in the global zone, including the SYSMOD entries and the MCS
entries:
//LIST EXEC SMPPROC
//SMPCNTL DD *
LIST /* List all entries. */.
/*
If you do not require a listing of the SYSMOD and MCS entries, you can use the
LIST operands that enable you to list only specific entry types. For additional
information, see “Listing by specific entry type.”
SMP/E also provides support to list all the entries for all the zones defined in the
GLOBALZONE entry. This enables you to display all data for all systems with one
SMP/E command. Here is an example of this option:
//LIST EXEC SMPPROC
//SMPCNTL DD *
ALLZONES /* for all zones. */.
/*
Notes:
1. The ALLZONES operand should be used with caution; it may produce a large
amount of output.
2. This function can also be qualified by other LIST operands to limit the entries
listed from each zone. For example, the following job will list only the
SYSMOD entries from all zones:
//LIST EXEC SMPPROC
//SMPCNTL DD *
LIST SYSMOD /* List the SYSMOD entries */
ALLZONES /* for all zones. */.
/*
Listing by specific entry type

At times, you may need to have a listing of all entries of a certain type. For
example:
v You may want to display all the DDDEF entries for a particular target zone or
distribution zone.
v You may want to list all the OPTIONS and UTILITY entries in the global zone
so you do not duplicate an existing entry.
v You may want to list all ORDER entries in the global zone.

LIST command
SMP/E supports various operands on the LIST command that you can use to list
all the entries for one or more entry types. The following job shows how to use the
DDDEF, OPTIONS, UTILITY, and ORDER operands on the LIST command to do
this:
//LIST EXEC SMPPROC
//SMPCNTL DD *
SET BDY(TGT1) /* Set to target zone */.
LIST DDDEF /* List all DDDEF entries. */
/* */.
SET BDY(DLIB1) /* Set to DLIB zone */.
LIST DDDEF /* List all DDDEF entries. */
/* */.
LIST OPTIONS /* List all options */
UTILITY /* and utility entries. */
ORDER(ORD000034,ORD000035) /* List details for orders */
/* ORD000034 and ORD000035 */
.
/*
For a complete list of all the operands corresponding to the various entry types,
see The LIST Command in SMP/E Commands.
Note: Not all entry types are valid for each zone type. For example, requesting a
listing of the OPTIONS entries in a target zone results in an error, because
OPTIONS entries exist only in the global zone. For a summary of the types
of entries contained in each type of zone, see Table 2 on page 59 and Table 3
on page 60.
Sometimes, the various entry-type operands can be further qualified to process

only a subset of all existing entries. The most common type for which this can be
done is the SYSMOD entries. There are numerous operands to qualify SYSMOD
entries. The LIST Command in SMP/E Commands describes all of them in detail.
One of the most common uses of this function is to determine the functions (or
FMIDs) that have been installed. The following job can be used to list:
v The function SYSMODs installed on TGT1
v Those PTFs that have been applied to TGT1 but not yet accepted to DLIB1
//LIST EXEC SMPPROC
//SMPCNTL DD *
FUNCTIONS /* for function SYSMODs. */
/* */.
PTFS /* for PTF type SYSMODs */
NOACCEPT(DLIB1) /* not yet accepted. */
/* */.
/*
Listing specific entries

When you encounter a problem in your system and contact the IBM Support
Center to resolve the problem, you may be asked to provide very specific
information. For example:
v What is the service level of the module in which the problem was reported?
v Are there any USERMODs for the module?
Chapter 11. Displaying the data managed by SMP/E: The LIST command 151
LIST command
v Do you have a specific PTF installed?
If you have a complete, current listing of all the entries in your system, you can
get this information from that listing. You can also get it through the SMP/E
dialogs while you are talking to the IBM Support Center.
SMP/E also provides additional LIST functions that you can use to display only
specified entries. This is done by allowing you to specify a list of entry names (in
parentheses) after each of the entry-type operands. For example, assume that you
need to know the function and service level for modules GIMMPDRV and
GIMMPIO and if SYSMOD UR12345 has been installed. The following job can be
used:
//LIST EXEC SMPPROC
//SMPCNTL DD *
LIST MOD(GIMMPDRV /* List these two modules */
GIMMPIO) /* */
SYSMOD(UR12345) /* and this SYSMOD. */
/* */.
/*
You receive a listing of the required information for the two modules. If SYSMOD
UR12345 was installed, it should be listed; otherwise, you receive a message saying
that the entry was not found (meaning it has not been installed).
Another common use for this function is to list the cover letters for specific PTFs.
The following job shows an example of a job for listing the cover letters for PTFs
UR00001, UR00002, and UR00003:
//LIST EXEC SMPPROC
//SMPCNTL DD *
LIST MCS( /* List cover letters */
UR00001 /* for these three PTFs. */
UR00002
UR00003
)
.
/*
Listing by FMID or FMIDSET

Frequently, you deal with one area of the system at a time and would like to see
all the information relating to that one area. You can use the FORFMID operand in
conjunction with the various entry-type operands to limit the entries processed.
Here is an example listing:
v All the entries associated with function HXY1100
v All the MAC entries associated with function HXZ2100
v All the SYSMOD and module entries associated with either function JXY1123 or
JXY1124
//LIST EXEC SMPPROC
//SMPCNTL DD *
FORFMID(HXY1100) /* for this FMID. */
/* */.
LIST MAC /* List only the macros */

LIST command
FORFMID(HXZ2100) /* for this FMID. */

/* */.
LIST SYSMOD /* List SYSMOD entries */
MOD /* and MOD entries */
FORFMID(JXY1123 /* for this FMID */
JXY1124) /* or this FMID. */
/* */.
/*
Note: The names within the FORFMID operand can be names of FMIDSET entries.
In this case, SMP/E lists all the entries associated with any of the FMIDs
defined in the FMIDSET entry.
Listing to compare two zones

If you have multiple zones, you may sometimes want to determine the functional
and service differences between them. The LIST command provides you with this
capability.
Note: You can also use the REPORT SYSMODS command to compare the
SYSMOD content of two zones. Besides telling you which SYSMODs are
installed in one zone but not in a second, REPORT SYSMODS also indicates
which of the uninstalled SYSMODs are applicable to the second zone and
generates commands you can run to install the SYSMODs in the second
zone. For more information, see Chapter 16, “Comparing the SYSMODs
installed in two zones: The REPORT SYSMODS command,” on page 167.
One possibility might be that you have two products at different service levels. The
product at the lower service level works, and the product at the higher service
level does not work. You might use LIST to compare the zones for the two systems
and to determine what is causing the problem.
This example compares two target zones, TGT1 and TGT2. The commands in the
following example perform the comparison for you:
//LIST EXEC SMPPROC
//SMPCNTL DD *
SET BDY(TGT1) /* Set to target zone TGT1. */.
LIST SYSMODS /* List the SYSMODs */
/* in zone TGT1 */
NOAPPLY(TGT2) /* that have not been */
/* applied to TGT2. */
/* */.
SET BDY(TGT2) /* Set to target zone TGT2. */.
LIST SYSMODS /* List the SYSMODs */
/* in zone TGT2 */
NOAPPLY(TGT1) /* that have not been */
/* applied to TGT1. */
/* */.
/*
By comparing the two resulting listings, you can see the differences between the
two zones.
In this second example, the same product is installed in different zones. You want
to compare the service to make sure both copies of the product are at the same
level. For example, assume that product PVT1102 is installed in two target zones
and two distribution zones:
Chapter 11. Displaying the data managed by SMP/E: The LIST command 153
LIST command
You want to make sure that PVT1102 is at the same service level in all the zones.
To do this, you can use the LIST command and compare which SYSMODs are
installed in which zones.
To compare the service levels of product PVT1102 in the two distribution zones,
you can use the commands in the following example:
//LIST EXEC SMPPROC
//SMPCNTL DD *
SET BDY(DLIB1) /* Set to DLIB1. */.
LIST SYSMOD /* List SYSMODs */
FORFMID (PVT1102) /* for PVT1102 */
NOACCEPT (DLIB2) /* in DLIB1, not DLIB2. */.
SET BDY(DLIB2) /* Set to DLIB2. */.
LIST SYSMOD /* List SYSMODs */
FORFMID (PVT1102) /* for PVT1102 */
NOACCEPT (DLIB1) /* in DLIB2, not DLIB1. */.
Similarly, to compare the service records for the target zone copies of PVT1102, you
can use LIST with the NOAPPLY operand.
Summary
The LIST command enables you to:
v Compare two target zones using the NOAPPLY operand.
v Compare two distribution zones using the NOACCEPT operand in place of the
NOAPPLY operand.
v Compare a target zone and a distribution zone using both the NOAPPLY and
NOACCEPT operands.
This gives you the ability to compare all combinations of zone types, keeping in
mind that the zone for the NOAPPLY operand must be a target zone, and that the
zone for the NOACCEPT operand must be a distribution zone.

Chapter 12. Changing the data SMP/E manages: The UCLIN
command
This chapter discusses the UCLIN command, with an emphasis on how and when
to use it.
Introduction
The SMPCSI and associated data sets contain basically two types of information:
v Information added as a result of installing SYSMODs. Generally, this type is
managed completely by SMP/E; that is, appropriate entries are added, changed,
or deleted as SYSMODs are installed. You need not make any modification to
the database to record this information. You may, however, need to make
changes to do the following:
– Record changes made outside SMP/E
– Delete information no longer required
– Recover from an SMP/E or system error
v Information added by you to control the installation of SYSMODs. You must
manually add this information to the database before processing any SYSMODs.
You may later need to modify the information to reflect new processing
information.
The UCLIN command helps you to make these changes.
To use UCLIN effectively, you should have a detailed understanding of how it

works and what it can do. The UCLIN Command in SMP/E Commands and SMP/E
data-set entries in the SMP/E Reference manual provide that level of detail. Read
those chapters before trying to run any UCLIN commands. The following sections
describe, at a very high level, what UCLIN is.
When to use UCLIN

UCLIN is a very powerful function that must be used with extreme caution. You
can use UCLIN to modify almost all the data in the SMP/E database. When you
are modifying an entry, SMP/E makes sure the data within that one entry is
consistent—that is, that the result could have occurred during normal SMP/E
processing. However, no checking is done to make sure the resulting entry is
consistent with other related entries in the database.
For example, you can use UCLIN to delete a UTILITY entry in the global zone
without SMP/E detecting any error condition. However, if there is an OPTIONS
entry within the global zone that refers to the deleted UTILITY entry, an error
occurs when you attempt to use that OPTIONS entry. This is a very simple
example of inconsistent data across entries that does not result in a serious error.
UCLIN modifications to other entries, such as element, LMOD, or SYSMOD, may
not be detected as error conditions during processing, but may cause incorrect
processing, such as failing to link a module, updating the wrong library, or
installing a SYSMOD that should not be installed.
In general, consider the following before making the UCLIN update:

1. Determine whether there is a better method of obtaining the same result.
Table 13 on page 156 shows where to find more information about alternatives
UCLIN command
to UCLIN.
Table 13. Alternatives to UCLIN
To do this without UCLIN: Look here for more information:
Change a common subentry in several The ZONEEDIT Command in SMP/E
DDDEF or UTILITY entries in the same zone Commands
Update the cross-zone subentries of the The ZONEEDIT Command in SMP/E
MOD, LMOD, and TARGETZONE entry Commands
Add, rename, or delete a zone SMP/E Commands: Chapters on the following
commands:
v Adding a zone: To add a zone, you can
use the following commands, depending
on the particular situation:
ZONECOPY and ZONERENAME
ZONEEXPORT, UCLIN for the
ZONEINDEX, and ZONEIMPORT
v Renaming a zone: To rename a zone, you
must use the following command:
ZONERENAME
v Deleting a zone: To delete a zone, you
can use the following commands,
depending on the particular situation:
ZONEDELETE or ZONEEXPORT
Change the structure of your system z/OS Packaging Rules: Section on how to
avoid UCLIN by using the appropriate
(For example, to add, delete, move, or MCSs and operands
rename elements or their aliases)
2. If you are not the originator of the UCLIN, make sure you understand exactly
what is being done and why. If you are not sure, find out before making the
update.
3. Make sure the UCLIN is being done in the correct sequence in the
process—before or after the installation of the SYSMOD.
4. Make sure all the data is correct.
5. List the entry before changing it. This makes sure you know what the original
entry looked like in case an error is reported during the UCLIN or the
modification causes an error.
6. After you have done all of these steps, if you have been given directions for
installing the UCLIN (for example, in the PTF cover letter or in the program
directory for a new function), follow those directions.
How to use UCLIN

UCLIN is used to update the entries in the SMP/E database, just as the SPZAP
utility is used to update the system libraries. That is, it enables you to:
v Add new information
v Delete existing information
v Replace existing information with new information
Some UCLIN functions will not work for certain entries or data sets. The chapters
on the UCLIN command in SMP/E Commands and SMP/E data-set entries in

UCLIN command
SMP/E Reference provide detailed information on which entries may be modified

for each data set, what data within each entry may be modified, and the exact
syntax for each entry and data item.
The general format for UCLIN statements is:

SET BDY(xxxxxxx) /* Set to correct zone. */.
UCLIN /* Marks start of UCLIN. */.
... /* UCL */
UCL statements /* statements */
... /* to make modifications. */
ENDUCL /* Marks end of UCLIN. */.
The general format of each UCL statement is as follows:

ADD|REP|DEL /* Type modification */
type(name) /* Entry type and name */
operand /* */
operand /* Optional */
operand /* operands */
operand /* */
/* End of one UCL statement */.
Where:
ADD|REP|DEL
specifies the action to be taken on the entry or operands specified.
In general, ADD means add the entry or operand only if it is not already
present; REP means replace the operand if it is already present, otherwise add
it; and DEL means delete the entry or operand if it exists. For a more detailed
description of the functions provided by ADD, REP, and DEL, see the chapter
on the UCLIN command in the SMP/E Commands manual.
type(name)
specifies the entry type (such as MOD, MAC, ORDER, SRC, data element type,
SYSMOD) and name of the entry (such as GIMMPDRV, HELP, ORD000034,
MYSRCMOD, MYCLIST, UR12345).
operand
specifies the individual data items in the entry that are to be modified.
The data items can be:
v Single operands, such as RENT or REUS or COPY
v Single value subentries, such as DISTLIB(AOS12), where only one value can
be placed within the parentheses
v Multiple value subentries, such as LMOD(LMOD01,LMOD02,LMOD3) or
MAC(MAC01,MAC02), where more than one value can be specified within
the parentheses
Chapter 12. Changing the data SMP/E manages: The UCLIN command 157
UCLIN command

Chapter 13. Identifying cross-zone requisites: The REPORT
CROSSZONE command
This chapter contains information about using the REPORT CROSSZONE
command to check for requisites between zones. It discusses these topics:
v How to define a ZONESET
v How to run the REPORT CROSSZONE command
v How to install the SYSMODs, using the output from the REPORT CROSSZONE
command
Introduction
Your system may contain products that are packaged and installed separately, but
which have service level or interface dependencies. For example, an interface error
in one product may require a change to another product that used the interface.
When this happens, a unique PTF is generated for each product. The relationship
between the PTFs can be specified in a conditional requisite (++IF) modification
control statement in the PTFs. If you have completed the steps listed in “Specifying
automatic cross-zone requisite checking” on page 78, SMP/E automatically checks
the requisites during APPLY, ACCEPT, and RESTORE processing, whether the
products are in the same zone or in separate zones. However, if you have not
completed these steps and the products are in separate zones, the requisites are not
automatically checked in any other zones. In this case, to make sure these
requisites are installed where they are needed, you must:
1. Define a set of zones to be checked for conditional requisites. This is done with
the ZONESET entry in the global zone.
Note: All the zones in the ZONESET must be defined by ZONEINDEX

subentries in the same global zone as the ZONESET entry.
2. Run the REPORT CROSSZONE command to get a list of the SYSMODs that
must be installed and the zones where they are needed.
3. Install the SYSMODs in the indicated zones.
Note: If you just want to check service levels for products, you should use the
REPORT SYSMODS command or the LIST command. See Chapter 16,
“Comparing the SYSMODs installed in two zones: The REPORT SYSMODS
command,” on page 167 and “Listing to compare two zones” on page 153
for more information.
Defining a ZONESET
To tell SMP/E which zones it should check for missing requisites, you must define
a global zone ZONESET entry. You may have one or more ZONESETs to describe
groups of products that might have dependencies on each other.
For example, assume you have a system that supports two products, ABC and
AYZ, that have dependencies on one another. You might have one zone, BASEABC,
for the base ABC function, and another zone, PRODABC, for a dependent function.
Likewise, you might have a zone BASEXYZ for the base XYZ function, and
another zone, PRODXYZ, for a dependent function. The dependent functions are

REPORT CROSSZONE command
different versions of the same product, and they must be synchronized with each
other and with their base functions. You can set up two ZONESETs to help keep
these products at the same service level.
These are the commands you can use to define the ZONESETs:
//UCL JOB ’accounting info’,MSGLEVEL=(1,1)
//UCL EXEC SMPPROC
//SMPCNTL DD *
UCLIN /* */.
ADD ZONESET(ZONEA) /* Create ZONESET ZONEA. */
ZONE(BASEABC, /* Include these zones. */
PRODABC, /* */
PRODXYZ) /* */
.
ADD ZONESET(ZONEX) /* Create ZONESET ZONEX. */
ZONE(BASEXYZ, /* Include these zones. */
PRODXYZ, /* */
PRODABC) /* */
.
ENDUCL /* */.
/*
When you define a ZONESET, remember:

v Each zone in a ZONESET must also be defined in the global zone.
v Each zone in a ZONESET must be defined in the same global zone. They cannot
be defined in global zones that are in different CSI data sets.
v A zone can be part of more than one ZONESET.
v A ZONESET can contain both target and distribution zones.
For more information on defining the ZONESET entry, see SMP/E Reference.
Running the REPORT CROSSZONE command

After you have defined the appropriate ZONESET, you can run the REPORT
CROSSZONE command to get a list of all the SYSMODs that must be installed and
which zones in the ZONESET need them. This list is the Cross-Zone Requisite
SYSMOD report. It identifies which of the needed SYSMODs must be received and
which SYSMODs caused the needed SYSMODs to be listed. Besides the report,
SMP/E writes commands to the SMPPUNCH data set. You can use them to install
the SYSMODs in the appropriate zones.
If all the zones in a ZONESET are of the same type, SMP/E determines the type of
report to be generated. For example, a ZONESET containing only target zones
results in a Cross-Zone Requisite SYSMOD report for APPLY processing. On the
other hand, your ZONESET can be mixed; that is, it may contain both target and
distribution zones. If this is the case, you need to specify which type of Cross-Zone
report you want generated.
You can limit which SYSMODs SMP/E reports on by specifying any combination
of these operands on the REPORT CROSSZONE command:
v FORZONE: to list only SYSMODs needed in specific zones in the ZONESET
v FORFMID: to list only SYSMODs needed for specific FMIDs in the ZONESET
zones
v DLIBZONE or TARGETZONE: to tell SMP/E which zones you want a report
on (if the zones contained in the zone set specified by ZONESET are mixed)

REPORT CROSSZONE command
Note: DLIBZONE and TARGETZONE are mutually exclusive operands.
For more information on the REPORT CROSSZONE command, see the SMP/E
Reference manual.
To continue the example for this chapter, assume you applied a number of
SYSMODs to the zones in ZONESET ZONEA. Some of these SYSMODs named
conditional requisites. To find out which zones are affected by these requisites, you
can use the following commands:
//REPORT JOB ’accounting info’,MSGLEVEL=(1,1)
//REPORT EXEC SMPPROC
//SMPCNTL DD *
REPORT CROSSZONE /* Report on */
ZONESET(ZONEA) /* ZONESET ZONEA. */.
/*
Installing the SYSMODs

Once you have the Cross-Zone Requisite SYSMOD report, you can install the
missing SYSMODs in the zones where they are needed. Follow these steps:
1. Receive any SYSMODs that have not yet been received.
2. Install the SYSMODs in the zones where they are needed. If you have the
SMPPUNCH output from the REPORT CROSSZONE command, you can use
that.
3. Rerun the REPORT command using the same ZONESET to check the results of
installing the new SYSMODs.
4. Receive and install any additional SYSMODs that are needed.
For the example in this chapter, follow these steps for ZONESET ZONEA and
ZONESET ZONEX, because both contain zone PRODXYZ. You can continue to run
the REPORT CROSSZONE command and install SYSMODs until all the needed
SYSMODs have been installed in both ZONESETs.
Chapter 13. Identifying cross-zone requisites: The REPORT CROSSZONE command 161
Chapter 14. Identifying installed SYSMODs affected by error
holds: The REPORT ERRSYSMODS command
This chapter contains information about using the REPORT ERRSYSMODS
command to check for installed SYSMODs affected by error holds that were
subsequently received. It discusses these topics:
v How to run the REPORT ERRSYSMODS command
v How to use output from the REPORT ERRSYSMODS command for installing the
SYSMODs
Introduction
In the course of maintaining your system, you can apply or accept service and
later receive HOLDDATA that affects that service. If any of that HOLDDATA was
for an error reason ID, you should install a resolving SYSMOD so that the error
does not occur on your system. However, SMP/E does not automatically write any
reports during RECEIVE processing to help you do this. To see if any installed
SYSMODs are affected by error holds that were subsequently received, you must:
1. Run the REPORT ERRSYSMODS command to get a list of the SYSMODs that
must be installed and the zones where they are needed.
2. Install the resolving SYSMODs in the indicated zones.
3. Determine whether any resolving SYSMODs are available for held SYSMODs.
Running the REPORT ERRSYSMODS command

After you have received HOLDDATA, you can run the REPORT ERRSYSMODS
command to get a list of all the SYSMODs that are affected by any unresolved
error holds. This list is the Exception SYSMOD report, which also tells whether any
resolving SYSMODs have been received for these holds and whether any of the
resolving SYSMODs have error holds against them. Besides the report, SMP/E
writes commands to the SMPPUNCH data set. You can use them to install the
resolving SYSMODs in the appropriate zones.
You can limit which HOLDDATA SMP/E reports on by specifying any

combination of these operands on the REPORT ERRSYSMODS command:
v BEGINDATE: to check HOLDDATA entries for error reason IDs that were
received by SMP/E on or after the specified date
v ENDDATE: to check HOLDDATA entries for error reason IDs that were received
by SMP/E on or before the specified date
v FORFMID: to list only SYSMODs owned by specific FMIDs
For more information on the REPORT ERRSYSMODS command, see SMP/E

Commands.
Here is an example of when you might want to use the REPORT ERRSYSMODS
command. Assume that you have just received some HOLDDATA, and you need
to know whether it affects any of the SYSMODs you have already accepted into
distribution zone DZONE1. You can use the following commands:

REPORT ERRSYSMODS command
//REPORT JOB ’accounting info’,MSGLEVEL=(1,1)

//REPORT EXEC SMPPROC
//SMPCNTL DD *
SET BDY(GLOBAL) /* process global zone */.
REPORT ERRSYSMODS /* report on exception */
ZONES(DZONE1) /* SYSMODs in this zone */
BEGINDATE(01 01 03) /* for HOLDDATA received */
ENDDATE(02 01 03) /* between these dates */.

Once you have the Exception SYSMOD report, you can install the resolving
SYSMODs in the zones where they are needed. Follow these steps:
1. If any resolving SYSMODs are held, run REPORT ERRSYSMODS, specifying
the global zone, to see if any SYSMODs have been received that resolve the
additional holds for the resolving SYSMODs.
2. If the Exception SYSMOD report for the global zone shows resolving SYSMODs
for the additional holds, edit the SMPPUNCH output to add the new resolving
SYSMODs and to update the selection list so the held resolving SYSMODs will
be processed.
3. Use the SMPPUNCH output to install the resolving SYSMODs in DZONE1.

Chapter 15. Listing the source IDs in a zone: The REPORT
SOURCEID command
This chapter contains information about using the REPORT SOURCEID command
to list the source IDs assigned to SYSMODs in a given zone or ZONESET. It
discusses these topics:
v How to run the REPORT SOURCEID command
v How to list SYSMODs using the output from the REPORT SOURCEID command
Introduction
In the course of maintaining your system, you may need to find out which source
IDs are assigned to SYSMODs in a given zone. For example, assume you install
service using CBPDOs, which assign source IDs to the service SYSMODs they
contain. You can use the REPORT SOURCEID command to determine the latest
service level you have installed in a particular zone. To determine the service level
based on source IDs, follow these steps:
1. Run the REPORT SOURCEID command to get a list of which source IDs are
assigned to SYSMODs in a given zone.
2. If desired, list the SYSMOD entries for the SYSMODs with those source IDs.
Running the REPORT SOURCEID command

You can use the REPORT SOURCEID command to get a list of all the source IDs
assigned to SYSMODs in a given zone. This list is the SOURCEID report, which
may also indicate which SYSMODs these source IDs are assigned to. Besides the
report, SMP/E writes commands to the SMPPUNCH data set, which you can use
to list the SYSMODs. For details on the REPORT SOURCEID command, see SMP/E
Commands.
Here is an example of when you might want to use the REPORT SOURCEID
command. Assume you want to find out which source IDs are associated with
SYSMODs in target zone TGT1, and you want to know which SYSMODs each
source ID is assigned to. You can use the following commands:
SET BDY(GLOBAL).
REPORT SOURCEID
ZONES(TGT1)
SYSMODIDS.
Listing the SYSMODs

If you want more information about the SYSMODs that are assigned the source
IDs shown in the SOURCEID report, you can list the related SYSMOD entries. The
SMPPUNCH output produced by the REPORT SOURCEID command contains the
LIST SYSMOD SOURCEID(...) commands needed to list the SYSMODs for the
source IDs in the SOURCEID report. You can tailor the SMPPUNCH output to list
the SYSMODs in which you are interested, and run the commands to list the
desired SYSMODs.

Chapter 16. Comparing the SYSMODs installed in two zones:
The REPORT SYSMODS command
This chapter contains information about using the REPORT SYSMODS command
to check if SYSMODs installed in one zone are applicable to and installed in a
second zone. It discusses these topics:
v How to run the REPORT SYSMODS command
v How to install SYSMODs using the output from the REPORT SYSMODS
command
Introduction
In the course of maintaining your system, you may need to compare the function
and service level of two zones. For example, if you have installed the same
products in different zones, you may want to make sure that both copies of these
products are at the same level. SMP/E does not do this kind of checking
automatically. To compare the SYSMODs installed in two zones, you must:
1. Run the REPORT SYSMODS command to get a list of the SYSMODs that are
installed in the first zone and that are applicable to the second zone but are not
yet installed there.
2. Install the applicable SYSMODs in the second zone.
Running the REPORT SYSMODS command

You can run the REPORT SYSMODS command to get a list of all the SYSMODs
that are installed in one zone and not in a second. This list is the SYSMOD
Comparison report, which also tells which of these SYSMODs are applicable to the
second zone, shows if the SYSMODs have been received, and lists any source IDs
associated with the SYSMODs. Besides the report, SMP/E writes commands to the
SMPPUNCH data set, which you can use to install the SYSMODs. For details on
the REPORT SYSMODS command, see SMP/E Commands.
Here is an example of when you might want to use the REPORT SYSMODS
command. Assume you have two z/OS systems. The target zones that control
these systems are TGZONE1 and TGZONE2, and they are serviced from the same
global zone. You want to determine which SYSMODs are installed in TGZONE1
and are not installed in, but are applicable to, TGZONE2. You can use the
following commands:
SET BDY(GLOBAL) /* process global zone */.
REPORT SYSMODS /* report on SYSMODs */
INZONE(TGZONE1) /* input zone TGZONE1 */
COMPAREDTO(TGZONE2) /* comparison zone TGZONE2 */.

Once you have the SYSMOD Comparison report, you can install the applicable
SYSMODs in the zone where they are needed. Follow these steps:
1. Research the report to determine which of the identified SYSMODs you want to
install into the comparison zone.

REPORT SYSMODS Command
2. Find and receive any applicable SYSMODs that were not available and that you
want to install. The source ID in the report identifies some possible sources for
obtaining the SYSMODs.
3. Tailor the SMPPUNCH output to install the set of SYSMODs that you deem
appropriate; then run the commands to install the desired SYSMODs.

Chapter 17. Building a user modification
This chapter discusses steps and considerations for building a user modification
(USERMOD). It provides the following information:
v How to choose between building a SYSMOD as a USERMOD and building it as
a function
v How to create modification control statements
v Examples of USERMODs
Choosing between a USERMOD and a function SYSMOD

Software products available from IBM provide you with many functions. However,
these functions may not always exactly meet your processing requirements. Many
of these products provide interfaces, such as user exit routines or dummy modules,
that you can use to customize the functions to your needs. Sometimes, however,
you may need to change a function substantially. You can do this by either:
v Constructing a user modification as USERMODs to an existing function
v Constructing an additional function SYSMOD
Although you might be able to make these changes without SMP/E, there are
advantages to creating them either as USERMODs or as function SYSMODs so that
SMP/E can install them.
When SMP/E installs the changes, it does the following:

v Keeps a record of the changes
v Reports any intersections with SYSMODs provided by IBM
v Ensures that the changes are not regressed
v Ensures that the changes are installed properly in the correct libraries
v Lets you remove the changes if there are problems
Before creating your changes, you must decide whether to build a USERMOD type
SYSMOD or a function SYSMOD. Table 14 lists considerations that will help you
make this decision.
Table 14. Comparison of USERMODs and function SYSMODs
USERMOD Function SYSMOD
Provides changes for elements owned by an Provides new elements or new element
existing function. versions for a new function.
May provide new elements for an existing

function.
SMP/E reports on changes attempted by SMP/E does not report on changes
PTFs, APARs, or USERMODs. attempted by PTFs, APARs, or USERMODs.
Other SYSMODs for the same function can Because the SYSMOD owns the element,
update the elements. SYSMODs for other functions cannot update
the element without also changing the
owner of the element.
Better for small changes that affect only a Better for major additions to the system.
few elements.

Building a USERMOD
Creating the MCSs

This section describes some of the considerations for building the MCSs for a
USERMOD SYSMOD. See z/OS Packaging Rules for guidelines on when to use
USERMODs and SMP/E Modification Control Statements in SMP/E Reference for
more information on MCS syntax.
The ++USERMOD MCS

The ++USERMOD statement identifies this SYSMOD as a USERMOD and assigns
a 7-character identifier to the SYSMOD.
The format of the ++USERMOD statement is:

++USERMOD(sysmod_id) /* */.
The ++VER MCS

The ++VER statement is necessary in all SYSMODs. It describes the environment
necessary for installing the SYSMOD.
The general format of the ++VER statement follows:

++VER /* Environment MCS */
(srel) /* System and release ID */
FMID(aaaaaaa) /* Functional area */
PRE ( /* */
bbbbbbb bbbbbbb /* Prerequisite PTFs */
bbbbbbb bbbbbbb /* */
) /* */
REQ ( /* */
ccccccc ccccccc /* Other related USERMODs */
ccccccc ccccccc /* */
) /* */
SUP ( /* */
ddddddd ddddddd /* Other USERMODs incorp- */
ddddddd ddddddd /* orated into this one */
) /* */.
Specifying the proper system release

The SREL value (srel) must be one of those defined as an SREL subentry in the
TARGETZONE entry. If the USERMOD is a change to an IBM product, the SREL
should correspond to the SREL value specified in the IBM product that currently
owns the elements within this SYSMOD.
Specifying the FMID value

If any element is owned by an FMID different from that specified in the ++VER
statement, that element is not selected for installation during APPLY or ACCEPT
processing, and message GIM45401W is issued. This condition is a SYSMOD
construction error. SMP/E supports a SYSMOD construction that assumes that this
condition occurs regularly and that the SYSMOD contains an ++IF statement
specifying another SYSMOD that supplies the proper functional version of the
element.
Note: As was explained earlier, it is a good idea to construct a USERMOD so that

each SYSMOD contains only one element. This construction method
eliminates this problem.
Specifying the proper requisites

When you specify requisite SYSMODs, you are defining two kinds of relationships:
v The relationship of your SYSMOD to previous versions of the element

Building a USERMOD
v The relationship of your SYSMOD to other SYSMODs currently on the system
The following text describes how you can define these relationships.
Relationships to earlier versions of the elements:

1. If the element entry in your target zone has an RMID value different from its
FMID value, ensure it is a prerequisite of the USERMOD fix; that is, make sure
the bbbbbbb value shown in the example is accurate. If the RMID and FMID
values are equal, the bbbbbbb value need not be specified.
2. If the element entry in your target zone has any UMID values, you should first
check to make sure the USERMOD itself was constructed so it works correctly
in that environment. You should then make sure each of the UMID values is
specified in the PRE operand in place of the ccccccc values shown in the
example. This is not an absolute requirement, but if it is omitted, SMP/E issues
warning messages during installation identifying that these SYSMODs may
have an intersection with the one you are installing and, therefore, may be
regressed. Putting the UMID values in the PRE list suppresses these messages.
3. If you do not specify the requisites that previously replaced an element, SMP/E
does not allow your USERMODs to be installed unless you specify BYPASS(ID)
on the APPLY or ACCEPT commands.
4. If you want this SYSMOD installed without any warning or error messages,
you must specify all the current UMID values of each element in the SYSMOD
in the PRE operand. This indicates to SMP/E that the SYSMOD was designed
to be installed on the current function and service level of the element,
including update level.
If you do not specify the requisites that previously updated an element, the
following occurs:
v If your SYSMOD contains an element replacement, SMP/E does not allow
the SYSMOD to be installed unless BYPASS(ID) is specified on the APPLY
and ACCEPT command.
v If your SYSMOD contained an element update, SMP/E allows the SYSMOD
to be installed, but issues a warning message for each requisite not specified
in the PRE list.
This means SMP/E is unable to determine if there is an intersection between
your update and those already on the element. It assumes that there is none,
and you should investigate both updates to verify this.
Relationships to Other SYSMODs: Your SYSMOD may depend on another

USERMOD or IBM PTF being installed, because you depend on the function
provided by that SYSMOD. You may want to indicate that this SYSMOD is part of
a set of USERMODs designed to provide some user function. Because each
SYSMOD contains only one element, you want to tell SMP/E that they should all
be installed together. This is done with the REQ operand (the ccccccc values in the
example).
Specifying superseded SYSMODs

If this SYSMOD is to fix multiple problems, each of the additional APARs that are
being fixed should be specified in the SUP operand (ddddddd values in the
example). This notifies SMP/E that it is not necessary to install those SYSMODs
after the current SYSMOD is installed.
Chapter 17. Building a user modification 171

Building a USERMOD
The ++JCLIN MCS

The ++JCLIN statement is necessary in all SYSMODs that add new modules or
change the link-edit characteristics or link-edit control cards for existing load
modules. The data following the ++JCLIN statement consists of the jobs necessary
for installing the new module or link-editing the affected load modules, as if that
JCL were actually going to build the load modules from scratch from members of
libraries.
The general format of the ++JCLIN statement is:
++JCLIN /* Installation data */.
++MOD and ++ZAP MCSs

The ++MOD and ++ZAP statements are used to identify a replacement and update
to a module in the distribution library and associated load modules.
The data following the ++MOD statement is an object deck. The general format of
the ++MOD statement is:
++MOD(modname) /* Distribution name */
DISTLIB(dlibname) /* DLIB ddname */
...
... Object deck for module
...
The data following the ++ZAP statement is a set of superzap control statements.
The general format of the ++ZAP statement is:
++ZAP(modname) /* Distribution name */
DISTLIB(dlibname) /* DLIB ddname */.
...
... Superzap control statement
...
The superzap control statements are the same as you would use if you were calling
the superzap utility. The only exception is that on the NAME statement, you
should specify only the CSECT name within the distribution library module, rather
than the load module name and the CSECT name. When SMP/E installs the
SYSMOD, it determines all the load modules into which this distribution module
was link-edited and then calls the superzap utility for each of these load modules,
modifying the NAME statement as appropriate.
++MAC and ++MACUPD MCSs

The ++MAC and ++MACUPD statements are used to identify a replacement and
update to a macro in the distribution library and associated target libraries.
The data following the ++MAC statement is the macro replacement. The general
format of the ++MAC statement is:
++MAC(macname) /* Macro name */
...
... Macro replacement
...
The data following the ++MACUPD statement is the control statements that would
have been used if you called the IEBUPDTE utility. The general format of the
++MACUPD statement is:

Building a USERMOD
++MACUPD(macname) /* Macro name */

...
... Update control statements
...
The following restrictions are enforced by SMP/E:

1. The first statement must be the “./ change name=macname” control statement.
2. The name specified on the change statement must be the same as on the
++MACUPD statement.
3. No insert or delete statement can be used. Inserting must be done by manually
assigning each line a number. Deleting must be done by commenting out the
line. This restriction enables SMP/E to merge the update control statement
when multiple SYSMODs modify the same macro.
++SRC and ++SRCUPD MCSs

The ++SRC and ++SRCUPD statements are used to identify a replacement and
update to source in the distribution library and associated target libraries. The
format of the MCSs and the data format and restrictions are similar to those for
macros.
The ++PROGRAM MCS

The ++PROGRAM statement is used to define replacements for program elements,
which are pre-built load modules or program objects. (For a complete description
of program elements, see the chapter on MCSs in SMP/E Reference.) You can add a
program element by packaging it in a USERMOD.
The ++PROGRAM MCS is immediately followed by the load module or program

object. The general format of the ++PROGRAM MCS is:
++PROGRAM(name) /* Data element type, name */
...
... Program element replacement
...
To be packaged inline, a program element must be unloaded along with its aliases
into a sequential data set and then transformed with GIMDTS into the required
fixed-block-80 record format before it is packaged. Later, when SMP/E installs the
element, it is changed back to its original format. For more information about
using GIMDTS, see SMP/E Reference.
Data element MCSs

Data element MCSs are used to define replacements for data elements, which are
elements other than macros, modules, program elements, and source code. These
include TSO CLISTs, help panels, ISPF dialog panels, and online publications
libraries. (For a complete description of data elements, see SMP/E Modification
Control Statements in SMP/E Reference.) You can add a data element by packaging
it in a USERMOD.
The data element MCS is immediately followed by the data element itself. The
general format of the data element MCS is:

Building a USERMOD
++element(name) /* Data element type, name */

...
... Data element replacement
...
To be packaged inline, a data element must contain fixed-block 80 records. If the

original format of the element is not fixed-block 80 records, you can use GIMDTS,
a service routine provided with SMP/E, to transform the element into the required
format before packaging it. Later, when SMP/E installs the element, it is
reconverted to its original format. For more information about using GIMDTS, see
SMP/E Reference.
Hierarchical file system element MCSs

The hierarchical file system element MCSs are used to define a replacement for an
element residing in a UNIX file system. You can add a hierarchical file system
element by packaging it in a USERMOD.
The hierarchical file system element MCS is immediately followed by the

hierarchical file system element itself. The general format of a hierarchical file
system element MCS is:
++hfs_element(hfsname) /* HFS name */
...
... Hierarchical file system element replacement
...
To be packaged inline, a hierarchical file system element must contain fixed-block

80 records. If the original format of the element is not fixed-block 80 records, you
can use GIMDTS, a service routine provided with SMP/E, to transform the
element into the required format before packaging it. Later, when SMP/E installs
the element, it is reconverted to its original format. For more information about
using GIMDTS, see SMP/E Reference.
Examples of USERMODs
This section contains examples of USERMODs that:
v Update a module
v Replace a module
v Add new modules
v Replace a macro or source code
v Update a macro or source code
v Add new source code
v Add source code that uses an IBM-supplied macro
v Add a module that uses an IBM-supplied macro
Example 1: Updating a module

This is an example of a USERMOD that updates module MODULEA.

Building a USERMOD
++USERMOD(USMP001) /* SYSMOD ID of USERMOD. */.

++VER(Z038) /* SREL for MVS. */
FMID(HMP1E00) /* Applicable to SMPE. */
. /* */
++ZAP(MODULEA) /* Name of module. */
DISTLIB(AOS12) /* ddname of DLIB. */
/* */.
NAME MODULEA
* Verify existing ddname.
VER 0050 404040404040404000
* Verify existing SYSOUT class.
VER 0058 40
* Verify existing terminal assignment.
VER 0059 00
* Replace with new data.
* M Y P R I N T = ddname
REP 0050 D4E8D7D9C9D5E3
* A = SYSOUT class
REP 0058 C1
Notes:
1. You should verify each location that you are going to update.
2. You can specify only one name in the NAME statement.
3. The changes made by the REP statements are explained by the preceding
comment lines.
Example 2: Replacing a module

This is an example of a USERMOD that replaces module MODULEB. After writing
and assembling MODULEB, package it in a USERMOD, as shown below:
++USERMOD(USMP002) /* SYSMOD is for USERMOD. */.
. /* */
++MOD(MODULEB) /* Name of module. */
DISTLIB(AOS12) /* ddname of DLIB. */
/* */.
...
... object module for MODULEB
...
Note: There are no PRE operands on the ++VER statement, because this module
has not been modified since its initial installation. Therefore, only the
++VER FMID operand is required.
Example 3: Adding new modules

This is an example of a USERMOD that adds new modules to create a new load
module. In this example, the new modules are SMPMOD01, SMPMOD02, and
SMPMOD03. These modules are link-edited to create load module SMPEXT01,
whose entry point is SMPMOD01. The target library for SMPEXT01 is
SYS1.LINKLIB. The following example shows how the new modules and load
modules can be packaged in a USERMOD.

Building a USERMOD
++USERMOD(USMP003) /* SYSMOD ID of USERMOD. */.

/* */.
++JCLIN /* JCLIN to install routine.*/.
//JOB1 JOB ’accounting info’,MSGLEVEL=(1,1)
//STEP1 EXEC PGM=IEWL
//PVTDLIB1 DD DSN=SYS1.PVTDLIB1,DISP=OLD
//SYSLMOD DD DSN=SYS1.LINKLIB,DISP=OLD
//SYSLIN DD *
INCLUDE PVTDLIB1(SMPMOD01)
INCLUDE PVTDLIB1(SMPMOD02,SMPMOD03)
ENTRY SMPMOD01
NAME SMPEXT01(R)
/*
++MOD(SMPMOD01) /* Customized exit routine. */
DISTLIB(PVTDLIB1) /* ddname of DLIB. */
/* */.
...
... object module for SMPMOD01
...
/* */.
...
...
/* */.
...
...
Notes:
1. The ++JCLIN statement is required because the SYSMOD provides new
modules and a new load module.
2. This SYSMOD could have been packaged as a ++FUNCTION, because each of
the modules is user-written.
3. This SYSMOD could have been packaged as three separate SYSMODs, each
containing one module. In that case, each SYSMOD would then need to specify
the ++VER REQ operand so the SYSMODs would be installed as a set.
Example 4: Replacing a macro or source code

This is an example of a USERMOD that adds a new macro. In this example, the
macro is installed in a target and a distribution library. In addition, source code
element USRSRC01 is to be assembled when the macro is installed.
++USERMOD(UJES004) /* SYSMOD ID of USERMOD. */.
FMID(HJE2102) /* Applicable to JES. */
/* */.
++MAC(USRMAC01) /* Name of new macro. */
DISTLIB(USRMACLB) /* ddname of DLIB. */
SYSLIB (MACLIB) /* ddname of system MACLIB. */
ASSEM (USRSRC01) /* Reassemble this source. */
/* */.
...
... macro replacement
...

Building a USERMOD
Notes:
1. No JCLIN is required to install a new macro; all necessary information can be
specified on the ++MAC statement.
2. You can follow this example to add new source code, with the following
exceptions:
v The ASSEM operand is not required for source code. If SMP/E understands
where source code is to be installed, it automatically tries to assemble the
new version of the source code after replacing the old version.
v To define how the source code should be installed, you should include a
++JCLIN statement followed by JCLIN data. The JCLIN data should be
similar to that in the example of adding a new module.
Example 5: Updating a macro or source code

The following is an example of a USERMOD to update a macro that was added by
a previous USERMOD.
++USERMOD(UJES005) /* SYSMOD ID of USERMOD. */.

FMID(HJE2102) /* Applicable to JES. */
PRE (UJES004) /* Previous replacement. */
/* */.
++MACUPD(USRMAC01) /* Update customized macro. */
DISTLIB(USRMACLB) /* ddname of DLIB. */
/* SMPE knows SYSLIB. */
ASSEM(USRSRC01) /* Reassemble this source. */
/* */.
./ CHANGE NAME=USRMAC01
...
... macro changed lines here with
... sequence numbers in columns 73–80
...
Notes:
1. The ++VER PRE operand specifies the previous SYSMOD that replaced the
macro.
2. You can follow this example to update source code, with the following
exceptions:
v The ASSEM operand is not used for source code.
v Because the SYSMOD adding the source code defined how the module
should be installed, there is no need to repeat that information on a ++JCLIN
statement in the USERMOD.
Example 6: Adding new source code

Assume that you have written new source code to be added to an existing product.
It is a member in one of your partitioned data sets. To be installed, it must be
assembled, then copied as a single-module load module into its target library.
Table 15 shows the information you need to specify and where to specify it.
Table 15. Information needed to add new source code
Information to provide: Where to specify it:
1 Name of source code element: v Element name on ++SRC MCS:

++SRC(IFBSRC01)
IFBSRC01

Building a USERMOD
Table 15. Information needed to add new source code (continued)

2 Name of object module produced by v JCLIN data, M= value on COPY SELECT statement:
assembly: COPY SELECT M=(IFBSRC01)
IFBSRC01 (same as source code
element)
3 Where input source code is v TXLIB value on ++SRC MCS:

provided: ++SRC(IFBSRC01) TXLIB(REPLACE)
PDS member v DDDEF entry or DD statement for APPLY and ACCEPT processing:
REPLACE DD DSN=...
4 Target library for source: v SYSLIB value on ++SRC MCS:

++SRC(IFBSRC01) SYSLIB(IFBSRC)
SYS1.IFBSRC
v JCLIN data, OUTDD value on COPY statement:
COPY INDD=inddval,OUTDD=IFBSRC TYPE=SRC
v DDDEF entry or DD statement for APPLY processing:
IFBSRC DD DSN=SYS1.IFBSRC
5 Target library for load module: v JCLIN data, OUTDD value on COPY statement:
COPY INDD=inddval,OUTDD=LPALIB TYPE=MOD
SYS1.LPALIB
v DD statement in JCLIN data and in SMP/E job for APPLY
processing:
LPALIB DD DSN=SYS1.LPALIB
6 Distribution library for source code: v DISTLIB value on ++SRC MCS:
++SRC(IFBSRC01) DISTLIB(AIFBSRC)
SYS1.AIFBSRC
v JCLIN data, INDD value on COPY statement:
COPY INDD=AIFBSRC,OUTDD=IFBSRC TYPE=SRC
v DD statement or DDDEF entry for ACCEPT processing:
AIFBSRC DD DSN=SYS1.AIFBSRC
7 Distribution library for assembled v DISTMOD value on ++SRC MCS:

object module: ++SRC(IFBSRC01) DISTMOD(AOS23)
SYS1.AOS23 v JCLIN data, INDD value on COPY statement:
COPY INDD=AOS23,OUTDD=LPALIB TYPE=MOD
v DD statement in JCLIN data and in SMP/E job during APPLY and
ACCEPT processing:
AOS23 DD DSN=SYS1.AOS23

Building a USERMOD
Table 15. Information needed to add new source code (continued)

8 How to install the source code: v ++SRC MCS automatically notifies SMP/E to assemble the module.
v JCLIN data, COPY steps:
Assemble, then copy
//JOB1 JOB ...
//STEP1 EXEC PGM=IEBCOPY
//AIFBSRC DD DSN=SYS1.AIFBSRC,...
//IFBSRC DD DSN=SYS1.IFBSRC,...
//AOS23 DD DSN=SYS1.AOS23,...
//LPALIB DD DSN=SYS1.LPALIB,...
//SYSIN DD *
COPY INDD=AIFBSRC,OUTDD=IFBSRC TYPE=SRC
SELECT M=(IFBSRC01)
COPY INDD=AOS23,OUTDD=LPALIB TYPE=MOD
SELECT M=(IFBSRC01)
/*
v The OPTIONS and UTILITY entries in effect for the APPLY or
ACCEPT command being processed specify the names of the
assembler and copy utilities to use and parameters to be passed to
them.
The following is a sample USERMOD that can be used to package the source code.
The numbers associate items in the SYSMOD with the information listed in
Table 15 on page 177.
++USERMOD(USR0001) /* My USERMOD. */.

++VER(Z038) FMID(MYFMID1). /* For my function. */.
8 ++JCLIN. /* Link object module. */.
//JOB1 JOB ...
//STEP1 EXEC PGM=IEBCOPY
6 //AIFBSRC DD DSN=SYS1.AIFBSRC,DISP=SHR
4 //IFBSRC DD DSN=SYS1.IFBSRC,DISP=SHR
7 //AOS23 DD DSN=SYS1.AOS23,DISP=SHR
5 //LPALIB DD DSN=SYS1.LPALIB,DISP=SHR
//SYSIN DD *
4,6 COPY INDD=AIFBSRC,OUTDD=IFBSRC TYPE=SRC
2 SELECT M=(IFBSRC01)
5,7 COPY INDD=AOS23,OUTDD=LPALIB TYPE=MOD
2 SELECT M=(IFBSRC01)
/*
1,2,8 ++SRC(IFBSRC01) /* My source module. */
3 TXLIB(REPLACE) /* Where source is. */
6 DISTLIB(AIFBSRC) /* DISTLIB for source. */
7 DISTMOD(AOS23) /* DISTLIB for object. */
4 SYSLIB(IFBSRC) /* SYSLIB for source. */.
Example 7: Adding new source code that uses an

IBM-supplied macro
Assume you are packaging one of your user-written routines as a USERMOD
(UMOD001). Your USERMOD includes assembler source (SRCPART), which is to
be included in load module LOADMOD1 and is packaged as a SRC element with
a ++SRC statement, as described in previous USERMOD examples. SRCPART
refers to an IBM-supplied macro (IBMMAC), which was packaged in its owning
product with a ++MAC statement. You want SRCPART to be automatically
reassembled every time IBMMAC is updated or replaced with service. To
accomplish this, your USERMOD must contain a JCLIN assembly step in addition
to the necessary SMP/E MCS statements. (This means you need to supply the

Building a USERMOD
same SRCPART definition as part of the JCLIN input, as well as after the ++SRC
statement.) Your USERMOD might look something like this:
++USERMOD(UMOD001).
++VER(srel) FMID(fmid).
++JCLIN.
//STEP1 EXEC PGM=IEUASM
//SYSPUNCH DD DSN=&&PUNCH(SRCPART),DISP=(PASS);
//SYSIN DD *
SRCPART CSECT
IBMMAC
BR 14
END
//LINK EXEC PGM=IEWL
//SYSLMOD DD DSN=SYS1.LINKLIB,DISP=SHR
//SYSLIN DD *
INCLUDE SYSPUNCH(SRCPART)
NAME LOADMOD1
/*
++SRC(SRCPART) SYSLIB(tgtlib) DISTLIB(dlib).
SRCPART CSECT
IBMMAC
BR 14
END
When the assembly step in the JCLIN is processed, a GENASM subentry is added
to the MAC entry for IBMMAC indicating SRCPART is to be assembled whenever
IBMMAC is updated. So, if you install a SYSMOD that updates IBMMAC, SMP/E
automatically reassembles SRCPART and link-edits the new copy into
LOADMOD1.
Note: Remember, when a link-edit step contains a SYSLMOD DD statement,

SMP/E uses the low-level qualifier of the data set name to determine the
ddname of the load module’s target library (and, by extension, the name of
the DDDEF entry to use when allocating this library). In this USERMOD, for
example, SMP/E looks for a DDDEF entry named LINKLIB in order to
allocate the SYSLMOD data set.
Example 8: Adding a new module that uses an IBM-Supplied

macro
Assume you are packaging one of your user-written routines as a USERMOD
(UMOD001). Your USERMOD includes an object module (SRCPART), which is to
be included in load module LOADMOD1 and is packaged as a MOD element with
a ++MOD statement, as described in previous USERMOD examples. SRCPART
refers to an IBM-supplied macro (IBMMAC), which was packaged in its owning
product with a ++MAC statement. You want to be notified whenever IBMMAC is
updated or replaced with service so you can update your module and reapply
your USERMOD. To accomplish this, your USERMOD must include the macro,
and must specify the last SYSMOD that replaced the macro (the RMID value in the
MAC entry) and all the SYSMODs that have updated the macro (the UMID values
in the MAC entry) as prerequisites. Your USERMOD might look something like
this:

Building a USERMOD
++USERMOD(UMOD001).
++VER(srel) FMID(fmid) PRE(macrmid).
++JCLIN.
//LINK EXEC PGM=IEWL
//SYSLMOD DD DSN=SYS1.LINKLIB,DISP=SHR
//SYSLIN DD *
INCLUDE DLIB1(SRCPART)
NAME LOADMOD1
/*
++MAC(IBMMAC).
...
macro source
...
++MOD(SRCPART) DISTLIB(dlib).
...
module object deck
...
Because you specified the macro’s RMID and UMIDs, when IBMMAC is serviced,
the APPLY will get a MODID error for the USERMOD. You will have to restore the
USERMOD to successfully apply the service. Then you can rework the USERMOD
and apply it again.
Note: Remember, when a link-edit step contains a SYSLMOD DD statement,

SMP/E uses the low-level qualifier of the data set name to determine the
ddname of the load module’s target library (and, by extension, the name of
the DDDEF entry to use when allocating this library). In this USERMOD, for
example, SMP/E looks for a DDDEF entry named LINKLIB in order to
allocate the SYSLMOD data set.

Building a USERMOD

Chapter 18. Determining which SYSMODs led others to fail:
The causer SYSMOD summary report
This chapter describes root cause analysis and provides examples of how to use
the causer SYSMOD information in the Causer SYSMOD Summary report and the
SYSMOD Status report to determine the cause of SYSMOD failure.
Introduction
A causer SYSMOD is a SYSMOD whose failure led to the failure of other
SYSMODs. To identify causer SYSMODs, SMP/E performs root cause analysis for
the ACCEPT, APPLY, and RESTORE commands. The types of errors SMP/E
analyzes to determine causer SYSMODs include the following:
v Held SYSMODs
v Missing requisite SYSMODs
v Utility program failures: copy, update, assembler, link, zap
v Out-of-space conditions: x37 ABENDs
v Missing DD statements and other allocation errors
v ID errors (a SYSMOD does not supersede or specify as a prerequisite an RMID
or a UMID of an element)
v JCLIN errors (syntax errors)
The results of SMP/E’s root cause analysis are presented in two reports:
v SYSMOD Status report
This report summarizes the processing done for each eligible SYSMOD. For
SYSMODs that failed, the report lists the causer SYSMODs.
After you check the SYSMOD Status report to determine the results of
processing, use the Causer SYSMOD Summary report to determine which errors
need to be corrected.
v Causer SYSMOD Summary report
This report lists the causer SYSMODs along with a brief summary of the related
messages, descriptions of the errors that caused the SYSMODs to fail, and, when
feasible, some causes for those errors.
Using causer SYSMOD information

How you use the causer SYSMOD information provided by the Causer SYSMOD
Summary report and the SYSMOD Status report depends on what you need to
install—all the SYSMODs specified on the command you are processing, or only
specific ones. This section describes some general scenarios and includes an
example of using these reports.
Resolving errors for all SYSMODs that failed

Suppose you are installing a CBPDO, but the reports show that several of the
SYSMODs failed. You want to resolve all the reported errors and install all the
SYSMODs in the CBPDO.
1. Go to the Causer SYSMOD Summary report to identify the causer SYSMODs
and determine how to resolve the errors.

Causer SYSMOD Summary Report
2. Resolve the errors.

3. Rerun the command that failed. If you find more errors on this pass, repeat the
process.
Resolving errors for a single SYSMOD that failed

Suppose you are installing a group of SYSMODs, but the reports show that several
of them have failed. You want to resolve the errors for one specific SYSMOD and
install it.
1. Use the SYSMOD Status report to determine the causer SYSMOD for the
SYSMOD that you need to install.
2. Go to the Causer SYSMOD Summary report and determine how to resolve the
errors for the causer SYSMOD.
3. Resolve the errors.
4. Rerun the command to install the SYSMOD that previously failed. If you find
more errors on this pass, repeat the process.
Example
Suppose you ran the following command:
APPLY S(UR00001) GEXT CHECK.
and the SYSMOD Status report included the results shown in Figure 36.
Page nnnn - NOW SET TO TARGET ZONE nnnnnnn DATE mm/dd/yy TIME hh:mm:ss SMP/E 34.nn SMPRPT OUTPUT
SYSMOD STATUS REPORT FOR APPLY PROCESSING SYSMODS APPLIED - 0
NOTE: ’-’ INDICATES THE REQUISITE SYSMOD OR HOLD CONDITION IS NOT SATISFIED
’*’ INDICATES THE NON SATISFIED REQUISITE SYSMOD OR HOLD CONDITION IS BYPASSED
’#’ INDICATES THE SUPERSEDING SYSMOD WAS NOT PROCESSED
SYSMOD STATUS TYPE FMID REQUISITE SYSMODS, SUPBY SYSMODS, HOLD REASON-IDS, AND CAUSER SYSMODS
AR00001 NOGO SUPBY #UR00002
UR00001 HELD PTF HROOTC1 HOLDE -AR00001

CAUSER UR00002
UR00002 HELD PTF HROOTC1 HOLDE -AR00003

CAUSER UR00002
Figure 36. SYSMOD Status Report: Sample Report for APPLY
In this case, the report indicates that APPLY processing failed for SYSMODs
UR00001 and UR00002 because of unresolved error holds. The causer SYSMOD for
both PTFs is UR00002. Next, you look up UR00002 in the Causer SYSMOD
Summary report, shown in Figure 37 on page 185.

Causer SYSMOD Summary Report
Page nnnn - NOW SET TO TARGET ZONE nnnnnnnn DATE mm/dd/yy TIME hh:mm:ss SMP/E 34.nn SMPRPT OUTPUT
CAUSER SYSMOD SUMMARY REPORT FOR APPLY PROCESSING
CAUSER FMID MESSAGE ID PAGE ERROR DESCRIPTION AND POSSIBLE CAUSES
UR00002 HROOTC1 GIM35901I 2 ERROR HOLD AR00003 WAS NOT RESOLVED.
Figure 37. Causer SYSMOD summary report: sample report for APPLY
That report shows that APPLY processing failed because error hold AR00003 was
not resolved for SYSMOD UR00002. By resolving this hold, you will fix the errors
listed in the SYSMOD Status report.
Chapter 18. Determining which SYSMODs led others to fail: The causer SYSMOD summary report 185
Chapter 19. Java archive update exploiter’s guide
This chapter is a guide to packaging Java Archive files (JAR files) and updates for
JAR files in SMP/E. The intended audience for this section is anyone who will be
exploiting this function, including product packagers and service teams. This guide
provides information and examples about packaging a product release using the
++JAR MCS and building subsequent PTFs to service that release using the
++JARUPD MCS.
JAR replacements in FMIDs

Suppose your product has a simple TicTacToe applet. The complete TicTacToe
applet (or Java application) is composed of a class file and audio and image files,
and is stored in a directory structure as follows:
/u/pezk/TicTacToe/TicTacToe.class
/audio/beep.au
/ding.au
/yahoo.au
/images/cross.gif
/not.gif
To build a JAR file element for the complete TicTacToe applet, you would first
make your working directory the TicTacToe directory where the applet files reside,
and then run the following jar command:
cd /u/pezk/TicTacToe/
jar cvf ABCTTT.jar *
Specified in this way, the jar command takes all files in the working directory
(/u/pezk/TicTacToe), and all files within subdirectories of the working directory,
and creates a JAR file named ABCTTT.jar.
Your product release, or FMID, can then contain this single JAR file, which is the
complete TicTacToe applet. You can do this using the ++JAR MCS as seen in the
following example:
++JAR(ABCTTT) DISTLIB(AABCBIN) SYSLIB(SABCBIN) RELFILE(2)
PARM(PATHMODE(0,6,4,4))
LINK(’../TicTacToe.jar’)
SYMLINK(’../../../../../usr/lib/TicTacToe.jar’)
SYMPATH(’../../usr/lpp/abc/bin/TicTacToe.jar’).
JAR updates in PTFs

Suppose some time after your FMID has been made available and is being used, a
defect is discovered (an APAR). The TicTacToe applet requires a change because of
the APAR. Specifically, you need to add another image file (images/new.gif), and
replace the class file (TicTacToe.class) with an updated copy of the class file.
Further, suppose the change team has placed the new image file and updated class
file in directories, as follows:
/u/apars/ow12345/TicTacToe/TicTacToe.class
/images/new.gif
You can instruct SMP/E to add new files to a JAR file, and replace files in an
existing JAR file by using the ++JARUPD MCS. If files TicTacToe.class and new.gif
were to be packaged and archived together in a JAR file of their own, then that

JAR Updates
file, in SMP/E terms, would be considered a JAR update file. For example, given
the directory structure above for the new and updated files, the jar command
could be used to create the JAR update file as follows:
cd /u/apars/ow12345/TicTacToe/
jar cvf ABCTTT.jarupd *
The resultant JAR file ABCTTT.jarupd would be packaged as a ++JARUPD in a

PTF as in the following example:
++PTF(UW12345).
++VER(Z038) FMID(fmid).
++JARUPD(ABCTTT)
PARM(PATHMODE(0,6,4,4)) JARPARM(0M)
Suppose now another defect (APAR) is discovered against the TicTacToe applet,
and you must update the /audio/beep.au file within the archive. Again, you can
use the ++JARUPD MCS to describe an update to the archive. Assuming the
replacement audio file resides in a directory as follows:
/u/apars/ow54321/TicTacToe/audio/beep.au
The following jar command could create the necessary JAR update:
cd /u/apars/ow54321/TicTacToe/
jar cvf ABCTTT.jarupd *
The resulting JAR file ABCTTT.jarupd would be packaged as a ++JARUPD in a

PTF as in the following example:
++PTF(UW54321).
++VER(Z038) FMID(fmid) PRE(UW12345).
++JARUPD(ABCTTT)
Notice the second PTF has a prerequisite for the first PTF. Such a relationship is
required by SMP/E because both PTFs update the same JAR file.
JAR replacements in PTFs

Just as replacements for JAR files can be delivered in an FMID, so can JAR file
replacements be delivered in PTFs. Suppose you decide to create a ″level-set″ PTF,
which incorporates the updates from several previous APARs and PTFs. In this
case, you want to replace the entire TicTacToe applet JAR file, rather then add or
replace files within the archive.
You create the JAR file for the applet using the jar command as discussed earlier
and then package it as a ++JAR element in a PTF, as in the following example:
++PTF(UW87654).
++VER(Z038) FMID(fmid) SUP(UW12345 UW54321).
++JAR(ABCTTT) DISTLIB(AABCBIN) SYSLIB(SABCBIN)

JAR Updates
Notice this PTF supersedes both PTFs that supplied updates for the JAR file.
Again, such relationships (either supersede or prerequisite) are required by SMP/E,
because both PTFs provided updates for the JAR file contained in the current PTF.
Chapter 19. Java archive update exploiter’s guide 189

Appendix A. Migration
Migration overview
Your plan for migrating to the new level of SMP/E should include information
from a variety of sources. These sources of information describe topics such as
coexistence, service, migration procedures, and interface changes.
The following documentation, which is supplied with your product order, provides
information about installing your z/OS system. In addition to specific information
about SMP/E, this documentation contains information about all of the z/OS
elements.
v z/OS and z/OS.e Planning for Installation
This book describes the installation requirements for z/OS at a system and
element level. It includes hardware, software, and service requirements for both
the driving and target systems. It also describes any coexistence considerations
and actions.
v SMP/E Program Directory
This document, which is provided with your SMP/E product order, leads you
through the specific installation steps for SMP/E.
v ServerPac Installing Your Order
This is the order-customized installation book for using the ServerPac
installation method. Be sure to review “Appendix A. Product Information”,
which describes data sets supplied, jobs or procedures that have been completed
for you, and product status. IBM may have run jobs or made updates to
PARMLIB or other system control data sets. These updates could affect your
migration.
Within this book, you can find information about the specific updates and
considerations that apply to this release of SMP/E.
v “SMP/E V3R4 overview” on page 193
This section describes the specific updates that were made to SMP/E for the
current release. For each item, this section provides an overview of the change, a
description of any migration and coexistence tasks that may be considered, and
where you can find more detailed information in the SMP/E library or other
element libraries.
v “Summary of interface changes” on page 213
This section provides a summary of the changes that are made to z/OS user and
programming interfaces.
Terms you need to know

This section describes some terms you may need to know as you use this section.
Migration Activities that relate to the installation of a new version or release
of a program to replace an earlier level. Completion of these
activities ensures that the applications and resources on your
system will function correctly at the new level.
Coexistence Two or more systems at different levels (for example, software,
service, or operational levels) that share resources. Coexistence
includes the ability of a system to respond in the following ways to

a new function that was introduced on another system with which
it shares resources: ignore a new function, terminate gracefully,
support a new function.
New releases of SMP/E generally add to or enhance the
information stored in SMP/E data sets. This new or enhanced data
is not always compatible with previous releases of SMP/E. Results
are unpredictable when a prior release of SMP/E encounters data
introduced by a later release. Therefore, IBM provides toleration
PTFs that allow selected prior releases to ignore changes
introduced by a later release. In some cases, a compatibility PTF or
small programming enhancement (SPE) is available to add a new
function to selected prior releases. z/OS and z/OS.e Planning for
Installation, GA22-7504, provides a list of the required toleration
and compatibility PTFs for each prior release of SMP/E.
SMP/E release levels

The SMP/E element of z/OS is usually not updated with every release of z/OS.
For example, the level of SMP/E that was shipped with z/OS Release 1 is identical
to that shipped with OS/390 V2R7.
Developing a migration strategy

The recommended steps for migrating to a new release of SMP/E are:
1. Become familiar with the supporting migration and installation documentation
for the release.
You should determine what updates are needed for products that are supplied
by IBM, system libraries, and non-IBM products. Review z/OS and z/OS.e
Planning for Installationand z/OS Introduction and Release Guidefor information
about SMP/E and other z/OS elements.
2. Develop a migration plan for your installation.
When planning to migrate to a new release of SMP/E, you must consider
high-level support requirements, such as machine and programming
restrictions, migration paths, and program compatibility.
3. Obtain and install any required program temporary fixes (PTFs) or updated
versions of the operating system.
Call the IBM Software Support Center to obtain the preventive service planning
(PSP) upgrade for SMP/E, which provides the most current information about
PTFs for SMP/E. Check RETAIN again just before testing SMP/E. For
information about how to request the PSP upgrade, refer to the SMP/E Program
Directory. Although the SMP/E Program Directory contains a list of the required
PTFs, the most current information is available from the IBM Software Support
Center.
4. Install the product using the SMP/E Program Directory or the ServerPac Installing
Your Order documentation.
5. Verify that any application programs that use the SMP/E API will continue to
run and, if necessary, make changes to ensure compatibility with the new
release.
6. Use the new release before initializing major new function.
7. If necessary, customize the new function for your installation.
8. Exercise the new functions.

Reviewing changes to SMP/E processing
As you define your installation’s migration plan, consider how the new and
changed SMP/E functions might affect the following areas of SMP/E processing.
for each item described in “SMP/E V3R4 overview,” you should review the
“Migration procedures” section to determine how, or if, the change affects the tasks
that are performed at your installation.
Customization You can customize some SMP/E functions to take
advantage of new support after the product is
installed. For example, you can tailor SMP/E
through the use of installation exit routines or
SMPPARM options. This section lists changes to
SMP/E that might require you to tailor SMP/E to
ensure that it runs as before.
Operations The new SMP/E release might introduce changes
to its operating characteristics, such as changed
commands, new or changed messages, or in the
methods of implementing new functions. This book
identifies those changes for which you should
provide user education before using this release of
SMP/E.
Programming To ensure that existing programs run as before,
programmers who develop programs that use the
GIMAPI, library change file records, or UNIX shell
scripts must be aware of any changes to those
interfaces introduced in a new release of SMP/E.
Reviewing changes to SMP/E interfaces

When defining your installation’s migration plan, also consider that SMP/E
interfaces may also be affected by the new or changed functions that are
introduced in this release. These interfaces include:
v Commands
v Exits
v Macros
v Messages
v Panels
v SYS1.SAMPLIB members
“Summary of interface changes” on page 213 provides a summary of the changes

that affect these interfaces for the release. This information is also listed in the
“What this change affects” section that is provided for each release enhancement in
“SMP/E V3R4 overview.”
SMP/E V3R4 overview

The following sections describe the new and changed SMP/E functions that are
introduced for SMP/E V3R4. The information about each item includes:
v Description
v Summary of the SMP/E tasks or interfaces that might be affected
v Coexistence considerations, if any, that are associated with the item
v Migration procedures, if any, that are associated with the item
v References to other publications that contain additional detailed information
Appendix A. Migration 193

Enhancement to the RECEIVE command
To support Internet Service Retrieval, the RECEIVE command has been enhanced
to enable you to request HOLDDATA or PTF orders directly from the IBM
Automated Delivery Request server, automatically download the resulting service
package and receive the HOLDDATA and PTFs contained in the service package.
Using the RECEIVE command with the new ORDER operand, you can request a
new HOLDDATA or PTF order based on the content criteria you specify. SMP/E
waits for the server to fulfill the order request. The order request is fulfilled by
building a package of PTFs and/or HOLDDATA that meets your content criteria.
When the resultant package is ready, SMP/E downloads the package to the local
z/OS host and performs traditional RECEIVE command operations on the contents
of the package.
In addition to submitting new orders, the RECEIVE ORDER command can also be
used to download orders that are in a pending state. If a requested PTF service
order cannot be fulfilled within the allowed time, the RECEIVE command
processing stops and SMP/E considers the order to be in the pending state. The
server continues fulfilling the order, however. You can use the RECEIVE ORDER
command to check on the status of the service package, and retrieve the package
when it is ready to be downloaded.
ORDERSERVER data set

The new ORDERSERVER data set provides the necessary information about the
server that will fulfill an SMP/E HOLDDATA or PTF order request.
CLIENT data set

The CLIENT data set has been changed to include information about the local
z/OS client system used for RECEIVE FROMNETWORK and RECEIVE ORDER
command processing, such as navigating an FTP firewall and an HTTP proxy
server.
Coexistence considerations
RECEIVE ORDER processing requires a new ORDERSERVER data set and new
operands on the RECEIVE command. SMP/E releases prior to SMP/E V3R4 cannot
process the new ORDERSERVER data set and do not recognize the new RECEIVE
command operands.
Impacts to SMP/E zone entries

This release introduces a new entry type, ORDER, as well as a new subentry type
for OPTIONS entries, ORDERRET.
ORDER entry in the global zone

The ORDER entry is a new entry in the global zone to describe a HOLDDATA or
PTF order initiated with the RECEIVE ORDER command. An ORDER entry is
created in the global zone when SMP/E sends an order request to the IBM
Automated Delivery Request server and the order is accepted by the server. This
ORDER entry is used to record information about the order so SMP/E can query
the server for status of orders that have not yet been completed.
ORDER RETENTION subentry on the OPTIONS entry

Indicates the retention period, in days, ORDER entries will be kept in the global
zone before being deleted. During RECEIVE ORDER command processing, the
ORDER entry will be deleted from the global zone if the ORDER RETENTION
subentry value has been exceeded.

When an ORDER entry is deleted from the global zone, SMP/E also deletes the
order package stored in the SMPNTS.
SMP/E releases prior to SMP/E V3R4 cannot process the new ORDER entry or the
new ORDERRET subentry of the OPTIONS entry. A toleration PTF provides:
v LIST, GZONEMERGE, GIMAPI and Query Dialog support for the ORDERRET
subentry and causes SMP/E to ignore the subentry in all other instances.
v GZONEMERGE, GIMAPI and Query Dialog support for the ORDER entry and
causes SMP/E to issue a warning message if ORDER entries exist in the zone
when the LIST command is requested. The ORDER entry will be ignored in all
other instances.
ICSF not required for GIMZIP and RECEIVE FROMNETWORK

The z/OS Integrated Cryptographic Services Facility (ICSF) is used by SMP/E to
calculate SHA-1 hash values. These hash values are calculated for files within a
GIMZIP package to verify the integrity of the data within the package. SMP/E has
been enhanced to use an alternate method to calculate SHA-1 hash values if ICSF
is not available for use.
Although ICSF is the preferred method, SMP/E no longer requires it for use by the
GIMZIP and GIMUNZIP service routines, nor for the RECEIVE FROMNETWORK
or RECEIVE FROMNTS command. If SMP/E detects that ICSF is not available,
SMP/E automatically uses an SMP/E Java application class to calculate SHA-1
hash values as an alternative. See “Options that affect Java” on page 94for the
details of the required setup to allow SMP/E to use the Java application class.
Improved load module build processing

The load module build phase of the APPLY, RESTORE and LINK LMODS
commands has been enhanced to be more tolerant of allocation errors for the
distribution libraries. This accommodates distribution libraries which may be
offline. In this case, SMP/E continues its search for a usable copy of the module
instead of immediately failing because of the error allocating the module’s
distribution library.
SMP/E administration dialog

The new ORDER Management dialog allows you to manage ORDER entries in the
global zone. ORDER entries are created by using the RECEIVE ORDER command
to request orders of HOLDDATA and PTFs from an IBM server. The ORDER
Management dialog allows you to
v See the status of all orders at a glance.
v View the details of individual orders.
v Delete the ORDER entry for selected orders.
SMP/E query dialog

Existing panel GIMQU1PO now allows you to specify the ORDER entry.

SMP/E V3R3 overview
v Description
GIMGTPKG service routine

The new GIMGTPKG service routine can be used to get GIMZIP packages from a
remote FTP server in a TCP/IP network and store the package on a local z/OS
host. GIMGTPKG performs the functions of the SMP/E RECEIVE
FROMNETWORK TRANSFERONLY command, but does so independently of
SMP/E. It uses FTP to transport the files of a GIMZIP package from a remote FTP
server to a local host, thus providing:
v Industry standard FTP protocol.
v Secure transmission using the capabilities of the z/OS FTP client.
v Ensured integrity of the transported files.
Enhancements to GIMZIP and GIMUNZIP service routines

Formerly GIMZIP could create and GIMUNZIP could process packages that
contained only sequential and partitioned data sets. Also, GIMUNZIP would only
extract data from an archive file into a new data set allocated directly by
GIMUNZIP. These service routines have been enhanced for SMP/E V3R3, as
follows:
v Packages may now contain VSAM data sets and UNIX files and directories.
v You may assign a unique id to an archive during GIMZIP processing and then
use that id to identify the archive that is to be unzipped during GIMUNZIP
processing.
v GIMUNZIP now allows GIMUNZIP operations into existing data sets.
GIMUNZIP determines if the data set specified on the <ARCHDEF> tag already
exists. If the data set already exists, GIMUNZIP copies the archive file into the
existing data set. If the data set does not already exist, GIMUNZIP allocates a
new data set and then copies the archive file into that new data set.
RECEIVE FROMNETWORK FTP interface enhancements

RECEIVE FROMNETWORK has been enhanced to:
v Allow user credentials and file data transferred between an FTP client and
server to be secured with respect to encryption, authentication, and data
integrity using the Transport Layer Security (TLS) enablement for FTP.
v Allow the z/OS FTP client to connect to FTP servers that reside beyond a
firewall that runs a SOCKS server.
v Make IPv6 connectivity possible for both the FTP client and server.
v Use the FTP.DATA configuration data set to allow the client to specify local site
parameters. The FTP.DATA configuration data set is optional, but must be used
by the client to specify the parameters for TLS security and SOCKS firewall
support.

Migration tasks
1. SMP/E will now use the FTP.DATA configuration data set to allow the client to
specify local site parameters. Two of the values specified in the FTP.DATA data
set are FWFriendly and FTPKEEPALIVE. These values correspond to the pasv
and keepalive attributes in the CLIENT data set. Therefore, these attributes
should no longer be specified in the CLIENT data set. If the pasv and keepalive
attributes are specified in the CLIENT data set, they will be ignored. If
required, these values must be specified in the FTP.DATA data set. Refer to
z/OS Communications Server: IP User’s Guide and Commands for more information
about the statements that can be coded in the FTP.DATA data set.
2. To enable TLS security, SOCKS firewall support, and IPv6 addressing, ensure
that z/OS Communications Server V1R2 (or higher) is installed.
3. If you previously specified FTP commands to navigate your local firewall using
the <FIRECMD> tag of the CLIENT data set, then you may need to update
them. Specifically, subcommands such as USER, PASS, and ACCT should no
longer be specified in <FIRECMD>. The commands you specify in the
<FIRECMD> tags should be the same as those you use with the z/OS
Communications Server FTP client, and should contain only the actual values
(or the appropriate substitution variables) for userid, password, and account.
REJECT CHECK command

A CHECK function has been added to the REJECT command. CHECK indicates
whether SMP/E should do a trial run of a command without actually updating
any libraries. This is a way to test for errors that might occur during actual
processing and to receive reports on the changes that would be made.
Extended RECEIVE SOURCEID processing

The RECEIVE command will now assign the source ID specified on the
SOURCEID operand of the command to SYSMODs found in the SMPPTFIN input
stream, even if the SYSMOD is already received.
SPCLCMOD and CMWA

SMP/E passes new default parameters (SPCLCMOD and CMWA=256K) to the
copy utility when copying modules, load modules, or programs.
SMP/E V3R2 overview

v Description
LINK LMODS command

The new LINK LMODS command can be used to refresh the callable services for
all load modules within a particular target zone. The LINK LMODS command can
also be used to refresh the callable services only for those load modules that have
a dependency on a particular set of CALLLIBS. The LINK LMODS command

replaces the REPORT CALLLIBS command, which has been removed from SMP/E
V3R2. For more information about the LINK LMODS command, refer to SMP/E
Commands.
REPORT CALLLIBS command removal

The REPORT CALLLIBS command has been removed from SMP/E V3R2. It has
been replaced by the LINK LMODS command.
UPGRADE command
New releases of SMP/E must sometimes make changes to SMP/E data sets that
cannot be properly processed by prior SMP/E releases. SMP/E usually makes
incompatible changes only when necessary to provide new and improved
capabilities. For example, a new type of element requires a new entry type in
SMPCSI data sets and these new entry types are typically not understood or
processed correctly by SMP/E levels that have not been specifically updated to do
so.
The UPGRADE command allows you to specify when SMP/E is permitted to

make incompatible changes to SMP/E data sets. This, in turn, allows you to make
the trade-off between exploiting new SMP/E functions and preserving
compatibility with prior SMP/E releases. For more information about the
UPGRADE command, refer to SMP/E Commands.
A toleration PTF will enable OS/390 V2R7 SMP/E, z/OS V1R2 SMP/E, and
SMP/E V3R1 to automatically check for incompatible changes made by a higher
level of SMP/E. A toleration PTF will also provide a warning message should a
user try to issue an UPGRADE command on a release of SMP/E prior to V3R2.
GIMXSID service routine

The GIMXSID service routine is used as part of the ShopzSeries offering. GIMXSID
creates a single data source required by ShopzSeries to place customized software
product and service orders. The data source created by GIMXSID, the software
inventory data, is a composite of three kinds of information as follows:
1. A list of FEATUREs found in the SMPCSI data sets. The Feature List is used by
ShopzSeries to perform product requisite checking and also to prime the order
checklist when ordering a ServerPac.
2. A list of the FMIDs found in the SMPCSI data sets. The FMID List is used by
ShopzSeries to scope service orders to the PTFs applicable solely to the user’s
desired configuration of target and global zones.
3. A bitmap representation of the PTFs found in the specified target zones and
global zones. The PTF Bitmap is used by ShopzSeries (CCSS) to produce service
packages that do not contain PTFs that are already present in the user’s
configuration.
GIMZIP: Archive segmentation

The GIMZIP service routine has been enhanced with a new SEGMENT option to
allow you to specify the maximum size for the network transportable objects
produced by GIMZIP. Very large objects will be divided into archive segments that
are within the specified size. The GIMUNZIP service routine and the RECEIVE
FROMNETWORK and RECEIVE FROMNTS commands will now accept network
packages that contain archive segments as input. For more information about the

GIMZIP and GIMUNZIP service routines, refer to SMP/E Reference. For more
information about the RECEIVE commands, refer to SMP/E Commands.
Also, you can now use the SMPWKDIR DD statement to specify a location for
temporary files produced during GIMZIP processing.
SMP/E releases prior to SMP/E V3R2 cannot process GIMZIP output that contains
segmented archive files. A toleration PTF will issue an error message should a user
try to process GIMZIP output that contains segmented archive files on a release of
SMP/E prior to V3R2.
GIMZIP: User defined subdirectories

Users of GIMZIP may now specify subdirectories in which to store documentation,
samples, readme files, and other files. This is done with the new subdir attribute
of the <FILEDEF> tag of the GIMZIP package control statement. The subdirectory
is created in a UNIX file system, within the parent directory pointed to by the
SMPDIR DD statement in the JCL used to invoke GIMZIP.
Unless the required PTF is installed, SMP/E releases prior to SMP/E V3R2 cannot
process GIMZIP packages that exploit user-defined subdirectories.
Java archive files

Many z/OS software products developed with Java use Java Archive (JAR) files as
the packaging format for Java application files. To better accommodate such
products, SMP/E V3R2 is introducing JAR file update support. For SMP/E, there
will be two forms of JAR files; JAR replacement files and JAR update files. JAR
replacement files are complete replacements for a JAR file and are treated simply
as files in the UNIX file system. SMP/E will copy replacement JAR files into the
UNIX file system, just as is done for hierarchical file system elements. JAR update
files are archive files themselves, but do not contain all of the component files that
make up the complete JAR file. A JAR update file contains only the new and
changed component files. These new and changed component files are archived
into a JAR file. SMP/E uses the JAR command and the contents of a JAR update
file to update an existing JAR file.
For information about the JAR command and JAR files, refer to the “Java
Developer Connection, Using JAR Files: The Basics”, at http://
developer.java.sun.com/developer/Books/JAR/basics/index.html
SMP/E releases prior to SMP/E V3R2 cannot process JAR replacement files or JAR
update files, nor can they process the data entries for these elements. A toleration
PTF will cause SMP/E to issue an error message if it encounters an unsupported
JAR element or data entry.
Smaller SMPLTS data set

The SMPLTS data set is used by SMP/E to save the base version of a product’s
load modules that use callable services. To reduce SMPLTS space requirements,
SMP/E now saves a base version of a load module in the SMPLTS data set only if
it contains both CALLLIBS and XZMOD subentries. If a load module contains
CALLLIBS subentries, but no XZMOD subentries, this load module is not saved in
the SMPLTS.

For more detailed information about these command changes, refer to SMP/E
Commands.
The SMPLTS data set used by SMP/E V3R2 may not contain the base version of
load modules with CALLLIBS subentries. Because of this, once the SMPLTS has
been modified by SMP/E V3R2, you cannot use certain commands from older
levels of SMP/E that depend on the base version of a load module being in the
SMPLTS data set. Specifically, if any of the following updates were made to a zone
with load modules containing CALLLIBS but no XZMOD subentries, the target
zone is marked:
v CLEANUP command is run against the zone
v GENERATE command is run against the zone
v APPLY or RESTORE command is run and the base version of the load module is
deleted from the SMPLTS
A toleration PTF against older releases of SMP/E allows certain commands to be

processed against a target zone that has been marked to indicate that the SMPLTS
data set can not be used. These commands do not depend on the base version of
load modules existing in the SMPLTS:
v APPLY
v BUILDMCS
v CLEANUP
v DEBUG
v GENERATE
v JCLIN
v LIST
v LOG
v RESETRC
v SET
v UCLIN
v UNLOAD
v ZONECOPY
v ZONEDELETE
v ZONEEDIT
v ZONEEXPORT
v ZONEIMPORT
v ZONEMERGE
The toleration PTF also prohibits the RESTORE or LINK MODULE command from
being run against a target zone that has been marked.
DUMMY data set for SYSDEFSD

The SYSDEFSD DD statement defines the location for the binder to write IMPORT
statements for all exported entries of a DLL load module. Whenever entries are to
be exported, the binder expects the SYSDEFSD data set and, if it is not present,
will issue a warning message to indicate the missing data set. Product developers
who supply DLLs do not always want or need the IMPORT statements associated
with a DLL retained. In these instances, the SYSDEFSD would be better defined as
either a temporary or DUMMY data set.
SMP/E is defining a new ddname called SMPDUMMY, which will always be

allocated as ’DD DUMMY’. Product packagers may now specify the SYSDEFSD
DD statement in the JCLIN input stream as any of the following:
v //SYSDEFSD DD DSN=SMPDUMMY,DISP=xxx

v //SYSDEFSD DD DSN=NULLFILE
v //SYSDEFSD DD DUMMY
In each case, the SIDE DECK LIBRARY subentry of the LMOD entry will be set to
SMPDUMMY. When needed for processing, SMPDUMMY will be dynamically
allocated by SMP/E as a DUMMY data set.
process SYSMOD input that uses the SYSDEFSD DUMMY enhancement, nor can
they process the data entries for these elements.
For more information about SYSDEFSD DUMMY, refer to SMP/E Reference.
SMP/E dialog customization

A new option, Option 0, has been added to the SMP/E Primary Option Menu
GIM@PRIM to implement the current SMP/E customization options. This new
option allows you to enter or change the values for the customization options that
were previously found in panel GIM@UPRM. When you select option 0 from the
GIM@PRIM panel, the panel GIM@PARM will appear. The options you then
specify are saved permanently in the ISPF profile pool for later use by other
SMP/E dialog processes.
Migration tasks
All dialog customization formerly specified on panel GIM@UPRM must now be
specified using Option 0 on the SMP/E Primary Option Menu. When you move to
a new release of SMP/E and continue to use the same ISPF profile data set, no
migration actions are required to use the options previously entered and saved.
For more information about SMP/E dialog customization, refer to the tutorial
panels that accompany the SMP/E dialogs.
GIMUTTBL removal
Module GIMUTTBL and load module GIMUTTBL are no longer supplied as part
of SMP/E. Macro GIMDFUT, which was used to replace the IBM-supplied copy of
GIMUTTBL, is also no longer supplied. GIMUTTBL was formerly used to specify
which utility programs SMP/E can call.
Migration tasks
You can specify which utility programs SMP/E can call by using the PROGRAM
class of the z/OS Security Server (RACF). Refer to z/OS Security Server RACF
Security Administrator’s Guide, SA22-7683 for more information about how to use
this function.
SMP/E V3R1 overview

v Description

Defining exit routines using SMPPARM member GIMEXITS
SMP/E V3R1 allows you to define exit routines that are to be given control during
SMP/E processing by specifying new GIMEXITS control statements in SMPPARM
member GIMEXITS. This replaces the previous method of updating module
GIMMPUXD. Putting the exit routine information in an SMPPARM member means
that the information is persistent and that you do not need to update module
GIMMPUXD every time a new release of SMP/E is installed.
Migration tasks
If you have existing exit routines defined in GIMMPUXD or wish to create new
exit routines, you must define them in a GIMEXITS member. Any exit routines
defined in GIMMPUXD will be ignored by SMP/E
For more detailed information about this change, refer to SMP/E Reference.
Dynamic allocation using SMPPARM member GIMDDALC

SMP/E V3R1 allows you to define data sets to be dynamically allocated by SMP/E
by specifying new GIMDDALC control statements in SMPPARM member
GIMDDALC, as well as by using DDDEF entries. This replaces the previous
method of updating module GIMMPDFT. Putting the dynamic allocation
information in an SMPPARM member means that the information is persistent and
that you do not need to update module GIMMPDFT every time a new release of
SMP/E is installed.
Migration tasks
If you have used GIMMPDFT to define data sets for dynamic allocation, you must
create new definitions in GIMDDALC. SMP/E will ignore any definitions in
GIMMPDFT.
For more information about this change, refer to SMP/E Reference.
Enhanced link name values

SMP/E V3R1 has increased the maximum length allowed for a hierarchical file
system element LINK value from 64 characters to 1023 characters. SMP/E has also
increased the maximum length allowed for the alias values on ALIAS link-edit
control statements from 64 characters to 1023 characters.
process a SYSMOD containing a hierarchical file system element MCS that specifies
a LINK value longer than 64 characters, nor can they process a hierarchical file
system element entry containing a LINK subentry, or an LMOD entry containing
an LMODALIAS subentry, created by SMP/E V3R1, or higher. If installed, the PTF
will update the SMP/E Query dialogs and the GIMAPI to retrieve and display
information about LINK or LMODALIAS subentries created by SMP/E V3R1, or
higher. For all other SMP/E processing, the PTF will cause SMP/E to issue an
error message if it encounters an unsupported LINK value or LINK or
LMODALIAS subentry.
Migration tasks
An application program that uses the SMP/E Alias Record Type 0 (A0) library
change record may need to be updated to handle alias and link values of up to
1023 characters to avoid truncating data. For more information about this change,
refer to the chapter on Library Change File Records in SMP/E Reference.

An application program that uses the GIMAPI to query the LINK subentry of a
hierarchical file system element entry or the LMODALIAS subentry of an LMOD
entry may need to be updated to allow for the longer length of these subentries.
For more information about this change, refer to the description of these subentries
in the GIMAPI chapter of SMP/E Reference.
Removal of function to create backup IEANUC01 load

modules
The ability to save the target system’s nucleus load module (IEANUC01) during
APPLY, LINK, and RESTORE command processing has been removed from z/OS
V1R2 SMP/E. The NUCID operand of the APPLY command and the NUCID
subentry are no longer supported and will be ignored by SMP/E if specified.
Migration tasks
An application program that uses the GIMAPI to query the NUCID subentry of an
OPTIONS entry must be updated because this subentry no longer exists. For more
information about the GIMAPI, refer to the GIMAPI chapter of SMP/E Reference.
Conditional JCLIN processing

SMP/E V3R1 allows a packager to use special JCL comments in the JCLIN input to
cause SMP/E to skip over parts of the JCLIN input based on the installation
environment. The parts of the JCLIN input that are skipped are not processed by
the JCLIN command and do not contribute to the structure information derived
from JCLIN processing.
For more information about this change, refer to the section on ″Conditional JCLIN
Comment Statements″ in SMP/E Commands, SA22-7771.
process JCLIN input that contains the special JCL comments used to skip over
parts of the JCLIN input. If installed, the PTF will cause SMP/E to issue an error
message if the unsupported JCL comments are encountered.
Network delivery of SMP/E input

SMP/E V3R1 can receive input from a network server, in addition to tape and
DASD. This enables the delivery of SMP/E-installable products and service over
the internet or an intranet. By installing software directly from a network source,
SMP/E enables a more seamless integration of electronic software delivery and
installation. This reduces the tasks and time required to install software delivered
electronically.
SMP/E also provides the GIMZIP and GIMUNZIP service routines to construct,
and then later unwrap, network transportable packages of software. This allows
you to create your own packages of SMP/E installable software, and then
distribute them within your own enterprise, or to other enterprises. Specifically, the
GIMZIP service routine will accept partitioned or sequential data sets as input and
will create a network transportable package as output. For more information on
the GIMZIP and GIMUNZIP service routines, see SMP/E Reference, SA22-7772.
Once a package is made accessible on an FTP server, you can use the SMP/E
RECEIVE command to transfer the package through a TCP/IP network directly
into an SMP/E environment. The RECEIVE command has been extended with new
DELETEPKG, FROMNETWORK, and FROMNTS operands to process these

network transportable packages. For more information on the RECEIVE command
changes, see SMP/E Commands, SA22-7771.
New CLIENT and SERVER data sets and SMPDIR and SMPNTS directories have
been created to support this new processing. For more information on CLIENT,
SERVER, SMPDIR, and SMPNTS, see SMP/E Reference, SA22-7772.
Network delivery of SMP/E input requires a new packaging format for that input
and new operands on the RECEIVE command. SMP/E releases prior to z/OS
SMP/E cannot process this new packaging format and do not recognize the new
RECEIVE command operands.
AMODE=64 and COMPAT=PM4 link edit parameters

SMP/E V3R1 recognizes and saves the AMODE=64 and COMPAT=PM4 link edit
parameters.
The AMODE option assigns the addressing mode for all of the entry points into a
program module (the main entry point, its true aliases, and all of the alternate
entry points). AMODE=64 instructs the binder to create AMODE 31/64 executables
with 8-byte adcons.
The COMPAT option identifies the compatibility level of the binder.
Selected SMP/E data sets may now reside in a UNIX file

system
SMP/E V3R1 allows the following data sets to reside in a UNIX file system:
v SMPCNTL
v SMPDEBUG
v SMPHOLD
v SMPJCLIN
v SMPLIST
v SMPOUT
v SMPPTFIN
v SMPPUNCH
v SMPRPT
For more information about this change, refer to the description of each of these
data sets in SMP/E Reference.
HFS data set identification

SMP/E V3R1 has enhanced the SMP/E File Allocation Report and SMP/E library
change file records to identify the physical HFS data sets where files in a UNIX file
system reside.
SMPPTS spill data sets

SMP/E RECEIVE processing can use SMPPTS spill data sets, if defined, to store
SYSMODs when the primary SMPPTS data set is full. Up to 99 spill data sets,
named SMPPTS1 through SMPPTS99, can be defined with DD statements or
DDDEFs. By eliminating the tasks involved when recovering from an overflowing

SMPPTS data set, the use of SMPPTS spill data sets can reduce the amount of
manual intervention and data set management required to install software service.
HOLDDATA summary reports

SMP/E V3R1 now provides additional HOLDDATA reports for APPLY and
ACCEPT processing. The new reports provide you with ++HOLD information in
the context of the APPLY or ACCEPT processing output. This frees you from
having to manually collect this information, thus saving you significant research
time.
SMP/E load modules and service routines moved to

SYS1.MIGLIB
The SMP/E load modules, service routines, and other SMP/E components that
formerly resided in the SYS1.LINKLIB library have been moved to SYS1.MIGLIB.
Putting SMP/E into SYS1.MIGLIB enables a driving system to STEPLIB to SMP/E
while ensuring that the STEPLIB does not expose the driving system to other
executables that are not at the correct level for the driving system.
GIMXTRX service routine

GIMXTRX is intended for use as part of an offering called ShopzSeries. It provides
two basic functions:
1. Generate a list of target zone names associated with a given GLOBAL zone
SMPCSI data set name.
2. Generate a BITMAP representation of FUNCTION and PTF SYSMODs found in
a given list of target zone names associated with a given GLOBAL zone
SMPCSI data set name.
OS/390 version 2 release 7 SMP/E overview

introduced for OS/390 Version 2 Release 7 (V2R7). (This level of SMP/E was also
supplied with OS/390 Version 2, Releases 8, 9, and 10.) The information about each
item includes:
v Description
SMP/E planning and migration assistant

OS/390 V2R7 (or later) SMP/E provided a Planning and Migration Assistant to
assist users in managing their existing system and planning to migrate to a new
OS/390 system.
Data element reformatting

SMP/E can install data elements during APPLY, ACCEPT, RESTORE, and
GENERATE into the output data sets based on their physical attributes.
Releases of SMP/E prior to OS/390 Version 2 Release 7 cannot process the
DEIINST and HFSINST jobs created by the GENERATE command of SMP/E V3R1

or z/OS SMP/E. Also, an HFSINST job created by the GENERATE command from
a release of SMP/E prior to OS/390 Release 7 cannot be processed by z/OS
SMP/E or by OS/390 V2R7 (or later) SMP/E.
Rerun the GENERATE job using the SMP/E release that will be used to process
the resulting JCL.
Description for a SYSMOD

OS/390 V2R7 (or later) SMP/E enabled product developers and packagers to
include additional descriptive information in a SYSMOD header MCS (that is, in a
++APAR, ++FUNCTION, ++PTF, or ++USERMOD statement).
Improved protection for UNIX file system files

Before manipulating files in a UNIX file system, SMP/E temporarily switches the
SMP/E user to superuser authority (UID=0) when manipulating files in the UNIX
file system and restores the user to the previous level of authority when the
SMP/E updates are done. This means that SMP/E users do not need to have
UID=0 (superuser) authority all the time, which reduces the chance of such users
accidentally erasing or damaging files in a UNIX file system while performing
non-SMP/E work.
The SMP/E user must be defined to the security class BPX.SUPERUSER for this
process to work properly.
Pre-built load module support

The ++PROGRAM MCS can be used to add, replace, or delete pre-built load
modules and program objects in PDS and PDSE data sets as complete entities in
functions and PTFs.
Product data
The ++PRODUCT and ++FEATURE MCS can be used to add, replace, or delete
additional product and feature data.
Sequential data set support

SMP/E can now install a data element into a sequential data set.
Releases of SMP/E prior to OS/390 Version 2 Release 7 cannot process the
DEIINST and HFSINST jobs created by the GENERATE command of SMP/E V3R1
or z/OS SMP/E. Also, an HFSINST job created by the GENERATE command from
a release of SMP/E prior to OS/390 Version 2 Release 7 cannot be processed by
z/OS SMP/E or by OS/390 V2R7 (or later) SMP/E.
Rerun the GENERATE job using the SMP/E release that will be used to process
the resulting JCL.
Shell script support

OS/390 V2R7 (or later) SMP/E enabled the execution of UNIX shell scripts during
SMP/E installation of code into the OS/390 UNIX Services environment. SMP/E
provides a generic interface to enable a packager to deliver a shell script that can
be executed during SMP/E installation, thus reducing the pre-install and
post-install requirements of OS/390 UNIX Services application programs.

Symbolic link support
Symbolic links can be specified on hierarchical file system MCS.

supplied with OS/390 Version 2 Release 6.) The information about each item
includes:
v Description
CBIPO dialogs
The CBIPO installation dialogs formerly included with SMP/E were removed from
SMP/E in OS/390 V2R5 SMP/E. Customers wishing to install a CBIPO on an
OS/390 system may still do so using bootstrap dialogs provided with the CBIPO.
Client code installation

OS/390 V2R5 (or later) SMP/E provides facilities to simplify the installation of
cooperative or client/server products (such as OS/2). This is done with a common
SMP/E packaging structure, a common S/390 server repository for client
components, and a server repository accessible from any client platform. These
facilities allow, for example, storing the client parts in a UNIX file system and
thereby allowing them to be packaged and installed along with the host parts,
rather than separately.
Enhanced hierarchical file system element MCS cannot be used by SMP/E releases
prior to OS/390 V2R5 SMP/E, unless the required PTF is installed.
Global zone merge

OS/390 V2R5 (or later) SMP/E provides a method to merge information from one
global zone into another global zone, which customers can use to reduce the
number of global zones that they must manage. The merged information includes:
v SYSMOD and HOLDDATA entries,
v SYSMOD members in the SMPPTS data set,
v OPTIONS, UTILITY, DDDEF, ZONESET, and FMIDSET entries
v Global zone entry information, such as zone indexes, FMID list, and SRELs.
This facility is useful to customers who use ServerPac.
Library change interface

OS/390 V2R5 (or later) SMP/E provides a programming interface that can be used
to obtain a synopsis of SMP/E APPLY and RESTORE processing at the library or
member level. Customers can use this interface to propagate the libraries and
members modified by SMP/E APPLY and RESTORE processing to other systems
requiring the changes, thereby facilitating the integration of SMP/E-managed
information in multisystem environments.

OPTIONS entries that contain the CHANGEFILE subentry cannot be used by
SMP/E releases prior to OS/390 V2R5 SMP/E, unless the required PTF is installed.
Improved load module build processing

OS/390 V2R5 (or later) SMP/E will not build a load module if SMP/E cannot
include all of the load module’s component modules that have been installed or
are being installed. If such a load module can not be completely built, SMP/E
terminates APPLY processing for all affected SYSMODs. In addition, OS/390 V2R5
(or later) SMP/E reduces the likelihood of termination owing to incomplete load
modules by expanding its search for the component modules to include copies of
modules from within previously installed SYSMODs in the SMPPTS data set.
Load module return code

OS/390 V2R5 (or later) SMP/E allows product packagers to provide information in
the JCLIN to identify the highest return code allowable for each load module. IBM
will provide toleration PTFs for this function for prior releases of OS/390 and
currently supported releases of SMP/E.
LMOD entries that contain the RETURN CODE subentry cannot be used by
SMP/E releases prior to OS/390 V2R5 SMP/E, unless the required PTF is installed.
Performance improvements
OS/390 V2R5 (or later) SMP/E provides for multitasking of link-edit operations
during APPLY, ACCEPT, and RESTORE processing.
PTF compaction in SMPPTS data set

OS/390 V2R5 (or later) SMP/E compacts the SYSMOD (PTF) data within the
SMPPTS data set to reduce its size. This ability to compact PTF data prior to
release by IBM also allows IBM to shrink the size of the OS/390 service stream,
thus reducing the amount of physical media (tapes) required to distribute service,
as well as shrink the size of service distributed through various electronic
mediums.
v Compacted SMPPTS data created by OS/390 V2R5 (or later) SMP/E cannot be
used by releases prior to OS/390 V2R5 SMP/E, unless the required PTF is
installed.
v OPTIONS entries that contain the COMPACT subentry cannot be used by
SMP/E releases prior to OS/390 V2R5 SMP/E, unless the required PTF is
installed.
Enhanced RECEIVE command processing

OS/390 V2R5 (or later) SMP/E enables users to prevent the RECEIVE command
from processing SYSMODs that are already applied or accepted. Users can specify
this with the OPTIONS entry, on the RECEIVE command, or both. This
enhancement reduces the need for the user to manually manage the SMPPTS with
REJECT commands.
OPTIONS entries that contain the RECZGRP and RECEXZGRP subentries cannot
be used by SMP/E releases prior to OS/390 V2R5 SMP/E.

Reduced SMP/E message output
OS/390 V2R5 (or later) SMP/E reduced the number of messages issued during
APPLY, ACCEPT, and RESTORE processing for easier identification of potential
problems. Also, users may specify that messages issued to SMPOUT be formatted
to an 80 character width, instead of the previous 120 character width, to make the
messages easier to view when displayed on a terminal screen.
OPTIONS entries that contain the MSGWIDTH and MSGFILTER subentries cannot
be used by SMP/E releases prior to OS/390 V2R5 SMP/E.
GIMAPI: All entries and subentries support

For OS/390 V2R5 (or later) SMP/E, an application program using the GIMAPI
QUERY command may specify an asterisk (*) on entry and subentry parameters to
retrieve the Consolidated Software Inventory (CSI) data for all entry types, all
subentries, or both.
GIMAPI: Version support

OS/390 V2R5 (or later) SMP/E can supply to an application program the version
of GIMAPI being executed to retrieve information from the CSI. This allows the
application program to determine whether information stored in the CSI is
supported with the level of the QUERY program that is being executed.
Application programs cannot use the VERSION command of the GIMAPI
programming interface on releases of SMP/E prior to OS/390 V2R5 (or later)
SMP/E, unless the required PTF is installed.

supplied with OS/390 Version 2 Release 4.) The information about each item
includes:
v Description
API for user access to the CSI

A programming interface is provided for read only access to SMP/E’s consolidated
software inventory (CSI) data. The data in CSI can be used to further automate
systems management tasks.
A program called GIMAPI is used to invoke the API. The function can be called
from different languages. Examples are provided for C/370 and PL/I.
The following commands are used with the GIMAPI call:

QUERY Request data from the SMP/E CSI and return it to the calling
program.
FREE Free storage allocated by invocations of the QUERY command.

Enhanced cross-zone requisite checking
Cross-zone requisite checking is enhanced. Immediate feedback from the APPLY,
ACCEPT, and RESTORE commands assists you in verifying that cross-zone
requisites are installed and satisfied.
Optional parameters with these commands provide you the flexibility to:
v Override SMP/E’s default method for determining which zones are checked for
cross-zone requisites
v Install unsatisfied cross-zone requisites into the set-to zone
v Lessen the severity of a missing cross-zone requisite to a warning versus a
terminating error
Releases of SMP/E prior to OS/390 V1R3 SMP/E cannot perform the cross-zone
requisite checking requested by the XZREQCHK(YES) subentry of a ZONESET
entry and will ignore the request.
Enhanced exception SYSMOD report

The enhanced Exception SYSMOD Report, available as a small programming
enhancement (SPE) for OS/390 V1R3 SMP/E, includes IBM OS/390 Enhanced
HOLDDATA that is provided in ++HOLD statements. The report is reformatted so
that it is ordered by FMID within each requested zone. (Previously, the report was
ordered by SYSMOD within each zone.) A summary section is placed at the end of
the report. The enhanced REPORT ERRSYSMODS command continues to work
with legacy HOLDDATA.
The REPORT ERRSYSMODS command has also been enhanced by this SPE to
handle held SYSMODs. Previously, a held, resolving SYSMOD was placed in the
SMPPUNCH output, but was commented out. The customer had to rerun REPORT
ERRSYSMODS command against the GLOBAL zone to determine if the held,
resolving SYSMOD had an available resolving SYSMOD. REPORT ERRSYSMODS
does this research for the customer and produces one SMPPUNCH output.
v The REPORT ERRSYSMODS command in SMP/E releases prior to OS/390 V1R3
cannot display IBM z/OS Enhanced HOLDDATA as intended. OS/390 V1R3 and
V2R4 require the installation of the appropriate SPE.
v The format of the HOLDDATA provided by the SMARTMVS service in Europe
or the Electronic HOLDDATA service in the U.S. is not compatible with z/OS
Enhanced HOLDDATA and does not take advantage of the enhanced REPORT
ERRSYSMODS command. Customers who currently use these services, and who
wish to make full use of the REPORT ERRSYSMODS command, must refresh
their CSI’s HOLDDATA with z/OS Enhanced HOLDDATA.
Enhanced ++IF FMID processing

z/OS SMP/E allows the ++IF MCS FMID operand to specify the same value as the
value of the FMID operand of the previous ++VER MCS (if the SYSMOD is not a
base function) or the name of the SYSMOD (if the SYSMOD is a base function).
Enhanced internal HOLD SYS processing

Analysis of internal HOLD information when one SYSMOD supersedes another is
simplified. When a SYSMOD has ++HOLD information and it is superseded by
another SYSMOD, the ++HOLD can be brought forward unchanged. The SYSMOD

ID on the ++HOLD need not change to that of the superseding SYSMOD. Even if
the SYSMOD ID on the ++HOLD is not the same as the containing SYSMOD, the
++HOLD is effective only against the SYSMOD that contains it. If the SYSMOD ID
on the ++HOLD is not the same as the containing SYSMOD, SMP/E can determine
if internal HOLDs are satisfied during APPLY and ACCEPT processing and thereby
eliminate manual analysis.
SYSMODs that contain a ++HOLD MCS that specifies a SYSMOD ID that is
superseded on a preceding ++VER MCS cannot be processed by previous releases
of SMP/E during RECEIVE processing, unless the appropriate PTF is installed for
the prior release.
Enhanced ZONEEDIT command

The ZONEEDIT command is enhanced to provide a simplified method of changing
path names. A PATH subentry is included on the unconditional CHANGE
statement of the ZONEEDIT DDDEF command.
An example of when you might want to use the PATH subentry on the CHANGE
statement is to modify path names of DDDEFs during the z/OS UNIX service
process.
Enhancements to the binder utility in DFSMS/MVS

SMP/E can use enhancements to the binder utility in DFSMS/MVS. The
enhancements to the binder include elimination of the LE/370 prelinker utility, and
building dynamic load library (DLL) program objects. SMP/E’s support includes:
v New link-edit parameters are recognized on the LEPARM operand of the
++MOD MCS and in JCLIN used to define a load module. The new parameters
are ALIASES, DYNAM, FILL, HOBSET, REUS(NONE|REFR|RENT|SERIAL),
RMODE=SPLIT, and UPCASE(YES|NO). All of these new parameters can be
specified in JCLIN and all except ALIASES and DYNAM can be specified on the
LEPARM operand.
v SMP/E supports the binder in dynamically building a definition side deck file
for DLL program objects when those program objects are installed. The library to
contain the definition side deck file is identified by a new side deck library
(SIDEDECKLIB) subentry in the LMOD entry.
v Load modules that use DLLs can reference the definition side deck files
associated with the DLLs. This is done by including the definition side deck files
during a link-edit operation. The LMOD entry will contain a new utility input
(UTIN) subentry list to record definition side deck files to be included during a
link-edit operation.
Releases of SMP/E prior to OS/390 V1R3 SMP/E cannot:
v Correctly install products and service that were developed for installation using
the INCLUDE statements in JCLIN that identify UTILITY INPUT for a load
module, the SYSDEFSD DD statements in JCLIN that identify the definition side
deck library for a load module, and the FILL, HOBSET, RMODE=SPLIT, and
EXITS link edit attributes on the LEPARM operand on the ++MOD MCS and in
JCLIN.
v Use target and distribution zones containing LMOD or MOD entries updated by
OS/390 V1R3 (or later) with FILL, HOBSET, RMODE=SPLIT, EXITS, ALIASES,

or DYNAM link edit parameters in the MOD and LMOD entries, the UTILITY
INPUT subentry list, or the SIDE DECK LIBRARY subentry in the LMOD
entries.
System/390 service update facility

The System/390 Service Update Facility (SUF) available as a small programming
enhancement (SPE) for OS/390 V1R3 SMP/E, provides, along with other
System/390 products, provides a common tool across multiple platforms to help
customers to maintain their systems with System/390 service facilities.

introduced for OS/390 Version 1 Release 2 (V1R2). The information about each
item includes:
v Description
BLOCKSIZE=8800 for SMP/E data sets

For the RELFILE data sets, target libraries, and distribution libraries containing
z/OS SMP/E, data sets that are allocated with RECFM=FB and LRECL=80 are
allocated with BLKSIZE=8800.
BUILDMCS command
The BUILDMCS command provides a process for copying products from one pair
of target and distribution zones and libraries, to another pair of target and
distribution zones and libraries. This command generates the MCS and JCLIN
required to reinstall the specified FMIDs.
SYSMOD input (modification control statements) created by the BUILDMCS
command cannot be processed by SMP/E releases prior to OS/390 V1R2 SMP/E,
unless the required PTF is installed.
Bypassing system holds for specific SYSMODs

For APPLY and ACCEPT processing, you can bypass a particular system hold for
specific SYSMODs, instead of for all SYSMODs held for that reason ID. For
example, a number of SYSMODs might be held because they require you to take
some required action before installing them. If you have completed the required
action for some (but not all) of the held SYSMODs, you can request SMP/E to
bypass that hold reason ID only for the SYSMODs you specify. All other SYSMODs
affected by that reason ID remain held.
FMIDSET selection
SMP/E provides additional granularity of FMIDSET specification on the SELECT
operand of the APPLY, ACCEPT, RESTORE, and RECEIVE commands to allow you
to install sets of FMIDs.

Receiving relative file data sets created from PDSEs
When allocating a new SMPTLIB data set during RECEIVE processing, SMP/E
checks the format of the associated relative file (RELFILE) data set, then uses the
appropriate data set type (LIBRARY or PDS) for the SMPTLIB data set. Here are
some benefits of this change:
v When packaging SYSMODs, you can ship program objects in RELFILEs, because
SMP/E can load RELFILEs that were created from PDSEs into SMPTLIB data
sets that are PDSEs.
v When receiving SYSMODs, you do not have to preallocate SMPTLIB data sets
with the appropriate data set type, because SMP/E can allocate the SMPTLIB
data set as PDS or LIBRARY, based on the format of the corresponding RELFILE
data set.
SMP/E dialogs: FIND command

You can use the FIND primary command in the SMP/E dialogs. The FIND
command makes it easier for you to quickly locate a specified character string in
the table display section of panels in the following dialogs:
v SYSMOD Management
v Query
v Receive
Panels that allow the FIND command state that you can use the command. The
help panels for these dialog panels explain how to use the FIND command.
SMP/E GIMOPCDE member moved from PARMLIB

The GIMOPCDE member, which SMP/E optionally uses to determine valid
OPCODES during the scanning of JCLIN, has been removed from PARMLIB.
Instead, a ready-to-use default set of OPCODE definitions is contained within
SMP/E.
You may optionally provide an SMPPARM data set, which may contain your own
OPCODE member to override the defaults supplied with SMP/E. The
user-provided OPCODE member is a text member that you store in a
user-allocated PDS named SMPPARM. You are not required to allocate the
SMPPARM data set, unless you want to supply your own OPCODE member. If
you provide an OPCODE member, it is used instead of SMP/E’s default set.
SMP/E also provides a sample text member, named GIMOPCDE, that you can use
as a starting point for creating your own OPCODE member.
Summary of interface changes

This section summarizes the new and changed interface components of z/OS
SMP/E.
Commands
Table 16 on page 214 lists the changes to the SMP/E commands for this release.
See SMP/E Commands, SA22-7771 for more detailed information about these
commands.

Table 16. Summary of changed commands
Command name Release Description Related support
APPLY SMP/E V3R1 Changed: NUCID operand removed “Removal of function to
create backup IEANUC01
load modules” on page 203
LINK LMODS SMP/E V3R2 New: for load module management “Smaller SMPLTS data set”
on page 199
RECEIVE SMP/E V3R1 Changed: New operands “Network delivery of
SMP/E input” on page 203
REJECT SMP/E V3R3 Changed: New CHECK operand “REJECT CHECK
command” on page 197
REPORT CALLLIBS SMP/E V3R2 Removed “Smaller SMPLTS data set”
on page 199
UPGRADE SMP/E V3R2 New: for change control “UPGRADE command” on
page 198
Data sets and files

Table 17 lists the changes to the SMP/E data sets and files for this release. See
SMP/E Reference, SA22-7772 for more detailed information about these data sets
and files.
Table 17. Summary of changed data sets
Data set Release Description Related support
SMPCLNT SMP/E V3R3 New: for GIMGTPKG service routine “GIMGTPKG service
processing routine” on page 196
SMPSRVR SMP/E V3R3 New: for GIMGTPKG service routine “GIMGTPKG service
processing routine” on page 196
SMPWKDIR SMP/E V3R2 New: for GIMZIP, GIMUNZIP, RECEIVE “GIMZIP: Archive
FROMNETWORK, and RECEIVE segmentation” on page
FROMNTS processing 198
SYSDEFSD SMP/E V3R2 Changed: to allocate a DUMMY data set “DUMMY data set for
SYSDEFSD” on page
200
CLIENT SMP/E V3R1 New: for RECEIVE FROMNETWORK “Network delivery of
processing SMP/E input” on page
203
SERVER SMP/E V3R1 New: for RECEIVE FROMNETWORK “Network delivery of
203
SMPDIR SMP/E V3R1 New: for GIMZIP and GIMUNZIP “Network delivery of
203
SMPNTS SMP/E V3R1 New: for RECEIVE FROMNETWORK “Network delivery of
and RECEIVE FROMNTS processing SMP/E input” on page
203
SMPPTS SMP/E V3R1 Changed: Spill data sets may be defined “SMPPTS spill data
sets” on page 204
various SMP/E V3R1 Changed: Selected data sets may reside “Selected SMP/E data
in a UNIX file system sets may now reside in
a UNIX file system” on
page 204

Exits
Although the methods by which SMP/E exits are invoked has changed (see the
entry on GIMEXITS in Table 23 on page 218), no changes have been made to the
SMP/E exits themselves. For more information on SMP/E exits, see SMP/E
Reference.
Macros
Table 18 lists the changes to the SMP/E macros for this release.
Table 18. Summary of Changed Macros
Macro Release Description Related support
GIMDFUT SMP/E V3R2 Deleted “SMP/E dialog
customization” on page 201
GIMMPUXD SMP/E V3R1 Deleted “Defining exit routines
using SMPPARM member
GIMEXITS” on page 202
Messages
This section describes the new and changed SMP/E messages. To determine if
updates are required, compare the message identifiers and the corresponding
message text with any automated operation procedures your installation uses.
See SMP/E Messages, Codes, and Diagnosis, GA22-7770 for detailed information
about these SMP/E messages. For information about other z/OS message changes
that may affect your installation, refer to z/OS Summary of Message Changes.
Panels
Table 19 lists the new and changed SMP/E panels. Some panels may be updated
by more than one release enhancement.
Table 19. Summary of new and changed panels
Panel name Release Description Related support
GIM@PARM SMP/E V3R2 New “SMP/E dialog
customization” on page
201
GIM@PRIM SMP/E V3R2 Updated “SMP/E dialog
201
GIM@UPRM SMP/E V3R2 Deleted “SMP/E dialog
201
GIMQIT28 SMP/E V3R1 Updated “Enhanced link name
values” on page 202
GIMCGRDI SMP/E V3R1 Updated “Network delivery of
SMP/E input” on page
203
GIMCGRVE SMP/E V3R1 Updated “Network delivery of
SMP/E input” on page
203

Table 19. Summary of new and changed panels (continued)
GIMCGAPA SMP/E V3R1 Updated “Removal of function to
load modules” on page
203
GIMDFOH SMP/E V3R1 Updated “Removal of function to
203
GIMHCAP SMP/E V3R1 Updated “Removal of function to
203
GIMHCNU SMP/E V3R1 Updated “Removal of function to
203
GIMHIPB0 SMP/E V3R1 Updated “Removal of function to
203
GIMHIPD1 SMP/E V3R1 Updated “Removal of function to
203
GIMHIPN SMP/E V3R1 Updated “Removal of function to
203
GIMHIP0P SMP/E V3R1 Updated “Removal of function to
203
GIMHOH0 SMP/E V3R1 Updated “Removal of function to
203
GIMHOO0 SMP/E V3R1 Updated “Removal of function to
203
GIMHQ011 SMP/E V3R1 Updated “Removal of function to
203
GIMHXA3B SMP/E V3R1 Updated “Removal of function to
203
GIMISAPB SMP/E V3R1 Updated “Removal of function to
203

Table 19. Summary of new and changed panels (continued)
GIMISAPD SMP/E V3R1 Updated “Removal of function to
203
GIMQIT11 SMP/E V3R1 Updated “Removal of function to
203
GIMQIX10 SMP/E V3R1 Updated “Removal of function to
203
Programming interfaces
Table 20 identifies changes to the SMP/E programming interfaces.
Table 20. Summary of SMP/E programming interfaces
Interface Release Description Related support
GIMMPDFT SMP/E V3R1 Deleted “Dynamic allocation using
SMPPARM member
GIMDDALC” on page 202
GIMMPUXD SMP/E V3R1 Deleted “Defining exit routines
using SMPPARM member
GIMUTTBL SMP/E V3R2 Deleted “GIMUTTBL removal” on
page 201
GIMAPI SMP/E V3R1 Changed: length of the LINK and “Enhanced link name
LMODALIAS subentries values” on page 202
GIMAPI SMP/E V3R1 Changed: NUCID subentry deleted “Removal of function to
203
Library Change File SMP/E V3R1 Changed: length of alias and link values in “Enhanced link name
Records A0 record values” on page 202
Link Edit Parameters SMP/E V3R1 Changed: additional values recognized for “AMODE=64 and
AMODE and COMPAT COMPAT=PM4 link edit
parameters” on page 204
Service routines
Table 21 identifies changes to the SMP/E service routines.
Table 21. Summary of SMP/E service routines
GIMGTPKG SMP/E V3R3 New: service routine to get GIMZIP “Network delivery of
packages SMP/E input” on page
203
GIMXSID SMP/E V3R2 New: service routine for use with “GIMXSID service
ShopzSeries routine” on page 198

Table 21. Summary of SMP/E service routines (continued)
GIMZIP SMP/E V3R2 New: subdir attribute on <FILEDEF> tag to “GIMZIP: User defined
create user-defined subdirectories subdirectories” on page
199
GIMZIP SMP/E V3R2 New: ARCHSEG tag to define an archive “GIMZIP: Archive
segment segmentation” on page
198
GIMZIP SMP/E V3R2 New: SEGMENT option to define the size of “GIMZIP: Archive
archive segments segmentation” on page
198
GIMUNZIP SMP/E V3R1 New: service routine to unzip network “Network delivery of
203
GIMXTRX SMP/E V3R1 New: service routine for use with “GIMXTRX service
ShopzSeries routine” on page 205
GIMZIP SMP/E V3R1 New: service routine to create network “Network delivery of
203
SMPPARM members
Table 22 identifies changes to the SMPPARM members.
Table 22. Summary of SMP/E changes to SYS1.SAMPLIB
Member name Release Description Related support
GIMEXITS SMP/E V3R1 New: Identifies exit routines “SMP/E dialog
customization” on page 201
SYS1.SAMPLIB members
Table 23 identifies changes to the SMP/E members of SYS1.SAMPLIB.
Table 23. Summary of SMP/E changes to SYS1.SAMPLIB
GIMASAMP OS/390 V1R3 Contains sample GIMAPI assembler “API for user access to the
application program CSI” on page 209
GIMCRSAM OS/390 V1R3 Contains sample C/370 ″clean-up FMIDs″
program using GIMAPI
GIMCSAMP OS/390 V1R3 Contains sample GIMAPI C/390 application “API for user access to the
program CSI” on page 209
GIMDDALC SMP/E V3R1 Contains sample statements for data set “Dynamic allocation using
allocation SMPPARM member
GIMDDALC” on page 202
GIMEXITS SMP/E V3R1 Contains sample statements for defining exit “Defining exit routines
routines using SMPPARM member
GIMOPCDE OS/390 V1R2 Contains sample ODCDE statements “SMP/E GIMOPCDE
member moved from
PARMLIB” on page 213
GIMPRSAM OS/390 V1R3 Contains sample PL/I ″clean-up FMIDs″
program using GIMAPI

Table 23. Summary of SMP/E changes to SYS1.SAMPLIB (continued)
GIMPSAMP OS/390 V1R3 Contains sample GIMAPI PL/I application “API for user access to the
program CSI” on page 209
GIMSAMPU pre-OS/390 Contains sample UCLIN statements
GIMSAMPU SMP/E V3R2 Updated: Contains steps to allocate SMPCSI
data sets and SMP/E operational data sets
(such as SMPPTS and SMPLOG) and
UCLIN statements to initialize the newly
allocated SMPCSI data sets

Appendix B. Recommended service upgrade (RSU)
Recommended Service Upgrade (RSU) is a preventive service philosophy that
applies to z/OS. RSU is intended to reduce the volume of PTFs customers must
apply for preventive maintenance and to reduce the chance of encountering a PTF
in error (PE), resulting in a more stable system.
IBM recommends that customers APPLY all RSU PTFs as preventive maintenance
on their z/OS systems. However, customers must make the final decision as to
what maintenance they will install.
The recommended service for the following products is tested in the Consolidated
Service Test (CST) cycle:
v z/OS and OS/390
v CICS Transaction Server for OS/390
v DB2 UDB for OS/390
v IMS
v MQSeries for OS/390
Recommended Service Upgrades, with an SMP/E SOURCEID of RSUyymm, are

available:
v Quarterly, with all PTFs that completed Consolidated Service Test (CST) testing
during the prior quarter, including severity 1 through severity 4 APARs.
v Monthly, with high impact or pervasive (HIPER) PTFs, PTF-in-error (PE) PTFs,
and other recommended PTFs (recommended because of new function,
serviceability, installability, security, or integrity) that have been CST tested.
For information about the latest recommended level of service, see the CST and
RSU web site:
http://www.ibm.com/servers/eserver/zseries/zos/servicetst/
Note that while all CST-tested PTFs become RSU PTFs, not all RSU PTFs are tested
in CST. Only the following are included in CST testing: z/OS, z/OS.e, OS/390,
CICS, DB2, IMS, and MQSeries.
RSU is provided for z/OS through ESO, CBPDO, and ServerPac.

Appendix C. Accessibility
Accessibility features help a user who has a physical disability, such as restricted
mobility or limited vision, to use software products successfully. The major
accessibility features in z/OS enable users to:
v Use assistive technologies such as screen readers and screen magnifier software
v Operate specific or equivalent features using only the keyboard
v Customize display attributes such as color, contrast, and font size
Using assistive technologies

Assistive technology products, such as screen readers, function with the user
interfaces found in z/OS. Consult the assistive technology documentation for
specific information when using such products to access z/OS interfaces.
Keyboard navigation of the user interface

Users can access z/OS user interfaces using TSO/E or ISPF. Refer to z/OS TSO/E
Primer, z/OS TSO/E User’s Guide, and z/OS ISPF User’s Guide Vol I for information
about accessing TSO/E and ISPF interfaces. These guides describe how to use
TSO/E and ISPF, including the use of keyboard shortcuts or function keys (PF
keys). Each guide includes the default settings for the PF keys and explains how to
modify their functions.
z/OS information
z/OS information is accessible using screen readers with the BookServer/Library
Server versions of z/OS books in the Internet library at:
www.ibm.com/servers/eserver/zseries/zos/bkserv/

Notices
This information was developed for products and services offered in the USA.
IBM may not offer the products, services, or features discussed in this document in
other countries. Consult your local IBM representative for information on the
products and services currently available in your area. Any reference to an IBM
product, program, or service is not intended to state or imply that only that IBM
product, program, or service may be used. Any functionally equivalent product,
program, or service that does not infringe any IBM intellectual property right may
be used instead. However, it is the user’s responsibility to evaluate and verify the
operation of any non-IBM product, program, or service.
IBM may have patents or pending patent applications covering subject matter
described in this document. The furnishing of this document does not give you
any license to these patents. You can send license inquiries, in writing, to:
IBM Director of Licensing
IBM Corporation
North Castle Drive
Armonk, NY 10504-1785
USA
For license inquiries regarding double-byte (DBCS) information, contact the IBM
Intellectual Property Department in your country or send inquiries, in writing, to:
IBM World Trade Asia Corporation
Licensing
2-31 Roppongi 3-chome, Minato-ku
Tokyo 106, Japan
The following paragraph does not apply to the United Kingdom or any other
country where such provisions are inconsistent with local law:
INTERNATIONAL BUSINESS MACHINES CORPORATION PROVIDES THIS
PUBLICATION “AS IS” WITHOUT WARRANTY OF ANY KIND, EITHER
EXPRESS OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
WARRANTIES OF NON-INFRINGEMENT, MERCHANTABILITY OR FITNESS
FOR A PARTICULAR PURPOSE. Some states do not allow disclaimer of express or
implied warranties in certain transactions, therefore, this statement may not apply
to you.
This information could include technical inaccuracies or typographical errors.

Changes are periodically made to the information herein; these changes will be
incorporated in new editions of the publication. IBM may make improvements
and/or changes in the product(s) and/or the program(s) described in this
publication at any time without notice.
Any references in this information to non-IBM Web sites are provided for
convenience only and do not in any manner serve as an endorsement of those Web
sites. The materials at those Web sites are not part of the materials for this IBM
product and use of those Web sites is at your own risk.
IBM may use or distribute any of the information you supply in any way it
believes appropriate without incurring any obligation to you.

Licensees of this program who wish to have information about it for the purpose
of enabling: (i) the exchange of information between independently created
programs and other programs (including this one) and (ii) the mutual use of the
information which has been exchanged, should contact:
IBM Corporation
Mail Station P300
2455 South Road
Poughkeepsie, NY 12601-5400
USA
Such information may be available, subject to appropriate terms and conditions,

including in some cases, payment of a fee.
The licensed program described in this information and all licensed material
available for it are provided by IBM under terms of the IBM Customer Agreement,
IBM International Program License Agreement, or any equivalent agreement
between us.
If you are viewing this information softcopy, the photographs and color
illustrations may not appear.
Trademarks
The following terms are trademarks of the IBM Corporation in the United States,
other countries, or both:
BookManager
CICS
CA-ACF2
CA-Top Secret
DATABASE 2
DB2
DFSMS
DFSMS/MVS
DFSMSdfp
DFSMSdss
eTrust
IBM
IBMLink
IMS
OS/2
OS/390
RACF
RETAIN
SP
System/390
SystemPac
VTAM
z/OS
GeoTrust is a registered trademark of GeoTrust, Inc.
Java and all Java-based trademarks are trademarks of Sun Microsystems, Inc. in the
United States, other countries, or both.

Microsoft, Windows, Windows NT, and the Windows logo are trademarks of
Microsoft Corporation in the United States, other countries, or both.
UNIX is a registered trademark of The Open Group in the United States and other
countries.
VeriSign is a registered trademark of VeriSign, Inc.
Other company, product, and service names may be trademarks or service marks
of others.
Notices 227
Glossary
This glossary defines technical terms and Synonym for. This indicates that the term has
abbreviations used in SMP/E documentation. If the same meaning as a preferred term, which
you do not find the term you are looking for, is defined in its proper place in the glossary.
refer to the index of the appropriate SMP/E Synonymous with. This is a backward
manual or view IBM Glossary of Computing Terms, reference from a defined term to all other
located at: http://www.ibm.com/networking/ terms that have the same meaning.
nsg/nsgmain.htm See. This refers you to multiple-word terms
that have the same last word.
Sequence of entries: For clarity and consistency See also. This refers the reader to related
of style, this glossary arranges the entries terms that have a related, but not
alphabetically on a letter-by-letter basis. In other synonymous, meaning.
words, only the letters of the alphabet are used to Deprecated term for or deprecated
determine sequence; special characters and spaces abbreviation for. This indicates that the term
between words are ignored. or abbreviation should not be used. It refers to
a preferred term, which is defined in its
Organization of entries: Each entry consists of a proper place in the glossary.
single-word or multiple-word term or the
abbreviation or acronym for a term, followed by a Selection of Terms: A term is a word or group of
commentary. A commentary includes one or more words to be defined. In this glossary, the singular
items (definitions or references) and is organized form of the noun and the infinitive form of the
as follows: verb are the terms most often selected to be
1. An item number, if the commentary contains defined. If the term may be abbreviated, the
two or more items. abbreviation is given in parentheses immediately
following the term. The abbreviation is also
2. A usage label, indicating the area of
defined in its proper place in the glossary.
application of the term, for example, “In
programming,” or “In SMP/E.” Absence of a
usage label implies that the term is generally A
applicable to SMP/E, to IBM, or to data
processing. ACCEPT. The SMP/E command used to install
SYSMODs in the distribution libraries.
3. A descriptive phrase, stating the basic
meaning of the term. The descriptive phrase is accept. In SMP/E, to install SYSMODs in the
assumed to be preceded by “the term is distribution libraries. This is done with the ACCEPT
defined as ...” The part of speech being command.
defined is indicated by the opening words of
accepted SYSMOD. A SYSMOD that has been
the descriptive phrase: “To ...” indicates a
successfully installed by the SMP/E ACCEPT
verb, and “Pertaining to ...” indicates a command. Accepted SYSMODs do not have the ERROR
modifier. Any other wording indicates a noun flag set and are found as SYSMOD entries in the
or noun phrase. distribution zone.
4. Annotative sentences, providing additional or
Access method services (AMS). The system utility
explanatory information.
program used to support VSAM data sets.
5. References, directing the reader to other
entries or items in the dictionary. AMS. Access method services.
APAR. Authorized program analysis report.

References: The following cross-references are
used in this glossary: APAR fix. A temporary correction of a defect in an
Contrast with. This refers to a term that has IBM system control program or licensed program that
an opposed or substantively different affects a specific user. An APAR fix is usually replaced
meaning. later by a permanent correction called a PTF. APAR
fixes are identified to SMP/E by the ++APAR
statement.

applied SYSMOD • conditional requisites
applied SYSMOD. A SYSMOD that has been base level system. The level of the target system
successfully processed by the SMP/E APPLY command. modules, macros, source, and DLIBs created by system
Applied SYSMODs do not have the ERROR flag set generation, to which function and service modifications
and are found as SYSMOD entries in the target zone. are applicable.
APPLY. The SMP/E command used to install base version of a load module. Some load modules
SYSMODs in the target libraries. include modules both explicitly (through INCLUDE
statements) and implicitly (through a SYSLIB
apply. In SMP/E, to install SYSMODs in the target allocation). The base version of such a load module
libraries. This is done with the APPLY command. includes only the explicitly-defined modules for the
load module. It is maintained by SMP/E in the
archive. An archive is a network transportable file SMPLTS data set. The base version of a load module is
containing two files; a file attribute file (FAF) and the used with the SYSLIB allocation as input to the
data file that the FAF describes. link-edit utility in order to build the load module in its
target libraries.
ASSEM entry. An SMP/E entry containing assembler
statements that can be assembled to create an object binder. A program that processes the output of
module. language translators and compilers into an executable
program (load module). It is part of the DFSMS/MVS
authorized program analysis report (APAR). A report
element of z/OS.
of a problem caused by a suspected defect in a current
unaltered release of a program. The correction is called bypass. In SMP/E, to circumvent errors that would
an APAR fix. otherwise cause SYSMOD processing to fail. This is
done by using the BYPASS operand on an SMP/E
B command.
BACKUP entries. A collection of SMP/E target zone

entries that are copied into the SMPSCDS data set
C
during APPLY processing before they are updated by causer SYSMOD. A SYSMOD identified by root cause
inline JCLIN, a ++MOVE MCS, or a ++RENAME MCS, analysis to be at the base of errors that caused other
or before they are deleted by an element MCS with the SYSMODs to fail. See root cause analysis.
DELETE operand.
BACKUP entries consist of: CBPDO. MVS Custom-Built Product Delivery
Offering.
v A SYSMOD entry indicating what entries have been
added, deleted, or updated CICS. Customer Information Control System.
v ASSEM entries for updated target zone ASSEM
entries CLEANUP. The SMP/E command used to delete
v LMOD entries for updated target zone LMOD entries entries from the SMPMTS, SMPSTS, and SMPSCDS
data sets after a SYSMOD has been accepted into the
v MAC entries for updated or deleted target zone related distribution zone.
MAC entries
v MOD entries for updated or deleted target zone CNTL. See SMPCNTL.
MOD entries
coexisting functions. Functions that meet these
v SRC entries for updated or deleted target zone SRC
requirements: (1) they are for the same system or
entries
subsystem and have the same SREL value, (2) they do
v Data element entries for deleted target zone data not delete or supersede each other and are not negative
element entries prerequisites, and (3) if they are base functions, they
v DLIB entries for updated target zone DLIB entries are for different products. See also conditionally
coexisting functions and unconditionally coexisting
BALR. Branch and link register commands. functions.
base function. A SYSMOD defining elements of the comment statements. Special control statements that
base system or other products that were not previously are coded as JCL comments and which are used to
present in the target libraries. Base functions are convey information to SMP/E during JCLIN
identified to SMP/E by the ++FUNCTION statement. processing.
SMP/E is an example of a base function.
conditional requisites. Requisites defined by an ++IF
statement. These are requisites that must be installed if
the functions specified on the ++IF statements are
installed.

conditionally coexisting functions • entry
conditionally coexisting functions. Functions that definition side deck. A file in a UNIX file system, a
coexist but do not have to be in the same zone. member of a partitioned data set, or a sequential data
set that contains binder IMPORT control statements.
consolidated software inventory. See SMPCSI.
deleted function. In SMP/E, a function that was
corequisite SYSMODs. SYSMODs each of which can removed from the system when another function was
be installed properly only if the other is present. installed. This is indicated by the DELBY subentry in
Corequisites are defined by the REQ operand on the the SYSMOD entry for the deleted function. See also
++VER statement. explicitly deleted function and implicitly deleted function.
corrective service. Any SYSMOD used to selectively deleting function. A function that removes other
fix a system problem. Generally, corrective service functions from the system. This is done by specifying
refers to APAR fixes. them on the DELETE operand of the ++VER statement.
cross-zone. (1) A target zone other than the set-to zone dependent function. A function that introduces new
that defines a load module containing modules from elements or redefines elements of the base level system
set-to zone. Also called a TIEDTO zone. The modules or other products. A dependent function cannot exist
were added to the load module through the SMP/E without a base function. Dependent functions are
LINK command. The relationship between the identified to SMP/E by the ++FUNCTION statement.
cross-zone and the set-to zone is established through
the TIEDTO subentry in their zone definition entries. dialog. The interactive support provided by SMP/E
See also set-to zone and TIEDTO relationship. (2) through ISPF. Instead of entering specific commands
Pertaining to relationships between zones, especially as and operands, you can use panels to specify the
a result of conditional requisites (++IF statements) or desired processing.
LINK processing. See also cross-zone requisite, cross-zone
load module, and cross-zone module. distribution library (DLIB). A library that contains
the master copy of all the elements in a system. A
cross-zone load module. A load module containing distribution library can be used to create or back up a
modules from a different zone as a result of LINK target library.
processing.
distribution zone. In SMP/E, a group of records in a
cross-zone module. A module included in a load CSI data set that describes the SYSMODs and elements
module from a different zone as a result of LINK in a distribution library.
processing.
DLIB. Distribution library.
cross-zone requisite. A conditional requisite that must
be installed in one zone because of another SYSMOD DLIB entry. An SMP/E entry describing a distribution
that is installed in a different zone. The REPORT library that has been totally copied into a target library.
command can be used to check information saved from
++IF statements and determine where any cross-zone DLIBZONE entry. An SMP/E entry containing
requisites should be installed. information used by SMP/E to process a specific
distribution zone and the associated distribution
CSI. Consolidated software inventory data set. See libraries.
SMPCSI.
DLL. Dynamic link library
D E
data element. An element that is not a macro, module,
or source—for example, a dialog panel or sample code. EC. Engineering change.
DDDEF entry. An SMP/E entry containing the element. In SMP/E, part of a product, such as a
information SMP/E needs in order to dynamically macro, module, dialog panel, or sample code.
allocate a particular data set.
element MCS. An MCS used to replace or update an
DEBUG. The SMP/E command used to obtain element.
additional information for problem determination—for
element selection. The process by which SMP/E
example, to trace messages or take dumps.
chooses the appropriate changes for an element affected
debug. In SMP/E, to obtain additional information for by several SYSMODs being installed at the same time.
problem determination—for example, to trace messages
entry. In SMP/E, a collection of records in a CSI data
or take dumps. This is done with the DEBUG
set. An entry can be created, updated, or deleted by use
command.
of UCL statements.
Glossary 231
environment • global zone
environment. The functions (FMIDs) installed on a FTP.DATA. Configuration data set used to change
particular system or subsystem (SREL). local site default values for the z/OS FTP Client.
ERROR indicator. In SMP/E, an indicator in a target FMID. Function modification identifier.

or distribution zone SYSMOD entry that shows that
SYSMOD processing failed. The ERROR indicator is set FMIDSET. A group of FMIDs to be used in
before SMP/E updates any libraries and is reset if processing an SMP/E command—for example, to
processing is successful. If processing fails, it remains indicate that SYSMODs applicable to certain functions
set to show that an error occurred. should be installed.
ESO. Expanded service options. FMIDSET entry. An SMP/E entry defining an

FMIDSET.
exception SYSMOD. A SYSMOD that is in error or
that requires special processing before it can be function. In SMP/E, a product (such as a system
installed. ++HOLD and ++RELEASE statements component or licensed program) that can be installed
identify exception SYSMODs. in a user’s system if desired. Functions are identified to
SMP/E by the ++FUNCTION statement. Each function
EXCP. Execute channel programs. must have a unique FMID.
expanded service options (ESO). A tape that includes function modification identifier (FMID). The
preventive service PTFs. Where available, it replaces SYSMOD ID of a function SYSMOD. It identifies the
PUTs as the vehicle for delivering preventive service. function that currently owns a given element.
An ESO contains PTFs and ++ASSIGN statements
assigning source IDs for the PTFs. In the United States, functionally higher SYSMOD. A SYSMOD that uses
this tape is available from the IBM Support Center and the function contained in an earlier SYSMOD (called
can be ordered either by subscription or as needed. the functionally lower SYSMOD) and contains additional
functions as well.
explicitly deleted function. A function deleted
because it was specified on the DELETE operand of a functionally lower SYSMOD. A SYSMOD whose
++VER statement in another SYSMOD. function is also contained in a later SYSMOD (called
the functionally higher SYSMOD).
exported zone. A zone copied into a sequential data
set by use of the SMP/E ZONEEXPORT command.
G
external HOLDDATA. ++HOLD statements contained
in SMPHOLD. Contrast with internal HOLDDATA. GENASM. A subentry in the MAC entry that lists the
ASSEM or SRC entries that must be assembled if the
macro is replaced or updated.
F
GENERATE. The SMP/E command used to create a
FAF. file attribute file. job stream that builds a set of target libraries from a set
of distribution libraries.
FE. Field engineering.
generate. In SMP/E, to create a job stream that builds
feature. See dependent function. a set of target libraries from a set of distribution
libraries. This is done with the GENERATE command.
file attribute file. A file attribute file (FAF) is a file
containing control statements that describe the GIMUNZIP. An SMP/E service routine used to
attributes of the file contained in an SMP/E network extract files contained in network transportable
transportable archive. The FAF is contained within the packages that were built using GIMZIP.
archive information about how the file was created.
GIMZIP. An SMP/E service routine used to produce
File Transfer Protocol. File Transfer Protocol (FTP) is a network transportable packages.
protocol that defines the interactions necessary between
a client and server to facilitate the exchange of binary global zone. A group of records in a CSI data set used
and textual data files. to record information about SYSMODs received for a
particular system. The global zone also contains
firewall. A firewall is an intermediate server that information that (1) enables SMP/E to access target and
functions to isolate a secure network from an insecure distribution zones in that system, and (2) enables you
network. to tailor aspects of SMP/E processing.
FTP. File Transfer Protocol.

GLOBALZONE entry • JAR
GLOBALZONE entry. An SMP/E entry containing imported zone. A zone copied from a sequential data
information that SMP/E uses to process the global set into another zone by use of the SMP/E
zone, the associated target and distribution zones, and ZONEIMPORT command.
SMPPTS.
IMS. Information Management System.
GTF. Generalized trace facility.
IMSGEN. IMS generation.
H indicator. See subentry indicator.
hashing. An operation that uses a one-way in effect. Having control over SMP/E processing. For
(irreversible) function on data, usually to reduce the example, an OPTIONS entry is in effect if (1) it is
length of the data and to provide a verifiable specified on the SET command or (2) it is defined as
authentication value (checksum) for the hashed data. the default OPTIONS entry for the set-to zone.
header MCS. An ++APAR, ++FUNCTION, ++PTF, or inline data. Information (such as utility control
++USERMOD statement. The header MCS indicates the statements or code for an element) that is packaged
type of SYSMOD. directly after the associated MCS, rather than in a
separate file or data set.
HFS. Hierarchical file system.
inline JCLIN. The JCL statements associated with a
hierarchical file system element. An element that has ++JCLIN statement. Inline JCLIN may immediately
a UNIX file system as its “target library”. follow the ++JCLIN statement, or it may be in the
RELFILE or TXLIB data set pointed to by the ++JCLIN
hierarchy. In SMP/E, the top-down structure of statement. Inline JCLIN is used to update the target
function and service SYSMODs, in which each zone when a SYSMOD is applied, or the distribution
SYSMOD is dependent on the one above it. zone when a SYSMOD is accepted. Contrast with JCLIN
input.
higher functional level. An element version that
contains all the functions of all other relevant versions inner macro. A macro invoked by another macro. In
of that element. particular, inner macros are those that SMP/E does not
detect during JCLIN processing of assembler job steps.
HOLDDATA. In SMP/E, MCSs used to indicate that
certain SYSMODs contain errors or require special install. In SMP/E, to apply a SYSMOD to the target
processing before they can be installed. ++HOLD and libraries or to accept a SYSMOD into the distribution
++RELEASE statements are used to define libraries.
HOLDDATA. SYSMODs affected by HOLDDATA are
called exception SYSMODs. internal HOLDDATA. ++HOLD statements contained
within a SYSMOD. Contrast with external HOLDDATA.
HOLDDATA entry. An SMP/E entry containing
++HOLD statements that either were received from I/O. Input or output.
SMPHOLD (external HOLDDATA) or were within a
SYSMOD that was received (internal HOLDDATA). IOGEN. Input/output device generation.
IPL. Initial program load.

I
IPv6. Internet Protocol Version 6.
ICSF. Integrated Cryptographic Service Facility.
ISMD. IBM Software Manufacturing and Delivery
IFREQ. A conditional requisite. Conditional requisites (formerly called PID).
are specified on the REQ operand of the ++IF
statement. ISPF. Interactive System Productivity Facility.
IMASPZAP. The system utility program used to ISPF/PDF. Interactive System Productivity
install superzaps, which are changes for modules, load Facility/Program Development Facility.
modules, or CSECTs within modules.
IVP. Installation verification procedure.
implicitly deleted function. A function deleted
because of its dependency on an explicitly deleted
function that is specified on the DELETE operand of
J
the ++VER statement. JAR. The SMP/E entry or MCS that describes a Java
ARchive (JAR) file. Also, the abbreviation for Java
ARchive (see Java ARchive(JAR)).
Glossary 233
JARUPD • modification identifier (MODID)
JARUPD. The SMP/E MCS used to describe an LMOD. In SMP/E, an abbreviation for load module.
update to a JAR element.
LMOD entry. An SMP/E entry containing all the
JCL. Job control language. information needed to replace or update a given load
module.
JCLIN. (1) The SMP/E command used to process data
from SMPJCLIN. (2) The ++JCLIN statement, which is load module. A computer program in a form suitable
associated with JCLIN data that is included in a for loading into main storage for execution. It is usually
SYSMOD. (3) SMPJCLIN. See SMPJCLIN. the output of a link-edit utility.
See also inline JCLIN and JCLIN data.
LOG. (1) The SMP/E command used to write
JCLIN data. The JCL statements associated with the user-supplied information to the SMPLOG data set. (2)
++JCLIN statement or saved in the SMPJCLIN data set. The SMPLOG data set. See SMPLOG.
They are used by SMP/E to update the target zone
lower functional level. An element version that is
when the SYSMOD is applied. Optionally, SMP/E can
contained in a later element version.
use JCLIN data to update the distribution zone when
the SYSMOD is accepted.
M
JCLIN input. The JCL statements contained in
SMPJCLIN and used as input for the JCLIN command. MAC. The SMP/E entry or MCS that describes a
Contrast with inline JCLIN. macro.
jar. The Java command used to invoke the Java macro. An instruction in a source language that is to
Archive Tool. The Java Archive Tool is used to perform be replaced by a defined sequence of instructions in the
operations on Java ARchive (JAR) files. same source language.
Java ARchive (JAR). An archive file format based on MACUPD. The SMP/E MCS used to update a macro.
the ZIP file format. Used for aggregating many Java
applet component files into one. mass-mode processing. In SMP/E, processing that
includes all eligible SYSMODs, regardless of whether
job control language (JCL). A problem-oriented they were individually selected.
language designed to express statements in a job that
are used to identify the job or describe its requirements master CSI. The CSI data set that contains the global
to an operating system. zone.
MCS. Modification control statement.

L
MCS entry. An SMP/E entry containing a copy of a
licensed program. A program that performs a function SYSMOD exactly as it was received from SMPPTFIN.
for the user, and usually interacts with and relies upon MCS entries are in SMPPTS, which is used to store
the system control program or some other SYSMODs.
IBM-provided control program. Generally, a licensed
program is a software package that can be ordered MOD. The SMP/E entry or MCS that describes an
from the program libraries, such as IBM Software object module or a single-module load module.
Distribution (ISMD). IMS and CICS are examples of
licensed programs. MODID. Modification identifier.
LINK LMODS. The SMP/E command used to relink modification. In SMP/E, an alteration or correction to
load modules that use CALLLIBS. a system control program, licensed program, or user
program. Synonymous with system modification
LINK MODULE. The SMP/E command used to link (SYSMOD).
modules in one zone with load modules in another
zone. modification control statement (MCS). An SMP/E
control statement used to package a SYSMOD. MCSs
link library (LKLIB). A data set containing link-edited describe the elements of a program and the
object modules. relationships that program has with other programs
that may be installed on the same system.
LIST. The SMP/E command used to display entries in
SMP/E data sets. modification identifier (MODID). A list of SYSMOD
IDs, including the last SYSMOD that totally replaced
list. In SMP/E, to display entries in SMP/E data sets. the element (RMID), any subsequent partial updates to
This is done with the LIST command.
LKLIB. Link library.

modification level • product version
the element (UMIDs), and the function that owns the module (for example, to prepare the module for
element (FMID). MODIDs are contained in element shipment in RELFILE format or in an LKLIB data set or
entries. as part of SMP/E ACCEPT processing).
modification level. A distribution of all temporary operating system. In SMP/E, the system updated by
fixes that have been issued since the previous APPLY and RESTORE processing. It consists of the
modification level. A change in modification level does target libraries. Also called the target system.
not add new functions or change the programming
support category of the release to which it applies. OPTIONS entry. An SMP/E entry defining
Contrast with release and version. processing options that are to be used by SMP/E.
Note: Whenever a new release of a program is P

shipped, the modification level is set to 0. When
the release is reshipped with the accumulated package attribute file. A package attribute file (PAF)
services changes incorporated, the modification is a file that contains control statements describing the
level is incremented by 1. contents of a network transportable package.
module. Synonym for object module or single-module packaging. Adding the appropriate MCS statements to
load module. elements to create a SYSMOD, then putting the
SYSMOD in the proper format on the distribution
MTS. Macro temporary storage data set. See
medium, such as a tape or direct access data sets.
SMPMTS.
PAF. package attribute file.
MTSMAC entry. An SMP/E entry that is a copy of a
macro that resides only in a distribution library but is partitioned data set extended (PDSE). A
needed temporarily during APPLY processing. system-managed data set containing an indexed
MTSMAC entries are in the SMPMTS data set. directory and members that are similar to the directory
and members of partitioned data sets. A PDSE can be
MVS. Multiple Virtual Storage.
used instead of a partitioned data set.
MVS Custom-Built Product Delivery Offering
PE. See program error PTF.
(CBPDO). A software delivery offering used to add
products or service to an existing MVS, NCP, CICS, or PE-PTF. See program error PTF.
IMS system.
PID. The former name for ISD.
N PRE. Prerequisite.
NCP. Network Control Program. prerequisite (PRE). In SMP/E, a SYSMOD that must
be installed before or along with another SYSMOD in
negative prerequisite (NPRE). In SMP/E, a function order for that other SYSMOD to be successfully
that is mutually exclusive with another function. It is installed. It is defined by the PRE operand on the
defined by the NPRE operand on the ++VER statement. ++VER statement.
NPRE. Negative prerequisite. preventive service. (1) The mass installation of PTFs
to avoid rediscoveries of the APARs fixed by those
O PTFs. (2) The SYSMODs delivered on the program
update tape.
object deck. Object module input to the link-edit
utility that is placed in the input stream, in card format. preventive service planning (PSP). Installation
recommendations and HOLDDATA for a product or a
object module. A module that is the output from a service level. PSP information can be obtained through
language translator (such as a compiler or an the IBM Support Center.
assembler). An object module is in relocatable format
with machine code that is not executable. Before an product. Generally, a software package, such as a
object module can be executed, it must be processed by licensed program or a user application. A product can
the link-edit utility. contain one or more functions and can consist of one or
more versions and releases.
When an object module is link-edited, a load module is
created. Several modules can be link-edited together to product version. All the releases for a given version of
create one load module (for example, as part of SMP/E a product.
APPLY processing), or an object module can be
link-edited by itself to create a single-module load
Glossary 235
program error PTF (PE-PTF) • REPORT
program error PTF (PE-PTF). A PTF that has been regression. In SMP/E, the condition that occurs when
found to contain an error. A PE-PTF is identified on a an element is changed by a SYSMOD that is not related
++HOLD ERROR statement, along with the APAR that to SYSMODs that previously modified the element.
first reported the error.
REJECT. The SMP/E command used to remove
program object. An executable program stored in a SYSMODs from the global zone and SMPPTS.
PDSE program library. It is similar to a load module,
but has fewer restrictions. For SMP/E purposes, reject. In SMP/E, to remove SYSMODs from the
program objects are referred to as load modules. global zone and SMPPTS and delete any related
SMPTLIB data sets. This is done with the REJECT
program packaging. See packaging. command.
program product. The former term for licensed related installation materials (RIMs). In IBM
program. custom-built offerings, task-oriented documentation,
jobs, sample exit routines, procedures, parameters, and
program temporary fix (PTF). A temporary solution examples developed by IBM.
or bypass of a problem that may affect all users and
that was diagnosed as the result of a defect in a current related SYSMOD. A SYSMOD associated with other
unaltered release of the program. In the absence of a SYSMODs by the FMID, PRE, REQ, or SUP operands.
new release of a system or component that incorporates
the correction, the fix is not temporary but is the related zone. The zone named in the RELATED
permanent and official correction mechanism. New subentry of a TARGETZONE or DLIBZONE entry. For
elements can also be defined in a PTF. PTFs are a target zone, the related zone is generally the
identified to SMP/E by the ++PTF statement. distribution zone for the libraries used to create the
target libraries. For a distribution zone, the related zone
program update tape (PUT). The former vehicle for is generally the target zone for the libraries built from
preventive service. See expanded service options. the distribution libraries.
PSP. Preventive service planning. relative file (RELFILE) format. A SYSMOD packaging
method where elements and JCLIN data are in separate
PSW. Program status word. relative files from the MCSs. When SYSMODs are
packaged in relative file format there is a file of MCSs
PTF. Program temporary fix. for one or more SYSMODs, and one or more relative
files containing unloaded source-code data sets and
PTS. PTF temporary store data set. See SMPPTS.
unloaded link-edited data sets containing executable
PTFIN. PTF input. See SMPPTFIN. modules. The relative files can be either unloaded files
in IEBCOPY format, or they can be partitioned data
PUT. See expanded service options. sets. Relative file format is the typical method used for
packaging function SYSMODs.
R relative files (RELFILEs). Unloaded files containing
modification text and JCL input data associated with a
RACF. Resource Access Control Facility. SYSMOD. These files are used to package a SYSMOD
in relative file format.
RECEIVE. The SMP/E command used to read in
SYSMODs and other data from SMPPTFIN and release. A distribution of a new product or new
SMPHOLD. function and APAR fixes for an existing product.
Contrast with modification level and version.
receive. In SMP/E, to read SYSMODs and other data
from SMPPTFIN and SMPHOLD and store them on the replacement modification identifier (RMID). The
global zone for subsequent SMP/E processing. This is SYSMOD ID of the last SYSMOD that completely
done with the RECEIVE command. replaced a given element.
regressed SYSMOD. A SYSMOD one or more of REPORT. The SMP/E command used to obtain
whose elements are modified by subsequent SYSMODs information about SYSMODs that have been installed.
that are not related to it. These are the types of REPORT commands:
regressing SYSMOD. A SYSMOD that causes v REPORT CROSSZONE: Lists conditional requisites
regression of previous modifications when it is that must be installed in certain zones because of
installed. SYSMODs installed in other zones.
v REPORT ERRSYSMODS: Determines whether any
SYSMODs already installed are now exception
SYSMODs.

requisite • SMPCNTL
v REPORT SOURCEID: Lists the source IDs associated v ID errors (a SYSMOD does not supersede or specify
with SYSMODs in the specified zones. as a prerequisite an RMID or a UMID)
v REPORT SYSMODS: Compares the SYSMODs v JCLIN failures (syntax errors)
installed in two target or distribution zones.
RPL. Request parameter list.
requisite. A SYSMOD that must be installed before or
at the same time as the SYSMOD being processed. RTM2WA. Recovery termination manager 2 work
There are several types of requisites: area.
v Prerequisites, which are specified by the PRE
operand on the SYSMOD’s ++VER statement S
v Corequisites, which are specified by the REQ
operand on the SYSMOD’s ++VER statement SCDS. Save control data set. See SMPSCDS.
v Conditional requisites, which are specified by the SCP. System control program.
REQ operand on the SYSMOD’s associated ++IF
statement select-mode processing. In SMP/E, processing that
includes individually selected SYSMODs.
requisite chain. A sequence of SYSMODs that are
directly or indirectly identified as requisites for a given service. PTFs and APAR fixes.
SYSMOD, (A SYSMOD may identify other SYSMODs
as requisites, which in turn may have requisites of their service level. The FMID, RMID, and UMID values for
own. The requisite chain extends from the initial an element. The service level identifies the owner of the
SYSMOD, through the hierarchy of requisites, until no element, the last SYSMOD to replace the element, and
more SYSMODs are found that have requisites.) See all the SYSMODs that have updated the element since
requisite. it was last replaced.
requisite set. The set of all SYSMODs on the requisite service order relationship. A relationship among
chain for a particular SYS service SYSMODs that is determined by the PRE and
SUP operands, and the type of SYSMOD.
RESETRC. The SMP/E command used to set the
return codes for the previous commands to zero, so service SYSMOD. Any SYSMOD identified by an
that SMP/E can process the current command. ++APAR or ++PTF statement.
RESTORE. The SMP/E command used to remove service update. The integration of available service
applied SYSMODs from the target libraries. into the current release of a function. Since this is not a
new release of the function, it does not change the
restore. In SMP/E, to remove applied SYSMODs from function’s FMID.
the target libraries by use of the RESTORE command.
SET. The SMP/E command used to indicate the zone
restore group. All the SYSMODs that have a direct or to be processed.
indirect relationship with a SYSMOD being restored by
use of the GROUP operand. set. In SMP/E, to indicate which zone should be
processed by the subsequent commands. This is done
RIM. Related installation material. with the SET command.
RMID. Replacement modification identifier. set-to zone. The zone that was specified on the
previous SET command and that is currently being
RMF. Resource measurement facility. processed. Contrast with cross-zone.
root cause analysis. Processing done by SMP/E for SHA-1. Secure Hash Algorithm 1.
the ACCEPT, APPLY, and RESTORE commands to
identify causer SYSMODs (SYSMODs whose failure has side deck. See definition side deck.
led to the failure of other SYSMODs). The types of
errors SMP/E analyzes to determine causer SYSMODs single-module load module. A load module created
include the following: by link-editing a single object module by itself—for
v Held SYSMODs example, to prepare the module for shipment in
RELFILE format or in an LKLIB data set or as part of
v Missing requisite SYSMODs
SMP/E ACCEPT processing.
v Utility program failures: copy, update, assembler,
link, zap SMPCNTL. The SMP/E data set or file in a UNIX file
v Out-of-space conditions: x37 abends system that contains the SMP/E commands to be
processed.
v Missing DD statements and other allocation errors
Glossary 237
SMPCSI • SMPRPT
SMPCSI. The SMP/E data set that contains SMPLTS. The SMP/E data set used as a target load
information about the structure of a user’s system as module library to maintain the base version of a load
well as information needed to install the operating module that specifies a SYSLIB allocation in order to
system on a user’s system. The SMPCSI DD statement implicitly include modules.
refers specifically to the CSI that contains the global
zone. This is also called the master CSI. SMPMTS. The SMP/E data set used as a target
library for macros that exist only in a distribution
SMPDEBUG. The SMP/E data set or file in a UNIX library, such as macros in SYS1.AMODGEN. The
file system that contains a dump requested by the SMPMTS enables the current version of these macros to
DEBUG command. Depending on the operands be used for assemblies during APPLY processing.
specified, it may contain (1) a dump of SMP/E control
blocks and storage areas associated with the specified SMPNTS. The SMPNTS is a directory structure and
dump points or (2) a dump of the VSAM RPL control associated files contained in a UNIX file system used
block for the specified SMP/E function. for temporary storage of network transported packages
that were received during SMP/E RECEIVE processing.
SMPDUMMY. The SMP/E data set used to define a
load module’s definition side deck library as a SMPOBJ. The SMP/E data set used for
DUMMY data set. SMPDUMMY is always allocated by source-maintained products. SMPOBJ contains
SMP/E as a DUMMY data set. preassembled modules that can be used to avoid
reassembling those modules. These modules must be in
SMP/E. A program product, or an element of OS/390 load module format—that is, in the same format as
or z/OS, used to install software and software changes modules residing in the distribution library.
on z/OS systems. SMP/E consolidates installation data,
allows more flexibility in selecting changes to be SMPOUT. The SMP/E data set or file in a UNIX file
installed, provides a dialog interface, and supports system that contains messages issued during SMP/E
dynamic allocation of data sets. processing. It may also contain a dump of the VSAM
RPL, if a dump was taken. In addition, it may contain
SMP/E commands. Commands defining the LIST output and reports if SMPLIST and SMPRPT are
processing to be done by SMP/E, such as RECEIVE. not defined.
SMP/E entry. An entry in an SMP/E data set—for SMPPARM. The data set that contains members to
example, a MOD entry in a CSI data set. define parameters, such as macros, assembler operation
codes, GIMDDALC control statements, and exit
SMPHOLD. SMPHOLD is the source for HOLDDATA routines.
(++HOLD and ++RELEASE statements) to be processed
by the RECEIVE command. SMPHOLD may be a tape SMPPTFIN. SMPPTFIN is the source of SYSMODs
file, a data set, or one or more files in a UNIX file and ++ASSIGN statements to be processed by the
system. RECEIVE command. SMPPTFIN may be a tape file, a
data set, or one or more files in a UNIX file system.
SMPJCLIN. The SMP/E data set or file in a UNIX file
system that contains a job stream of assembly, link-edit, SMPPTS. The SMP/E data set that contains
and copy job steps. This data is typically the stage 1 SYSMODs received from SMPPTFIN. SMPPTS is the
output from the most recent full or partial system source of SYSMODs that are installed in the target and
generation. However, it may also be other data in a distribution libraries.
similar format, such as the output of the GENERATE
command. This job stream is used as input to the SMPPTS spill data sets. Optional SMP/E data sets
JCLIN command to update or create entries in a target that can be used to store SYSMODs when the SMPPTS
zone. data set becomes full.
SMPLIST. The SMP/E data set or file in a UNIX file SMPPUNCH. The SMP/E data set or file in a UNIX
system that contains the output of all LIST commands. file system that contains output from various SMP/E
commands. This output generally consists of
SMPLOG. The SMP/E data set that contains commands or control statements.
time-stamped records of SMP/E processing. The v GENERATE: A job stream for building target libraries
records in this data set can be written automatically by
v REPORT: Commands for installing or listing
SMP/E or added by the user through the LOG
SYSMODs
command.
v UNLOAD: UCLIN statements for recreating the
SMPLOGA. A secondary log data set for SMP/E entries that were unloaded
processing. If SMPLOGA is defined, it is automatically
used when the SMPLOG data set is full. SMPRPT. The SMP/E data set or file in a UNIX file
system that contains the reports produced during
SMP/E processing.

SMPSCDS • subentry
SMPSCDS. The SMP/E data set that contains backup source. The source statements that constitute the input
copies of target zone entries that are created during to a language translator for a particular translation.
APPLY processing. These backup copies are made
before the entries are (1) changed by inline JCLIN, a source ID. A 1- to 8-character identifier that indicates
++MOVE MCS, or a ++RENAME MCS, or (2) deleted how a SYSMOD was obtained—for example, from a
by an element MCS with the DELETE operand. The particular service level in an ESO. A source ID is
backup copies are used during RESTORE processing to associated with a specific SYSMOD by the RECEIVE
return the entries to the way they were before APPLY command or the ++ASSIGN statement.
processing.
SOCKS. A networking proxy protocol that enables
SMPSNAP. The SMP/E data set that is used for snap hosts on one side of a SOCKS server to gain full access
dump output. When a severe error such as an abend or to hosts on the other side of the SOCKS server without
severe VSAM return code occurs, SMP/E requests a requiring direct IP-reachability.
snap dump of its storage before doing any error
recovery. In addition, the DEBUG command can SOURCEID. The operand used to refer to a source ID.
request a snap dump of SMP/E storage when specified
source module. The source statements that constitute
messages are issued, or can request a snap dump of
the input to a language translator, such as a compiler
control blocks and storage areas associated with a
or an assembler, for a particular translation.
specified dump point.
SRC. The SMP/E entry or MCS statement that
SMPSTS. The SMP/E data set used as a target library
describes a source.
for source that exists only in a distribution library. The
SMPSTS enables the current version of this source to be SRCUPD. The MCS used to update a source.
used for assemblies during APPLY processing.
SREL. System release identifier.
SMPTLIB. The SMP/E data sets used as temporary
storage for relative files loaded from SMPPTFIN during Storage Management Subsystem (SMS). A
RECEIVE processing. The SMPTLIB data sets are DFSMS/MVS facility used to automate and centralize
deleted when the associated SYSMOD is deleted by the management of storage. Using SMS, a storage
REJECT, RESTORE, or ACCEPT processing. administrator describes data allocation characteristics,
performance and availability goals, backup and
SMPWKDIR. An optional directory in a UNIX file retention requirements, and storage requirements to the
system used for temporary work files. system through data class, storage class, management
class, storage group, and ACS routine definitions.
SMPWRK1. The SMP/E data set used as temporary
storage for macro updates and replacements that will STS. Source temporary store data set. See SMPSTS.
be processed by an update or copy utility program.
SMP/E places the input in SMPWRK1 during APPLY STSSRC entry. An SMP/E entry that is a copy of
and ACCEPT processing before calling the utility. source that resides only in a distribution library but is
needed temporarily during APPLY processing. STSSRC
SMPWRK2. The SMP/E data set used as temporary entries are in the SMPSTS data set.
storage for source updates and source replacements
that will be processed by an update or copy utility stub entry. An element entry or LMOD entry that
program. SMP/E places the input in SMPWRK2 during does not contain the basic information SMP/E requires
APPLY and ACCEPT processing before calling the in order to process the element or load module (such as
utility. FMID, RMID, or library names), but does contain other
information, such as subentries describing cross-zone
SMPWRK3. The SMP/E data set used as temporary relationships.
storage for object modules supplied by a SYSMOD,
object modules created by assemblies, and zap utility stub load module. A load module that does not
input following ++ZAP statements. contain the modules needed to perform its basic
functions, but does contain other modules, such as
SMPWRK4. The SMP/E data set used as temporary cross-zone modules.
storage for zap utility or link-edit utility input that
contains EXPAND control statements. subentry. A field in an SMP/E entry. Each subentry
has associated with it a type and a value. The same
SMPWRK6. The SMP/E data set used during subentry type may occur several times in a single entry,
ACCEPT and APPLY processing as temporary storage each time with a different value. For example, the
for inline replacements for data elements. SMP/E modules supplied by a PTF are saved as MOD
places the input in this data set so that it can be subentries in the PTF’s SYSMOD entry. Some subentries
directly accessed and installed by the copy utility or occur only once within an entry, such as the FMID
SMP/E. subentry in a target zone MOD entry.
Glossary 239
subentry indicator • TCP/IP
subentry indicator. A subentry that does not have a SYSPUNCH. The temporary data set containing object
data value associated with it. An example of an modules assembled by running the job stream
indicator is the ERROR indicator in the SYSMOD entry. produced by system generation or the GENERATE
An indicator is either on or off. command. These modules are not installed in the
distribution libraries at ACCEPT time.
subentry list. Multiple occurrences of the same
subentry type in an entry, each with a different value. system control program (SCP). IBM-supplied
For example, the modules supplied by a PTF are saved programming that is fundamental to the operation and
as names in the MOD subentry list within the SYSMOD maintenance of the system. It serves as an interface
entry for that PTF. with licensed programs and user programs and is
available without additional charge.
SUP. Supersede.
system generation (SYSGEN). The process of
superseded-only SYSMOD. A SYSMOD that has not selecting optional parts of an operating system and of
been installed, but that has been superseded by another creating a particular operating system tailored to the
SYSMOD that has been installed. requirements of a data processing installation.
superseded SYSMOD. In SMP/E, a SYSMOD that is system modification (SYSMOD). The input data to
contained in or replaced by the SYSMOD or requisite SMP/E that defines the introduction, replacement, or
set of SYSMODs currently being processed. This is update of elements in the operating system and
indicated by the SUPBY subentry in the SYSMOD entry associated distribution libraries to be installed under
for the superseded SYSMOD. A superseded SYSMOD is the control of SMP/E. A system modification is defined
functionally lower than the SYSMOD that superseded by a set of MCS.
it.
system modification identifier (SYSMOD ID). The
superseding SYSMOD. In SMP/E, a SYSMOD that name that SMP/E associates with a system
contains all the functions in another SYSMOD and is modification. It is specified on the ++APAR,
recognized as the equivalent of that other SYSMOD. ++FUNCTION, ++PTF, or ++USERMOD statement.
The superseding SYSMOD uses SUP operand on its
++VER statement to specify the superseded SYSMOD. system release identifier (SREL). A 4-byte value
representing the system or subsystem, such as Z038 for
superzap. A generic term for the process performed MVS-based products.
by IMASPZAP. It can also refer to the module updates
processed by IMASPZAP. SYSUT1, SYSUT2, SYSUT3. Scratch data sets for
SMP/E and the utilities it calls.
SVC. Supervised call.
SYSUT4. A data set that is used instead of the SYSIN
SVRB. Supervisor request block. data sets when certain utilities are called.
SYSGEN. System generation.
T
SYSLIB. (1) A subentry used to identify the target
library in which an element is installed. (2) A target library. A library containing the executable
concatenation of macro libraries to be used by the code that makes up a system.
assembler. (3) A set of routines used by the link-edit
utility to resolve unresolved external references. target system. The system updated during APPLY and
RESTORE processing, also referred to as the operating
SYSMOD. System modification. system. See also target libraries.
SYSMOD entry. An SMP/E entry containing target zone. In SMP/E, a group of records in a CSI
information about a SYSMOD that has been received data set that describes the SYSMODs, elements, and
into SMPPTS, accepted into the distribution libraries, or load modules in a target library.
applied to the target libraries.
TARGETZONE entry. An SMP/E entry containing
SYSMOD ID. System modification identifier. information used by SMP/E to process a specific target
zone and the associated target libraries.
SYSMOD packaging. See packaging.
TCP/IP. Transmission Control Protocol/Internet
SYSMOD selection. The process of determining Protocol (TCP/IP) is a hardware independent
which SYSMODs are eligible to be processed. communication protocol used between physically
separated computers. It was designed to facilitate
SYSPRINT. The data set that contains output from the
communication between computers located on different
utilities called by SMP/E.
physical networks.

temporary data set • ZONEDELETE
temporary data set. A work data set unload. In SMP/E, to copy data out of SMP/E data
(SMPWRK1–SMPWRK6) or utility data set set entries in the form of UCL statements, by use of the
(SYSUT1–SYSUT4). Temporary data sets are allocated UNLOAD command.
when processing for an SMP/E command begins, and
deleted when processing is finished. update. In SMP/E, to change an existing element
without replacing it.
text library (TXLIB). A data set containing JCLIN
input or replacements for macros, source, or object update modification identifier (UMID). The
modules that have not been link-edited. It is used when SYSMOD ID of a SYSMOD that updated the last
the JCLIN or elements are provided in partitioned data replacement of a given element.
sets rather than inline or in relative files.
user modification (USERMOD). A change constructed
TGTLIB. Target library. by a user to modify an existing function, add to an
existing function, or add a user-defined function.
TIEDTO relationship. A cross-zone relationship USERMODs are identified to SMP/E by the
between two target zones created when the LINK ++USERMOD statement.
command updates a load module in one of the zones to
include modules from the other zone. This relationship USERMOD. User modification.
is established through the TIEDTO subentry in the zone
definition entries for each of the zones. UTILITY entry. An SMP/E entry containing
information used by SMP/E to invoke a particular
TIEDTO zone. See cross-zone. system utility program.
TLIB. Temporary library. See SMPTLIB.

V
transformed data. Data processed by the GIMDTS
service routine so that it can be packaged inline in VERSION. An operand on the ++VER or element
fixed-block 80 records. statement. VERSION specifies one or more SYSMODs
containing elements that are functionally lower than
Transport Layer Security (TLS). A protocol that elements in the SYSMOD that specifies the operand.
provides communications privacy over the Internet. The VERSION operand is also used to change
ownership of elements.
TSO. Time-sharing option.
version. A separate licensed program that is based on
TXLIB. Text library. an existing licensed program and that usually has
significant new code or new functions. Contrast with
U release and modification level.
versioned element. An element that is part of more

UCL. Update control language. than one function—for example, one that is part of a
base function and a dependent function.
UCL statement. An SMP/E control statement used to
define or change information in an SMP/E data set VSAM. Virtual Storage Access Method.
entry. UCL statements are coded between the UCLIN
and ENDUCL commands. The UCL statement specifies VTOC. Volume table of contents.
the action to be taken (ADD, REP, or DEL), the entry to
be modified, and any indicators and subentries to be
changed. Z
UCLIN. The SMP/E command used to mark the ZAP. (1) The SMP/E MCS used to package an update
beginning of UCL statements, which are used to make for an object module. (2) The superzap control
changes to entries in SMP/E data sets. statement used to update an object module. (3) A
shortened name for the superzap utility, which is used
UMID. Update modification identifier. to install these updates. See IMASPZAP.
unconditionally coexisting functions. Functions that zone. A partition in a CSI data set.
coexist and must be in the same zone.
ZONECOPY. The SMP/E command used to copy a
UNLOAD. The SMP/E command used to copy data zone from one CSI data set to another.
out of SMP/E data set entries in the form of UCL
statements. ZONEDELETE. The SMP/E command used to delete
a zone from a CSI data set.
Glossary 241
ZONEEDIT • z/OS UNIX System Services (z/OS UNIX)
ZONEEDIT. The SMP/E command used to change

the values for a subentry in all the DDDEF or UTILITY
entries in a given zone.
ZONEEXPORT. The SMP/E command used to copy a

zone into a sequential data set.
ZONEIMPORT. The SMP/E command used to load

an exported zone from a sequential data set into
another zone.
ZONEMERGE. The SMP/E command used to copy

one zone into another, or to merge two zones into one.
ZONERENAME. The SMP/E command used to

change the name of a zone.
ZONESET. A group of zones to be used when

processing an SMP/E command. For example, it may
define the zones that the REPORT command is to check
for cross-zone requisites. A ZONESET may also define
a group of zones to be checked or ignored by the
REJECT command.
ZONESET entry. An SMP/E entry defining a

ZONESET.
z/OS UNIX System Services (z/OS UNIX). The set of

functions provided by the Shell and Utilities, kernel,
debugger, file system, C/C++ Run-Time Library,
Language Environment, and other elements of the
z/OS operating system that allow users to write and
run application programs that conform to UNIX
standards.

Index
Special characters accessibility 223
alias defined for user catalog 56, 58
catalogs
alias defined for user catalog 56, 58
++element MCS allocating data sets for CSI 56
USERMODs 173 DDDEF entry 63 listing 61
++HOLD MCS dynamic allocation 62 causer SYSMODs 183
coexistence considerations 211 PTS 62 CBPDO
operands 135 SCDS 62 Recommended Service Upgrade 221
++IF MCS SMPCSI 56 CBPDO tape
cross-zone requisite checking 159 ALLZONES 150 format 110
++JAR MCS AMODE=64 link edit parameter 204 functions, source of 101
using 187, 188 AMS utility HOLDDATA
++JARUPD MCS allocating the CSI 56 processing 141
using 187 default values 65 source of 139
++JCLIN MCS reorganizing a CSI 61 preventive service, source of 109
USERMODs 172 APAR fixes CHANGEFILE
++MAC MCS defining 38 coexistence considerations 208
USERMODs 172 APAR SYSMODs 5 CHECK operand
++MACUPD MCS APPLY CHECK command on REJECT command 197
USERMODs 172 See also APPLY command CISIZE
++MOD MCS corrective service 127 for CSI data set 57
USERMODs 172 functions 104 CLEANUP command 46
++PROGRAM MCS preventive service 114 CLIENT data set
USERMODs 173 USERMODs 132 changes to 194
++RELEASE MCS APPLY command migration tasks 197
operands 135 corrective service 127 CLIST data set for SMP/E dialogs,
++SRC MCS description 13 LIBDEF restrictions 72
USERMODs 173 examples 21 CMWA
++SRCUPD MCS exception SYSMOD processing 137 copy utility parameter 197
USERMODs 173 functions 106 commands
++USERMOD MCS preventive service 117 ACCEPT 44
building USERMODs 170 processing 19 APPLY 43
examples 174 reports produced 23 CLEANUP 46
++VER MCS summary 43 DEBUG 47
coexistence considerations 211 USERMODs 133 GENERATE 44
USERMODs 170 ASMA90 utility JCLIN 45
++ZAP MCS See assembler utility LINK LMODS 47
USERMODs 172 assembler utility LINK MODULE 47
keepalive attribute default values 65 LIST 44
migration tasks 197 Automated Service Delivery Package LOG 46
pasv attribute HOLDDATA RECEIVE 42
migration tasks 197 source of 140 RECEIVE ORDER 42
automatic call libraries 147 REJECT 45
automatic cross-zone requisite checking REPORT CROSSZONE 44
A specifying 78 REPORT ERRSYSMODS 44
ACCEPT CHECK command REPORT SOURCEID 44, 165
See also ACCEPT command REPORT SYSMODS 44, 167
corrective service 128 B RESETRC 47
RESTORE 44
functions 107 BACKUP entry 20
preventive service 119 SET 42
base functions 38
ACCEPT command UCLIN 45
binder
corrective service 129 ZONECOPY 46
See link-edit utility
description 13 ZONEDELETE 46
BPX.SUPERUSER security class
examples 29 ZONEEDIT 45
coexistence considerations 206
exception SYSMOD processing 138 ZONEEXPORT 46
functions 108 ZONEIMPORT 46
preventive service 120 ZONEMERGE 46
processing 27 C ZONERENAME 47
reports produced 30 CALL COMPACT
summary 44 effect of CALLLIBS subentry on 66 coexistence considerations 208
Access Method Services (AMS) calling SMP/E 81 comparing two zones
See AMS utility cataloged procedure for SMP/E 82 LIST command 153

comparing two zones (continued) CSI (continued) dialogs (continued)
REPORT CROSSZONE structures, examples of LIBDEF restrictions 72
command 159 multiple-CSI structure 53 logon procedure for SMP/E 74
COMPAT=PM4 link edit parameter 204 separate CSI for each SREL, query 195
compress utility combining target and DLIB required programs 71
default values 65 zones 55 restrictions 72
compressing a CSI 61 separate CSI for each zone 54 saving JCL generated by SMP/E
concatenating dialog libraries 72 separate global zones 52 dialogs 72
conditional JCLIN processing single-CSI structure 52 specifying defaults for 75
coexistence considerations 203 summary 49 table data sets 72
Consolidated Software Inventory CUM tape target libraries
See SMPCSI See cumulative service tape in logon procedure for SMP/E 74
consolidated software inventory data set customizing disability 223
(SMPCSI) 11 an element (USERMOD SYSMODs) 6 displaying SMP/E data
CONTROLINTERVALSIZE JOB statement 76 API for 35
for CSI data set 57 SMP/E dialogs 75 LIST command 33
copy utility CYLINDERS Query dialogs 31
default values 65 for CSI data set 57 REPORT commands 34
copy utility parameter distribution libraries (DLIBs)
CMWA 197 description of 39
SPCLCMOD 197
corrective service
D zones for 40, 58
distribution zone
data element MCS
description of 38 defining 58
USERMODs 173
installation description of 11, 40
data set
ACCEPT CHECK processing 128 SYSMOD entries 28
ORDERSERVER 194
ACCEPT processing 129 DLIBZONE entry 58
data sets
APPLY CHECK process 127 DYNAM
CLIENT 194
APPLY processing 127 coexistence considerations 212
dynamically allocating 62
construct the fix 124 dynamic allocation
DATE parameter, EXEC statement for
deciding whether to accept 128 CSI parameter on EXEC statement for
GIMSMP 81
prepare 125 GIMSMP 81
DDDEF entry
RECEIVE ORDER processing 126 DDDEF entries 64
instead of DD statements in cataloged
RECEIVE processing 125 DDDEF entry 63
procedure 83
research the ACCEPT CHECK GIMDDALC 63
used for dynamic allocation 63
reports 128 migration tasks 202
DEBUG command 47
research the APPLY CHECK SMPPARM 63
default utilities used by SMP/E 65
reports 127 sources of information
default zone group
summary 123 DDDEF entries 63
defining 78
test 128 standard defaults 64
defaults
corrective service (APAR SYSMODs) 5 summary 62
for SMP/E dialogs 75
cover letters, listing 152
SMPOUT 64
cross-product load modules
SYSPRINT 64
example 148
cross-zone load modules
defining data sets E
DDDEF entry 63 END
example 148
dynamic allocation 62 EXEC statement parameter for
cross-zone requisite checking 159
PTS 62 GIMSMP 82
Cross-Zone Requisite SYSMOD
SCDS 62 enhanced link name values
report 160
SMPCSI 49 coexistence considerations 202
cross-zone requisites
DEIINST job migration tasks 202
bypassing unsatisfied 80
coexistence considerations 205, 206 entries in CSI data sets
checking for 79
deleting SYSMODs from your system DLIBZONE entry 58
resolving 80
(RESTORE command) 13, 23 GLOBALZONE entry 58
unsatisfied 79
dependent functions 38 TARGETZONE entry 58
CSI
dialogs ESO
See also SMPCSI
administration 195 format 110
API for 14
concatenating libraries 72 HOLDDATA
cataloging considerations 56
connecting to ISPF master application processing 142
defining entries
menu 74 source of 139
sample UCL statements 59
customizing 75 exception data (HOLDDATA) 12
defining zones 58
migration tasks 201 exception SYSMOD data
importing 61
overview 201 See also HOLDDATA
master CSI
default values Automated Service Delivery Package
definition of 52
panel GIM@PARM 75 source of 140
parameter
SMP/E dialogs 75 CBPDO tape
EXEC statement for GIMSMP 81
distribution libraries source of 139
reorganizing 61
in logon procedure for SMP/E 74 exception SYSMOD management
editing dialog JCL 72 ++HOLD MCS 135

exception SYSMOD management GIM@PRIM (SMP/E Primary Option HOLDDATA (continued)
(continued) Menu) 75, 215 PSP files
operands 135 GIM@UPRM processing 143
++RELEASE MCS 135 removal of 201 source of 140
operands 135 GIMDFOG summary reports 205
categories of HOLDDATA 135 ORDER RETENTION subentry 195 HOLDDATA entries
examples of 135 GIMDFUT created during RECEIVE 137
HOLDDATA removal of 201 in the global zone 16
CBPDO tape, processing 141 GIMGTPKG service routine 196
ESO, processing 142 GIMMPDFT
obtaining 139
processing example 144
migration tasks 202
GIMMPUXD
I
IBMLINK, source of HOLDDATA 140
PSP files, processing 143 migration tasks 202
IDCAMS utility
introduction 135 GIMSAMPU 59
See AMS utility
processing GIMSMP 81, 82
IEANUC01 load module 203
ACCEPT 138 GIMUNZIP 203
IEBCOPY utility
APPLY 137 GIMUTTBL
See copy utility
RECEIVE 136 migration tasks 201
IEBUPDTE utility
REJECT 139 removal of 201
See update utility
RESTORE 139 GIMXSID 198
IEWBLINK utility
exception SYSMOD report 163 GIMXTRX 205
EXEC statement GIMZIP
IEWL utility
CSI=dsname 81 archive segmentation
DATE=date 81 coexistence considerations 199
IMASPZAP utility
PARM 81, 82 overview 198
See superzap utility
PROCESS=END 82 network delivery of SMP/E input
implicitly including modules from
PROCESS=WAIT 82 coexistence considerations 204
another product 147
exit routines 86 overview 203
installation methods
migration tasks 202 user defined subdirectories
RECEIVE-APPLY-ACCEPT 101
Expanded Service Option (ESO) coexistence considerations 199
installation procedures
Recommended Service Upgrade 221 overview 199
corrective service 123
exporting a CSI data set 61 global zone
functions 101
defining 58
preventive service 109
description of 11, 40
USERMODs 131
F HOLDDATA entries 16
ORDER entry 194
installing SMP/E
FILL connecting the dialogs 71
SYSMOD entries 16, 20, 25, 28
coexistence considerations 211 installing SYSMODs
GLOBALZONE entry 58
fixing an element 5 distribution libraries (ACCEPT
function SYSMODs 3 command) 13, 27
installation target libraries (APPLY
ACCEPT CHECK processing 107 H command) 13, 19
ACCEPT processing 108 HEWLH096 utility introducing an element 3
APPLY CHECK processing 104 See link-edit utility invoking SMP/E 81
APPLY processing 106 HFSINST job ISPCTL1 72
choosing the update mode 104 coexistence considerations 205, 206 ISPCTL2 72
get additional SYSMODs 106 hierarchical file system ISPF (Interactive System Productivity
preparation 102 copy utility Facility)
RECEIVE function 103 default values 65 concatenating libraries 72
RECEIVE processing 103 hierarchical file system element MCS connecting dialogs 71, 74
research the ACCEPT CHECK coexistence considerations 207 ISPF/PDF (Interactive System
reports 108 USERMODs 174 Productivity Facility/Program
research the APPLY CHECK HOBSET Development Facility) 71
reports 105 coexistence considerations 211
summary 101 HOLDDATA 139, 140
test the new function 106
summary 37
See also exception SYSMOD
management
J
Java Archive (JAR) files
functions in a system 2 CBPDO tape
building 187
processing 141
checking effect on installed SYSMODs
in FMIDs 187
G (REPORT ERRSYSMODS) 163
ESO
in PTFs 187, 188
GENERATE command URL for 199
processing 112, 127, 142
summary 44 using 187
source of 139
generated JCL, saving for SMP/E JCL generated by SMP/E dialogs,
from IBMLINK 140
dialogs 72 saving 72
from the IBM Support Center 127
GIM@PARM (SMP/E Dialog JCLIN command
provided for SYSMODs 12
Settings) 75, 201, 215 summary 45
Index 245
JOB statement migration (continued) preventive service 109 (continued)
customizing 76 OS/390 V2R7 SMP/E summary 205 PTF SYSMODs 4
overview 191 RECEIVE ORDER requests 111
recommended steps for 192 PROCESS parameter, EXEC statement for
K SMP/E V3R1 summary 201
terminology 191
GIMSMP 82
product reinstall
keyboard 223
modification control statement (MCS) 3, See GENERATE
KEYS
15 products (function SYSMODs) 3
for CSI data set 57
modification identifiers program element MCS
function (FMID) 9 USERMODs 173
replacement (RMID) 9 program temporary fix (PTF) 4
L update (UMID) 9 programming control facility (PCF) 71
LIBDEF restrictions 72 MSGFILTER PSP files
link edit utility output coexistence considerations 209 HOLDDATA
recommended DDDEF entries 77 MSGWIDTH processing 143
LINK LMODS command coexistence considerations 209 source of 140
overview of 197 multiple-CSI zone structure 50, 53 PTF
summary 47 See also corrective service
LINK MODULE command See also preventive service
description 147
example 148
N introduction 109
Recommended Service Upgrade 221
NCAL
summary 47 summary 38
effect of CALLLIBS subentry on 66
when to use 147 PTF cover letters, listing 152
network delivery of SMP/E input
link-edit utility PTF SYSMODs 4
default values 65 PTS, summary 62
notices 225
link-editing object modules into load PUT
NUCID
modules 2 See ESO
migration tasks 203
LIST command
operand of APPLY command 203
comparing two zones 153
subentry 203
cover letters 152
description 14
Q
Query dialog
examples 33
listing a specific entry type 150 O description 14
example 31
listing an FMID or FMIDSET 152 object module, link-editing into a load
listing specific entries 151 module 2
reports produced 34 OPTIONS entry
summary 44, 149 ORDER RETENTION subentry 194 R
LISTCAT job 61 ORDER entry RACF (z/OS Security Server)
listing SMP/E data global zone 194 regulating SMP/E utility programs
API for 35 ORDER RETENTION with 201
LIST command 33 OPTIONS entry subentry 194 RC
Query dialogs 31 ORDERSERVER coexistence considerations 208
REPORT commands 34 data set 194 reason IDs 135
load module data set for SMP/E dialogs, RECEIVE command 194
LIBDEF restrictions 72 assigning source IDs to
load module, created by link-editing
object modules 2
P SYSMODs 197
PCF (programming control facility) 71
LOG command 46 description 13
PE PTF
logon procedure (TSO) 74 entries created
See exception SYSMODs
logon procedure, sample 73 HOLDDATA entry 137
prerequisites for SYSMODs 7
MCS entry 136
SYSMOD entry 137
CBPDO tapes 109
M description of 38
examples 17
functions 103
master application menu for ISPF 74 ESO tapes 110
master catalog, alias for user catalog 56, installation
processing 15
58 ACCEPT CHECK processing 119
reports produced 18
master CSI ACCEPT processing 120
summary 42
definition of 52 APPLY CHECK processing 114
USERMODs 131
specified on DD statement 83 APPLY processing 117
RECEIVE ORDER command
specified on EXEC statement 81, 83 get additional SYSMODs 117
MCS entry, created during preparation 112
summary 42
RECEIVE 136 RECEIVE processing 112
RECEIVE ORDER request
methods of installation 101 research the ACCEPT CHECK
preventive service, source of 111
migration reports 120
RECEXZGRP
OS/390 V1R2 SMP/E summary 212 research the APPLY CHECK
OS/390 V1R3 SMP/E summary 209 reports 115
OS/390 V2R5 SMP/E summary 207 test 118

Recommended Service Upgrade RMODE=SPLIT SMPDUMMY ddname (continued)
(RSU) 221 coexistence considerations 211 coexistence considerations 201
RECORDSIZE root cause analysis 183 overview 200
for CSI data set 57 RSU (Recommended Service SMPHOLD 113
recovering from utility errors 69 Upgrade) 221 SMPLTS
RECZGRP coexistence considerations 200
coexistence considerations 208 use of 41, 199
REJECT command
CHECK operand 197
S SMPMTS
use of 41
sample logon procedure 73
exception SYSMOD processing 139 SMPOUT
saving JCL generated by SMP/E
summary 45 default 64
dialogs 72
removing SYSMODs from your system SMPPROC (cataloged procedure for
SCDS, summary 62
(RESTORE command) 13, 23 SMP/E) 82
separate CSI for each SREL, combining
reorganizing a CSI 61 SMPPTS
target and DLIB zones 55
REPORT CALLLIBS command coexistence considerations 208
separate CSI for each zone 54
removal of 198 spill data sets 204
ServerPac
REPORT command use of 40
Recommended Service Upgrade 221
description 14 SMPPUNCH output
servicing a product
example 35 REPORT CROSSZONE
corrective service (APAR
reports produced 35 command 160
SYSMODs) 5
REPORT CROSSZONE command 80 REPORT ERRSYSMODS
preventive service (PTF SYSMODs) 4
introduction 159 command 163
SET command 13, 42
summary 44 SMPSCDS
SHAREOPTIONS
REPORT ERRSYSMODS command use of 41
for CSI data set 57
HOLDDATA for installed SMPSTS
shortcut keys 223
SYSMODs 163 use of 41
SIDEDECKLIB
introduction 163 SMPTABL
summary 44 description 72
single-CSI structure 50, 52
REPORT SOURCEID command space allocation 73
SMP/E cataloged procedure 82
introduction 165 source code, assembling to create an
SMP/E commands 12, 41
summary 44 object module 2
SMP/E data
REPORT SYSMODS command SPCLCMOD
changing 155
introduction 167 copy utility parameter 197
listing 149
summary 44 specifying the zone to be updated (SET
SMP/E data sets
reports command) 13
residing in UNIX file system 204
ACCEPT CHECK reports standard method of installation 101
SMPCSI 11
corrective service 128 superzap utility
SMPPTS 15
functions 108 default values 65
SMPSCDS 20
preventive service 120 SYS1.LINKLIB 205
SMPTLIB 15
APPLY CHECK reports SYS1.LPALIB 3
SMP/E dialog libraries,
corrective service 127 SYS1.MACLIB 62
concatenating 72
functions 105 SYS1.MIGLIB 3, 205
SMP/E dialogs
preventive service 115 SYS1.SAMPLIB 218
administration 195
USERMODs 132 SYS1.SVCLIB 3
customizing
Causer SYSMOD Summary SYSDEFSD
migration tasks 201
report 183 DUMMY data set for
overview 201
SYSMOD Status report 183, 184 coexistence considerations 201
query 195
RESETRC command 47 overview 200
SMP/E reports
RESTORE command SYSLIB
ACCEPT CHECK reports
description 13 concatenation 85
examples 25 SYSMOD
functions 108
exception SYSMOD processing 139 assigning source IDs to 197
processing 23 SYSMOD entries
APPLY CHECK reports
reports produced 26 created during RECEIVE 137
summary 44 distribution zone 28
functions 105
restrictions global zone 16, 20, 25, 28
CLIST data set for SMP/E dialogs 72 target zone 20, 24
USERMODs 132
dialogs 72 SYSMODs
SMP/E summary 37
LIBDEF 72 APAR 5
SMPCSI
load module data set for SMP/E definition of 37
allocating 56
dialogs 72 description 2
master CSI 83
retry processing 69 function
multiple-CSI structure 50
retry utility base 4
single-CSI structure 50
default values 65 dependent 4
summary 49
RMODE=ALIASES hierarchy 37
zones, description of 40
coexistence considerations 212 listing
SMPDUMMY ddname
REPORT SOURCEID 165
allocation rules 64
Index 247
SYSMODs (continued) USERMOD (continued) ZONEDELETE command (continued)
listing (continued) MCS statements (continued) summary 46
SMPPUNCH 165 ++USERMOD 170 ZONEEDIT command
prerequisites 7 ++VER 170 alternative to UCLIN 156
PTF 4 ++ZAP 172 summary 45
summary 37 hierarchical file system ZONEEXPORT command
USERMOD 6 element 174 alternative to UCLIN 156
SYSPRINT preventing ACCEPT processing 133 summary 46
default 64 summary 38 ZONEIMPORT command
system modifications USERMOD SYSMODs 6 alternative to UCLIN 156
See SYSMODs utility errors, recovery from 69 summary 46
utility programs ZONEINDEX
default values for zone group 79
T access method services (AMS) 65
assembler 65
ZONEMERGE command
summary 46
table data sets for dialogs 72
compress 65 ZONERENAME command
target libraries, description of 39
copy 65 alternative to UCLIN 156
target zone
hierarchical file system copy 65 summary 47
defining 58
link-edit utility 65 zones
description of 11, 40
retry 65 comparing
SYSMOD entries 20, 24
superzap 65 LIST command 153
TARGETZONE entry 58
update 65 REPORT CROSSZONE
temporary fix (APAR SYSMODs) 6
specifying which utility programs command 159
SMP/E can call 201 REPORT SYSMODS
UTIN command 167
U coexistence considerations 212 description of 40
UCLIN command zones in the SMPCSI data set
general syntax 156 See also distribution zone
introduction 155
samples in GIMSAMPU 59
V See also global zone
See target zone
VERSION command
summary 45 ZONESET entry
UNIX file system cross-zone processing 159
identification of data sets 204 defining 78
SMP/E data sets residing in 204
update utility W
default values 65 WAIT
UPGRADE command EXEC statement parameter for
coexistence considerations 198 GIMSMP 82
overview of 198
user catalog
alias in master catalog 56, 58
for CSI 56
X
XZREQCHK subentry
user-defined subdirectories
use of 78
USERMOD 133
See also user modification
creating 169
examples 174 Z
installation z/OS Security Server (RACF)
APPLY CHECK process 132 regulating SMP/E utility programs
APPLY process 133 with 201
prepare 131 zone entries
RECEIVE processing 131 impacts 194
research APPLY CHECK zone group
reports 132 default 78
summary 131 defining 78
test 133 specifying on command 78
MCS statements ZONEINDEX for 79
++element (for data zone structures
elements) 173 examples 52
++JCLIN 172 multiple CSIs 50
++MAC 172 single CSI 50
++MACUPD 172 ZONECOPY command
++MOD 172 alternative to UCLIN 156
++PROGRAM 173 summary 46
++SRC 173 ZONEDELETE command
++SRCUPD 173 alternative to UCLIN 156

Readers’ Comments — We’d Like to Hear from You
IBM SMP/E for z/OS
User’s Guide
Publication No. SA22-7773-11
We appreciate your comments about this publication. Please comment on specific errors or omissions, accuracy,
organization, subject matter, or completeness of this book. The comments you send should pertain to only the
information in this manual or product and the way in which the information is presented.
For technical questions and information about products and prices, please contact your IBM branch office, your
IBM business partner, or your authorized remarketer.
When you send comments to IBM, you grant IBM a nonexclusive right to use or distribute your comments in any
way it believes appropriate without incurring any obligation to you. IBM or any other organizations will only use
the personal information that you supply to contact you about the issues that you state on this form.
Comments:
Thank you for your support.

Submit your comments using one of these channels:
v Send your comments to the address on the reverse side of this form.
v Send your comments via e-mail to: mhvrcfs@us.ibm.com
If you would like a response from IBM, please fill in the following information:
Name Address
Company or Organization
Phone No. E-mail address

___________________________________________________________________________________________________
Readers’ Comments — We’d Like to Hear from You Cut or Fold
SA22-7773-11 Along Line
_ _ _ _ _ _ _Fold
_ _ _and
_ _ _Tape
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _Please
_ _ _ _ _do
_ _not
_ _ staple
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _Fold
_ _ _and
_ _ Tape
______
NO POSTAGE
NECESSARY
IF MAILED IN THE
UNITED STATES
BUSINESS REPLY MAIL

FIRST-CLASS MAIL PERMIT NO. 40 ARMONK, NEW YORK
POSTAGE WILL BE PAID BY ADDRESSEE
IBM Corporation
MHVRCFS, Mail Station P181
2455 South Road
Poughkeepsie, NY
12601-5400
_________________________________________________________________________________________
Fold and Tape Please do not staple Fold and Tape
Cut or Fold
SA22-7773-11 Along Line

Program Number: 5655-G44
Printed in USA
SA22-7773-11

SMPE

Uploaded by

Copyright:

Available Formats

SMPE

Uploaded by

Document Information

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

SMPE

Uploaded by

Copyright:

Available Formats

IBM SMP/E for z/OS

Twelfth Edition, April 2007

FAX (United States & Canada): 1+845+432+9405

IBMLink (United States customers only): IBMUSM10(MHVRCFS)

© Copyright IBM Corp. 1986, 2007 iii

iv SMP/E V3R4.0 User’s Guide

vi SMP/E V3R4.0 User’s Guide

© Copyright IBM Corp. 1986, 2007 vii

© Copyright IBM Corp. 1986, 2007 ix

Who should read this publication

© Copyright IBM Corp. 1986, 2007 xi

xii SMP/E V3R4.0 User’s Guide

This revision reflects the deletion, addition, or modification of information to

This revision reflects the deletion, addition, or modification of information to

Section ″ORDER RETENTION Subentry on the OPTIONS Entry″ in Appendix A,

This revision reflects the deletion, addition, or modification of information to

Chapter 4, ″Preparing to use internet service retrieval″, has been rewritten to

© Copyright IBM Corp. 1986, 2007 xiii

This revision reflects the deletion, addition, or modification of information to

This revision reflects the deletion, addition, or modification of information to

This revision reflects the deletion, addition, or modification of information to

This revision reflects the deletion, addition, or modification of information to

xiv SMP/E V3R4.0 User’s Guide

This revision reflects the deletion, addition, or modification of information to

xvi SMP/E V3R4.0 User’s Guide

The book contains information previously presented in SA22-7773-00, which

This book contains terminology, maintenance, and editorial changes. Technical

Summary of changes xvii

What is SMP/E, and why should I use it?

Understanding your system

© Copyright IBM Corp. 1986, 2007 1

v System Display and Search Facility (SDSF)

Each system function is composed of one or more load modules. In a z/OS

Figure 1. Load module creation

Changing the elements of the system

2 SMP/E V3R4.0 User’s Guide

A SYSMOD is the actual package containing information SMP/E needs to install

Introducing an element—the function SYSMOD

When we refer to installing a function SYSMOD, we are referring to the placing of

Chapter 1. SMP/E primer 3

Figure 2. Introducing an element

In this figure, the installation of a function SYSMOD link-edits object modules

There are two types of function SYSMODs:

Here’s an example of a simple function SYSMOD that introduces four elements:

Preventing or fixing problems with an element—the PTF

4 SMP/E V3R4.0 User’s Guide

Figure 3. Preventing problems with an element

In Figure 3, we see a previously installed load module, LMOD2. If we want to

Here is an example of a simple PTF SYSMOD:

Fixing problems with an element—the APAR SYSMOD

Chapter 1. SMP/E primer 5

In Figure 4, the shaded section shows an area of MOD2 containing an error.

Figure 4. Fixing problems with an element

Here is an example of a simple APAR SYSMOD:

The APAR SYSMOD always has the installation of a function SYSMOD as a

Customizing an element—the USERMOD SYSMOD

In Figure 5, MOD3 has been updated through the installation of a USERMOD

6 SMP/E V3R4.0 User’s Guide

Figure 5. Customizing an element

Here is an example of a simple USERMOD SYSMOD: