z/OS V1R8 DFSMS Technical Update

An IBM Redbook Publication
IBM Redbook Form Number: SG24-7435-00
ISBN: 073848573X
ISBN: 9780738485737
Publication Date: 23-Apr-2008
Last Update Date: 30-Apr-2008
Find Similar Download

Related People

Mary Lovelace - Author [+5] [-5]
Anthony Fletcher - Author
Daniel Perkin - Author
Norbert Schlumberger - Author
Andre Coelho - Author
Gerhard Weisshaar - Author

Abstract

Each release of DFSMS builds upon the previous version to provide enhanced storage management, data access, device support, program management, and distributed data access for the z/OS platform in a system-managed storage environment.

This IBM Redbooks publication provides a technical overview of the functions and enhancements in z/OS V1R8 DFSMS. It provides you with the information that you need to understand and evaluate the content of this DFSMS release, along with practical implementation hints and tips. Also included are enhancements that were made available through an enabling PTF that has been integrated into z/OS DFSMS V1R8.

This book was written for storage professionals and system programmers who have experience with the components of DFSMS. It provides sufficient information so that you can start prioritizing the implementation of new functions and evaluating their applicability in your DFSMS environment.

Language

English

Table of Content

Chapter 1. Something old, something new
Chapter 2. Useful information to get you started
Chapter 3. DFSMSdfp V1R8 enhancements
Chapter 4. OAM enhancements
Chapter 5. DFSMSdss enhancements
Chapter 6. DFSMShsm enhancements
Chapter 7. DFSMShsm fast replication
Chapter 8. DFSMSrmm enhancements
Chapter 9. Tape security
Appendix A. APAR text
Appendix B. Code samples
ibm.com/redbooks
Front cover
z/OS V1R8 DFSMS
Technical Update
Mary Lovelace
Anthony Fletcher
Daniel Perkin
Norbert Schlumberger
Andre Coelho
Gerhard Weisshaar
Understand the features and functions
in z/OS DFSMS V1.8
Read implementation hints and
tips
Learn DFSMS trends and
directions


International Technical Support Organization
z/OS V1R8 DFSMS Technical Update
April 2008
SG24-7435-00

© Copyright International Business Machines Corporation 2008. All rights reserved.
Note to U.S. Government Users Restricted Rights -- Use, duplication or disclosure restricted by GSA ADP Schedule
Contract with IBM Corp.
First Edition (April 2008)
This edition applies to Version 1, Release 8 of z/OS DFSMS (product number 5694-A01).
Note: Before using this information and the product it supports, read the information in “Notices” on
page ix.

© Copyright IBM Corp. 2008. All rights reserved.
iii
Contents
Notices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix
Trademarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .x
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
The team that wrote this book . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
Become a published author. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .xii
Comments welcome. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii
Chapter 1. Something old, something new. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 What was new in z/OS V1R7 DFSMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1.1 DFSMSdfp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1.2 DFSMSdss . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1.3 DFSMShsm. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.1.4 DFSMSrmm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.1.5 OAM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.2 What is new in z/OS V1R8 DFSMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.2.1 DFSMSdfp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.2.2 DFSMSdfp OAM enhancements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2.3 DFSMSdss enhancements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2.4 DFSMShsm enhancements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2.5 DFSMSrmm enhancements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Chapter 2. Useful information to get you started. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.1 Cheating with OAM. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.2 HSM fast migration reconnect users. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.3 Users of Fast Subsequent Migration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.4 Tape security. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.5 RECYCLE error - potential data loss . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.6 Management class retention limit and tape data sets. . . . . . . . . . . . . . . . . . . . . . . . . . 11
Chapter 3. DFSMSdfp V1R8 enhancements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.1 Catalog enhancements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.1.1 LISTCAT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.1.2 Dynamic service count . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.2 SMS volume selection performance enhancement. . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.3 COPY SCDS to ACDS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
3.4 SMS serviceability. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.5 VSAM code modernization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.6 RLS updates. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.6.1 RLS enhanced recovery. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
3.6.2 VSAM RLS DIAG command on z/OS V1R8 system. . . . . . . . . . . . . . . . . . . . . . . 26
3.6.3 VSAM RLS DIAG command on pre-z/OS V1R8 systems. . . . . . . . . . . . . . . . . . . 27
3.6.4 VSAM RLS performance enhancement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
3.6.5 VSAM RLS use of 64-bit data buffers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.6.6 DSS LOGICAL COPY WARNING. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
3.6.7 Monitoring the VSAM RLS use of 64-bit data buffers. . . . . . . . . . . . . . . . . . . . . . 30
3.6.8 SMF Record type 42 Subtype 16 information example. . . . . . . . . . . . . . . . . . . . . 30
3.6.9 SMF Record type 42 Subtype 18 information example. . . . . . . . . . . . . . . . . . . . . 36
3.6.10 SMF Record type 42 Subtype 19 information example. . . . . . . . . . . . . . . . . . . . 44

iv
z/OS V1R8 DFSMS Technical Update
3.7 Device manager enhancements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
3.7.1 Rapid index rebuild. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
3.7.2 New MODIFY DEVMAN parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
3.8 PDSE enhancements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
3.8.1 PDSE PO5 module format compatibility on pre-z/OS V1R8 systems. . . . . . . . . . 55
3.8.2 PDSE 64-bit virtual storage use enablement . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
3.8.3 PDSE 64-bit virtual storage option setting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
3.8.4 Retain buffers beyond PDSE close. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
3.8.5 PDSE retain buffers beyond PDSE close enablement . . . . . . . . . . . . . . . . . . . . . 62
3.8.6 DFSMS V1R8 PDSE enhancements in storage requirements. . . . . . . . . . . . . . . 66
3.9 PDSE buffer management statistics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
3.9.1 BMF data capture preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
3.9.2 BMF analysis preparation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
3.9.3 SMF statistics interpretation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
3.10 Integrated Catalog Forward Recovery Utility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
3.10.1 ICFRU system flow. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
3.10.2 ICFRU installation readiness overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
3.10.3 Installation readiness preparation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
3.10.4 Installation readiness verification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
3.10.5 ICFRU implementation final steps. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
Chapter 4. OAM enhancements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
4.1 Binary large object support. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
4.1.1 Implementing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
4.1.2 Validating . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
4.2 Immediate backup copy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
4.2.1 Implementing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
4.2.2 Validating . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
4.3 Automated selection of RECYCLE volumes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
4.3.1 Implementing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
4.3.2 Validating . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
4.4 GLOBAL display keyword. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
4.4.1 Usage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
4.5 Update from z/OS V1.7. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
Chapter 5. DFSMSdss enhancements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
5.1 Defining logical and physical processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
5.2 Physical and logical data set difference summary . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
5.3 Physical and logical data set copy specification changes. . . . . . . . . . . . . . . . . . . . . . 136
5.3.1 Logical copy operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
5.3.2 DSS LOGICAL COPY warning. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
5.3.3 Physical copy operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
Chapter 6. DFSMShsm enhancements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
6.1 Error handling on alternate duplex tapes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
6.2 Recycle SYNCDEV at intervals. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
6.3 Migration scratch queue for non-VSAM data set . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
6.3.1 Individual data set restore for ARECOVER processing . . . . . . . . . . . . . . . . . . . 152
6.4 New command ALTERPRI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
Chapter 7. DFSMShsm fast replication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
7.1 Fast replication overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
7.1.1 FlashCopy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
7.1.2 DFSMShsm fast replication. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174

Contents
v
7.1.3 Preparing for fast replication. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178
7.2 Backup and recovery of copy pools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
7.2.1 Creating a fast replication backup copy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
7.2.2 Using fast replication backups for recovery of copy pools and volumes. . . . . . . 188
7.3 Tape support. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
7.3.1 Dump of target volumes during autodump . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
7.3.2 FRBACKUP DUMP. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
7.3.3 FRBACKUP DUMPONLY. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
7.3.4 Copy pools that request NOCOPY type FlashCopy processing. . . . . . . . . . . . . 213
7.3.5 Two kinds of dump copies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218
7.3.6 Recovery from fast replication dumps. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222
7.4 Data set recovery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224
7.4.1 Data set filtering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225
7.4.2 Multi-volume data sets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228
7.4.3 Data set recovery from fast replication dumps . . . . . . . . . . . . . . . . . . . . . . . . . . 236
7.4.4 Hints and tips for the recovery of data sets that no longer exist. . . . . . . . . . . . . 239
7.5 Reporting on the DFSMShsm fast replication environment . . . . . . . . . . . . . . . . . . . . 241
7.5.1 Statistic records . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
7.5.2 REPORT command . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
7.5.3 DSR records. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244
7.5.4 Updated VSR records. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244
7.5.5 New fields in FSR records. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244
7.6 Security for DFSMShsm fast replication. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251
7.7 Audit and error recovery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253
7.7.1 FRDELETE considerations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253
7.7.2 Decreasing copy pool backup volumes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255
7.7.3 Reuse of invalid backup versions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255
7.7.4 FIXCDS display and repair . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255
7.7.5 AUDIT COPYPOOLCONTROLS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259
Chapter 8. DFSMSrmm enhancements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
8.1 Support true e-mail address for the RMM NOTIFY function. . . . . . . . . . . . . . . . . . . . 264
8.2 Support true e-mail address implementation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265
8.2.1 Basic setup. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265
8.2.2 Defining an owner’s e-mail address . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265
8.2.3 SMTP server. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266
8.2.4 E-mail message configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268
8.3 Setting up DFSMSrmm common time support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270
8.3.1 DFSMSrmm - using the date and time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271
8.3.2 Date and time in a DFSMSrmm client server environment. . . . . . . . . . . . . . . . . 271
8.3.3 Enable common time support. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272
8.3.4 Potential problems using local time. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275
8.4 DFSMSrmm VRS policy management simplification . . . . . . . . . . . . . . . . . . . . . . . . . 276
8.4.1 Separation of Data Set Name Mask from the Policy. . . . . . . . . . . . . . . . . . . . . . 276
8.4.2 Release options applied if VRS matched . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
8.4.3 Special ABEND and OPEN via DSNAME match . . . . . . . . . . . . . . . . . . . . . . . . 279
8.4.4 Find unused VRSs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280
8.4.5 Incomplete VRS chains - dummy VRS *broken*. . . . . . . . . . . . . . . . . . . . . . . . . 281
8.4.6 Toleration and removal of old functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282
8.4.7 Conversion to DFSMSrmm from other tape management systems . . . . . . . . . . 284
8.5 DFSMSrmm usability items. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286
8.5.1 Updates to RMM TSO SEARCHVOLUME subcommand. . . . . . . . . . . . . . . . . . 286
8.5.2 ISPF lists show retention information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288

vi
z/OS V1R8 DFSMS Technical Update
8.5.3 SELECT primary command in RMM dialog search results. . . . . . . . . . . . . . . . . 291
8.5.4 RMM TSO CHANGEVRS subcommand. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
8.5.5 RMM TSO SEARCHOWNER subcommand. . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
8.5.6 Rexx variable constraint relief. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 300
8.6 Enabling ISPF Data Set List (DSLIST) support. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303
8.6.1 Implementation steps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303
8.6.2 Use the ISPF Configuration Utility. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304
8.6.3 Using the ISPF Data Set List Utility support . . . . . . . . . . . . . . . . . . . . . . . . . . . . 310
8.6.4 Move the ISPCFIGU module to the SISPLPA library (optional) . . . . . . . . . . . . . 313
8.7 Prepare for future releases. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313
8.7.1 Set a DFSMSrmm control data set ID. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314
8.7.2 Re-allocate your DFSMSrmm control data set . . . . . . . . . . . . . . . . . . . . . . . . . . 315
8.7.3 Update LRECL for REPORT, BACKUP, and JRNLBKUP DD . . . . . . . . . . . . . . 317
8.7.4 Migrate from VRSEL(OLD) to VRSEL(NEW) . . . . . . . . . . . . . . . . . . . . . . . . . . . 319
Chapter 9. Tape security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 331
9.1 Tape data set authorization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332
9.1.1 Recommendations for tape security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332
9.1.2 Overview of the TAPEVOL and TAPEDSN processing . . . . . . . . . . . . . . . . . . . 333
9.1.3 How the DFSMS V1.8 tape data set authority checking works. . . . . . . . . . . . . . 335
9.2 Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 338
9.2.1 Check all high-level qualifiers on tape. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339
9.2.2 Update DEVSUPnn PARMLIB member . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 340
9.3 Removing TAPEVOL and TAPEDSN processing. . . . . . . . . . . . . . . . . . . . . . . . . . . . 342
9.3.1 Check and modify your RACF settings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 342
9.3.2 Check and modify your DFSMShsm settings . . . . . . . . . . . . . . . . . . . . . . . . . . . 346
9.3.3 Clean up your TAPEVOL profiles using DFSMSrmm settings . . . . . . . . . . . . . . 347
9.3.4 Clean up your TAPEVOL profiles using commands. . . . . . . . . . . . . . . . . . . . . . 349
9.4 Error messages. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351
9.5 Testing various security settings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353
9.5.1 Test case 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355
9.5.2 Test case 2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 358
9.5.3 Test case 3. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 360
9.5.4 Test case 4. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 363
9.5.5 Test case 5. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 365
9.5.6 Test case 6. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 368
9.5.7 Test case 7. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 371
9.5.8 Test case 8. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374
9.5.9 Test case 9. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 377
9.5.10 Test case 10. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 379
9.5.11 Test case 11. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 382
9.5.12 Test case 12. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 385
9.5.13 Test case 13. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 387
9.5.14 Test case 14. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 390
9.5.15 Test case 15. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 392
9.5.16 Test case 16. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 395
9.5.17 Test case 17. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 399
9.5.18 Test case 18. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 402
9.5.19 Test case 19. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 405
9.5.20 Test case 20. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 409
9.5.21 Test case 21. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 410
9.5.22 Test case 22. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 413
9.5.23 Test case 23. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 416

Contents
vii
9.5.24 Test case 24. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 419
9.5.25 Test case 25. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 422
9.5.26 Test case 26. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 424
9.5.27 Test case 27. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 427
9.5.28 Test case 28. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 430
9.5.29 Test case 29. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 432
9.5.30 Test case 30. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 435
9.5.31 Test case 31. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 438
9.5.32 Test case 32. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 440
9.5.33 Test case 33. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 444
9.5.34 Test case 34. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 446
9.5.35 Test case 35. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 448
9.5.36 Test case 36. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 451
Appendix A. APAR text. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 455
APARs referenced in the book. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 456
OA17704. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 456
OA11708. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 457
OA13332. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 458
OA14666. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 459
OA16191. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 459
OA16192. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 460
OA18319. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 462
OA18465. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 462
OA19493. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 464
OA20170. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 464
OA20242. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 465
OA09928. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 465
OW45264 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 466
OW43224 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 466
OA17734. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 467
OA16372. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 468
OA17415. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 469
OA20293. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 471
OA17011. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 474
Appendix B. Code samples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 477
SMF record type 42 sub type 1 data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 478
SMF record type 85 subtype 38 data display program . . . . . . . . . . . . . . . . . . . . . . . . . . . 484
SMF record type 85 subtype 39 data display program . . . . . . . . . . . . . . . . . . . . . . . . . . . 490
SMF record type 85 subtype 40 data display program . . . . . . . . . . . . . . . . . . . . . . . . . . . 496
SMF record type 85 subtype 32-35 data display program. . . . . . . . . . . . . . . . . . . . . . . . . 502
SMF record type 85 subtype 1-7 data display program. . . . . . . . . . . . . . . . . . . . . . . . . . . 513
SMF record type 42 subtype 16 data display program . . . . . . . . . . . . . . . . . . . . . . . . . . . 522
SMF Record type 42 subtype 18 data display program. . . . . . . . . . . . . . . . . . . . . . . . . . . 534
SMF record type 42 subtype 19 data display program . . . . . . . . . . . . . . . . . . . . . . . . . . . 547
Related publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 561
IBM Redbooks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 561
Other publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 561
Online resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 562
How to get IBM Redbooks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 562
Help from IBM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 562

viii
z/OS V1R8 DFSMS Technical Update
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 563

© Copyright IBM Corp. 2008. All rights reserved.
ix
Notices
This information was developed for products and services offered in the U.S.A.
IBM may not offer the products, services, or features discussed in this document in other countries. Consult
your local IBM representative for information about 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 U.S.A.
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.
Information concerning non-IBM products was obtained from the suppliers of those products, their published
announcements or other publicly available sources. IBM has not tested those products and cannot confirm the
accuracy of performance, compatibility or any other claims related to non-IBM products. Questions on the
capabilities of non-IBM products should be addressed to the suppliers of those products.
This information contains examples of data and reports used in daily business operations. To illustrate them
as completely as possible, the examples include the names of individuals, companies, brands, and products.
All of these names are fictitious and any similarity to the names and addresses used by an actual business
enterprise is entirely coincidental.
COPYRIGHT LICENSE:
This information contains sample application programs in source language, which illustrate programming
techniques on various operating platforms. You may copy, modify, and distribute these sample programs in
any form without payment to IBM, for the purposes of developing, using, marketing or distributing application
programs conforming to the application programming interface for the operating platform for which the sample
programs are written. These examples have not been thoroughly tested under all conditions. IBM, therefore,
cannot guarantee or imply reliability, serviceability, or function of these programs.

x
z/OS V1R8 DFSMS Technical Update
Trademarks
The following terms are trademarks of the International Business Machines Corporation in the United States,
other countries, or both:
Redbooks (logo) ®
z/OS®
CICS®
DB2®
DFSMS™
DFSMSdfp™
DFSMSdss™
DFSMShsm™
DFSMSrmm™
DFSORT™
DS8000™
Enterprise Storage Server®
FlashCopy®
IBM®
MVS™
OS/390®
Redbooks®
RACF®
RAMAC®
REXX™
RMF™
Sysplex Timer®
System Storage™
TotalStorage®
The following terms are trademarks of other companies:
Shared Virtual Array, and all Java-based trademarks are trademarks of Sun Microsystems, Inc. in the United
States, other countries, or both.
FRx, and the Windows logo are trademarks of Microsoft Corporation in the United States, other countries, or
both.
Other company, product, or service names may be trademarks or service marks of others.

© Copyright IBM Corp. 2008. All rights reserved.
xi
Preface
Each release of DFSMS™ builds upon the previous version to provide enhanced storage
management, data access, device support, program management, and distributed data
access for the z/OS® platform in a system-managed storage environment.
This IBM® Redbooks® publication provides a technical overview of the functions and
enhancements in z/OS V1R8 DFSMS and follow-on releases. It provides you with the
information that you need to understand and evaluate the content of these DFSMS releases,
along with practical implementation hints and tips. Also included are enhancements that were
made available through an enabling PTF that have been integrated into z/OS V1R7 DFSMS.
This publication is written for storage professionals and system programmers who have
experience with the components of DFSMS. It provides sufficient information so that you can
start prioritizing the implementation of new functions and evaluate their applicability in your
DFSMS environment.
The team that wrote this book
This book was produced by a team of specialists from around the world working at the
International Technical Support Organization, San Jose Center.
Mary Lovelace is a Consulting IT Specialist at the International Technical Support
Organization. She has more than 20 years of experience with IBM in large systems, storage
and storage networking product education, system engineering and consultancy, and
systems support. She has written many IBM Redbooks on TotalStorage® Productivity Center
and z/OS storage products.
Anthony Fletcher is an IT Specialist working with the Global Services Delivery (GSD) z/OS
software platform of IBM Global Services On Demand Infrastructure Services, based in New
Zealand but working for New Zealand and Australia. He has 36 years of experience in z/OS
and OS/390® and their predecessors and related components both as a customer of IBM and
with IBM Global Services. He is a Team Leader for the mainframe operations of four diverse
clients in the banking, airline, investment, and telecommunications industries. He holds a
degree in Electrical Engineering from SALFORD University, Lancashire, UK. His main areas
of expertise include DFSMS, DFSMSrmm™, and DFSMShsm™, and he has a working
knowledge of RACF®. He also has experience in installing non-IBM products for the GSD
platform.
Daniel Perkin is a Software Engineer with IBM DASD test organization in Tucson, Arizona.
He has nine years of experience in large systems and storage. His main areas of expertise
include z/OS Systems Programming, DASD, and copy services.
Norbert Schlumberger is an IT Architect with IBM Germany. He has 30 years of experience
in storage software and storage management for IBM and customer systems, including 18
years of experience in DFSMSrmm. He also has experience in DFSMShsm and a good
knowledge of RACF. Norbert's areas of expertise include performing conversions from
vendor tape management products to DFSMSrmm, new DFSMSrmm implementations, and
marketing support for DFSMSrmm including IBM 3494/IBM 3495 ATLs, VTSs, and vendor
robotics. He also has experience in DFSMShsm and a good knowledge of RACF. He has
worked at IBM for 33 years.

xii
z/OS V1R8 DFSMS Technical Update
Andre Coelho is an IT Storage Specialist in Rio de Janeiro, Brazil, but working for IBM West
Internal Accounts, Boulder, Colorado. He has 22 years of experience in z/OS and
predecessors. His areas of expertise include DFSMSdfp™, DFSMShsm, and DFSMSrmm.
Gerhard Weisshaar is a Senior Consultant working with Empalis group in Germany. He has
more than 20 years of experience in z/OS and its predecessors. His areas of expertise
include implementation of storage and storage networking software and hardware, as well as
z/OS Security Server. He has written extensively on the chapters in this book related to
DFSMShsm and DFSMSrmm.
Thanks to the following people for their contributions to this project:
Sangam Racherla
Emma Jacobs
International Technical Support Organization, San Jose Center
Bob Haimowitz
International Technical Support Organization, Raleigh Center
Stephen Branch
IBM San Jose
Philip Chauvet
Victor Liang
Brian Corkill
Terri Menendez
Miguel Perez
Helen Witter
Art Bariska
Sherrie Neiderbrach
Jeff Suarez
Mike Wood
IBM UK
Bernd-Rainer Bresser
Reinhard Siegel
IBM Germany
Become a published author
Join us for a two- to six-week residency program! Help write an IBM Redbook dealing with
specific products or solutions, while getting hands-on experience with leading-edge
technologies. You will have the opportunity to team with IBM technical professionals,
Business Partners, and Clients.
Your efforts will help increase product acceptance and customer satisfaction. As a bonus, you
will develop a network of contacts in IBM development labs, and increase your productivity
and marketability.
Find out more about the residency program, browse the residency index, and apply online at:
ibm.com/redbooks/residencies.html

Preface
xiii
Comments welcome
Your comments are important to us!
We want our Redbooks to be as helpful as possible. Send us your comments about this or
other Redbooks in one of the following ways:
Use the online Contact us review book form found at:
ibm.com/redbooks
Send your comments in an e-mail to:
redbooks@us.ibm.com
Mail your comments to:
IBM Corporation, International Technical Support Organization
Dept. HYTD Mail Station P099
2455 South Road
Poughkeepsie, NY 12601-5400

xiv
z/OS V1R8 DFSMS Technical Update

© Copyright IBM Corp. 2008. All rights reserved.
1
Chapter 1.
Something old, something new
This chapter provides:
A summary of the function provided by z/OS V1R7 DFSMS in “What was new in z/OS
V1R7 DFSMS” on page 2
An overview of what is new in z/OS V1R8 DFSMS in “What is new in z/OS V1R8 DFSMS”
on page 3
1

2
z/OS V1R8 DFSMS Technical Update
1.1 What was new in z/OS V1R7 DFSMS
The followings section summarize the enhancements made in various components in z/OS
V1R7 DFSMS. For details on these enhancements refer to z/OS V1R7 DFSMS Technical
Update, SG24-7225.
1.1.1 DFSMSdfp
DFSMSdfp V1R7 provided enhancements in the following areas:
Implementation of large format data sets
Implementation of the device support address space (DEVMAN)
Support for processor multiple subchannels
Provision of REPRO MERGECAT Fromkey/Tokey option
Catalog enhancements
– VVDS implicit size specification
– Automatic Catalog access tuning
VSAM data set extent constraint removal
VSAM RLS 64-bit data buffers
SMS Volume and ACS allocation test enhancements
– SMS volume status change from NOTCON by VARY command
– SMS ACS message processing enhancements
– SMS ACS environment enhancements
Extended Remote Copy Plus
DEVSERV QLIB command
PDSE restartable address space (This was introduced with V1R6 but has significant
changes by way of APAR fixes since the initial release.)
Function removals - although not enhancements in the usual sense
– ISAM removal
– JOBCAT and STEPCAT removal reminder
– VSAM Attributes removal reminder
1.1.2 DFSMSdss
DFSMSdss™ provided enhancements to support the new large format data sets.
Large format sequential data sets are now supported for the following DFSMSdss functions:
Logical COPY
Logical and physical DUMP and RESTORE
Stand-Alone RESTORE
Logical and physical RELEASE
PRINT
DEFRAG

Chapter 1. Something old, something new
3
DFSMSdss also supports the use of large format data sets for the following:
Output from logical and physical DUMP
Input to logical, physical, and Stand-Alone RESTORE
Input or output for COPYDUMP.\
Large format data sets are processed in exactly the same way as basic format data sets by
most of the commands above. The remainder of this chapter highlights how large format data
sets are processed where there is a difference between the processing of large format data
sets and basic format data sets.
1.1.3 DFSMShsm
DFSMShsm V1R7 provided enhancements to support the following:
Support for large format data sets
Fast subsequent migration improvements
Extended tape table of contents (TTOC)
Removal of ABARS requirement for INCLUDE statement
Cancellation of individual HSM tasks
Using wild cards with HMIGRATE
Saving LRECL of migrated data sets in the MCD
New recycle processing options for connected sets
Audit Media Controls Resume
Dump encryption support
1.1.4 DFSMSrmm
DFSMSrmm V1R7 provided the following enhancements:
Facility to issue DFSMSrmm TSO commands from the operator console
Improved security control over DFSMSrmm functions
Enterprise Enablement
Support for the use of large format data sets
1.1.5 OAM
DFSMSdfp V1R7 provided the following enhancements to OAM:
Tape dispatcher display
Immediate recall to DB2®
Clear old location option
Return to MVS™ scratch exit
Enhanced MOVEVOL utility
1.2 What is new in z/OS V1R8 DFSMS
Details of the new functions and enhancements in z/OS V1R8 DFSMS are provided in later
chapters, as listed below. The following sections are summaries of what is covered in the
remainder of this book.

4
z/OS V1R8 DFSMS Technical Update
1.2.1 DFSMSdfp
DFSMSdfp V1R8 has been enhanced in the following areas:
Improvements in tape data set security by new options in DEVSUPxx member
Catalog
SMS Fast Path Volume Select
SMS SCDS create ACDS
SMS Serviceability
VSAM code modernization
DFSMS RLS Enhanced recovery
This is an internal improvement that occurs automatically.
DFSMS RLS
– Problem diagnosis - See 3.6.2, “VSAM RLS DIAG command on z/OS V1R8 system” on
page 26.
– Performance enhancement - See 3.6.4, “VSAM RLS performance enhancement” on
page 27.
– SMF records for analysis - See 3.6.7, “Monitoring the VSAM RLS use of 64-bit data
buffers” on page 30.
PDSE Program Object format 5 (PO5) module format compatibility on pre-z/OS V1R8
systems
Refer to 3.8.1, “PDSE PO5 module format compatibility on pre-z/OS V1R8 systems” on
page 55, for details.
PDSE larger buffer specification
See 3.8.2, “PDSE 64-bit virtual storage use enablement” on page 55.
– PDSE_HSP-SIZE
– PDSE1_HSP_SIZE
– PDSE_DIRECTORY_SIZE
– PDSE1_DIRECTORY_SIZE
PDSE buffer hold beyond close
See 3.8.4, “Retain buffers beyond PDSE close” on page 62.
– PDSE_BUFFER_BEYOND_CLOSE
– PDSE1_BUFFER_BEYOND_CLOSE
DEVMAN Rapid Index rebuild
ICFRU
This was added to DFSMS V1R7 but not documented in z/OS V1R7 DFSMS Technical
Update, SG24-7225. For an ICFRU Readiness review example see 3.10, “Integrated
Catalog Forward Recovery Utility” on page 73.

Chapter 1. Something old, something new
5
1.2.2 DFSMSdfp OAM enhancements
DFSMS OAM V1R8 has been enhanced in the following areas:
Binary large object support
Immediate backup copy
Automated selection of RECYCLE volumes
Global display keyword
1.2.3 DFSMSdss enhancements
DFSMSdss V1R8 has been enhanced in the following areas (see 5.1, “Defining logical and
physical processing” on page 134):
DSS LOGICAL COPY
DSS extended keywords to differentiate between logical and physical operations
1.2.4 DFSMShsm enhancements
DFSMShsm V1R8 has been enhanced in the following areas:
Fast replication tape support
Data set recover from fast replication backup and dump versions
ARECOVER individual data set restore
Better handling of errors on alternate duplex tapes
Recycle SYNCHDEV at intervals
Migration scratch queue for non-VSAM data sets
New command ALTERPRI
1.2.5 DFSMSrmm enhancements
DFSMSrmm V1R8 has been enhanced in the following areas:
Improvements in tape data set security
Enterprise RMM CIMOM enhancements

6
z/OS V1R8 DFSMS Technical Update

© Copyright IBM Corp. 2008. All rights reserved.
7
Chapter 2.
Useful information to get you
started
This chapter contains useful information that we found by trial and error, including common
errors and undocumented tidbits of information.
2

8
z/OS V1R8 DFSMS Technical Update
2.1 Cheating with OAM
After implementing the immediate backup and automated selection of RECYCLE volumes
functions, you may want to verify that it works, but you might not have any tape volumes that
are actually full. Here is one method that could serve to verify both new functions:
1.Set SMS up so that an immediate backup occurs to tape.
2.Run OSREQ STORE to store an object.
3.Observe the tape mount and the immediate back up of data to the first backup.
4.For the volume where the backup data was written, issue the command:
MODIFY OAM,UPDATE,VOLUME,xxxxxx,FULL,Y
5.Run OSREQ STORE to store a second object.
6.Observe the tape mount and the immediate back up of data.
7.Issue the command MODIFY OAM,START,RECYCLE,(ALLBK1),DISPLAY,PV=100.
You will see the volume you marked as full.
8.Issue the command MODIFY OAM,UPDATE,VOLUME,xxxxxx,FULL,N for the volume you
had previously marked full.
2.2 HSM fast migration reconnect users
There is a problem affecting users of the HSM fast reconnect to migrated copies of data sets
that in certain circumstances can lead to loss of data. Refer to “OA19493” on page 464, for
further information.
2.3 Users of Fast Subsequent Migration
There is a possibility of data loss if DFSMSdss and DFSMShsm with Fast Subsequent
Migration (FSM) is used on any level of OS390 V2.10 or z/OS V1.1 to V1.6.
From z/OS V1.7 or later there is no problem if all use of DFSMSdss is through the API. This
means that DFSMShsm use and any other use through the API is not a problem, and a
problem only arises if DFSMSdss is used through JCL/BATCH.
If you fall into the affected group, or if in doubt, you should consider the following:
Users of DFSMSdss together with DFSMShsm FSM should refer to the current version of
Information APAR OA20117.
This APAR describes a situation where a data set's change bit could be turned off if
DFSMSdss is used to restore the data set outside the control of DFSMShsm, and
DFSMShsm is configured to use FSM.
This is because DFSMSdss, when run using JCL to invoke it, always turns the data set
changed bit off on any data set restored.

Chapter 2. Useful information to get you started
9
This could result in what should be a new migration being skipped when DFSMShsm
considers the data set for migration (see Figure 2-1). If the bit is turned off when the data
set is evaluated, and a migrated copy of the data set is found in the DFSMShsm MCDS, it
may consider that to be a suitable migrated copy, and will revalidate the entry and delete
the data set from DASD. A subsequent recall would then find an old copy of the data set.
Figure 2-1 FSM potential data loss scenario
Using DFSMShsm to manage data set restores manages the situation properly even if, as
is the case, the migrated copy was moved by DFSMSdss.
If in doubt as to whether all data set restores are under the control of DFSMShsm, you
should consider suspending the use of DFSMShsm FSM until resolution is available.
Apar OA20907 has been opened to provide a patch option for DFSMSdss that changes
the behavior of DFSMSdss to
not
reset the changed bit if it was found to be on in the
restored copy. This makes its behavior consistent with the invocation of DFSMSdss via
DFSMShsm. You should monitor the status of APAR OA20907.
2.4 Tape security
DFSMS support introduces new options for securing tape data sets using the System
Authorization Facility (SAF). These are designed to allow you to define profiles to protect data
sets on tape using the DATASET class without the need to activate the TAPEDSN option or
the TAPEVOL class. DFSMS also provides options that you can use to specify that all data
sets on a tape volume should have common authorization and that users are authorized to
overwrite existing files on a tape volume.

10
z/OS V1R8 DFSMS Technical Update
The new options were introduced as a result of customer requirements. So, it is time to
rethink the way that you are currently protecting data on tape.
There are new options available for use in the DEVSUPxx member of PARMLIB to change
the behavior of open data set security checking in case of tape data sets. Refer to Chapter 9,
“Tape security” on page 331, for details.

Chapter 2. Useful information to get you started
11
2.5 RECYCLE error - potential data loss
There is a red alert about potential data loss after a RECYCLE error on DFSMShsm V1R8.
Figure 2-2 contains the description of the red alert. The APAR for this problem is APAR
OA18465. Refer to “OA18465” on page 462 for additional information.
Figure 2-2 Output about DFSMShsm red alert
2.6 Management class retention limit and tape data sets
After OA17011 is applied, tape volumes are prematurely released during RMM housekeeping
because the special expiration date explicitly specified in the JCL by the user is overridden
RED ALERT: V1R8 DFSMShsm potential data loss after RECYCLE error.
ABSTRACT:
POTENTIAL DATA LOSS AFTER RECYCLE ERROR.
DESCRIPTION:
A problem was detected in the V1R8 DFSMShsm RECYCLE function which under
certain circumstances, can potentially result in data loss. The problem occurs
only when the recycle function encounters certain errors with its input tape.
The most common type of error would be the failure to mount the input tape,
resulting in the failure of the recycle function with message ARC0833I rc31.
The problem detected causes DFSMShsm to lose knowledge of the logical EOF on
the output tape. As a result, the next time that DFSMShsm attempts to write to
the recycle output tape, it will write from load point, overwriting data
previously written to the tape. Attempts to recall/recover the overwritten data
sets will fail with ARC1001I rc68 reas16.
THE PTF for APAR OA18465/UA29904 will cause the recycle processing to maintain
the logical EOF information for its output tape. Error scenarios that would
have caused DFSMShsm to lose knowledge of the logical EOF on the recycle output
tape will now result in an abend of the recycle task. Although the attempt to
recycle the input tape will fail, the potential for the output tape being
subsequently overwritten is eliminated.
V1R8 customers are strongly advised to avoid running the RECYCLE function on
V1R8 DFSMShsm until the PTF for APAR OA18465/UA29904 has been installed. Please
see APAR OA18465 for more information.
The DFSMShsm development team has created a tape assessment tool for those
customers that have run RECYCLE already. Information on the tool can be found
in APAR OA18465.
RECOMMENDED ACTIONS:
Do not run RECYCLE on a V1R8 DFSMShsm prior to installing PTF UA29904 for
OA18465. All installations who have run RECYCLE on a V1R8 DFSMShsm are
requested to mark any partial ML2 and backup tapes full using the DELVOL volser
MIGRATION(MARKFULL) or DELVOL volser BACKUP(MARKFULL) command, as appropriate.
NOTE: DFSMShsm RECYCLE can be run safely on systems with z/OS V1R7 and below.

12
z/OS V1R8 DFSMS Technical Update
when the management class assigned to the tape data set specifies Retention Limit = 0.
Management class attributes should not apply to tape data sets.
This error is addressed by APAR OA20293. Refer to “OA17011” on page 474 and “OA20293”
on page 471 for details.

© Copyright IBM Corp. 2008. All rights reserved.
13
Chapter 3.
DFSMSdfp V1R8 enhancements
In this chapter we discuss new and changed functions in DFSMSdfp. The following topics are
covered:
Catalog enhancements
SMS volume selection performance enhancement
Copy SCDS to ACDS
SMS serviceability
VSAM code modernization
RLS updates
Device manager enhancements
PDSE enhancements
PDSE buffer management statistics
ICFRU
3

14
z/OS V1R8 DFSMS Technical Update
3.1 Catalog enhancements
z/OS V1R8 introduces improvements in LISTCAT processing and output as well as the ability
to specify the number of catalog address space requests available for user requests.
3.1.1 LISTCAT
IDCAMS LISTCAT processing in z/OS V1R8 has been enhanced to provide better
performance, especially for large catalogs. The performance improvement is automatic in
V1R8, and no action needs to be taken in order to exploit the new function.
Examples
A new header line appears on each page of LISTCAT output. Additionally, there have been
changes in the way LISTCAT LEVEL processing works for GDGs and ALIASes. Figure 3-1
shows an example of pre-z/OS V1R8 LISTCAT LEVEL output, specifying the ALIAS name in
the LISTCAT LEVEL command.
Figure 3-1 Pre-z/OS V1R8 LISTCAT output
Important: The output from LISTCAT commands has also changed, which might affect
products that process the output.
LISTCAT LEVEL(SYSDOC)
GDG BASE ------ SYSDOC.CIMN.SYSLOGD
IN-CAT --- COMCIC.ICFCAT
NONVSAM ------- SYSDOC.CIMN.SYSLOGD.G0832V00
IN-CAT --- COMCIC.ICFCAT
.
.
.
IDCAMS SYSTEM SERVICES TIME:
17:06:24
THE NUMBER OF ENTRIES PROCESSED WAS:
AIX -------------------0
ALIAS -----------------0
CLUSTER ---------------0

Chapter 3. DFSMSdfp V1R8 enhancements
15
Figure 3-2 shows an example of z/OS V1R8 LISTCAT LEVEL output illustrating the header
and ALIAS changes.
Figure 3-2 z/OS V1R8 LISTCAT output
Migration and coexistence
Information APAR II14250 contains details about the differences in IDCAMS LISTCAT
processing in z/OS V1R8. This is generally not a consideration unless, for example, you have
a vendor product that processes LISTCAT output.
Maintenance
We recommend that you have the PTFs for APARs OA16912, OA18720, and OA20169
applied prior to using the new support. In addition, the PTF for APAR OA16372 should be
applied.
OA16912 and OA18720 represent the roll-up APARs for this new support. APAR OA20169 is
the APAR resulting from OA18184’s PTF being marked PE. It is not closed at the time of this
writing, but is important because it fixes the PE introduced with OA18184 as well as the issue
OA18184 was intended to fix, where LISTCAT was not issuing error messages for offline
volumes.
OA16372 represents VSAM GET/PUT requests failing when accessing a VSAM data set and
the data set has more than three volumes that have been recataloged out of the original
extend order. Refer to “OA16372” on page 468 for a full description.
3.1.2 Dynamic service count
z/OS V1R8 introduces the ability to alter the number of catalog address space service tasks
available for user requests. An insufficient number of service tasks is usually indicated by
running a MODIFY CATALOG,REPORT command and observing a HIGHEST # OF
SERVICE TASKS value approaching the SERVICE TASK UPPER LIMIT value. Exceeding
the limit results in performance problems.
LISTCAT LEVEL(SYSDOC)
LISTING FROM CATALOG -- COMCIC.ICFCAT
GDG BASE ------ SYSDOC.CIMN.SYSLOGD
IN-CAT --- COMCIC.ICFCAT
NONVSAM ------- SYSDOC.CIMN.SYSLOGD.G0832V00
IN-CAT --- COMCIC.ICFCAT
.
.
.
IDCAMS SYSTEM SERVICES TIME:
17:04:40
LISTING FROM CATALOG -- SCI18A.DASDPLEX.CATALOG
ALIAS --------- SYSDOC
IN-CAT --- SCI18A.DASDPLEX.CATALOG
IDCAMS SYSTEM SERVICES TIME:
17:04:40
LISTING FROM SELECTED CATALOGS
THE NUMBER OF ENTRIES PROCESSED WAS:
AIX -------------------0
ALIAS -----------------1
CLUSTER ---------------0

16
z/OS V1R8 DFSMS Technical Update
The default number of service tasks is 200 and the maximum potential number is 999. Of
these, 90% are used as the maximum number of user catalog requests that can be
processed concurrently. Figure 3-3 shows an example of the MODIFY CATALOG,REPORT
command using the default value. It shows a value of 180 for the SERVICE TASK UPPER
LIMIT, which is 90% of 200. In this case the HIGHEST # OF SERVICE TASKS value is only
16, so there is no need to change the default upper limit value.
Figure 3-3 CATALOG REPORT showing default value
If the HIGHEST # OF SERVICE TASKS does approach or exceed the 180 boundary, we use
this new function to change the upper limit value.
Implementing
The SYSCAT statements in the LOADxx member of PARMLIB cannot be used to change the
default. A SYSCATxx member in SYS1.NUCLEUS must be used. Many installations
presently use the LOADxx member for SYSCAT statements, so a new SYSCATxx member
might have to be allocated. We have provided an example of a job to allocate the SYSCATxx
member in Figure 3-4. The number of service tasks defined is specified in columns 65–67 of
the SYSCATxx member. This example changes the service task count to 999.
Figure 3-4 Example SYSCATxx member
More information about the format and use of the SYSCATxx member can be found in
DFSMS Managing Catalogs, SC26-7409.
MODIFY CATALOG,REPORT
IEC351I CATALOG ADDRESS SPACE MODIFY COMMAND ACTIVE
IEC359I CATALOG REPORT OUTPUT 831
*CAS************************************************************
* CATALOG COMPONENT LEVEL = HDZ1180 *
* CATALOG ADDRESS SPACE ASN = 0034 *
* SERVICE TASK UPPER LIMIT = 180 *
* SERVICE TASK LOWER LIMIT = 60 *
* HIGHEST # SERVICE TASKS = 16 *
* CURRENT # SERVICE TASKS = 16 *
* MAXIMUM # OPEN CATALOGS = 1,024 *
----+----1----+----2----+----3----+----4----+----5----+----6----+----7--
********************************* Top of Data **************************
//SYSCAT JOB CLASS=A,MSGCLASS=T,NOTIFY=&SYSUID,REGION=6M
//SYSCAT EXEC PGM=IEBGENER
//SYSPRINT DD SYSOUT=*
//SYSIN DD DUMMY
//SYSUT2 DD DSN=SYS1.NUCLEUS(SYSCATLG),DISP=SHR,DCB=(RECFM=U)
//SYSUT1 DD *
SBOX0011 MCAT.SANDBOX.Z18.SBOX00 Y 999
//*
******************************** Bottom of Data ************************

Chapter 3. DFSMSdfp V1R8 enhancements
17
After we have removed the SYSCAT statements from LOADxx and IPLed using the new
SYSCATxx member we added to PARMLIB, we issue another MODIFY CATALOG,REPORT
command. Figure 3-5 is a catalog report that now shows 900, or 90% of the 999 figure we
specified in SYSCATxx.
Figure 3-5 CATALOG REPORT showing maximum value
Migration and coexistence
If you decide to IPL a pre-V1R8 level after making these changes, you first need to undo the
changes you made to SYSCATxx and LOADxx when implementing this V1R8 support.
3.2 SMS volume selection performance enhancement
SMS has been enhanced with a new function to speed up volume selection. In storage
groups with hundreds or thousands of candidate volumes, it can take an unacceptably long
time for SMS to determine a target for placement of a new data set.
Turning fast path volume selection on causes SMS to select volumes normally until DADSM
rejects 100 volumes for insufficient free space. It then excludes all volumes that do not have
sufficient free space in the current SMS volume statistics, which has the effect of reducing the
overall number of candidate volumes and therefore the number of attempts SMS must make
in order to find a suitable target. This is not a function that is normally exercised often. It is a
performance improvement for the worst-case scenario when you have a lot of volumes and
the majority of them are nearly full. This worst-case scenario is more likely to occur when you
have a large number of volumes in a single storage group. That type of environment benefits
the most from this new function.
MODIFY CATALOG,REPORT
IEC351I CATALOG ADDRESS SPACE MODIFY COMMAND ACTIVE
IEC359I CATALOG REPORT OUTPUT 998
*CAS************************************************************
* CATALOG COMPONENT LEVEL = HDZ1180 *
* CATALOG ADDRESS SPACE ASN = 0034 *
* SERVICE TASK UPPER LIMIT = 900 *
* SERVICE TASK LOWER LIMIT = 60 *
* HIGHEST # SERVICE TASKS = 21 *
* CURRENT # SERVICE TASKS = 21 *
* MAXIMUM # OPEN CATALOGS = 1,024 *

18
z/OS V1R8 DFSMS Technical Update
There are two limits to using fast volume selection:
It does not apply to striping allocation. Striping allocation has already excluded the
volumes above the high threshold from volume selection.
It may inadvertently exclude volumes that have sufficient free space but for which the
SMS volume statistics indicate that they do not, as described in the MVS Initialization and
Tuning Reference, SA22-7592, and shown in Figure 3-6.
Figure 3-6 Occasions for bypass due to SMS volume statistics
Implementing
Fast volume selection is requested by using the new FAST_VOLSEL parameter. This can be
set on or off by:
Using the IGDSMSxx member of PARMLIB
Issuing the SETSMS FAST_VOLSEL command
An example of setting fast volume selection to on using the IGDSMSxx member of PARMLIB
is provided in Figure 3-7:
Figure 3-7 Placing the FAST_VOLSEL parameter in PARMLIB
Invalid volume statistics can occur as a result of the following events:

The VTOC index is broken.

OEM products bypass CVAF processing.

In an SMSplex when the SMS synchronization time interval has not
yet been driven to update the SMS configuration with the most
current space statistics. These statistics are based on updates
that can occur on another system in the SMSplex.
SYS1.PARMLIB(IGDSMS03) - 01.04
===>
***************************** Top of Data ********
SMS ACDS(SYS1.SMS.ACDS)
COMMDS(SYS1.SMS.COMMDS)
INTERVAL(15)
DINTERVAL(150)
DEADLOCK_DETECTION(15,4)
FAST_VOLSEL(ON)
SMF_TIME(YES)
CF_TIME(1800)
RLSINIT(YES)
RLS_MAX_POOL_SIZE(100)
REVERIFY(NO)

Chapter 3. DFSMSdfp V1R8 enhancements
19
An example of setting fast volume selection on using the SETSMS FAST_VOLSEL command
is provided in Figure 3-8. The command does not come back with a message indicating that it
is complete. You can verify its present state by issuing the
D SMS,OPTIONS command, which is also shown in Figure 3-8.
Figure 3-8 Turning FAST_VOLSEL(ON) with the SETSMS command and verifying state
SETSMS FAST_VOLSEL(ON)
.
.
.
D SMS,OPTIONS
IGD002I 17:01:47 DISPLAY SMS 253
ACDS = SYS1.SMS.ACDS
COMMDS = SYS1.SMS.COMMDS
INTERVAL = 15 DINTERVAL = 150
SMF_TIME = YES CACHETIME = 3600
CF_TIME = 1800 PDSE_RESTARTABLE_AS = YES
PDSE_BMFTIME = 3600 PDSE1_BMFTIME = 3600
PDSE_LRUTIME = 60 PDSE1_LRUTIME = 50
PDSE_LRUCYCLES = 15 PDSE1_LRUCYCLES = 200
LOCAL_DEADLOCK = 15 GLOBAL_DEADLOCK = 4
REVERIFY = NO DSNTYPE = PDS
.
.
.
PDSE_BUFFER_BEYOND_CLOSE = NO
PDSE1_BUFFER_BEYOND_CLOSE = NO
GDS_RECLAIM = YES DSSTIMEOUT = 0
BLOCKTOKENSIZE = NOREQUIRE FAST_VOLSEL = ON
IGD002I 17:01:47 DISPLAY SMS
TRACE = ON SIZE = 128K TYPE = ALL
JOBNAME = * ASID = *
TRACING EVENTS:
MODULE = ON SMSSJF = ON SMSSSI = ON ACSINT = ON
OPCMD = ON CONFC = ON CDSC = ON CONFS = ON
MSG = ON ERR = ON CONFR = ON CONFA = ON
ACSPRO = ON IDAX = ON DISP = ON CATG = ON
VOLREF = ON SCHEDP = ON SCHEDS = ON VTOCL = ON
VTOCD = ON VTOCR = ON VTOCC = ON VTOCA = ON
RCD = ON DCF = ON DPN = ON TVR = ON
DSTACK = ON UAFF = ON DEBUG = ON
VOLSELMSG = (OFF,0) TYPE = ALL JOBNAME = *
ASID = * STEPNAME = *
DSNAME = *

20
z/OS V1R8 DFSMS Technical Update
Once FAST_VOLSEL is turned on, you see a new IGD17294I message when fast volume
selection is used. An example of this appears in Figure 3-9, which is the output from an
IEFBR14 job allocating a data set in a space-constrained storage group.
Figure 3-9 Fast volume selection in action
Migration and coexistence
If you IPL a pre-V1R8 driver you need to remove FAST_VOLSEL from the IGDSMSxx
member of PARMLIB if you included it while implementing z/OS V1R8.
3.3 COPY SCDS to ACDS
This enhancement allows users to create an ACDS from any valid SCDS without first having
to activate the ACDS. An example of its use is be to create an ACDS for use at a separate
disaster recovery location.
Preallocating the ACDS
One of the requirements for this command is that the target ACDS for the COPYSCDS
command must be pre-allocated. In Figure 3-10 we provide an example of an IDCAMS job to
allocate a 60-track ACDS. Check the size of your existing SCDS to determine how many
tracks your target ACDS should be.
Figure 3-10 Pre-allocating your ACDS
Using the new COPYSCDS command
The requirements are:
The ACDS must be pre-allocated.
The source SCDS should be validated before being copied.
IEF236I ALLOC. FOR IEFBR14 D301
IGD17294I FAST VOLUME SELECTION IS USED TO ALLOCATE DATA SET
MHLRES3.MHLSMS.XLD301
IGD101I SMS ALLOCATED TO DDNAME (SEQOUTG )
DSN (MHLRES3.MHLSMS.XLD301 )
STORCLAS (MHLSMS) MGMTCLAS ( ) DATACLAS ( )
VOL SER NOS= SLD14D
IEF142I IEFBR14 D301 - STEP WAS EXECUTED - COND CODE 0000
IGD104I MHLRES3.MHLSMS.XLD301 RETAINED,
DDNAME=SEQOUTG
IEF373I STEP/D301 /START 2007052.1126
IEF374I STEP/D301 /STOP 2007052.1126 CPU 0MIN 00.00SEC SRB 0MIN
00.00S
//ALLCACDS JOB CLASS=A,MSGCLASS=T,NOTIFY=&SYSUID,REGION=6M
//STEP EXEC PGM=IDCAMS
//SYSUDUMP DD SYSOUT=*
//SYSPRINT DD SYSOUT=*
//SYSIN DD *
DEFINE CLUSTER(NAME(MHLRES3.TEST.ACDS) LINEAR VOL(SBOX11) -
TRK(60 60) SHAREOPTIONS(2,3)) -
DATA(NAME(MHLRES3.TEST.ACDS.DATA)REUSE)

Chapter 3. DFSMSdfp V1R8 enhancements
21
SMS automatically verifies that:
The acds_dsn is not the currently active ACDS.
The scds_dsn is a valid SCDS.
If you do break any of these rules, SMS issues an IGD088I message, as in the case where
we attempt to copy our SCDS into the active ACDS, as seen in Figure 3-11.
Figure 3-11 Example of the IGD088I message with COPYSCDS failure
The command format for the new COPYSCDS command is:
SETSMS COPYSCDS(scds_dsn,acds_dsn)
Where scds_dsn is the name of the SCDS that is copied to the target ACDS acds_dsn.
Figure 3-12 is an example of the SYSLOG output that you should expect to see after
successfully issuing this command.
Figure 3-12 SYSLOG output from a successful COPYSCDS command
Migration and coexistence
There are no migration or coexistence considerations for the COPYSCDS support.
3.4 SMS serviceability
A new DEBUG parameter has been added to the SETSMS TRACE command to add trace
points at strategic locations and reduce the size of SMS trace data and therefore the amount
of time needed to analyze the data. This serves to reduce the total time necessary for
problem resolution.
This parameter should only be used at the direction of the system programmer when
engaged in problem determination.
SETSMS COPYSCDS(SYS1.SMS.SCDS,SYS1.SMS.ACDS)
IGD088I COPYSCDS COMMAND FAILED SCDS SYS1.SMS.ACDS 313
- REASON CODE 6135
SETSMS COPYSCDS(SYS1.SMS.SCDS,MHLRES3.TEST.ACDS)
IEF196I IEF237I 831A ALLOCATED TO SYS00016
IEF196I IEF237I D30C ALLOCATED TO SYS00017
IGD019I SCDS SYS1.SMS.SCDS SUCCESSFULLY COPIED AS ACDS TO 243
MHLRES3.TEST.ACDS
IEF196I IGD104I MHLRES3.TEST.ACDS RETAINED,
IEF196I DDNAME=SYS00017
IEF196I IGD104I SYS1.SMS.SCDS RETAINED,
IEF196I DDNAME=SYS00016

22
z/OS V1R8 DFSMS Technical Update
Using the DEBUG parameter
The new DEBUG parameter is set with the SETSMS TRACE command and is displayed with
the D SMS,TRACE command. IBM provides the specific commands at problem determination
time. An example is provided in Figure 3-13.
Figure 3-13 Turning the DEBUG parameter ON and displaying its status
Migration and coexistence
There are no migration or coexistence considerations when using this new function.
3.5 VSAM code modernization
VSAM code has been modernized so that it automatically dumps for errors that normally
indicate a system problem or invalid data set. No action is required to implement the
automatic dumps.
Additionally, the current procedure for collecting data for a specific VSAM return code is
complex and prone to errors. New parameters for the MODIFY CATALOG command have
been added to assist with the first-time data capture of user-specified VSAM problems.
SETSMS DESELECT(ALL)
IEE712I SETSMS PROCESSING COMPLETE
SETSMS TRACE(ON) TYPE(ALL) SIZE(1M) SELECT(DEBUG)
IEE712I SETSMS PROCESSING COMPLETE
D SMS,TRACE
IGD002I 12:23:29 DISPLAY SMS 020
TRACE = ON SIZE = 1024K TYPE = ALL
JOBNAME = * ASID = *
TRACING EVENTS:
MODULE = OFF SMSSJF = OFF SMSSSI = OFF ACSINT = OFF
OPCMD = OFF CONFC = OFF CDSC = OFF CONFS = OFF
MSG = OFF ERR = OFF CONFR = OFF CONFA = OFF
ACSPRO = OFF IDAX = OFF DISP = OFF CATG = OFF
VOLREF = OFF SCHEDP = OFF SCHEDS = OFF VTOCL = OFF
VTOCD = OFF VTOCR = OFF VTOCC = OFF VTOCA = OFF
RCD = OFF DCF = OFF DPN = OFF TVR = OFF
DSTACK = OFF UAFF = OFF DEBUG = ON
VOLSELMSG = (OFF,0) TYPE = ALL JOBNAME = *
ASID = * STEPNAME = *
DSNAME = *

Chapter 3. DFSMSdfp V1R8 enhancements
23
Usage
The new VDUMPON and VDUMPOFF parameters for the MODIFY CATALOG command, as
described in DFSMS Managing Catalogs, SC26-7409, can be seen in Figure 3-14.
Figure 3-14 The new VDUMPON and VDUMPOFF parameters
Example
In order to demonstrate use of this new function, we wrote an IDCAMS job that attempts to
delete a non-existant VSAM data set, which results in a RC 8 ERROR 16 VSAM RECORD
NOT FOUND failure. The example job is shown in Figure 3-15 and the SYSLOG output for
this is shown in Figure 3-16 on page 24.
Figure 3-15 Deleting a non-existant VSAM data set
MODIFY CATALOG,VDUMPON(pdf,rc,compid,error)
pdf Specifies the VSAM Problem Determination Function code (one
to three characters from 0 to 255), or * (asterisk).
rc Specifies the VSAM return code in decimal format (one to
three characters from 0 to 255), or * (asterisk).
compid Specifies the component code (0 - 5), or * (asterisk).
error Specifies the VSAM error code in decimal format (one to
three characters from 0 to 255), or * (asterisk).
Notes:

1. Specifying an asterisk (*) for any parameter indicates that a wild
card search is to be done for that value. Up to three asterisks
(without intervening spaces) may be specified where a single
asterisk is allowed, but the extra asterisks have no effect on the
command's output.
2. If a parameter is to be omitted, it must be specified as an
asterisk (*). For example, VDUMPON(*,*,*,2) is syntactically
valid, while VDUMPON(2), VDUMPON( , , ,2), and VDUMPON(2, ,2) are
all invalid.
3. At least one of the parameters must be specified with a value
other than asterisks. For example, VDUMPON(*,*,*,*) is not
allowed.
4. User-initiated VSAM dynamic dumping does not occur unless you
specify VDUMPON.
5. You can set only one VDUMPON at a time. Each entry overwrites the
previous information. After a match occurs, the information is
cleared and no further user-initiated dumps will be taken.
.
.
.
Issuing MODIFY CATALOG,VDUMPOFF will clear the VDUMP options.
//DELVSM JOB MSGCLASS=H,REGION=0M
//DELSTG EXEC PGM=IDCAMS
//SYSPRINT DD SYSOUT=*
//SYSIN DD *
DELETE SLT.DATASET.DOESNT.EXIST CLUSTER PURGE

24
z/OS V1R8 DFSMS Technical Update
Figure 3-16 Example VSAM dump for VSAM record not found
Migration and coexistence
There are no migration or coexistence considerations when using this new function.
3.6 RLS updates
In z/OS V1R8 the following VSAM RLS features were introduced:
The ability for users to detect VSAM RLS latch contention has been introduced with the
VSAM RLS DIAG command (see “VSAM RLS DIAG command on z/OS V1R8 system” on
page 26).
A performance enhancement is provided to speed up the process of terminating the
SMSVSAM address spaces across the members of a SYSPLEX (see “VSAM RLS
performance enhancement” on page 27).
z/OS V1R7 64-bit (2 GB bar) implementation continuation: The SMF recording of use of
64-bit buffers that was intended for delivery in z/OS V1R7 was held over and is now available
in V1R8. See 3.6.5, “VSAM RLS use of 64-bit data buffers” on page 28.
MODIFY CATALOG,VDUMPON(*,8,*,16)
IEC351I CATALOG ADDRESS SPACE MODIFY COMMAND ACTIVE
IEC359I CATALOG VSAM DUMP OPTIONS 957
*CAS*************************************************************
* STATUS FUNC CODE RETURN CODE COMPONENT ERROR CODE *
* ON *** 008 *** 016 *
*CAS*************************************************************
IEC352I CATALOG ADDRESS SPACE MODIFY COMMAND COMPLETED
$HASP100 DELVSM ON INTRDR FROM TSU13819
MHLRES3
IRR010I USERID MHLRES3 IS ASSIGNED TO THIS JOB.
ICH70001I MHLRES3 LAST ACCESS AT 11:22:32 ON MONDAY, MARCH 12, 2007
$HASP373 DELVSM STARTED - INIT 1 - CLASS B - SYS CIMN
IEA794I SVC DUMP HAS CAPTURED: 963
DUMPID=001 REQUESTED BY JOB (CATALOG )
DUMP TITLE=VSAM DYNAMIC RPL DUMP - IDA019RA+1D8A FEEDBACK CODE:
0A080010
.
.
.
IEA611I COMPLETE DUMP ON SYSDOC.DUMP.DA.D070312.T112257.CIMN.S00001 975
DUMPID=001 REQUESTED BY JOB (CATALOG )
Important: Before using VSAM RLS in a DFSMS V1.8 environment you should review
OA17415. All users of VSAM RLS reading or writing to a multi-volume VSAM data set
where the volumes have been recataloged in a different order than originally extended to
are affected. Refer to “OA17415” on page 469 for more details.

Chapter 3. DFSMSdfp V1R8 enhancements
25
3.6.1 RLS enhanced recovery
In order to talk about RLS enhanced recovery, we should discuss VSAM record level sharing
(RLS). It is a data set access mode that allows multiple address spaces, CICS® application
owning regions on multiple z/OS systems, DFSMShsm control data sets access, DFSMStvs,
and jobs to access data at the same time. VSAM RLS requires that the data sets be System
Managed Storage (SMS) data sets.
The purpose of the RLS function is to enhance access and recovery capabilities to existing
VSAM files. One particular area of enhancement is in the area of reliability, availability, and
serviceability (RAS). You can divide that into two separate but related areas:
First time data capture - to document internal logic errors at the earliest point in time
possible. It consists of any system failure introduced by coding errors or
bugs
.
System error recovery - to recover from an identified system error. It can be identified by a
health check or by error detecting features on the base operating system. It will try to
recover as much as possible, minimizing the impact to the customer. The highest priority is
to prevent damage to customer data. After that, it can minimize a failing function and
automate steps to restore the system to full functionality.
RLS works using a VSAM address space named SMSVSAM. Using it, the requesting
program gains control through cross memory and data access.
The original RLS design included extensive first time data capture. Each of the RLS
subcomponents added many health checks, as well as local recovery routines, which detect
and document internal logic errors. If a health check identified an error, an SVC D command
is issued with a 0F4 completion code along with a unique return/reason code identifying the
error. The local recovery routine would then receive control and issue an SDUMP macro to
obtain an SVC dump. This design has proven very successful and has enabled many internal
logic errors to be solved with the first occurrence of the problem.
The original RLS design also included system error recovery. Following an internal logic error,
most of the RLS subcomponents opted to terminate the SMSVSAM address space in order to
prevent possible data integrity problems and correct environmental errors. By terminating the
SMSVSAM address space, the
damaged
system environment is cleared and a new
clean

SMSVSAM address space is created. Additionally, the process of terminating and restarting
the SMSVSAM address space was automated so that the customer does not have to
manually perform these steps. The first exploiter of this resource, CICS, had added some
automation to this specific case, receiving system signs of individual regions, so it can act
accordingly, closing and reopening files on the failing system.
This idea of terminating the SMSVSAM address space because of environmental problems
has not proven to be the best approach. It provided protection, but at the same time,
produced availability problems. A logic error that occurs during a single read/write request to
a single data set can result in a system outage that impacts all files opened on the system.
And as most of the subcomponents adopted the same strategy, it started to terminate the
SMSVSAM address space for reasons other than data integrity. Even a display command has
the potential for terminating the SMSVSAM address space.
With z/OS V1.R8 the development team redesigned the system error recovery area. Instead
of terminating the SMSVSAM address space, due to internal logic errors, it remains up, and
the individual function with error is fails. The first time data capture design remains as before.
See “OA17734” on page 467 for more detailed information.
Tip: We recommend that you read Chapters 5 and 6 of the IBM Redbooks publication
VSAM Demystified, SG24-6105, for an introduction to this subject.

26
z/OS V1R8 DFSMS Technical Update
Coexistence and migration
See Table 3-1 for information about coexistence and migration for the redesigns of system
error recovery area.
Table 3-1 Coexistence table about the redesigned system error recovery area
3.6.2 VSAM RLS DIAG command on z/OS V1R8 system
A situation that can arise when using VSAM RLS is that one or more tasks using RLS appear
to hang. One of the causes for this is that latches are in contention. Use the following
command to display latch contention:
D SMS,SMSVSAM,DIAG(CONTENTION)
If there are latches in contention the user is informed with detailed information that is intended
to allow the problem to be resolved.
In Figure 3-17 we show the result of issuing the command on a system that has no latches in
contention, which means that message IGW342I is issued.
Figure 3-17 D SMS, SMSVSAM,DIAG(CONTENTION) command and output no latches
If there are latches in contention, message IGW343I is issued.
Element or feature DFSMSdfp
When change was introduced z/OS V1R8.0
Applies to migration from None
Timing After the first IPL of z/OS V1R8
Target system hardware requirements None
Target system software requirements None
Other system (coexistence or fallback)
requirements
None
Restrictions None
System impacts None
D SMS,SMSVSAM,DIAG(CONTENTION)
IGW342I VSAM RLS DIAG STATUS (V.01) 335
NO CONTENTION BY REGISTERED RESOURCES EXISTS

Chapter 3. DFSMSdfp V1R8 enhancements
27
In Figure 3-18 we show sample data resulting from the command when a latch was found to
be in contention. This information may be used to identify the holder from the ASID shown,
which may lead to a decision as to whether the holder could be cancelled or whether it would
be necessary to wait.
Figure 3-18 Example of information available if RLS latch contention is detected
3.6.3 VSAM RLS DIAG command on pre-z/OS V1R8 systems
There is no support for the VSAM RLS DIAG command on lower level systems, but toleration
maintenance is available to handle the command if it should be issued. In Figure 3-17 on
page 26 we show the toleration maintenance available for systems prior to z/OS V1R8.
Table 3-2 RLS DIAG command toleration maintenance
In Figure 3-19 we show the result of issuing the command on a z/OS V1R7 system that has
VSAM RLS enabled and has the PTF for APAR OA14568 installed.
Figure 3-19 D SMS, SMSVSAM,DIAG(CONTENTION) command and output on z/OS V1R7
3.6.4 VSAM RLS performance enhancement
z/OS V1R8 introduced a performance enhancement that skips cleaning up the XES lock table
when the SMSVSAM address space is terminated. This function is automatically available,
but can be triggered by using the following command if necessary:
V SMS,SMSVSAM,TERMINATESERVER
If the command is issued on a z/OS V1R8 system, since RLS is a sysplex-wide function, it
affects other systems, and toleration maintenance is required for this.
IGW343I VSAM RLS DIAG STATUS (V.01) 035
|----RESOURCE---| |------ WAITER ------| |--HOLDER---| ELAPSED
TYPE ID JOB NAME ASID TASK ASID TASK TIME
-------- -------- -------- ---- -------- ---- -------- --------
LATCH 7F38B928 SMSVSAM 000A 007CBB58 000A 007CBE88 00:01:07
DESCRIPTION: IDAVTCP - VTCPage Storage Pool
APAR Description Effect
OA14568 Eliminates confusing message relating to another
internal function
Generates message IGW495I D
SMS,SMSVSAM,DIAG COMMAND FUNCTION
HAS FOUND THE DIAG TABLE HAS NO
ENTRIES
D SMS,SMSVSAM,DIAG(CONTENTION)
IGW495I D SMS,SMSVSAM,DIAG COMMAND FUNCTION HAS FOUND 074
THE DIAG TABLE HAS NO ENTRIES.

28
z/OS V1R8 DFSMS Technical Update
Toleration on pre-z/OS V1R8 systems
There is no provision for the performance improvement on lower level systems, but toleration
maintenance is available. In Figure 3-3 on page 16 we show the toleration maintenance
available for systems prior to z/OS V1R8.
Table 3-3 RLS XES cleanup performance improvement toleration maintenance
3.6.5 VSAM RLS use of 64-bit data buffers
VSAM RLS has specific requirements for use of 64-bit buffers (over the 2 GB bar). These are
discussed in detail in z/OS V1R7 DFSMS Technical Update, SG24-7225, but an update to
what is published there is needed.
The basic requirements are:
Any data set to be used for RLS above the bar buffers must be associated with a SMS
data class with the RLS ABOVE THE BAR attribute set to Y.
SYS1.PARMLIB member IGDSMSxx must include a non-zero value for
RlsAboveThebarMaxPoolSize, as shown in Figure 3-20 on page 29:
RlsAboveThebarMaxPoolSize = 500
SYS1.PARMLIB member IGDSMSxx must include Rls_MaxCfFeatureLevel = A. This is
what has changed from the comment in z/OS V1R7 DFSMS Technical Update,
SG24-7225, where a value of Z was shown as allowable.
SYS1.PARMLIB member IGDSMSxx may include a non-zero value for RlsFixedPoolSize.
Figure 3-20 on page 29 shows RlsFixedPoolSize = 50
In Figure 3-20 on page 29 we show the output from the D SMS,OPTIONS command
showing the RLS requirements set.
Note: The current status of these and related APARs should be used.
APAR Description Effect
OA13332 - See “OA13332” on
page 458.
Prerequisite for OA11708 Adds SYSPLEX support
required for OA11708
OA11708 - See
“OA11708” on page 457.
Reduces termination delays when V
SMS,SMSVSAM,TERMINATESERVER
command is issued
Manages locks more rapidly
OA14666 -
See “OA14666” on page 459.
Post requisite for OA11708 Completes OA11708

Chapter 3. DFSMSdfp V1R8 enhancements
29
Figure 3-20 D SMS,OPTIONS showing RLS above 2 GB bar requirements set
3.6.6 DSS LOGICAL COPY WARNING
A VSAM KSDS with data and index components on different volumes may lose its SMS
RLSDATA information during DSS COPY. This may result in the correct attributes for VSAM
D SMS,OPTIONS
IGD002I 19:54:14 DISPLAY SMS 857
ACDS = SYS1.SMS.ACDS
COMMDS = SYS1.SMS.COMMDS
INTERVAL = 15 DINTERVAL = 150
SMF_TIME = YES CACHETIME = 3600
CF_TIME = 1800 PDSE_RESTARTABLE_AS = YES
PDSE_BMFTIME = 3600 PDSE1_BMFTIME = 3600
PDSE_LRUTIME = 60 PDSE1_LRUTIME = 50
PDSE_LRUCYCLES = 15 PDSE1_LRUCYCLES = 200
LOCAL_DEADLOCK = 15 GLOBAL_DEADLOCK = 4
REVERIFY = NO DSNTYPE = PDS
ACSDEFAULTS = NO PDSESHARING = EXTENDED
OVRD_EXPDT = NO SYSTEMS = 8
PDSE_HSP_SIZE = 0MB PDSE1_HSP_SIZE = 256MB
USE_RESOWNER = YES RLS_MAX_POOL_SIZE = 100MB
RLSINIT = YES RLSTMOUT = 0
COMPRESS = GENERIC LOG_OF_LOGS = IGWTVS.LOG.OF.LOGS
QTIMEOUT = 300 TVSNAME = 064
AKP = 1000 TV_START_TYPE = WARM
MAXLOCKS = (0,0)
CICSVR_INIT = YES CICSVR_DSNAME_PREFIX = DWWUSER.V3R1M0
CICSVR_RCDS_PREFIX = DWW
CICSVR_GRPNAME_SUFFIX = PROD
CICSVR_ZZVALUE_PARM =
CICSVR_UNDOLOG_CONTROL =
CICSVR_UNDOLOG_PREFIX = DWW
CICSVR_BACKOUT_CONTROL =
CICSVR_GENERAL_CONTROL =
Rls_MaxCfFeatureLevel = A
RlsAboveThebarMaxPoolSize = 500
RlsFixedPoolSize = 50
PDSE_MONITOR = (YES,0,0) PDSE1_MONITOR = (YES,0,0)
PDSE_DIRECTORY_STORAGE = 2000M
PDSE_BUFFER_BEYOND_CLOSE = NO
PDSE1_BUFFER_BEYOND_CLOSE = NO
GDS_RECLAIM = YES DSSTIMEOUT = 0
BLOCKTOKENSIZE = NOREQUIRE FAST_VOLSEL = ON
IGD002I 19:54:14 DISPLAY SMS
TRACE = OFF SIZE = 1024K TYPE = ALL
JOBNAME = * ASID = *
TRACING EVENTS:
MODULE = OFF SMSSJF = OFF SMSSSI = OFF ACSINT = OFF
OPCMD = OFF CONFC = OFF CDSC = OFF CONFS = OFF
MSG = OFF ERR = OFF CONFR = OFF CONFA = OFF
ACSPRO = OFF IDAX = OFF DISP = OFF CATG = OFF
VOLREF = OFF SCHEDP = OFF SCHEDS = OFF VTOCL = OFF
VTOCD = OFF VTOCR = OFF VTOCC = OFF VTOCA = OFF
RCD = OFF DCF = OFF DPN = OFF TVR = OFF
DSTACK = OFF UAFF = OFF DEBUG = ON
VOLSELMSG = (OFF,0) TYPE = ALL JOBNAME = *
ASID = * STEPNAME = *
PDSE1_BUFFER_BEYOND_CLOSE = NO
GDS_RECLAIM = YES DSSTIMEOUT = 0
BLOCKTOKENSIZE = NOREQUIRE FAST_VOLSEL = ON
IGD002I 19:54:14 DISPLAY SMS
TRACE = OFF SIZE = 1024K TYPE = ALL
JOBNAME = * ASID = *
TRACING EVENTS:
MODULE = OFF SMSSJF = OFF SMSSSI = OFF ACSINT = OFF
OPCMD = OFF CONFC = OFF CDSC = OFF CONFS = OFF
MSG = OFF ERR = OFF CONFR = OFF CONFA = OFF
ACSPRO = OFF IDAX = OFF DISP = OFF CATG = OFF
VOLREF = OFF SCHEDP = OFF SCHEDS = OFF VTOCL = OFF
VTOCD = OFF VTOCR = OFF VTOCC = OFF VTOCA = OFF
RCD = OFF DCF = OFF DPN = OFF TVR = OFF
DSTACK = OFF UAFF = OFF DEBUG = ON

VOLSELMSG = (OFF,0) TYPE = ALL JOBNAME = *
ASID = * STEPNAME = *
DSNAME = *

30
z/OS V1R8 DFSMS Technical Update
RLS use in general, and the 64-bit requirements in particular being dropped.
Refer to “OA18319” on page 462 for further information.
3.6.7 Monitoring the VSAM RLS use of 64-bit data buffers
You can monitor the use of VSAM RLS 64-bit data buffers, and check on the status of data
classes eligible and which data sets are eligible.
You can set up recording of SMF type 42 records then check the data returned in the
appropriate subtype records as listed below. For more information refer to Chapter 7 under
“Monitoring the use of 64-bit data buffers,” in the manual DFSMS Using the New Functions,
SC26-7473.
If you have access to IBM RMF™ then it is preferable to use it to look at the various statistics
provided. If you do not have RMF then the sample programs provided in the appendix and
illustrated below can be used to gain insight into what is happening on the system.
SMF records for analysis
The SMF records for analysis are:
SMF Record type 42 Subtype 16 (Data Set Summary) indicates whether a data set is
enabled for 64-bit data buffers and whether it uses storage above 2-gigabytes. Any data
set that you want SMF to monitor usage for must be nominated by means of the
command:
V SMS,MONDS(datasetname),ON
Where datasetname is a fully qualified name or one with one trailing asterisk (*).
For details on these SMF records for analysis refer to 3.6.8, “SMF Record type 42 Subtype
16 information example” on page 30.
SMF Record type 42 Subtype 18 (CF Cache Partition Usage) provides statistics on the
use of coupling facility cache storage classes. For details on these SMF records for
analysis see “SMF Record type 42 Subtype 18 information example” on page 36.
SMF Record type 42 Subtype 19 (LRU record) provides statistics for data buffers above
and below 2 gigabytes. For details on these SMF records for analysis see “SMF Record
type 42 Subtype 19 information example” on page 44.
3.6.8 SMF Record type 42 Subtype 16 information example
SMS writes SMF Record type 42 subtype 16 to document VSAM RLS data set activity.
We wrote a simple program called SMF42TG to scan the SMF records and summarize
activity. The program itself and how to construct it are documented in “SMF record type 42
subtype 16 data display program” on page 522.
Note: Before using any of the SMF records for analysis, review the current status of APAR
OA17704, as shown in “OA11708” on page 457.
Note: Before attempting to use SMF type 42 subtype 16 records in relation to 64-bit buffer
usage, APAR OA17704 (see “OA17704” on page 456) should be installed.

Chapter 3. DFSMSdfp V1R8 enhancements
31
In Figure 3-21 we show the JCL to extract the SMF records and run the program.
Figure 3-21 JCL to extract SMF records and run program SMF42TG
There are four sets of records contained in the SMF records type 42 subtype 16, two sections
relating to below the 2 GB bar and two relating to above the 2 GB bar. Hence, a large amount
of output could be produced. Samples of each type of output are presented from a run of the
program. The ability to use buffers above the 2 GB bar was made available with z/OS V1R7,
but the SMF data about activity was not available until z/OS V1R8.
Program SMF42TG output uses headings for the output that are taken from the SMF records
that it was obtained from. To assist with interpretation of the output, we provide extracts from
parts of the SMF mapping macro IGWSMF. In order to distinguish between the SMF record
type 42 subtype 16 fields from below and above the 16 GB bar, those relating to below the
2 GB bar are prefixed SMF42, and those relating to above the 2 GB bar are prefixed SMF2A
(2A is hexidecimal for 42).
The records (in each group) consist of SYSPLEX-wide data and Data set/MVS system
summary data.
Note: This JCL assumes that an extract is being taken from the active SMF data set, which
is then passed to the SMFT42TG program. The JCL could be changed to make a
permanent extract of the SMF data, or to read of an already created SMF data extract.
//MHLRES1L JOB (999,POK),MSGLEVEL=1,NOTIFY=MHLRES1
// EXEC PGM=IFASMFDP
//SYSPRINT DD SYSOUT=*
//DUMPIN DD DISP=SHR,DSN=SYS1.SC64.MAN2
//*DUMPIN DD DISP=SHR,DSN=SMFDATA.ALLRECS.G3196V00
//OUTDD DD DSN=&SMFT85,
// SPACE=(CYL,(10,5)),
// RECFM=VB,LRECL=32760,
// DISP=(,PASS,DELETE),
// UNIT=SYSDA
//SYSIN DD *
INDD(DUMPIN,OPTIONS(DUMP))
OUTDD(OUTDD,TYPE(42(16)))
/*
// EXEC PGM=SMF42TG
//STEPLIB DD DISP=SHR,DSN=MHLRES1.SMF42TG.LOAD
//SYSUDUMP DD SYSOUT=A
//SMFIN DD DISP=SHR,DCB=BFTEK=A,
// DSN=&SMFT85
//PRINT DD SYSOUT=A,RECFM=UA
Note: As a result of the problems described by APAR OA17704 (see “OA17704” on
page 456) there may be abnormal displays in the output. The fact that a record is not
correctly built in the SMF records is usually indicated by the cycle time records being zero
or abnormally large.

32
z/OS V1R8 DFSMS Technical Update
For below the 2 GB bar, the SYSPLEX-wide records come in group SMF42GA1 and the MVS
system records come in group SMF42G1A. Similar groups for above the 2 GB bar are
SMF2AGA1 and SMF2AG1A.
Example from group SMF42GA1 - SYSPLEX-wide summary
In Figure 3-22 we show part of the output from running SMF42GA1.
Figure 3-22 SMF42TG output for group SMF41GA1
A number of sets of output may be produced depending on the configuration of the
SYSPLEX.
The specific meanings of the fields can be found in the manual MVS System Management
Facilities (SMF), SA22-7630, or can be interpreted from the IGWSMF macro in
SYS1.MACLIB on the system, which may be more up to date than the version in the manual.
In Figure 3-23 on page 33 we show part of the IGWSMF macro that relates to group
SMF42GA1 records.
In Figure 3-22:
Fields SMF42GAA, SMF42GAB, and SMF42GAC show, respectively, the interval time,
data set name, and VSAM sphere name.
The SMF42GAA value can be used as an indicator that the record is valid. A value of 0 or
a very high number is likely to indicate that the SMF record may be suspect.
Fields SMF42GAE, SMF42GAF, SMF42GAH, SMF42GAI, and SMF42GAJ show,
respectively, the Storage Class Name, the CACHESET name, the DFP CACHE
STRUCTURE name, the indicator of the component being processed, and CF processing
status indictors.
The SMF42GAI field may no longer be supported. All test runs show it as a field of
x’00000000’).
The remaining fields can be interpreted from the IGWSMF macro.
Information is provided in Appendix B, “Code samples” on page 477, where the program is
described on how to alter it to process other SMF fields from those initially provided.
SMF42GA1 SYSPLEX D/S RESPONSE SUMMARY SET #: 0000001
YYYYDDD HH:MM:SS SMF42GAA SMF42GAB SMF42GAC
2007080 20:30:00 0000600 HSM.BCDS.RLS.DATA HSM.BCDS.RLS
SMF42GAE SMF42GAF SMF42GAH SMF42GAI SMF42GAJ
SC54GRT CSERLS RLS_CACHE UNKNOWN GT4KNOTACT
SMF42GAP SMF42GCA SMF42GCB SMF42GCC SMF42GCD SMF42GCE SMF42GCF SMF42GCI SMF42GCL SMF42GCM SMF42GCN
0000009 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000
.
.
.

Chapter 3. DFSMSdfp V1R8 enhancements
33
Figure 3-23 IGWSMF record extract for group SMF42GA1
* -----------------------------------------------------------------
* CF DATASET RESPONSE TIME SUMMARY (SMF42 SUBTYPE 16)
* -----------------------------------------------------------------
SMF420GA DSECT ,
SMF42GA1 DS 0CL1472 SYSPLEX WIDE SUMMARY @21A
SMF42GAA DS 1FL4 INTERVAL LENGTH. THIS IS THE
* TOTAL TIME OF THE MEASUREMENT
* PERIOD. (IN SECONDS) @10A
SMF42A00 DS CL12 RESERVED @MDK
*
SMF42GAB DS CL44 DATA SET NAME @10A
SMF42A01 DS 1FL4 RESERVED @10A
*
SMF42GAC DS CL44 VSAM SPHERE NAME @10A
SMF42A02 DS 1FL4 RESERVED @10A
*
SMF42GAD DS 1FL2 LENGTH OF THE STORAGE CLASS
* NAME @10A
SMF42GAE DS CL30 STORAGE CLASS NAME @10A
*
SMF42A03 DS 1FL2 CACHESETNAME LENGTH @10A
SMF42GAF DS CL30 CACHESET NAME @10A
*
SMF42GAG DS CL2 RESERVED @10A
SMF42GAH DS CL30 DFP CACHE STRUCTURE NAME @10A
*
SMF42GAI DS 1FL4 INDICATOR OF COMPONENT BEING
* PROCESSED @10A
SMF42GAI0 EQU X'80000000' DATA COMPONENT @10A
SMF42GAI1 EQU X'40000000' INDEX COMPONENT @10A
* @10A
SMF42GAJ DS CL12 INDICATES DFSMS GREATER @P5A
* THAN 4K CF CACHING STATUS @P5A
* VALUES ARE ALL, NONE, @P5A
* UPDATESONLY, GT4KNOTACT @P5A
SMF42GAK DS 1FL4 NUMBER OF LOCK REQUESTS
* PROCESSED @10A
SMF42GAL DS 1FL4 NUMBER OF TRUE CONTENTION
* LOCK REQUESTS @10A
SMF42GAM DS 1FL4 NUMBER OF FALSE CONTENTION
* LOCK REQUESTS @10A
SMF42GZ1 DS 1FL2 SMSDIRECTWEIGHT @10A
SMF42GZ2 DS 1FL2 SMSSEQUENTIALWEIGHT @10A
.
.
.
.

34
z/OS V1R8 DFSMS Technical Update
Example from group SMF42G1A - data set/MVS summary
In Figure 3-24 we show part of the output from running SMF42GA1.
Figure 3-24 SMF42TG output for group SMF41G1A
A number of sets of output may be produced depending on the configuration of the
SYSPLEX.
The specific meaning of the fields can be found in the manual MVS System Management
Facilities (SMF), SA22-7630, or can be interpreted from the IGWSMF macro in
SYS1.MACLIB on the system, which may be more up to date than the version in the manual.
SMF42G1A DATA SET/MVS SYSTEM SUMMARY SET #: 0000001
YYYYDDD HH:MM:SS SMF42GBA SMF42GBB SMF42GBC
2007080 20:30:00 0000600 HSM.BCDS.RLS.DATA HSM.BCDS.RLS
SMF42GBE SMF42GBF SMF42GBH SMF42GBI SMF42AO9
SC54GRT CSERLS SC64 UNKNOWN GT4KNOTACT
SMF42GBP SMF42GIA SMF42GIB SMF42GIC SMF42GID SMF42GIE SMF42GIF SMF42GIL SMF42GIR SMF42GIS
0000009 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 .
.
.

Chapter 3. DFSMSdfp V1R8 enhancements
35
In Figure 3-25 we show part of the IGWSMF macro that relates to group SMF42G1A records.
Figure 3-25 IGWSMF record extract for group SMF42G1A
In Figure 3-25:
Fields SMF42GBA, SMF42GBB, and SMF42GBC show, respectively, the interval time,
data set name, and VSAM sphere name.
The SMF42GBA value can be used as an indicator that the record is valid. A value of 0 or
a very high number is likely to indicate that the SMF record may be suspect.
Fields SMF42GBE, SMF42GBF, SMF42GBH, SMF42GBI, and SMF42A09 show,
respectively, the Storage Class Name, the CACHESET name, the DFP CACHE
SMF42G1A DS 0CL1488 DATA SET/MVS SYSTEM SUMMARY @21A
* SECTION @21C
SMF42GBA DS 1FL4 INTERVAL LENGTH. THIS IS THE
* TOTAL TIME OF THE MEASUREMENT
* PERIOD. (IN SECONDS) @10A
SMF42A05 DS CL12 RESERVED @10A
*
SMF42GBB DS CL44 DATA SET NAME @10A
SMF42A06 DS 1FL4 RESERVED @10A
*
SMF42GBC DS CL44 VSAM SPHERE NAME @10A
SMF42A07 DS 1FL4 RESERVED @10A
*
SMF42GBD DS 1FL2 LENGTH OF THE STORAGE CLASS
* NAME @10A
SMF42GBE DS CL30 STORAGE CLASS NAME @10A
*
SMF42A08 DS 1FL2 CACHESETNAME LENGTH @10A
SMF42GBF DS CL30 CACHESET NAME @10A
*
SMF42A12 DS 1FL2 RESERVED @10A
SMF42GBG DS CL30 DFP CACHE STRUCTURE NAME @10A
*
SMF42GBH DS CL8 MVS SYSTEM NAME @10A
SMF42A11 DS CL8 RESERVED @10A
*
SMF42GBI DS 1FL4 INDICATOR OF COMPONENT BEING
* PROCESSED @10A
SMF42GBI0 EQU X'80000000' DATA COMPONENT @10A
SMF42GBI1 EQU X'40000000' INDEX COMPONENT @10A
SMF42A09 DS CL12 INDICATES DFSMS GREATER @P5A
* THAN 4K CF CACHING STATUS @P5A
* VALUES ARE ALL, NONE, @P5A
* UPDATESONLY, GT4KNOTACT @P5A
SMF42GBK DS 1FL4 NUMBER OF LOCK REQUESTS
* PROCESSED @10A
SMF42GBL DS 1FL4 NUMBER OF TRUE CONTENTION
* LOCK REQUESTS @10A
SMF42GBM DS 1FL4 NUMBER OF FALSE CONTENTION
* LOCK REQUESTS @10A
SMF42GZ8 DS 1FL2 SMSDIRECTWEIGHT @10A
SMF42GZ9 DS 1FL2 SMSSEQUENTIALWEIGHT @10A
*
.

36
z/OS V1R8 DFSMS Technical Update
STRUCTURE name, the indicator of the component being processed, and CF processing
status indictors.
The SMF42GBI field may no longer be supported. All test runs show it as a field of
x’00000000’).
The remaining fields can be interpreted from the IGWSMF macro.
Information is provided in Appendix B, “Code samples” on page 477, where the program is
described on how to alter it to process other SMF fields from those initially provided.
Example from group SMF2AGA1 - SYSPLEX-wide summary
Because of the problems with the content of the SMF records, as described in 3.6.8, “SMF
Record type 42 Subtype 16 information example” on page 30, actual output from the program
for above the 2 GB bar is not shown. If the program is run after the fixes have been
implemented, the output appears the same as from the below the 2 GB bar examples above,
but with 2A substituted for 42.
Example from group SMF2AG1A - data set/MVS summary
Because of the problems with the content of the SMF records, as described in 3.6.8, “SMF
Record type 42 Subtype 16 information example” on page 30, actual output from the program
for above the 2 GB bar is not shown. If the program is run after the fixes have been
implemented, the output appears the same as from the below the 2 GB bar examples above,
but with 2A substituted for 42.
3.6.9 SMF Record type 42 Subtype 18 information example
SMS writes SMF Record type 42 subtype 18 to document cache use by VSAM RLS.
We have written a sample program called SMF42TI to scan the SMF records and summarize
activity. The program itself and how to construct it is documented in “SMF Record type 42
subtype 18 data display program” on page 534.

Chapter 3. DFSMSdfp V1R8 enhancements
37
In Figure 3-26 we show the JCL to extract the SMF records and run the program.
Figure 3-26 JCL to extract SMF records and run program SMF42TI
Note: This JCL assumes that an extract is being taken from the active SMF data set, which
is then passed to the SMFT42TI program. The JCL could be changed to make a
permanent extract of the SMF data, or to read an already created SMF data extract.
//MHLRES1L JOB (999,POK),MSGLEVEL=1,NOTIFY=MHLRES1
// EXEC PGM=IFASMFDP
//SYSPRINT DD SYSOUT=*
//DUMPIN DD DISP=SHR,DSN=SYS1.SC64.MAN1
//*DUMPIN DD DISP=SHR,DSN=SMFDATA.ALLRECS.G3196V00
//OUTDD DD DSN=&SMFT85,
// SPACE=(CYL,(10,5)),
// RECFM=VB,LRECL=32760,
// DISP=(,PASS,DELETE),
// UNIT=SYSDA
//SYSIN DD *
INDD(DUMPIN,OPTIONS(DUMP))
OUTDD(OUTDD,TYPE(42(18)))
/*
// EXEC PGM=SMF42TI
//STEPLIB DD DISP=SHR,DSN=MHLRES1.SMF42TI.LOAD
//SYSUDUMP DD SYSOUT=A
//SMFIN DD DISP=SHR,DCB=BFTEK=A,
// DSN=&SMFT85
//PRINT DD SYSOUT=A,RECFM=UA

38
z/OS V1R8 DFSMS Technical Update
There are three different sets of output processed by SMF42TI if data is found in the records.
In Table 3-4 we show the relationship between the three types of data reported on by
SMF42TI.
In Figure 3-27 on page 39 we show example output from running program SMF42TI.
Table 3-4 SMF42TI report summary information
Data description SMF structure name Report identifier
Coupling Facility Cache
Partition Summary (SMF42
subtype 18) - data for this
partition across all cache
structures
SMF420IA SMF41IA1
Coupling Facility Cache
Partition Summary (SMF42
subtype 18) - data for this
partition for a singe cache
structure
SMF420IC SMF42IC1
Coupling Facility Cache
Partition Summary (SMF42
subtype 18) - directory/element
ratio data for this cache
structure
SMF420IE SMF42IE1

Chapter 3. DFSMSdfp V1R8 enhancements
39
Figure 3-27 SMF42TI program sample output
Program SMF42TI output uses headings for the output that is taken from the SMF records
that it was obtained from.
The specific meaning of the fields can be found in MVS System Management Facilities
(SMF), SA22-7630, or can be interpreted from the IGWSMF macro in SYS1.MACLIB on the
system, which may be more up to date than the version in the manual.
In Figure 3-27 the output is shown in three groups. The meanings of the fields are similar, and
we recommend understanding the output review of the records in SMFMAN or in the
IGWSMF macro, as the values presented depend on the system configuration.
In groups SMF41IA1, SMF42IC1, and SMF42IE1, respectively, fields SMF42IAA,
SMF42IAC, and SMF42IAE show the interval time. The value shown can be used as an
indicator that the record is valid. A value of 0 or a very high number is likely to indicate that
the SMF record may be suspect.
SMF TYPE 42 S/TYPE 18 RECS. COLS USE SMF NAMES
SMF42IA1 TOTALS SET #: 0000001
HH:MM:SS YYYYDDD SMF42IAA SMF42IBG SMF42IBH SMF42I01 SMF42IAD SMF42IAF SMF42IAG SMF42IAH SMF42IAI SMF42IAJ
11:00:00 2007074 0000599 DIRECT 0000001 PARTITION # 0000001 0000000 0000000 0000000 0000000 0000000
SMF42IAK SMF42IAL SMF42IAM SMF42IAN SMF42IAO SMF42IAP SMF42IAQ SMF42IAR SMF42IAS SMF42IAT SMF42IAU SMF42IAV
0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000
SMF42IAW SMF42IAX SMF42IAY SMF42IAZ SMF42IBA SMF42IBB SMF42IBC SMF42IBD SMF42IBE SMF42IBF
0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000

SMF42IA1 TOTALS SET #: 0000002
HH:MM:SS YYYYDDD SMF42IAA SMF42IBG SMF42IBH SMF42I01 SMF42IAD SMF42IAF SMF42IAG SMF42IAH SMF42IAI SMF42IAJ
11:00:00 2007074 0000599 SEQUENTIAL 0000001 PARTITION # 0000002 0000000 0000000 0000000 0000000 0000000
SMF42IAK SMF42IAL SMF42IAM SMF42IAN SMF42IAO SMF42IAP SMF42IAQ SMF42IAR SMF42IAS SMF42IAT SMF42IAU SMF42IAV
0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000
SMF42IAW SMF42IAX SMF42IAY SMF42IAZ SMF42IBA SMF42IBB SMF42IBC SMF42IBD SMF42IBE SMF42IBF
0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000
.
.
SMF42IC1 TOTALS SET #: 0000001
HH:MM:SS YYYYDDD SMF42ICA SMF42ICB SMF42IDG SMF42IDH SMF42ICD SMF42ICF SMF42ICG SMF42ICH
11:00:00 2007074 0000599 RLS_CACHE DIRECT 0000001 0000001 0000000 0000000 0000000
SMF42ICI SMF42ICJ SMF42ICK SMF42ICL SMF42ICM SMF42ICN SMF42ICO SMF42ICP SMF42ICQ SMF42ICR SMF42ICS
0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000
SMF42ICT SMF42ICU SMF42ICV SMF42ICW SMF42ICX SMF42ICY SMF42ICZ SMF42IDA SMF42IDB SMF42IDC SMF42IDD SMF42IDE
0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000

SMF42IC1 TOTALS SET #: 0000002
HH:MM:SS YYYYDDD SMF42ICA SMF42ICB SMF42IDG SMF42IDH SMF42ICD SMF42ICF SMF42ICG SMF42ICH
11:00:00 2007074 0000599 RLS_CACHE SEQUENTIAL 0000001 0000002 0000000 0000000 0000000
SMF42ICI SMF42ICJ SMF42ICK SMF42ICL SMF42ICM SMF42ICN SMF42ICO SMF42ICP SMF42ICQ SMF42ICR SMF42ICS
0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000
SMF42ICT SMF42ICU SMF42ICV SMF42ICW SMF42ICX SMF42ICY SMF42ICZ SMF42IDA SMF42IDB SMF42IDC SMF42IDD SMF42IDE
0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000
.
.
SMF42IE1 TOTALS SET #: 0000001
HH:MM:SS YYYYDDD SMF42IEA SMF42IEB SMF42IEC SMF42IED SMF42IEE SMF42IEF SMF42IEG SMF42IEH SMF42IEI SMF42IEJ
14:00:00 2007074 0000600 RLS_CACHE 0000000 0000000 0000252 0000255 0000252 0000255 0000252 0000255

SMF42IE1 TOTALS SET #: 0000002
HH:MM:SS YYYYDDD SMF42IEA SMF42IEB SMF42IEC SMF42IED SMF42IEE SMF42IEF SMF42IEG SMF42IEH SMF42IEI SMF42IEJ
14:00:00 2007074 0000600 RLS_CACHE 0000000 0000000 0000252 0000255 0000252 0000255 0000252 0000255
.
.

40
z/OS V1R8 DFSMS Technical Update
For group SMF42IA1 in Table 3-5 we show the interpretation of the field names as taken
from the IGWSMF macro.
Table 3-5 SMF42IA1 field interpretation
For group SMF42IC1 in Table 3-6 we show the interpretation of the field names as taken
from the IGWSMF macro.
Table 3-6 SMF42IC1 field interpretation
For group SMF42IE1 in Table 3-7 we show the interpretation of the field names as taken
from the IGWSMF macro.
Table 3-7 SMF42IE1 field interpretation
Field
name
Meaning
SMF42IBG Partition type
SMF42IBH SMS specified cache weight
SMF42IO1 Partition name word
SMF42IAD CF cache partition number
Field name Meaning
SMF42ICB cache structure name
SMF42IDG partition type
SMF42IDH SMS specified cache weight
SMF42ICD CF cache partition number
Field name Meaning
SMF42IEB SMS cache structure name
SMF42IEC Number of times the
directory portion of the
dir/ele ratio changes
SMF42IED Number of times the
element portion of the
dir/ele ratio changes
SMF42IEE Low ratio value in this
interval - directory part
SMF42IEF Low ratio value in this
interval - element part
SMF42IEG High ratio value in this
interval - directory part
SMF42IEH High ratio value in this
interval - element part
SMF42IEI Current ratio value in this
interval - directory part

Chapter 3. DFSMSdfp V1R8 enhancements
41
The remaining fields can be interpreted from the IGWSMF macro.
Information is provided in Appendix B, “Code samples” on page 477, where the program is
described for how to alter it to process other SMF fields from those initially provided.
To assist with interpretation of the output in this document, we provide extracts from parts of
the SMF mapping macro IGWSMF (Figure 3-28 on page 42 through to Figure 3-29 on
page 43).
SMF42IEJ Current ratio value in this
interval - element part
Note: Changes may be made to the SMF mapping macro to correspond with changes to
the data recorded, so for accurate analysis, the actual IGWSMF macro should be used.
Field name Meaning

42
z/OS V1R8 DFSMS Technical Update
Figure 3-28 IGWSMF macro part of DSECT SMF420IA section relating to subtype 18
* -----------------------------------------------------------------
* COUPLING FACILITY CACHE PARTITION SUMMARY (SMF42 SUBTYPE 18)
* DATA FOR THIS PARTITION ACROSS ALL CACHE STRUCTURES.
* -----------------------------------------------------------------
SMF420IA DSECT ,
SMF42IA1 DS 0D CF CACHE PARTITION TOTALS
* SECTION @10A
SMF42IAC DS 0CL196 CACHE BUFFER POOLS
* THERE ARE 23 POOLS:
* 11 SEQUENTIAL, 11 DIRECT,
* AND 1 DEFAULT
* THERE IS A SEPERATE SECTION FOR
* EACH PARTITION. THIS MEANS THAT
* THERE WILL BE A MINUMUM OF 23
* TOTALS SECTIONS AND 23 SUMMARY
* SECTIONS.
SMF42IAA DS 1FL4 INTERVAL LENGTH. THIS IS THE
* TOTAL TIME OF THE MEASUREMENT
* PERIOD. (IN SECONDS) @10A
SMF42I00 DS CL12 RESERVED @10A
*
SMF42IAB DS CL32 RESERVED @10A
*
SMF42IBG DS CL16 PARTITION TYPE @10A
* 'SEQUENTIAL' DATA ONLY @10A
* 'DIRECT' DATA ONLY @10A
* 'COMBINED' DATA @10A
*
SMF42IBH DS 1FL4 SMS SPECIFIED CACHE WEIGHT @10A
SMF42I01 DS CL12 PARTITIONNAMEWORD @10A
SMF42IAD DS 1FL4 COUPLE FACILITY CACHE
* PARTITION NUMBER @10A
SMF42IAE DS 1FL4 STATUS CONDITIONS @10A
SMF42IAF DS 1FL4 READ HIT COUNTER @10A
SMF42IAG DS 1FL4 READ MISS DIRECTORY HIT
* COUNTER @10A
SMF42IAH DS 1FL4 READ MISS ASSIGN SUPP
* COUNTER @10A
SMF42IAI DS 1FL4 READ MISS, NAME ASSIGNED
* COUNTER @10A
SMF42IAJ DS 1FL4 READ MISS TARGET STORAGE
* CLASS FULL COUNTER @10A
SMF42IAK DS 1FL4 WRITE HIT CHANGE BIT 0
* COUNTER @10A
SMF42IAL DS 1FL4 WRITE HIT CHANGE BIT 1
* COUNTER @10A
SMF42IAM DS 1FL4 WRITE MISS NOT REGISTERED
* COUNTER @10A
SMF42IAN DS 1FL4 WRITE MISS INVALID STATE
* COUNTER @10A
SMF42IAO DS 1FL4 WRITE MISS TARGET STORAGE CLASS
* FULL COUNTER @10A

Chapter 3. DFSMSdfp V1R8 enhancements
43
Figure 3-29 IGWSMF macro part of DSECT SMF420IC section relating to subtype 18
* -----------------------------------------------------------------
* COUPLING FACILITY CACHE PARTITION SUMMARY (SMF42 SUBTYPE 18)
* DATA FOR THIS PARTITION FOR A SINGLE CACHE STRUCTURE
* -----------------------------------------------------------------
SMF420IC DSECT ,
SMF42IC1 DS 0D CF CACHE PARTITION SUMMARY
* SECTION @10A
SMF42ICC DS 0CL196 CACHE BUFFER POOLS
* THERE ARE 23 POOLS:
* 11 SEQUENTIAL, 11 DIRECT,
* AND 1 DEFAULT
* THERE IS A SEPERATE SECTION FOR
* EACH PARTITION. THIS MEANS THAT
* THERE WILL BE A MINUMUM OF 23
* TOTALS SECTIONS AND 23 SUMMARY
* SECTIONS.
SMF42ICA DS 1FL4 INTERVAL LENGTH. THIS IS THE
* TOTAL TIME OF THE MEASUREMENT
* PERIOD. (IN SECONDS) @10A
SMF42I02 DS CL12 RESERVED @10A
*
SMF42I04 DS 1FL2 RESERVED @10A
SMF42ICB DS CL30 CACHE STRUCTURE NAME @10A
*
SMF42IDG DS CL16 PARTITION TYPE @10A
* 'SEQUENTIAL' DATA ONLY @10A
* 'DIRECT' DATA ONLY @10A
* 'COMBINED' DATA @10A
*
SMF42IDH DS 1F SMS SPECIFIED CACHE WEIGHT @10A
SMF42I03 DS CL12 RESERVED @10A
*
SMF42ICD DS 1FL4 COUPLE FACILITY CACHE
* PARTITION NUMBER @10A
SMF42ICE DS 1FL4 STATUS CONDITIONS @10A
SMF42ICF DS 1FL4 READ HIT COUNTER @10A
SMF42ICG DS 1FL4 READ MISS DIRECTORY HIT
* COUNTER @10A
SMF42ICH DS 1FL4 READ MISS ASSIGN SUPP
* COUNTER @10A
SMF42ICI DS 1FL4 READ MISS, NAME ASSIGNED
* COUNTER @10A
SMF42ICJ DS 1FL4 READ MISS TARGET STORAGE
* CLASS FULL COUNTER @10A
SMF42ICK DS 1FL4 WRITE HIT CHANGE BIT 0
* COUNTER @10A
SMF42ICL DS 1FL4 WRITE HIT CHANGE BIT 1
* COUNTER @10A
SMF42ICM DS 1FL4 WRITE MISS NOT REGISTERED
* COUNTER @10A
SMF42ICN DS 1FL4 WRITE MISS INVALID STATE
* COUNTER @10A
SMF42ICO DS 1FL4 WRITE MISS TARGET STORAGE CLASS
* FULL COUNTER @10A
SMF42ICP DS 1FL4 DIRECTORY ENTRY RECLAIM
* COUNTER @10A

44
z/OS V1R8 DFSMS Technical Update
Figure 3-30 IGWSMF macro DSECT SMF420IE section relating to subtype 18
3.6.10 SMF Record type 42 Subtype 19 information example
SMS can be configured to write SMF Record type 42 subtype 19 to document cache use.
We have written a simple program called SMF42TJ to scan the SMF records and summarize
activity. The program itself and how to construct it is documented in “SMF record type 42
subtype 19 data display program” on page 547.
* -------------------------------------------------------------------
* Coupling Facility Cache Partition Summary (SMF42 subtype 18)
* Directory/Element ratio data for this cache structure
* -------------------------------------------------------------------
SMF420IE DSECT , @11A
SMF42IE1 DS 0CL80 CF Cache Directory/element @11A
* ratio section @11A
SMF42IEA DS 1FL4 Interval length @11A
SMF42I06 DS CL12 reserved @11A
SMF42I07 DS CL2 reserved @11A
SMF42IEB DS CL30 DFSMS cache structure name @11A
SMF42IEC DS 1FL4 Number of times the @11A
* directory portion of the @11A
* dir/ele ratio changes @11A
SMF42IED DS 1FL4 Number of times the @11A
* Element portion of the @11A
* dir/ele ratio changes @11A
SMF42IEE DS 1FL4 Low ratio value in this @11A
* interval - directory part @11A
SMF42IEF DS 1FL4 Low ratio value in this @11A
* interval - Element part @11A
SMF42IEG DS 1FL4 High ratio value in this @11A
* interval - directory part @11A
SMF42IEH DS 1FL4 High ratio value in this @11A
* interval - Element part @11A
SMF42IEI DS 1FL4 Current ratio value in this @11A
* interval - directory part @11A
SMF42IEJ DS 1FL4 Current ratio value in this @11A
* interval - element part @11A
SMF420IE_LEN EQU *-SMF420IE Length of section @11A
*

Chapter 3. DFSMSdfp V1R8 enhancements
45
In Figure 3-31 we show the JCL to extract the SMF records and run the program.
Figure 3-31 JCL to extract SMF records and run program SMF42TJ
Note: This JCL assumes that an extract is being taken from the active SMF data set, which
is then passed to the SMFT42TJ program. The JCL could be changed to make a
permanent extract of the SMF data, or to read from an already created SMF data extract.
//MHLRES1L JOB (999,POK),MSGLEVEL=1,NOTIFY=MHLRES1
// EXEC PGM=IFASMFDP
//SYSPRINT DD SYSOUT=*
//DUMPIN DD DISP=SHR,DSN=SYS1.SC64.MAN1
//*DUMPIN DD DISP=SHR,DSN=SMFDATA.ALLRECS.G3196V00
//OUTDD DD DSN=&SMFT85,
// SPACE=(CYL,(10,5)),
// RECFM=VB,LRECL=32760,
// DISP=(,PASS,DELETE),
// UNIT=SYSDA
//SYSIN DD *
INDD(DUMPIN,OPTIONS(DUMP))
OUTDD(OUTDD,TYPE(42(19)))
/*
// EXEC PGM=SMF42TJ
//STEPLIB DD DISP=SHR,DSN=MHLRES1.SMF42TJJ.PDS
//SYSUDUMP DD SYSOUT=A
//SMFIN DD DISP=SHR,DCB=BFTEK=A,
// DSN=&SMFT85
//PRINT DD SYSOUT=A,RECFM=UA

46
z/OS V1R8 DFSMS Technical Update
In Figure 3-32 we show example output from running program SMF42TJ.
Figure 3-32 SMF42TJ program sample output
The fields prefixed SMF42 are below the 2 GB bar, and those prefixed SMF2A are above the
2 GB bar. The sample program and resulting output do not cover all the SMF records that are
available. The program as documented in the appendix can be adapted to extract and display
alternative records, or could be extended to cover more of the records.
The significance of the contents of these records depends on the particular environment
being used. These examples demonstrate that SMF was producing records for both below
and above the 2 GB bar activity, but as it was done on a low usage system (as far as RLS
was concerned), many values show as zero.
Interpretation of the field meaning can be done by following the examples for subtypes 16 and
18 above.
Program SMF42TJ output uses headings for the output that are taken from the SMF records
that it was obtained from. To assist with interpretation of the output, we provide extracts from
parts of the SMF mapping macro IGWSMF (Figure 3-33 on page 47 and Figure 3-34 on
page 49).
SMF42JNA SYSPLEX SET #: 0000001
HH:MM:SS YYYYDDD SMF42JNA SMF42JN7 SMF42JNG SMF42JNH SMF42JNI SMF42JNJ SMF42JNK SMF42JNL SMF42JNM SMF42JNN
17:30:00 2007073 0000599 0000000 0000040 0000160 0000000 0000000 0000000 0000000 0000000 0000000

SMF42JPA LPAR SET #: 0000001
HH:MM:SS YYYYDDD SMF42JPA SMF42JPB SMF42JP6 SMF42JPG SMF42JPH SMF42JPI SMF42JP2 SMF42JPJ SMF42JPK SMF42JPL SMF42JPM SMF42JPN
17:30:00 2007073 0000600 SC64 0000000 0000040 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000


SMF42JPA LPAR SET #: 0000002
HH:MM:SS YYYYDDD SMF42JPA SMF42JPB SMF42JP6 SMF42JPG SMF42JPH SMF42JPI SMF42JP2 SMF42JPJ SMF42JPK SMF42JPL SMF42JPM SMF42JPN
17:30:00 2007073 0000600 SC70 0000000 0000040 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000

SMF42JPA LPAR SET #: 0000003
HH:MM:SS YYYYDDD SMF42JPA SMF42JPB SMF42JP6 SMF42JPG SMF42JPH SMF42JPI SMF42JP2 SMF42JPJ SMF42JPK SMF42JPL SMF42JPM SMF42JPN
17:30:00 2007073 0000600 SC65 0000000 0000040 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000

SMF42JPA LPAR SET #: 0000004
HH:MM:SS YYYYDDD SMF42JPA SMF42JPB SMF42JP6 SMF42JPG SMF42JPH SMF42JPI SMF42JP2 SMF42JPJ SMF42JPK SMF42JPL SMF42JPM SMF42JPN
17:30:00 2007073 0000599 SC63 0000000 0000040 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000


SMF2AJNA SYSPLEX SET #: 0000001
HH:MM:SS YYYYDDD SMF2AJNA SMF2AJN7 SMF2AJNG SMF2AJNH SMF2AJNI SMF2AJNJ SMF2AJNK SMF2AJNL SMF2AJNM SMF2AJNN
17:30:00 2007073 0000599 0000000 0000020 0000080 0000000 0000000 0000000 0000000 0000000 0000000

SMF2AJPA LPAR SET #: 0000001
HH:MM:SS YYYYDDD SMF2AJPA SMF2AJPB SMF2AJP6 SMF2AJPG SMF2AJPH SMF2AJPI SMF2AJP2 SMF2AJPJ SMF2AJPK SMF2AJPL SMF2AJPM SMF2AJPN
17:30:00 2007073 0000600 SC64 0000000 0000040 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000

SMF2AJPA LPAR SET #: 0000002
HH:MM:SS YYYYDDD SMF2AJPA SMF2AJPB SMF2AJP6 SMF2AJPG SMF2AJPH SMF2AJPI SMF2AJP2 SMF2AJPJ SMF2AJPK SMF2AJPL SMF2AJPM SMF2AJPN
17:30:00 2007073 0000600 SC70 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000

SMF2AJPA LPAR SET #: 0000003
HH:MM:SS YYYYDDD SMF2AJPA SMF2AJPB SMF2AJP6 SMF2AJPG SMF2AJPH SMF2AJPI SMF2AJP2 SMF2AJPJ SMF2AJPK SMF2AJPL SMF2AJPM SMF2AJPN
17:30:00 2007073 0000600 SC65 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000

SMF2AJPA LPAR SET #: 0000004
HH:MM:SS YYYYDDD SMF2AJPA SMF2AJPB SMF2AJP6 SMF2AJPG SMF2AJPH SMF2AJPI SMF2AJP2 SMF2AJPJ SMF2AJPK SMF2AJPL SMF2AJPM SMF2AJPN
17:30:00 2007073 0000599 SC63 0000000 0000040 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000
Note: Changes may be made to the SMF mapping macro to correspond with changes to
the data recorded, so for accurate analysis, the actual IGWSMF macro should be used for
reference.

Chapter 3. DFSMSdfp V1R8 enhancements
47
Figure 3-33 IGWSMF macro relating to SMF type 42 subtype 19 (parts of the SMF420JA DSECT)
* --------------------------------------------------------------
* Coupling Facility Local Buffer Manager (SMF42 subtype 19)
* Local Buffer Manager LRU Statistics Summary. sysplex totals
* --------------------------------------------------------------
SMF420JA DSECT ,
SMF42JA1 DS 0D sysplex totals section
*
SMF42JAC DS 0CL1384
*
SMF42JNA DS 1FL4 Interval length. This is the
* total time of the measurement
* period. (in seconds) total
SMF42J00 DS CL12 reserved
*
SMF42JNB DS CL16 reserved
SMF42JND DS CL8 Reserved
*
SMF42JNE DS 2F
* Average Cpu time for all
* systems in the sysplex. This
* is the amount of CPU time
* (on average) that each system
* consumed while processing the
* Local Buffer Manager LRU
* alogrithm.
*
SMF42JNF DS 2F
* Total CPU time for this
* record (in milli-seconds)
* (sysplex totals)
*
SMF42JN0 DS CL4 Reserved
SMF42JN7 DS 1FL4 Total number of write
* requests
* (sysplex totals)
SMF42JNG DS 1FL4 Average number of buffer
* manager LRU intervals
* processed during this period.
* (sysplex totals)
*
SMF42JNH DS 1FL4 Total number of buffer
* manager LRU intervals
* processed during this period.
* (across the sysplex)
*
SMF42JNI DS 1FL4 Average
* Number of buffer manager
* LRU intervals processed
* where BMF was over the goal
* and normal algorithms were
* bypassed to reclaim buffers.
*
SMF42JNJ DS 1FL4 Total
* Number of buffer manager
* LRU intervals processed
* where BMF was over the goal
* and normal algorithms were

48
z/OS V1R8 DFSMS Technical Update
The above example shows the records for below the 2 GB bar records where they are
prefixed SMF42xxx. The IGWSMF macro contains corresponding definitions for the above
the bar records where they are prefixed SMF2Axxx.

Chapter 3. DFSMSdfp V1R8 enhancements
49
Figure 3-34 GWSMF macro section SMF type 42 subtype 19 (parts of the SMF420PA DSECT)
* --------------------------------------------------------------
* Coupling Facility Local Buffer Manager (SMF42 subtype 19)
* Local Buffer Manager LRU Statistics Summary. single system
* --------------------------------------------------------------
SMF420PA DSECT ,
SMF42PA1 DS 0D
*
SMF42PAC DS 0CL800
*
SMF42JPA DS 1FL4 Interval length. This is the
* total time of the measurement
* period. (in seconds) total
*
SMF42J01 DS CL12 Reserved
*
SMF42JPB DS CL8 MVS system name.
SMF42JPC DS CL8 Reserved
SMF42JPD DS CL8 Reserved
*
SMF42JPE DS CL8 Average CPU time spent by
* BMF LRU processing during
* each cycle (in milli-seconds)
*
SMF42JPF DS CL8 Total CPU time for this
* record (in milli-seconds)
*
SMF42JP1 DS CL12 Reserved
SMF42JP6 DS 1FL4 Total number of write
* requests
SMF42JPG DS 1FL4 Number of buffer manager
* LRU intervals processed
* during this period.
*
SMF42JPH DS 1FL4 Number of buffer manager
* LRU intervals processed
* where BMF was over the goal
* and normal algorithms were
* bypassed to reclaim buffers.
*
SMF42JPI DS 1FL4 Total number of times that BMF
* was called in this interval
*
SMF42JP2 DS 1FL4 Number of buffer manager
* LRU intervals processed
* where BMF was over the goal,
* accelerated the aging, but
* did not go into panic mode.
SMF42JPJ DS 1FL4 Buffer manager number of
* hits during this interval.
SMF42JPK DS 1FL4 Buffer manager number of
* 'hits' current percentage
* during this interval. Value
* for the last LRU cycle before
* the SMF record was processed

50
z/OS V1R8 DFSMS Technical Update
The above example shows the records for below the 2 GB bar records where they are
prefixed SMF42xxx. The IGWSMF macro contains corresponding definitions for the above
the bar records where they are prefixed SMF2Axxx.
3.7 Device manager enhancements
z/OS V1R8 introduces several new enhancements to DEVMAN, the device manager address
space. These expand upon the DEVMAN CTRACE support and parameters introduced in
z/OS V1R7.
3.7.1 Rapid index rebuild
This enhancement significantly improves the performance of the convert routine used when
rebuilding a VTOC index. It has been implemented by moving the responsibility for this
function into the DEVMAN address space, which now reads the entire contents of the
volume’s VTOC into a data space with as little as one channel program. Prior to this
enhancement, DADSM read and processed each track in the VTOC one at a time, causing
significant I/O overhead due to multiple start subchannel operations. The original DADSM
code is retained so that if the DEVMAN address space is unavailable the rebuild can still take
place.
Maintenance
In order to use this support, APAR PK19625 must be installed for ICKDSF R17. Without this
APAR support, the old DADSM method is used. You must also have the PTF for APAR
OA17478 installed. It fixes a situation where a BUILDIX failure can occur as well as an
ABEND0C4 problem.
Implementing
With DFSMS V1R8 the VTOC convert routine automatically uses DEVMAN if it is available.
ICKDSF R17 is used as before with no change to the command format, but it must have
maintenance installed to test for and call the DEVMAN routines. ICKDSF R17 can be
common to lower levels of DFSMS where, if the enabling APAR PK19625 is installed, it does
not find DEVMAN or may find DEVMAN but not the convert routine, in which case the original
DADSM routines are used.

Chapter 3. DFSMSdfp V1R8 enhancements
51
Examples
In Figure 3-35 we show sample JCL to rebuild the indexed VTOC on volume SLD000. This
job first converts our existing indexed VTOC to OSFORMAT and then converts back to an
indexed VTOC.
Figure 3-35 ICKDSF BUILDIX job
Figure 3-36 is an example of the IEC604I and ICKDSF messages that are seen in SYSLOG
as a result of executing the JCL provided in Figure 3-35.
Figure 3-36 SYSLOG output from the indexed VTOC build JCL
Message IEC604I has been updated to show when DEVMAN is being used by the addition of
DEVMAN to the end of the message.
A typical example of message IEC604I in SYSLOG without using DEVMAN would be as
shown in Figure 3-37.
Figure 3-37 IEC604I VTOC convert message not using DEVMAN
A typical example of message IEC604I in SYSLOG using DEVMAN is as shown in
Figure 3-38.
Figure 3-38 IEC604I VTOC convert message using DEVMAN
//INITJ03 JOB SLT,MSGCLASS=H,REGION=1024K
/*JOBPARM SYSAFF=*
//*
//DSF EXEC PGM=ICKDSF,PARM=NOREPLYU
//SYSPRINT DD SYSOUT=*
//DISK DD UNIT=3390,VOL=SER=SLD000,DISP=OLD
//SYSIN DD *
BUILDIX DDNAME(DISK) OS
BUILDIX DDNAME(DISK) IX
$HASP373 INITJ03 STARTED - INIT 1 - CLASS A - SYS CIMN
ICK502I BUILDIX FUNCTION STARTED
ICK503I D000 REQUEST RECEIVED TO CONVERT VTOC TO OSFORMAT
ICK504I D000 VTOC FORMAT IS CURRENTLY IXFORMAT, REQUEST ACCEPTED
IEC604I VTOC CONVERT ROUTINE ENTERED ON D000,SLD000,DOS,DEVMAN
ICK513I D000 BUILDIX PROCESSING COMPLETED: VTOC IS NOW IN OSFORMAT
ICK502I BUILDIX FUNCTION STARTED
ICK503I D000 REQUEST RECEIVED TO CONVERT VTOC TO IXFORMAT
ICK504I D000 VTOC FORMAT IS CURRENTLY OSFORMAT, REQUEST ACCEPTED
ICK513I D000 BUILDIX PROCESSING COMPLETED: VTOC IS NOW IN IXFORMAT
$HASP395 INITJ03 ENDED
IEC604I VTOC CONVERT ROUTINE ENTERED ON D000,SLD000,DOS
IEC604I VTOC CONVERT ROUTINE ENTERED ON D000,SLD000,DOS,DEVMAN

52
z/OS V1R8 DFSMS Technical Update
3.7.2 New MODIFY DEVMAN parameters
Three new MODIFY DEVMAN parameters have been added. Their functions as described in
MVS System Commands, SA22-7627, and are shown in Figure 3-39.
Figure 3-39 New DEVMAN parameters
There are two primary considerations when using the MODIFY DEVMAN,RESTART
command:
You can use RESTART to avoid IPL when you install software. You can install most device
manager APARs by refreshing LLA (F LLA,REFRESH) and then restarting the device
manager (F DEVMAN,RESTART).
You can end and not restart the device manager address space by using the CANCEL
DEVMAN command. When you end the address space in this way, you must restart the
device manager with the DEVMAN cataloged procedure.
Examples
These commands should only be issued at the direction of the system programmer.
F DEVMAN,{DUMP}
{REPORT}
{RESTART}
The following are brief descriptions of the parameters.

DUMP
Captures a diagnostic dump of the device manager address space,
including the dataspace that contains device manager CTRACE records.

Note: The device manager CTRACE component name is SYSDMO. To connect
device manager to an output writer, use the command TRACE
CT,ON,COMP=SYSDMO.

REPORT
Provides basic information about the current activity and module
levels for the device manager address space.

RESTART
Terminates the device manager address space and restarts the device
manager in a new address space. The system allows any subtasks that
are active in the device manager address space at the time of the
restart to finish processing. The time allowed for subtask completion
is determined by using the average time taken by previous subtasks.
The system abnormally ends any subtasks that do not complete in time
before it restarts the address space.

Chapter 3. DFSMSdfp V1R8 enhancements
53
In Figure 3-40 we show an example of the MODIFY DEVMAN,DUMP command.
Figure 3-40 Dumping the device manager address space
The REPORT parameter shows the current status of the device manager address space.
Figure 3-41 is a report from a system where we would expect no device manager specific
functions to be executing.
Figure 3-41 DEVMAN report with no subtasks active
If issued at the precise moment a VTOC index is being rebuilt, it shows DEVMANs current
status, as seen in Figure 3-42.
Figure 3-42 DEVMAN report with ICKDSF BUILDIX job running
MODIFY DEVMAN,DUMP
IEA794I SVC DUMP HAS CAPTURED: 857
DUMPID=019 REQUESTED BY JOB (DEVMAN )
DUMP TITLE=COMPONENT=DEVICE MANAGER,COMPID=DF133,ISSUER=DMOVS00
1,JOBNAME=DEVMAN
IEF196I IGD101I SMS ALLOCATED TO DDNAME (SYS00020)
IEF196I DSN (SYSDOC.DUMP.DA.D070218.T195615.CIMN.S00019 )
IEF196I STORCLAS (DUMPSC) MGMTCLAS ( ) DATACLAS (
IEF196I )
IEF196I VOL SER NOS= DMPDP3
IEF196I IGD104I SYSDOC.DUMP.DA.D070218.T195615.CIMN.S00019 RETAINED,
IEF196I DDNAME=SYS00020
IEA611I COMPLETE DUMP ON SYSDOC.DUMP.DA.D070218.T195615.CIMN.S00019 865
DUMPID=019 REQUESTED BY JOB (DEVMAN )
FOR ASID (0419)
MODIFY DEVMAN,REPORT
DMO0030I DEVICE MANAGER REPORT 341
**** DEVMAN ****************************************************
* FMID: HDZ1180 *
* APARS: UA28529 DMOAT002 *
* NO SUBTASKS ARE ACTIVE *
**** DEVMAN ****************************************************
MODIFY DEVMAN,REPORT
DMO0030I DEVICE MANAGER REPORT 818
**** DEVMAN ****************************************************
* FMID: HDZ1180 *
* APARS: UA28529 DMOAT002 *
* SUBTASKS: *
* JOBNAME STARTED SERVICE UNIT STATUS *
* ------------------------------------------------------------ *
* INITJ03 17.52.22 READVTOC D000 SUBTASK RUNNING *
**** DEVMAN ****************************************************

54
z/OS V1R8 DFSMS Technical Update
The device manager address space can be restarted with the MODIFY DEVMAN,RESTART
command. Figure 3-43 shows an example of this.
Figure 3-43 Restarting the DEVMAN address space
Migration and coexistence
There are no migration or coexistence considerations when using the new DEVMAN support.
3.8 PDSE enhancements
In z/OS V1R8 PDSE there are the following enhancements:
Program objects built may use Program Object format 5 (PO5).
In prior releases of DFSMS, PDSE cached directory buffers in the SYSBMFDS data
space. With z/OS V1R8, PDSE has completely eliminated the 31-bit SYSBMF data space
and now uses 64-bit buffers. Directory pages are now cached in 64-bit storage in the
SMSPDSE and SMSPDSE1 address space. Several of the PDSE control block cell pools
have been moved from 31-bit SMSPDSE and SMSPDSE1 address space storage to
64-bit address space storage.
In this section we cover the enhancements to Partitioned Data Set Extended (PDSE) HSP
and directory storage size that are available now and the option to retain buffers beyond
close.
Significant changes to PDSE were introduced with z/OS V1R6, and we recommend that you
refer to the IBM Redbooks publication Partitioned Data Set Extended Usage Guide,
SG24-6106, for background information.
MODIFY DEVMAN,RESTART
IEF352I ADDRESS SPACE UNAVAILABLE
$HASP395 DEVMAN ENDED
IEA989I SLIP TRAP ID=X33E MATCHED. JOBNAME=*UNAVAIL, ASID=0419.
$HASP250 DEVMAN PURGED -- (JOB KEY WAS C04EEE54)
IEF196I 1 //IEESYSAS JOB TIME=1440,
IEF196I // MSGLEVEL=1
IEF196I 2 //DEVMAN EXEC IEESYSAS,PROG=DMOVS001
IEF196I STMT NO. MESSAGE
IEF196I 2 IEFC001I PROCEDURE IEESYSAS WAS EXPANDED USING SYSTEM
IEF196I LIBRARY SYS1.IBM.PROCLIB
IEF196I 3 XXIEESYSAS PROC PROG=IEFBR14
IEF196I 00050000
IEF196I 4 XXIEFPROC EXEC PGM=&PROG
IEF196I 00100000
IEF196I XX* THE IEESYSAS PROCEDURE IS SPECIFIED IN THE
IEF196I 00150000
IEF196I XX* PARAMETER LIST TO IEEMB881 BY MVS COMPONENTS
IEF196I 00200000
IEF196I XX* STARTING FULL FUNCTION SYSTEM ADDRESS SPACES.
IEF196I 00250000
IEF196I IEFC653I SUBSTITUTION JCL - PGM=DMOVS001
DMO0010I DEVICE MANAGER INITIALIZATION STARTED
DMO0000I DEVICE MANAGER INITIALIZATION COMPLETE

Chapter 3. DFSMSdfp V1R8 enhancements
55
3.8.1 PDSE PO5 module format compatibility on pre-z/OS V1R8 systems
There is support for use of the PDSE Program Object format 5 (PO5) module format on prior
releases. In Table 3-8 we show the toleration maintenance available for systems prior to z/OS
V1R8.
Table 3-8 PDSE PO5 format toleration maintenance
3.8.2 PDSE 64-bit virtual storage use enablement
Prior to z/OS V1R8, the SMSPDSE and SMSPDSE1 address spaces were 31-bit
implementations. This limited the number of concurrently opened PDSE members that could
exist in a system to approximately one million.
With z/OS V1R8, both address spaces are implemented in 64-bit mode. Now you can specify
up to 16 GB, which allows a much larger number of PDSE members to be open.
The options that enabled by use of 64-bit storage are:
PDSE_DIRECTORY_STORAGE (up to 16 GB) - related to SMSPDSE address space
PDSE1_DIRECTORY_STORAGE (up to 16 GB) - related to SMSPDSE1 address space
PDSE_HSP-SIZE (up to 2047 MB) - related to SMSPDSE address space
PDSE1_HSP-SIZE (up to 2047 MB) - related to SMSPDSE1 address space
We recommend that you refer to the options available for member IGDSMSxx of PARMLIB in
MVS Initialization and Tuning Guide, SA22-7591, and MVS Initialization and Tuning
Reference, SA22-7592.
3.8.3 PDSE 64-bit virtual storage option setting
For the PDSE1 options, you can implement changes by using the SETSMS command (in
which case they may revert to prior values at an IPL unless PARMLIB is also updated).
For PDSE and PDSE1 options you can update the appropriate values in the SYS1.PARMLIB
member IGDSMSxx, in which case they take effect at an IPL.
You can issue the SET SMS=xx command to have the system implement the options
contained in SYS1.PARMLIB member IGDSMSxx (where xx is a user-specified value). For
some options the address space may have to be restarted (applies to SMSPDSE1 only).
To verify the effect of making changes, we did the following:
Displayed the current options
Changed the specific options of interest
Displayed the options to verify that the change had taken effect
In Figure 3-44 on page 57 we show the SMS options before making changes. The options
that we will modify are highlighted bold.
APAR Description
OA13525 Allows use of load modules built on z/OS V1R8 to be used on prior systems

56
z/OS V1R8 DFSMS Technical Update
Note: The system default for both PDSE_DIRECTORY_STORAGE(n) and
PDSE1_DIRECTORY_STORAGE(n) is 2047, and in the display these both show as
2000M.
The system default for PDSE_HSP_SIZE(n) and for PDSE1_HSP_SIZE(n) are also 0,
which indicates that the respective hiperspace should not be started. However, in the
display PDSE1_HSP_SIZE shows as 256 MB, indicating that the value has been set rather
than defaulted at system IPL. In this case, the value was set in SYS1.PARMLIB as
PDSE1_HSP_SIZE(256). If we change it by command it reverts to 256 MB at the next IPL.

Chapter 3. DFSMSdfp V1R8 enhancements
57
Figure 3-44 Current IGDSMSxx option display
D SMS,OPTIONS
IGD002I 13:26:24 DISPLAY SMS 566
ACDS = SYS1.SMS.ACDS
COMMDS = SYS1.SMS.COMMDS
INTERVAL = 15 DINTERVAL = 150
SMF_TIME = YES CACHETIME = 3600
CF_TIME = 1800 PDSE_RESTARTABLE_AS = YES
PDSE_BMFTIME = 3600 PDSE1_BMFTIME = 3600
PDSE_LRUTIME = 60 PDSE1_LRUTIME = 50
PDSE_LRUCYCLES = 15 PDSE1_LRUCYCLES = 200
LOCAL_DEADLOCK = 15 GLOBAL_DEADLOCK = 4
REVERIFY = NO DSNTYPE = PDS
ACSDEFAULTS = NO PDSESHARING = EXTENDED
OVRD_EXPDT = NO SYSTEMS = 8
PDSE_HSP_SIZE = 0MB PDSE1_HSP_SIZE = 256MB
USE_RESOWNER = YES RLS_MAX_POOL_SIZE = 100MB
RLSINIT = YES RLSTMOUT = 0
COMPRESS = GENERIC LOG_OF_LOGS = IGWTVS.LOG.OF.LOGS
QTIMEOUT = 300 TVSNAME = 064
AKP = 1000 TV_START_TYPE = WARM
MAXLOCKS = (0,0)
CICSVR_INIT = YES CICSVR_DSNAME_PREFIX = DWWUSER.V3R1M0
CICSVR_RCDS_PREFIX = DWW
CICSVR_GRPNAME_SUFFIX = PROD
CICSVR_ZZVALUE_PARM =
CICSVR_UNDOLOG_CONTROL =
CICSVR_UNDOLOG_PREFIX = DWW
CICSVR_BACKOUT_CONTROL =
CICSVR_GENERAL_CONTROL =
Rls_MaxCfFeatureLevel = Z
RlsAboveThebarMaxPoolSize = 0
RlsFixedPoolSize = 0
PDSE_MONITOR = (YES,0,0) PDSE1_MONITOR = (YES,0,0)
PDSE_DIRECTORY_STORAGE = 2000M
PDSE1_DIRECTORY_STORAGE = 2000M
PDSE_BUFFER_BEYOND_CLOSE = NO
PDSE1_BUFFER_BEYOND_CLOSE = NO
GDS_RECLAIM = YES DSSTIMEOUT = 0
GDS_RECLAIM = YES DSSTIMEOUT = 0
BLOCKTOKENSIZE = NOREQUIRE FAST_VOLSEL = OFF
IGD002I 13:26:24 DISPLAY SMS

58
z/OS V1R8 DFSMS Technical Update
SETSMS commands issued and results
In Figure 3-45 we show the command
SETSMS PDSE_DIRECTORY_STORAGE(2000M) being issued, and the response that
indicates that it is invalid (because it is for the PDSE environment).
Figure 3-45 SETSMS PDSE_DIRECTORY_STORAGE(2000M) showing error
In Figure 3-46 we show the command
SETSMS PDSE1_DIRECTORY_STORAGE(2000M) being issued. In this case the command
is accepted, but without any confirmation of acceptance, so we follow the SETSMS command
with part of the output from the D SMS,OPTIONS command to show that it has been
accepted.
Figure 3-46 SETSMS PDSE1_DIRECTORY_STORAGE(2500M) and D SMS,OPTIONS commands
SETSMS PDSE_DIRECTORY_STORAGE(2000M)
IGD029I ERROR FOR SETSMS COMMAND 531
ERROR IS INVALID KEYWORD: PDSE_DIRECTORY_STORAGE
SETSMS PDSE1_DIRECTORY_STORAGE(2500M)
D SMS,OPTIONS
IGD002I 16:11:03 DISPLAY SMS 578
ACDS = SYS1.SMS.ACDS
COMMDS = SYS1.SMS.COMMDS
INTERVAL = 15 DINTERVAL = 150
SMF_TIME = YES CACHETIME = 3600
CF_TIME = 1800 PDSE_RESTARTABLE_AS = YES
PDSE_BMFTIME = 3600 PDSE1_BMFTIME = 3600
PDSE_LRUTIME = 60 PDSE1_LRUTIME = 50
PDSE_LRUCYCLES = 15 PDSE1_LRUCYCLES = 200
LOCAL_DEADLOCK = 15 GLOBAL_DEADLOCK = 4
REVERIFY = NO DSNTYPE = PDS
ACSDEFAULTS = NO PDSESHARING = EXTENDED
OVRD_EXPDT = NO SYSTEMS = 8
PDSE_HSP_SIZE = 0MB PDSE1_HSP_SIZE = 256MB
USE_RESOWNER = YES RLS_MAX_POOL_SIZE = 100MB
RLSINIT = YES RLSTMOUT = 0
COMPRESS = GENERIC LOG_OF_LOGS = IGWTVS.LOG.OF.LOGS
QTIMEOUT = 300 TVSNAME = 064
AKP = 1000 TV_START_TYPE = WARM
MAXLOCKS = (0,0)
CICSVR_GENERAL_CONTROL =
Rls_MaxCfFeatureLevel = Z
RlsAboveThebarMaxPoolSize = 0
RlsFixedPoolSize = 0
PDSE_MONITOR = (YES,0,0) PDSE1_MONITOR = (YES,0,0)
PDSE_DIRECTORY_STORAGE = 2000M
PDSE1_DIRECTORY_STORAGE = 2500M
PDSE_BUFFER_BEYOND_CLOSE = NO
PDSE1_BUFFER_BEYOND_CLOSE = NO

Chapter 3. DFSMSdfp V1R8 enhancements
59
In Figure 3-47 we show the command
SETSMS PDSE1_HSP_SIZE(512) being issued. In this case the command is accepted, but
without any confirmation of acceptance, so we follow the SETSMS command with part of the
output from the D SMS,OPTIONS command to show that it has been accepted.
Figure 3-47 SETSMS PDSE1_HSP_SIZE(512) and D SMS,OPTIONS commands
SET SMS=xx command method of updating SMS options
In order to use the SET SMS=xx command, a member of SYS1.PARMLIB must be available.
For this demonstration we created IGDSMS88 as a copy of the member used for the last
system IPL.
SETSMS PDSE1_HSP_SIZE(512)
D SMS,OPTIONS
IGD002I 17:39:30 DISPLAY SMS 646
ACDS = SYS1.SMS.ACDS
COMMDS = SYS1.SMS.COMMDS
INTERVAL = 15 DINTERVAL = 150
SMF_TIME = YES CACHETIME = 3600
CF_TIME = 1800 PDSE_RESTARTABLE_AS = YES
PDSE_BMFTIME = 3600 PDSE1_BMFTIME = 3600
PDSE_LRUTIME = 60 PDSE1_LRUTIME = 50
PDSE_LRUCYCLES = 15 PDSE1_LRUCYCLES = 200
LOCAL_DEADLOCK = 15 GLOBAL_DEADLOCK = 4
REVERIFY = NO DSNTYPE = PDS
ACSDEFAULTS = NO PDSESHARING = EXTENDED
OVRD_EXPDT = NO SYSTEMS = 8
PDSE_HSP_SIZE = 0MB PDSE1_HSP_SIZE = 512MB

60
z/OS V1R8 DFSMS Technical Update
In Figure 3-48 we show part of the contents of IGDSMS88 with the lines being changed
highlighted in bold.
Figure 3-48 IGDSMS88 to be used with SET SMS=88 command
SMS ACDS(SYS1.SMS.ACDS)
COMMDS(SYS1.SMS.COMMDS)
INTERVAL(15)
DINTERVAL(150)
DEADLOCK_DETECTION(15,4)
SMF_TIME(YES)
CF_TIME(1800)
RLSINIT(YES)
RLS_MAX_POOL_SIZE(100)
REVERIFY(NO)
ACSDEFAULTS(NO)
PDSESHARING(EXTENDED)
PDSE_RESTARTABLE_AS(YES)
PDSE1_MONITOR(YES)
PDSE1_LRUCYCLES(200)
PDSE1_LRUTIME(50)
PDSE_HSP_SIZE(256)
PDSE1_HSP_SIZE(512)
PDSE1_BMFTIME(3600)
PDSE_DIRECTORY_STORAGE(2400M)
PDSE1_DIRECTORY_STORAGE(2500M)

Chapter 3. DFSMSdfp V1R8 enhancements
61
In Figure 3-49 we show the command SET SMS=88 being issued and the results as
automatically listed by the SET SMS command.
The new values for the following can be seen highlighted bold:
PDSE_HSP_SIZE
PDSE1_HSP_SIZE
PDSE_DIRECTORY_STORAGE(2400M)
PDSE1_DIRECTORY_STORAGE(2400M)
Figure 3-49 SET SMS=88 command and listing of results
Note: In order for the new value for PDSE1_HSP_SIZE to take effect it is necessary to
restart SMSPDSE1. The new value for PDSE_HSP_SIZE does not take effect as a result
of the update. That only occurs with an IPL. However, the contents of IGDSMS88 have
been validated by issuing the SET SMS=88 command.
SET SMS=88
IEE252I MEMBER IGDSMS88 FOUND IN SYS1.PARMLIB
IGD031I SMS PARAMETERS 775
ACDS = SYS1.SMS.ACDS
COMMDS = SYS1.SMS.COMMDS
INTERVAL = 15 DINTERVAL = 150
CACHETIME = 3600 SMF_TIME = YES
CF-TIME = 1800 PDSE_RESTARTABLE_AS = YES
PDSE_BMFTIME = 3600 PDSE1_BMFTIME = 3600
PDSE_LRUTIME = 60 PDSE1_LRUTIME = 50
PDSE_LRUCYCLES = 15 PDSE1_LRUCYCLES = 200
LOCAL_DEADLOCK = 15 GLOBAL_DEADLOCK = 4
REVERIFY = NO ACSDEFAULTS = NO
USE_RESOWNER = YES DSNTYPE = PDS
GDS_RECLAIM = YES PDSESHARING = EXTENDED
OVRD_EXPDT = NO RLS_MAX_POOL_SIZE = 100MB
SYSTEMS = 8 COMPRESS = GENERIC
PDSE_HSP_SIZE = 256MB PDSE1_HSP_SIZE = 512MB
RLSINIT = YES RLSTMOUT = 0
CICSVR_INIT = NO CICSVR_DSNAME_PREFIX = DWW.
CICSVR_RCDS_PREFIX = DWW
CICSVR_GRPNAME_SUFFIX = PROD
CICSVR_ZZVALUE_PARM =
CICSVR_UNDOLOG_CONTROL =
CICSVR_UNDOLOG_PREFIX = DWW
CICSVR_BACKOUT_CONTROL =
CICSVR_GENERAL_CONTROL =
Rls_MaxCfFeatureLevel = Z
RlsAboveTheBarMaxPoolSize = 0
RlsFixedPoolSize = 0
DSSTIMEOUT = 0 FAST_VOLSEL = OFF
PDSE_MONITOR = (YES,0,0)
PDSE_DIRECTORY_STORAGE = 2400M
PDSE1_DIRECTORY_STORAGE = 2500M
PDSE_BUFFER_BEYOND_CLOSE = NO
PDSE1_BUFFER_BEYOND_CLOSE = NO
BLOCKTOKENSIZE = NOREQUIRE

62
z/OS V1R8 DFSMS Technical Update
3.8.4 Retain buffers beyond PDSE close
This enhancement can provide improved performance when using PDSE members that are
opened and closed very often. Before z/OS V1R8, when a PDSE was closed, the buffers for
directory and member data were purged. If the member was opened again, the buffers had to
be acquired again. Performance can be improved by specifying that the buffers be retained
after the PDSE member is closed.
This option applies to all PDSE members, so if used it results in more storage in use for the
PDSE memory cache.
PDSE extended sharing
You can define the extent of PDSE sharing. PDSEs can be shared between users of a single
processor only (normal) or shared between users on any processor in a multisystem complex
(extended). You select the type of sharing by using the PDSESHARING keyword in the
IGDSMSxx member of SYS1.PARMLIB.
Extended sharing is available only when PDSESHARING(EXTENDED) has been specified in
the IGDSMSxx member of SYS1.PARMLIB. Extended sharing applies when a PDSE is
allocated with DISP=SHR. It cannot be used with DISP=OLD.
With extended sharing, any number of users or systems can concurrently share a PDSE or
members within it for input (read) or output (write). While multiple members can be updated
by different users on the same system, only one user can update a member at a given time.
In addition, only one system can access the PDSE when the update is being done.
3.8.5 PDSE retain buffers beyond PDSE close enablement
For the PDSE1 options, you can implement changes by using the SETSMS command (in
which case they may revert to prior values at an IPL unless PARMLIB is also updated).
SMSPDSE1 must be restarted after the SETSMS command has been issued.
For PDSE and PDSE1 options you can update the appropriate values in the SYS1.PARMLIB
member IGDSMSxx, in which case they take effect at an IPL.
You can issue the SET SMS=xx command to have the system implement the options
contained in SYS1.PARMLIB member IGDSMSxx (where xx is a user-specified value).
SMSPDSE1 must be restarted after the SETSMS command has been issued.
To verify the effect of making changes, we did the following:
Displayed the current options.
Changed the specific options of interest.
Displayed the options to verify that the change had taken effect.
Restriction: PDSE extended sharing must be active for the buffer beyond PDSE close
enhancement.

Chapter 3. DFSMSdfp V1R8 enhancements
63
In Figure 3-50 we show the SMS options before making changes. The options that we modify
are highlighted bold. The options that relate to caching of PDSE buffers after close are:
PDSE_BUFFER_BEYOND_CLOSE(YES|NO)
PDSE1_BUFFER_BEYOND_CLOSE(YES|NO)
The default in both cases is NO.
Figure 3-50 Display SMS OPTIONS before BUFFER BEYOND CLOSE changes
D SMS,OPTIONS
IGD002I 19:08:05 DISPLAY SMS 930
ACDS = SYS1.SMS.ACDS
COMMDS = SYS1.SMS.COMMDS
INTERVAL = 15 DINTERVAL = 150
SMF_TIME = YES CACHETIME = 3600
CF_TIME = 1800 PDSE_RESTARTABLE_AS = YES
PDSE_BMFTIME = 3600 PDSE1_BMFTIME = 3600
PDSE_LRUTIME = 60 PDSE1_LRUTIME = 50
PDSE_LRUCYCLES = 15 PDSE1_LRUCYCLES = 200
LOCAL_DEADLOCK = 15 GLOBAL_DEADLOCK = 4
REVERIFY = NO DSNTYPE = PDS
ACSDEFAULTS = NO PDSESHARING = EXTENDED
OVRD_EXPDT = NO SYSTEMS = 8
PDSE_HSP_SIZE = 0MB PDSE1_HSP_SIZE = 256MB
USE_RESOWNER = YES RLS_MAX_POOL_SIZE = 100MB
RLSINIT = YES RLSTMOUT = 0
COMPRESS = GENERIC LOG_OF_LOGS = IGWTVS.LOG.OF.LOGS
QTIMEOUT = 300 TVSNAME = 064
AKP = 1000 TV_START_TYPE = WARM
MAXLOCKS = (0,0)
CICSVR_INIT = YES CICSVR_DSNAME_PREFIX = DWWUSER.V3R1M0
CICSVR_RCDS_PREFIX = DWW
CICSVR_GRPNAME_SUFFIX = PROD
CICSVR_ZZVALUE_PARM =
CICSVR_UNDOLOG_CONTROL =
CICSVR_UNDOLOG_PREFIX = DWW
CICSVR_BACKOUT_CONTROL =
CICSVR_GENERAL_CONTROL =
Rls_MaxCfFeatureLevel = Z
RlsAboveThebarMaxPoolSize = 0
RlsFixedPoolSize = 0
PDSE_MONITOR = (YES,0,0) PDSE1_MONITOR = (YES,0,0)
PDSE_DIRECTORY_STORAGE = 2000M
PDSE1_DIRECTORY_STORAGE = 2000M
PDSE_BUFFER_BEYOND_CLOSE = NO
PDSE1_BUFFER_BEYOND_CLOSE = NO
GDS_RECLAIM = YES DSSTIMEOUT = 0
BLOCKTOKENSIZE = NOREQUIRE FAST_VOLSEL = OFF

64
z/OS V1R8 DFSMS Technical Update
In Figure 3-51 we show the effect of issuing the following command and the resulting error
since it is not supported:
SETSMS PDSE_BUFFER_BEYOND_CLOSE(YES)
Figure 3-51 SETSMS PDSE_BUFFERBEYOND_CLOSE command and error message
In Figure 3-52 we show the effect of issuing the following command and confirmation that it
was accepted. In this case the PDSE environment is still set OFF. The SMSPDSE1 address
space needs to be restarted to take advantage of the changed setting.
SETSMS PDSE1_BUFFER_BEYOND_CLOSE(YES)
Figure 3-52 SETSMS PDSE1_BUFFER_BEYOND_CLOSE command and confirmation
SET SMS=xx command method of updating SMS options
In order to use the SET SMS=xx command, a member of SYS1.PARMLIB must be available.
For this demonstration we created IGDSMS89 as a copy of the member used for the last
system IPL.
In Figure 3-53 we show part of the contents of IGDSMS89 with the lines being changed
highlighted bold.
Figure 3-53 GDSMS89 to be used with SET SMS=89 command
SETSMS PDSE_BUFFER_BEYOND_CLOSE(YES)
IGD029I ERROR FOR SETSMS COMMAND 934
ERROR IS INVALID KEYWORD: PDSE_BUFFER_BEYOND_CLOSE
SETSMS PDSE1_BUFFER_BEYOND_CLOSE(YES)
IGW040I Buffer past close not active for SMSPDSE
IGW040I Buffer past close active for SMSPDSE1
SMS ACDS(SYS1.SMS.ACDS)
COMMDS(SYS1.SMS.COMMDS)
PDSESHARING(EXTENDED)
PDSE_RESTARTABLE_AS(YES)
PDSE1_MONITOR(YES)
PDSE1_LRUCYCLES(200)
PDSE1_LRUTIME(50)
PDSE1_HSP_SIZE(256)
PDSE1_BMFTIME(3600)
PDSE_BUFFER_BEYOND_CLOSE(YES)
PDSE1_BUFFER_BEYOND_CLOSE(YES)

Chapter 3. DFSMSdfp V1R8 enhancements
65
In Figure 3-54 we show the following command and the results as listed automatically that
show that the changes have been made, in this case for both PDSE and PDSE1.
SET SMS=89
Figure 3-54 SET SMS=89 command and listing of results
For PDSE this has verified that the syntax of member IGDSMS89 is good, but the change
does not take effect until an IPL is done using this member.
For PDSE1, the change takes place with an IPL using this member, but the SMSPDSE1
address space uses the new setting if it was restarted.
SET SMS=89
IEE252I MEMBER IGDSMS89 FOUND IN SYS1.PARMLIB
IGD031I SMS PARAMETERS 985
ACDS = SYS1.SMS.ACDS
COMMDS = SYS1.SMS.COMMDS
INTERVAL = 15 DINTERVAL = 150
CACHETIME = 3600 SMF_TIME = YES
CF-TIME = 3600 PDSE_RESTARTABLE_AS = YES
PDSE_BMFTIME = 3600 PDSE1_BMFTIME = 3600
PDSE_LRUTIME = 60 PDSE1_LRUTIME = 50
PDSE_LRUCYCLES = 15 PDSE1_LRUCYCLES = 200
LOCAL_DEADLOCK = 15 GLOBAL_DEADLOCK = 4
REVERIFY = NO ACSDEFAULTS = NO
USE_RESOWNER = YES DSNTYPE = PDS
GDS_RECLAIM = YES PDSESHARING = EXTENDED
OVRD_EXPDT = NO RLS_MAX_POOL_SIZE = 100MB
SYSTEMS = 8 COMPRESS = GENERIC
PDSE_HSP_SIZE = 0MB PDSE1_HSP_SIZE = 256MB
RLSINIT = NO RLSTMOUT = 0
CICSVR_INIT = NO CICSVR_DSNAME_PREFIX = DWW.
CICSVR_RCDS_PREFIX = DWW
CICSVR_GRPNAME_SUFFIX = PROD
CICSVR_ZZVALUE_PARM =
CICSVR_UNDOLOG_CONTROL =
CICSVR_UNDOLOG_PREFIX = DWW
CICSVR_BACKOUT_CONTROL =
CICSVR_GENERAL_CONTROL =
Rls_MaxCfFeatureLevel = Z
RlsAboveTheBarMaxPoolSize = 0
RlsFixedPoolSize = 0
DSSTIMEOUT = 0 FAST_VOLSEL = OFF
PDSE_MONITOR = (YES,0,0)
PDSE1_MONITOR = (YES,0,0)
PDSE_DIRECTORY_STORAGE = 2000M
PDSE1_DIRECTORY_STORAGE = 2000M
PDSE_BUFFER_BEYOND_CLOSE = YES
PDSE1_BUFFER_BEYOND_CLOSE = YES
BLOCKTOKENSIZE = NOREQUIRE

66
z/OS V1R8 DFSMS Technical Update
3.8.6 DFSMS V1R8 PDSE enhancements in storage requirements
Implementation of PDSE_BUFFER_BEYOND_CLOSE or
PDSE1_BUFFER_BEYOND_CLOSE results in a requirement for more storage than before,
so the ability to specify higher values for PDSE_HSP_SIZE, PDSE1_HSP_SIZE,
PDSE_DIRECTORY_STORAGE, or PDSE1_DIRECTORY_STORAGE may be important.
There is no new instrumentation provided to specifically address effects on the Buffer
Management Facility (BMF) of these higher values because the PDSE address spaces are
always using 64-bit storage for buffers.
However, the techniques presented in Section 9.8 of Partitioned Data Set Extended Usage
Guide, SG24-6106, remain valid. The SMF42T1 program referenced in that section must be
re-assembled using the V1R8 version of the macros.
A simplified version of the information in Partitioned Data Set Extended Usage Guide,
SG24-6106, is presented in 3.9, “PDSE buffer management statistics” on page 66.
3.9 PDSE buffer management statistics
SMF is used to record data that may be useful in determining how the PDSE buffer
management function is performing, and may allow values to be tuned to improve
performance or reduce over commitment of real storage. BMF is responsible for managing
access to and from the PDSE hiperspace, and other than through SMF there is no way to
determine whether hiperspace is being used.
SMF record type 42, subtype 1 provides overall buffer use totals for SMSPDSE and, if
implemented, separately for SMSPDSE1, and totals for each storage class being used by the
SMSPDSE[1] address spaces and their hiperspaces.
3.9.1 BMF data capture preparation
In order to have SMF capture BMF data, the following items must be addressed:
PDSE_BMFTIME and, if SMSPDSE1 is in use, PDSE1_BMFTIME must be set to
appropriate values. The default setting of 3600 seconds (1 hour) may be too long for
meaningful analysis. Note that if no PDSE eligible activity occurs, no BMF data is captured
because hiperspace cache is not used. After eligible PDSE data sets are accessed, the
BMF data recording function starts, and it continues even after the triggering PDSE is no
longer in use.
Note that there is no eye catcher in the data to distinguish BMF data associated with
SMSPDSE from that associated with SMSPDSE1. However, as each address space has
its own LRUCYCLES definition, making these different does provide a way of inferring
which is which. The default setting for SMSPDSE is 15, and for SMSPDSE1 is 50. Note
that LRUCYCLES is one of the variables that may need to be adjusted to limit CPU or real
storage use, so if they are changed, be sure to maintain some difference between the two.
3.9.2 BMF analysis preparation
We provide a sample program that formats the information from what was recorded by SMF.
The program needs to be assembled and linked into an appropriate data set before use.
Refer to “SMF record type 42 sub type 1 data” on page 478 for information about building the
program.

Chapter 3. DFSMSdfp V1R8 enhancements
67
So that the user can relate the program output to the SMF manuals and other sources of
information, the field names as defined in the SMF macros are used.
The SMF records that are collected from the BMF totals are shown in Figure 3-55.
Figure 3-55 Extract of the BMF SMF data fields used
For the same SMF record as above, the data collected for the storage classes are shown in
Figure 3-56.
Figure 3-56 Extract of the BMF storage class fields used
Depending on the structure of the system, there may be one or more BMF totals sections,
which are shown in the program output as BMF TOTALS SET #: nnnnnnn, where nnnnnnn is
a numeric value, and there may be one or more storage class sections that are identified by
their names as defined to SMS. Normally, there is one set of SMF records for SMSPDSE and
one for SMSPDSE1.
The extract program assumes that the data it is processing is an extract from the SMF data
prepared by the SMF-supplied extraction program IFASMFDP. The SMFPRMxx member of
SYS1.PARMLIB should be checked to ensure that SMF record type 42 is being selected or
that it is not being suppressed.
The extract may be taken from the running copy of the VSAM ‘SYS1.MAN*’ data sets or from
a previously extracted sequential data set. IFASMFDP manages the correct handing of the
input data set, and regardless of the form of the input, produces a sequential data set. In
order to limit the amount of data extracted, the recommended control statement input to
IFASMFDP limits the output to select only records of type 42 subtype 1.
SMF4201A DSECT ,
SMF42BMF DS 0CL0036 Description of BMF totals section
SMF42TNA DS 1FL4 Total number of storage classes
SMF42TMT DS 1FL4 Interval length (total time of measurement
period)
SMF42TRT DS 1FL4 Total number of member data page reads
SMF42TRH DS 1FL4 Total number of member data page read hits
(found in BMF)
SMF42TDT DS 1FL4 Total number of directory data page reads
SMF42TDH DS 1FL4 Total number of directory data page read hits
(found in BMF)
SMF42BUF DS 1FL4 Total number of active BMF 4K buffers @18A
SMF42BMX DS 1FL4 High water mark of BMF buffers @18A
SMF42LRU DS 1FL2 BMF LRU interval time @18A
SMF42UIC DS 1FL2 BMF LRU cycles before buffer cast out @18A
SMF4201B DSECT ,
SMF42SC DS 0CL0048 Description of Stor Class Sum section @OW00573
SMF42PNA DS 0CL0032 Storage class name
SMF42PNL DS 1FL2 Storage class name length
SMF42PNN DS 1CL0030 Storage class name
SMF42SRT DS 1FL4 Total number of member data page reads
SMF42SRH DS 1FL4 Total number of member data page read hits X
(found in BMF)
SMF42SDT DS 1FL4 Total number of directory data page reads
SMF42SDH DS 1FL4 Total number of directory data page read hits X
(found in BMF)

68
z/OS V1R8 DFSMS Technical Update
If the IFASMFDP is set to extract all type 42 records, all those that are not type 42 subtype 1
are discarded by the SMF42T1.
Listings of the program source and examples of JCL to assemble and link the program are
shown in “SMF record type 42 sub type 1 data” on page 478. Assembly is required only once
for initial implementation of a given level of the operating system, then if maintenance is
applied to the IGWSMF macro.
An example of the job required to extract the SMF records then run the formatting program is
shown in Figure 3-57. This assumes that the formatting program is called SMF42T1 and that
it has previously been assembled and stored in the data set PDSERES.SMF42T1J.PDS (an
example name only). The IFASMFDP selection specification is part of the OUTDD statement
OUTDD(OUTDD,TYPE(42(1))).
Figure 3-57 Example of JCL to format SMF record type 42 subtype 1 records
//MHLRES1S JOB (999,POK),MSGLEVEL=1,NOTIFY=MHLRES1
//EXTRACT EXEC PGM=IFASMFDP
//SYSPRINT DD SYSOUT=A
//ADUPRINT DD SYSOUT=A
//*DUMPIN DD DISP=SHR,DSN=SMFDATA.ALLRECS
//DUMPIN DD DISP=SHR,DSN=SYS1.SC64.MAN1
//DUMPOUT DD DUMMY
//OUTDD DD DSN=&T1,SPACE=(CYL,(10,5)),RECFM=VB,LRECL=5096,
// DISP=(,CATLG,DELETE),
// UNIT=SYSDA
//SYSIN DD *
INDD(DUMPIN,OPTIONS(DUMP))
OUTDD(OUTDD,TYPE(42(1)))
/*
//FORMAT EXEC PGM=SMF42T1
//STEPLIB DD DISP=SHR,DSN=PDSERES.SMF42T1J.PDS
//SYSUDUMP DD SYSOUT=A
//SMFIN DD DISP=SHR,DSN=&T1
//PRINT DD SYSOUT=A,RECFM=UA

Chapter 3. DFSMSdfp V1R8 enhancements
69
An example of output from the IFASMFDP program is shown in Figure 3-58. The full output
shows all the records found, and the right-most column shows the number of records
selected. In this partial output example, no records were selected for types 41 and 43, and 41
records were found for the type 42 record, which is consistent with the selection criteria
specified on the OUTDD control statement.
Figure 3-58 Example of part of the output from IFASMFDP showing selection of type 42 records
3.9.3 SMF statistics interpretation
There are no absolute values that are correct or wrong for the tuning parameters. The SMF
data is useful to determine what is happening, then after a change verify that the change had
some effect.
An example of output from the SMF type 42 record formatting program is shown in
Figure 3-59. Note that the value for SMF42TNA of 1corresponds to the one SCLASS entries
that follows it. By checking the output from the d sms,options command we infer from the
value of 0006000 for SMF42LRU that this is data from SMSPDSE.
Figure 3-59 Example 1 of output from program formatting SMF type 42 subtype 1 records
IFA010I SMF DUMP PARAMETERS
IFA010I END(2400) -- DEFAULT
IFA010I START(0000) -- DEFAULT
IFA010I DATE(1900000,2099366) -- DEFAULT
IFA010I OUTDD(OUTDD,TYPE(42(1))) -- SYSIN
IFA010I INDD(DUMPIN,OPTIONS(DUMP)) -- SYSIN
IFA020I OUTDD -- SYS07053.T123632.RA000.MHLRES1S.T1.H01
IFA020I DUMPIN -- SMFDATA.ALLRECS.G3142V00
SUMMARY ACTIVITY REPORT
START DATE-TIME 02/21/2007-18:40:00 END DATE-TIME 02/21/2007-20:10:00
RECORD RECORDS PERCENT AVG. RECORD MIN. RECORD MAX. RECORD RECORDS
TYPE READ OF TOTAL LENGTH LENGTH LENGTH WRITTEN
2 1 .01 % 18.00 18 18 1
.
.
40 216 2.98 % 74.00 74 74 0
41 6 .08 % 332.00 332 332 0
42 651 8.98 % 605.10 176 32,620 2
SMF TYPE 42 S/TYPE 1 RECORDS. COLS USE SMF NAMES
BMF TOTALS SET #: 0000001
HH:MM:SS YYYYDDD SMF42TNA SMF42TMT SMF42TRT SMF42TRH SMF42TDT SMF42TDH SMF42BUF SMF42BMX SMF42LRU SMF42UIC
19:32:51 2007052 0000001 0003599 0000000 0000000 0000000 0000000 0000010 0000121 0006000 0000015

SMF42PNN (SCLASS): **NONE**
HH:MM:SS YYYYDDD SMF42SRT SMF42SRH SMF42SDT SMF42SDH
19:32:51 2007052 0000000 0000000 0000000 0000000

70
z/OS V1R8 DFSMS Technical Update
A further example of output from the SMF type 42 record formatting program is shown in
Figure 3-60. Note that the value for SMF42TNA of 1 corresponds to the one SCLASS entry
that follows it. By checking the output from the d sms,options command we infer from the
value of 0005000 for SMF42LRU that this is data from SMSPDSE1.
Figure 3-60 Example 2 of output from program formatting SMF type 42 subtype 1 records
For reference, some of the data reported shows what the current settings are, and there
would not be any change in these unless some specific action was taken, but the various
counts can be expected to change.
For each PDSE data space in use, that is always one and optionally two, there will be data for
the BMF and also for the processing as it relates to the storage classes involved.
The data that is recorded by SMF and ultimately returned in the SMF type 42 subtype 1
records are as shown in Figure 3-55 on page 67 and Figure 3-56 on page 67.
The following comments are presented in the order shown in the collection data fields in the
examples mentioned above.
BMF totals data
There is no specific data in the BMF totals data section to indicate which of the SMSPDSE or
SMSPDSE1 data spaces it applies to. In the SMF42T1 program output, the sections are
identified in the output in the program header as:
BMF TOTALS SET #: 000000n
Where n is replaced by the relative number of the BMF totals section found.
The default system settings for LRUCYCLES and BMFTIME are different for the SMSPDSE
and SMSPDSE1 data sets, so as long as a difference is retained in any site settings between
one or another of these, you can determine which one is for SMSPDSE and which for
SMSPDSE1. Example 3-1 shows two lines from a sample report. You can see the values
0005000 and 000200 under the SMF42LRU and SMF42UIC columns, respectively. From the
output from the command d sms,options you can see both the LRUCYCLES and BMFTIME
values, and from that determine which set of BMF totals is which. Refer to the SMF42LRU
explanation below for comments on the apparent multiplication factor applied over what was
specified in the IGDSMSxx parameters.
Example 3-1 Sample program showing SMF42LRU and SMF42UIC columns
YYYYDDD SMF42TNA SMF42TMT SMF42TRT SMF42TRH SMF42TDT SMF42TDH SMF42BUF SMF42BMX SMF42LRU SMF42UIC
2004327 0000002 0003599 0000000 0000000 0000010 0000008 0000011 0001080 0005000 0000200
BMF TOTALS SET #: 0000001
HH:MM:SS YYYYDDD SMF42TNA SMF42TMT SMF42TRT SMF42TRH SMF42TDT SMF42TDH SMF42BUF SMF42BMX SMF42LRU SMF42UIC
19:32:51 2007052 0000003 0003599 0000001 0000000 0000056 0000047 0000017 0001399 0005000 0000200

SMF42PNN (SCLASS): STANDARD
HH:MM:SS YYYYDDD SMF42SRT SMF42SRH SMF42SDT SMF42SDH
19:32:51 2007052 0000000 0000000 0000000 0000000

Chapter 3. DFSMSdfp V1R8 enhancements
71
The columns are:
SMF42TNA - total number of storage classes
Number of storage classes defined in the system. This is useful to correlate the number of
storage classes reported on in the storage class data report.
SMF42TMT - interval length for data gathering
This is the value defined in the BMF_TIME fields and may not be what is being used,
because this value is documented as overridden by the SMF INTVAL if the SMS OPTION
SMS_TIME is set to YES.
SMF42TRT - total number of member data page reads
This is the overall number of member data page reads, including those found in BMF.
SMF42TRH - total number of member data page reads found in BMF
This is the number of member data page reads satisfied from BMF. When BMF is
functioning in support of a PDSE, we expect to see that SMF42TRH is not zero, and that it
is less than SMF42TRT. Whether or not member data is cached by BMF depends on the
MSR value specified in the storage class associated with a particular data set being low.
SMF42TDT - total number of directory data page reads
This is the overall number of directory data page reads, including those found in BMF.
SMF42TDH - total number of directory data page reads found in BMF
This is the number of directory data page reads satisfied from BMF. When BMF is
functioning in support of a PDSE, we expect to see that SMF42TDH is not zero, and that it
is less than SMF42TDT. Directory data page reads should be managed by BMF whether
or not the data set is associated with a storage class.
SMF42BUF - total number of active BMF buffers
This number should correspond to the number of 4-K pages that will fit in the hiperspace
as defined by the HSP_SIZE.
SMF42BMX - high water mark of BMF buffers
This number represents the largest number of BMF buffers used from the total number of
BMF buffers available. If the number presented here is consistently significantly lower that
the number of buffers available, this may be an indication that the HSP_SIZE can safely
be reduced. Likewise, if the number presented here is consistently close to the total
number of buffers available, it may be necessary to consider increasing the HSP_SIZE.
However, the number of buffers in use is influenced by the values used for LRUCYCLES
and LRUTIME, and it may be that adjusting these should be done before considering
changing HSP_SIZE, as that requires a system IPL to implement.
SMF42LRU - BMF LRU interval time
This value corresponds to the PDSE[1]_LRUTIME value as specified or as defaulted in
the IGDSMSxx PARMLIB options member, or as subsequently altered by SETSMS
command. Note that the value as present is shown as multiplied by 100.
SMF42UIC - BMF LRU cycles before buffer cast out
This value corresponds to the PDSE[1]_LRUCYCLES value as specified in the
IGDSMSxx options member, or as subsequently altered by SETSMS command. Note that
Note: The number assigned to the BMF totals heading represents the relative order that
the data appears in the SMF type 42 subtype 1 records. For the life of a given IPL, it is
likely that the order will remain the same if there is more than one, but it is possible that
after a subsequent IPL, the order will have changed.

72
z/OS V1R8 DFSMS Technical Update
LRUTIME is not an actual time value, but an indication of how many times a buffer may
remain in the buffer but not allocated when buffer management cycle runs, which is
controlled by LRUCYCLES.
Storage class summary data
There is one of these sections for each storage class in the system related to each of the
SMSPDSE or SMSPDSE1 address spaces. For ease of interpretation, storage classes
intended to manage the PDSE buffer management processes should be named according to
a convention that indicates that they are intended for PDSE data set management, and
indicate whether the management intended was to have the PDSE cached in the PDSE
hiperspaces.
SMF42PNL - storage class data section name length
Reference information
SMF42PNN - Storage class name
Reference information
SMF42SRT - total number of member data page reads
This is the overall number of members read.
SMF42SRH - total number of member data pages read from BMF
This is the number of member data page reads supplied from the BMF buffers.
When a PDSE associated with this storage class is being serviced by BMF, the
SMF42SRH value should not be zero, but it will be less than the total number of member
data reads. This shows zero if the storage class does not have a low MSR value.
SMF42SDT - total number of directory data page reads
This is the total number of data pages read using this storage class.
SMF42SDH - total number of directory data page reads serviced by BMF.
This is the total number of directory data pages read that were found in the BMF buffers.
Directory pages are eligible for BMF processing whether or not the PDSE is associated
with an eligible storage class.
The value in SMF42SDH should not be zero, and will likely be less than the value in
SMF42SDT if the BMF function is managing the data to and from hiperspace.
Recommendation for further analysis
If the SMF42SRH results are significantly lower than SMF42SRT, it is likely to be an
indication that for some reason few PDSE data sets are being cached to hiperspace. If the
PDSE data sets in use are being used by single or very few users at a time, then there may
be no concern. However, if a PDSE is intended to be shared, then the benefits of using
hiperspace should be investigated.
One comparatively simple way to measure whether a particular data set comes into that
category is to assign it its own storage class for test purposes. That requires the creation of
Attention: The higher that LRUCYCLES is set the more CPU time is used running the
BMF process, and the lower that LRUTIME is set the more real storage remains allocated.
Tuning LRUCYCLES and LRUTIME for each of the SMSPDSE and SMSPDSE1 address
spaces can be carried out without a system IPL. Any changes should be carefully
monitored in conjunction with measurements of the committed CPU and real storage
frames. Any sudden increase in CPU or REAL storage commitment may reflect the
introduction of a PDSE into the system.

Chapter 3. DFSMSdfp V1R8 enhancements
73
the storage class if not already available, and probably requires an adjustment to the ACS
routines to ensure that the data set is assigned, by itself, to the test storage class. Once
implemented, the SMF type 42 subtype 1 records automatically pick up the additional storage
class, and the effect of changes to the MSR can be monitored.
Once the data set is being managed by BMF, any specific changes made to improve the BMF
member data hit ratio can then be fitted back into the original storage class.
3.10 Integrated Catalog Forward Recovery Utility
The Integrated Catalog Forward Recovery Utility (ICFRU) was originally available as a
separate product. With z/OS V1R7, it was integrated into the base product. With z/OS V1R8
the documentation was expanded to include details of a technique to verify readiness of the
product for use in a real disaster.
In order to use ICFRU a mechanism must exist to restore a valid version of a given catalog,
so a backup process must exist, and the correct SMF records need to be recorded and saved
in appropriate data sets.
The full description of ICFRU is contained in Appendix A of DFSMS Managing Catalogs,
SC26-7409.
3.10.1 ICFRU system flow
In summary, ICFRU is a pair of programs that are used to update a valid catalog from activity
records that have been recorded in the system SMF data sets.
Program CRURRSV is Integrated Catalog Forward Recovery Record Selection and
Validation, and CRURRAP is Integrated Catalog Forward Recovery Record Analysis and
Processing.
You can control the execution of ICFRU programs through the specification of execution
parameters (the same parameters for CRURRSV and CRURRAP) and by providing the
appropriate SMF record data and an IDCAMS EXPORT of the original catalog.
To use the ICFRU you need:
The name of the catalog to be recovered
Recovery start date and time - the date and time at which the backup to be used for
recovery was made
Recovery stop date and time - the date and time that correspond to the closing of the
catalog (or the time after which no updates were possible)
An SMF gap time - the (approximate) interval just smaller than the minimum time used to
fill (or switch) an SMF recording data set
A multi-system clock difference (approximate) maximum difference between the TOD
clocks of any two systems sharing the catalog
All SMF dump data sets spanning the recovery period
A sort utility and appropriate sort control
The IDCAMS EXPORT data set to be used as the basis for recovery

74
z/OS V1R8 DFSMS Technical Update
In Figure 3-61 we show the overall flow of the ICFRU process.
Figure 3-61 ICFRU Process flow
3.10.2 ICFRU installation readiness overview
This section provides a practical example based on Appendix A, section “Confirming
Installation Readiness,” of the manual DFSMS Managing Catalogs, SC26-7409.
SYSPRINT
REPORTS
(SYSLOG)
REPORTS
(SYSLOG)
SMF DUMP DATA SETS
IDCAMS
IMPORT
SORT
CRURRSV
PARAMETERS
PARAMETERS
SYSPRINT
REPORTS
(SYSPRINT)
MESSAGES
(SYSLOG)
SYSIN
SELECTED
SMF CATALOG
RECORDS
NEW
CATALOG
EXPORT COPY
RECOVERED
CATALOG
OLD CATALOG
EXPORT COPY
SORTED SMF
CATALOG
RECORDS
SMFIN
CRURRAP
SORTIN
SORTOUT
EXPIN SMFIN
INFILE
OUTDATASET
PARAMETERS
Note: This is a simulation of the process that allows you to verify that all items required are
available. You must set up the complete process for recovery that runs automatically to
ensure that you have all the data necessary to restore catalogs in the event of an
unexpected failure.

Chapter 3. DFSMSdfp V1R8 enhancements
75
The simulation is done by carrying out the following phases:
Phase 1
One-time preparation steps that need only be run once. See 3.10.3, “Installation readiness
preparation” on page 75.
Phase 2
Execution steps that may be repeated as necessary. See 3.10.4, “Installation readiness
verification” on page 77.
3.10.3 Installation readiness preparation
In this section we discuss installation readiness preparation.
Step 1 - verify that SMF is recording record types 61, 65, and 66
In Figure 3-62 on page 76 we show command D SMS,OPTIONS being issued and the
results. The options of interest are bolded. The TYPE entries for both SUBSYS(STC..) and
SYS.. must both cover the required records. In this case all records other than record type 99
are being collected. If the display had shown that any of record types 61, 65, or 66 were not
included, the SMFPRMxx PARMLIB member must be updated and implemented.
Note: It is necessary to carry out this check on all systems that share the same catalogs.

76
z/OS V1R8 DFSMS Technical Update
Figure 3-62 D SMF,OPTIONS command and results
D SMF,O
IEE967I 18.47.14 SMF PARAMETERS 451
MEMBER = SMFPRM00
MULCFUNC -- DEFAULT
BUFUSEWARN(25) -- DEFAULT
BUFSIZMAX(0128M) -- DEFAULT
SYNCVAL(00) -- DEFAULT
DUMPABND(RETRY) -- DEFAULT
SUBSYS(STC,NOINTERVAL) -- SYS
SUBSYS(STC,NODETAIL) -- SYS
SUBSYS(STC,EXITS(IEFUSO)) -- PARMLIB
SUBSYS(STC,EXITS(IEFUJP)) -- PARMLIB
SUBSYS(STC,EXITS(IEFUJI)) -- PARMLIB
SUBSYS(STC,EXITS(IEFACTRT)) -- PARMLIB
SUBSYS(STC,EXITS(IEFU85)) -- PARMLIB
SUBSYS(STC,EXITS(IEFU84)) -- PARMLIB
SUBSYS(STC,EXITS(IEFU83)) -- PARMLIB
SUBSYS(STC,EXITS(IEFU29)) -- PARMLIB
SUBSYS(STC,TYPE(0:98,100:255)) -- PARMLIB
SYS(NODETAIL) -- PARMLIB
SYS(NOINTERVAL) -- PARMLIB
SYS(EXITS(IEFU29)) -- PARMLIB
SYS(EXITS(IEFUTL)) -- PARMLIB
SYS(EXITS(IEFUJI)) -- PARMLIB
SYS(EXITS(IEFUSO)) -- PARMLIB
SYS(EXITS(IEFUJP)) -- PARMLIB
SYS(EXITS(IEFUSI)) -- PARMLIB
SYS(EXITS(IEFUJV)) -- PARMLIB
SYS(EXITS(IEFACTRT)) -- PARMLIB
SYS(EXITS(IEFU85)) -- PARMLIB
SYS(EXITS(IEFU84)) -- PARMLIB
SYS(EXITS(IEFU83)) -- PARMLIB
SYS(TYPE(0:98,100:255)) -- PARMLIB
NOBUFFS(MSG) -- PARMLIB
LASTDS(MSG) -- PARMLIB
LISTDSN -- PARMLIB
SID(SC64) -- PARMLIB
DDCONS(NO) -- PARMLIB
JWT(2400) -- PARMLIB
MEMLIMIT(NOLIMIT) -- PARMLIB
STATUS(010000) -- PARMLIB
MAXDORM(3000) -- PARMLIB
INTVAL(10) -- PARMLIB
REC(PERM) -- PARMLIB
NOPROMPT -- PARMLIB
DSNAME(SYS1.SC64.MAN3) -- PARMLIB
DSNAME(SYS1.SC64.MAN2) -- PARMLIB
DSNAME(SYS1.SC64.MAN1) -- PARMLIB

Chapter 3. DFSMSdfp V1R8 enhancements
77
Step 2 - create initial IDCAMS EXPORT data set and GDG
For the purposes of this simulation it is advisable to ensure that the catalog is free of errors
since any errors may affect the final compare. In a real disaster situation, whatever state the
catalog is in when it is backed up (for example, EXPORTED) is all you have to deal with.
You need to retain information about the status of the catalog and the EXPORT process, so
data sets are defined to capture this information. In this example Generation Data Group
(GDG) data sets are used, but any method that provides unique data set names can be used.
You need to establish data set and GDG information one time, then the same structure for
subsequent runs.
In Figure 3-63 we show the job to create the data sets and GDGs. The DELETE command for
the GDGs can be omitted the first time the job is run.
Figure 3-63 ICFRU readiness check - set up data sets and GDGs
3.10.4 Installation readiness verification
To verify:
1.Create an IDCAMS EXPORT. This is similar to the job that is run regularly.
2.Wait a while.
3.Create an IDCAMS EXPORT. This is similar to the job that is run regularly after the one
from step 1.
//MHLRES1I JOB (999,POK),'MHLRES2',CLASS=A,MSGCLASS=T,
// NOTIFY=&SYSUID,TIME=1440,REGION=6M
/*JOBPARM L=999,SYSAFF=*
//*
//*************************************************************
//* THE GDGS AND MODEL DSCBS FOR THE DATA SETS ASSOCIATED WITH
//* EXPORTING CATALOG UCAT.VSBOX01 ARE DEFINED. BY CONVENTION
//* ALL BACKUP DATA SETS WILL START WITH MHLRES1.BCAT .
//*************************************************************
//SETUPDS EXEC PGM=IDCAMS
//LIST DD DSN=MHLRES1.BCAT.LIST.DCB,DISP=(NEW,CATLG),
// VOL=SER=SBOX01,UNIT=SYSDA,SPACE=(0,0),
// DCB=(RECFM=VBA,LRECL=125,BLKSIZE=4250)
//EXPORT DD DSN=MHLRES1.BCAT.EXPORT.DCB,DISP=(NEW,CATLG),
// VOL=SER=SBOX01,UNIT=SYSDA,SPACE=(0,0),
// DCB=(RECFM=VBS,LRECL=32404)
//SYSPRINT DD SYSOUT=*
//SYSIN DD *
DELETE MHLRES1.BCAT.CATALOG.LISTING GDG
DELETE MHLRES1.BCAT.CATALOG.BACKUP GDG
DELETE MHLRES1.BCAT.PROBLEM.LISTING GDG
DEFINE GDG (NAME(MHLRES1.BCAT.CATALOG.LISTING) LIM(7) NEMP SCR -
FOR(9999))
DEFINE GDG (NAME(MHLRES1.BCAT.CATALOG.BACKUP) LIM(7) NEMP SCR -
FOR(9999))
DEFINE GDG (NAME(MHLRES1.BCAT.PROBLEM.LISTING) LIM(7) NEMP SCR -
FOR(9999))
LISTC LEVEL(MHLRES1.BCAT) ALL

78
z/OS V1R8 DFSMS Technical Update
4.Create an EXPORT data set from the EXPORT created in step 1 together with SMF
records that cover the period between steps 1 and 3.
5.Compare the EXPORT created in step 4 with that taken at step 3.
Note: In a real situation, at step 5 the EXPORT from step 4 is what will be used as an
IMPORT to create the replacement catalog rather then be used for comparison processes,
as shown here.

Chapter 3. DFSMSdfp V1R8 enhancements
79
In Figure 3-64 we show the job to run the catalog diagnosis and make the first export of
catalog UCAT.FLETCHER.
Figure 3-64 ICFRU readiness check IDCAMS diagnose and first EXPORT job
//MHLRES1I JOB (999,POK),'MHLRES2',CLASS=A,MSGCLASS=T,
// NOTIFY=&SYSUID,TIME=1440,REGION=6M
/*JOBPARM L=999,SYSAFF=*
//*****************************************************************
//* DIAGNOSE THE CATALOG
//* LIST ITS ALIASES
//* LIST ITS SELF-DESCRIBING ENTRY COMPLETELY
//* LIST ITS ENTRIES WITH JUST THE VOLUME INFORMATION
//DIAGLIST EXEC PGM=IDCAMS
//SETBKDS DD DSN=MHLRES1.BCAT.CATALOG.BACKUP(+1),DISP=(NEW,PASS),
// DCB=MHLRES1.BCAT.EXPORT.DCB,
// UNIT=SYSALLDA,SPACE=(TRK,(15,15),RLSE)
//SYSPRINT DD DSN=MHLRES1.BCAT.CATALOG.LISTING(+1),DISP=(MOD,PASS),
// DCB=MHLRES1.BCAT.LIST.DCB,
// UNIT=SYSALLDA,SPACE=(TRK,(15,15),RLSE)
//SYSIN DD *
DIAG ICFCAT INDATASET(UCAT.FLETCHER)
LISTCAT ENTRY(UCAT.FLETCHER) ALL
LISTCAT ENTRY(UCAT.FLETCHER) ALL CAT(UCAT.FLETCHER)
LISTCAT VOLUME CAT(UCAT.FLETCHER)
/*
//*****************************************************************
//* EXPORT THE CATALOG IF THE DIAGNOSTICS WERE OKAY
//EXPCAT EXEC PGM=IDCAMS,COND=(8,LE)
//CATBACK DD DSN=MHLRES1.BCAT.CATALOG.BACKUP(+1),DISP=(OLD,PASS),
// DCB=MHLRES1.BCAT.EXPORT.DCB,
// UNIT=SYSALLDA,SPACE=(TRK,(15,15),RLSE)
//SYSPRINT DD DSN=MHLRES1.BCAT.CATALOG.LISTING(+1),DISP=(MOD,PASS),
// DCB=MHLRES1.BCAT.LIST.DCB,
// UNIT=SYSALLDA,SPACE=(TRK,(15,15),RLSE)
//SYSIN DD *
EXPORT UCAT.FLETCHER OFILE(CATBACK) TEMP
/*
//*****************************************************************
//* CATALOG THE BACKUP AND LISTING IF EXPORT WAS OK
//CATAL EXEC PGM=IEFBR14,COND=(8,LE)
//CATBACK DD DSN=MHLRES1.BCAT.CATALOG.BACKUP(+1),DISP=(OLD,CATLG)
//SYSLIST DD DSN=MHLRES1.BCAT.CATALOG.LISTING(+1),DISP=(OLD,CATLG)
//*****************************************************************
//* COPY THE LISTING AND DISCARD THE BACKUP IF EXPORT FAILED
//REPOUT EXEC PGM=IDCAMS,COND=(0,EQ,CATAL)
//SYSPRINT DD SYSOUT=*
//CATBACK DD DSN=MHLRES1.BCAT.CATALOG.BACKUP(+1),DISP=(OLD,DELETE)
//CATLIST DD DSN=MHLRES1.BCAT.CATALOG.LISTING(+1),DISP=(OLD,DELETE)
//CATPROB DD DSN=MHLRES1.BCAT.PROBLEM.LISTING(+1),DISP=(NEW,CATLG),
// DCB=MHLRES1.BCAT.LIST.DCB,
// UNIT=SYSALLDA,SPACE=(TRK,(15,15),RLSE)
//SYSIN DD *
REPRO INFILE(CATLIST) OUTFILE(CATPROB)
/*

80
z/OS V1R8 DFSMS Technical Update
In Figure 3-65 and Figure 3-66 on page 81 we show the JES2 output from running the job
listed in Figure 3-64 on page 79.
Figure 3-65 ICFRU Readiness check IDCAMS Diagnose and first EXPORT job JES2 output (1 of 2)
J E S 2 J O B L O G -- S Y S T E M S C 6 4 -- N O D E W T S C P L X 2

12.39.14 JOB21908 ---- FRIDAY, 16 MAR 2007 ----
12.39.14 JOB21908 IRR010I USERID MHLRES1 IS ASSIGNED TO THIS JOB.
12.39.14 JOB21908 ICH70001I MHLRES1 LAST ACCESS AT 12:33:00 ON FRIDAY, MARCH 16, 2007
12.39.14 JOB21908 $HASP373 MHLRES1I STARTED - INIT 1 - CLASS A - SYS SC64
12.39.14 JOB21908 IEF403I MHLRES1I - STARTED - TIME=12.39.14 - ASID=002A - SC64
12.39.15 JOB21908 - --TIMINGS (MINS.)-- ----PAGING COUNTS---
12.39.15 JOB21908 -JOBNAME STEPNAME PROCSTEP RC EXCP CPU SRB CLOCK SERV PG PAGE SWAP VIO SWAPS STEPNO
12.39.15 JOB21908 -MHLRES1I DIAGLIST 00 99 .00 .00 .00 554 0 0 0 0 0 1
12.39.15 JOB21908 -MHLRES1I EXPCAT 00 75 .00 .00 .00 920 0 0 0 0 0 2
12.39.15 JOB21908 -MHLRES1I CATAL 00 10 .00 .00 .00 25 0 0 0 0 0 3
12.39.15 JOB21908 -MHLRES1I REPOUT FLUSH 0 .00 .00 .00 0 0 0 0 0 0 4
12.39.15 JOB21908 IEF404I MHLRES1I - ENDED - TIME=12.39.15 - ASID=002A - SC64
12.39.15 JOB21908 -MHLRES1I ENDED. NAME-MHLRES2 TOTAL CPU TIME= .00 TOTAL ELAPSED TIME= .00
12.39.15 JOB21908 $HASP395 MHLRES1I ENDED
------ JES2 JOB STATISTICS ------
16 MAR 2007 JOB EXECUTION DATE
50 CARDS READ
124 SYSOUT PRINT RECORDS
0 SYSOUT PUNCH RECORDS
8 SYSOUT SPOOL KBYTES
0.00 MINUTES EXECUTION TIME
1 //MHLRES1I JOB (999,POK),'MHLRES2',CLASS=A,MSGCLASS=T, JOB21908
// NOTIFY=&SYSUID,TIME=1440,REGION=6M 00020000
/*JOBPARM L=999,SYSAFF=* 00030000
//*****************************************************************
//* DIAGNOSE THE CATALOG
//* LIST ITS ALIASES
//* LIST ITS SELF-DESCRIBING ENTRY COMPLETELY
//* LIST ITS ENTRIES WITH JUST THE VOLUME INFORMATION
IEFC653I SUBSTITUTION JCL - (999,POK),'MHLRES2',CLASS=A,MSGCLASS=T,NOTIFY=MHLRES1,TIME=1440,REGION=6M

2 //DIAGLIST EXEC PGM=IDCAMS
3 //SETBKDS DD DSN=MHLRES1.BCAT.CATALOG.BACKUP(+1),DISP=(NEW,PASS),
// DCB=MHLRES1.BCAT.EXPORT.DCB,
// UNIT=SYSALLDA,SPACE=(TRK,(15,15),RLSE)
4 //SYSPRINT DD DSN=MHLRES1.BCAT.CATALOG.LISTING(+1),DISP=(MOD,PASS),
// DCB=MHLRES1.BCAT.LIST.DCB,
// UNIT=SYSALLDA,SPACE=(TRK,(15,15),RLSE)
5 //SYSIN DD *
//*****************************************************************
//* EXPORT THE CATALOG IF THE DIAGNOSTICS WERE OKAY
6 //EXPCAT EXEC PGM=IDCAMS,COND=(8,LE)
7 //CATBACK DD DSN=MHLRES1.BCAT.CATALOG.BACKUP(+1),DISP=(OLD,PASS),
// DCB=MHLRES1.BCAT.EXPORT.DCB,
// UNIT=SYSALLDA,SPACE=(TRK,(15,15),RLSE)
8 //SYSPRINT DD DSN=MHLRES1.BCAT.CATALOG.LISTING(+1),DISP=(MOD,PASS),
// DCB=MHLRES1.BCAT.LIST.DCB,
// UNIT=SYSALLDA,SPACE=(TRK,(15,15),RLSE)
9 //SYSIN DD *


//*****************************************************************
//* CATALOG THE BACKUP AND LISTING IF EXPORT WAS OK
10 //CATAL EXEC PGM=IEFBR14,COND=(8,LE)
11 //CATBACK DD DSN=MHLRES1.BCAT.CATALOG.BACKUP(+1),DISP=(OLD,CATLG)
12 //SYSLIST DD DSN=MHLRES1.BCAT.CATALOG.LISTING(+1),DISP=(OLD,CATLG)
//*****************************************************************
//* COPY THE LISTING AND DISCARD THE BACKUP IF EXPORT FAILED
13 //REPOUT EXEC PGM=IDCAMS,COND=(0,EQ,CATAL)
14 //SYSPRINT DD SYSOUT=*
15 //CATBACK DD DSN=MHLRES1.BCAT.CATALOG.BACKUP(+1),DISP=(OLD,DELETE)
16 //CATLIST DD DSN=MHLRES1.BCAT.CATALOG.LISTING(+1),DISP=(OLD,DELETE)
17 //CATPROB DD DSN=MHLRES1.BCAT.PROBLEM.LISTING(+1),DISP=(NEW,CATLG),
// DCB=MHLRES1.BCAT.LIST.DCB,
// UNIT=SYSALLDA,SPACE=(TRK,(15,15),RLSE)
18 //SYSIN DD *

Chapter 3. DFSMSdfp V1R8 enhancements
81
Figure 3-66 ICFRU Readiness check IDCAMS Diagnose and first EXPORT job JES2 output (2 of 2)
This was not the first run of this job since the initial data sets and GDGs were set up, so the
relative GDG number is G0005V00 rather than G0001V00 as might be expected.
The significant data sets from this run are MHLRES1.BCAT.CATALOG.BACKUP.G0005V00
and MHLRES1.BCAT.CATALOG.LISTING.G0005V00.
The output from this job, as captured in the MHLRES1.BCAT.CATALOG.LISTING.* output,
should be checked for errors, and if necessary any errors found should be corrected and the
job run again.
STMT NO. MESSAGE
-
ICH70001I MHLRES1 LAST ACCESS AT 12:33:00 ON FRIDAY, MARCH 16, 2007
IEF236I ALLOC. FOR MHLRES1I DIAGLIST
IGD101I SMS ALLOCATED TO DDNAME (SETBKDS )
DSN (MHLRES1.BCAT.CATALOG.BACKUP.G0005V00 )
STORCLAS (STANDARD) MGMTCLAS (MCDB22) DATACLAS ( )
VOL SER NOS= MLD00C
IGD101I SMS ALLOCATED TO DDNAME (SYSPRINT)
DSN (MHLRES1.BCAT.CATALOG.LISTING.G0005V00 )
STORCLAS (STANDARD) MGMTCLAS (MCDB22) DATACLAS ( )
VOL SER NOS= MLD10C
IEF237I JES2 ALLOCATED TO SYSIN
IEF237I 8019 ALLOCATED TO SYS00001
IGD104I UCAT.FLETCHER RETAINED, DDNAME=SYS00001
IEF142I MHLRES1I DIAGLIST - STEP WAS EXECUTED - COND CODE 0000
IGD106I MHLRES1.BCAT.CATALOG.BACKUP.G0005V00 PASSED, DDNAME=SETBKDS
IGD106I MHLRES1.BCAT.CATALOG.LISTING.G0005V00 PASSED, DDNAME=SYSPRINT
IEF285I MHLRES1.MHLRES1I.JOB21908.D0000101.? SYSIN
IEF373I STEP/DIAGLIST/START 2007075.1239
IEF374I STEP/DIAGLIST/STOP 2007075.1239 CPU 0MIN 00.02SEC SRB 0MIN 00.00SEC VIRT 1236K SYS 360K EXT 144K SYS
IEF236I ALLOC. FOR MHLRES1I EXPCAT
IGD103I SMS ALLOCATED TO DDNAME CATBACK
IGD103I SMS ALLOCATED TO DDNAME SYSPRINT
IEF237I JES2 ALLOCATED TO SYSIN
IEF237I 8019 ALLOCATED TO SYS00002
IGD104I UCAT.FLETCHER RETAINED, DDNAME=SYS00002
IEF142I MHLRES1I EXPCAT - STEP WAS EXECUTED - COND CODE 0000
IGD106I MHLRES1.BCAT.CATALOG.BACKUP.G0005V00 PASSED, DDNAME=CATBACK

IGD106I MHLRES1.BCAT.CATALOG.LISTING.G0005V00 PASSED, DDNAME=SYSPRINT
IEF285I MHLRES1.MHLRES1I.JOB21908.D0000102.? SYSIN
IEF373I STEP/EXPCAT /START 2007075.1239
IEF374I STEP/EXPCAT /STOP 2007075.1239 CPU 0MIN 00.03SEC SRB 0MIN 00.00SEC VIRT 444K SYS 360K EXT 1752K SYS
IEF236I ALLOC. FOR MHLRES1I CATAL
IGD103I SMS ALLOCATED TO DDNAME CATBACK
IGD103I SMS ALLOCATED TO DDNAME SYSLIST
IEF142I MHLRES1I CATAL - STEP WAS EXECUTED - COND CODE 0000
IGD107I MHLRES1.BCAT.CATALOG.BACKUP.G0005V00 ROLLED IN, DDNAME=CATBACK
IGD107I MHLRES1.BCAT.CATALOG.LISTING.G0005V00 ROLLED IN, DDNAME=SYSLIST
IEF373I STEP/CATAL /START 2007075.1239
IEF374I STEP/CATAL /STOP 2007075.1239 CPU 0MIN 00.00SEC SRB 0MIN 00.00SEC VIRT 4K SYS 316K EXT 0K SYS
IEF202I MHLRES1I REPOUT - STEP WAS NOT RUN BECAUSE OF CONDITION CODES
IEF272I MHLRES1I REPOUT - STEP WAS NOT EXECUTED.
IEF373I STEP/REPOUT /START 2007075.1239
IEF374I STEP/REPOUT /STOP 2007075.1239 CPU 0MIN 00.00SEC SRB 0MIN 00.00SEC VIRT 0K SYS 0K EXT 0K SYS
IEF375I JOB/MHLRES1I/START 2007075.1239
IEF376I JOB/MHLRES1I/STOP 2007075.1239 CPU 0MIN 00.05SEC SRB 0MIN 00.00SEC

82
z/OS V1R8 DFSMS Technical Update
Once the run is clean, as in this case, the significant part of the output is the last part of
Figure 3-67.
Figure 3-67 ICFRU Readiness check IDCAMS Diagnose and first EXPORT job captured output - last part
For the purposes of simulating a catalog recovery, we noted the time and date that the
EXPORT was created, as listed in the message:
IDC0594I PORTABLE DATA SET CREATED SUCCESSFULLY ON 03/16/07 AT 12:39:15
In order to guarantee that there has been some catalog activity since the initial EXPORT was
created, we created a job to define sixteen data sets.
EXPORT UCAT.FLETCHER OFILE(CATBACK) TEMP
IDC0005I NUMBER OF RECORDS PROCESSED WAS 6
IDC0594I PORTABLE DATA SET CREATED SUCCESSFULLY ON 03/16/07 AT 12:39:15
IDC1147I IT IS RECOMMENDED THAT DIAGNOSE AND EXAMINE BE RUN BEFORE
IDC1147I IMPORT OF CATALOG
IDC0001I FUNCTION COMPLETED, HIGHEST CONDITION CODE WAS 0

IDC0002I IDCAMS PROCESSING COMPLETE. MAXIMUM CONDITION CODE WAS 0

Chapter 3. DFSMSdfp V1R8 enhancements
83
In Figure 3-68 we show the job to allocate these data sets.
Figure 3-68 ICFRU Readiness check - allocate test data sets
We then ran the same job as used to create the first EXPORT — the job shown in Figure 3-64
on page 79. The output was basically the same as when run the first time except that the data
sets generated were suffixed G0006V00 instead of G0005V00.
//MHLRES1I JOB (999,POK),'MHLRES2',CLASS=A,MSGCLASS=T,
// NOTIFY=&SYSUID,TIME=1440,REGION=6M
/*JOBPARM L=999,SYSAFF=*
//*****************************************************************
//* CATALOG A NUMBER OF DATASETS TO ENSURE THERE HAS BEEN SOME
//* CATALOG ACTIVITY
//* N.B. RUN THE CORRESPONDING DELETE JOB AFTER THE SECOND
//* EXPORT HAS BEEN CREATED
//DEFDS EXEC PGM=IDCAMS
//SYSPRINT DD SYSOUT=*
//DD01 DD DISP=(,CATLG),SPACE=(TRK,(1,1)),
// DSN=MHLCAT9.BCAT.TEST01,UNIT=SYSALLDA
//DD02 DD DISP=(,CATLG),SPACE=(TRK,(1,1)),
// DSN=MHLCAT9.BCAT.TEST02,UNIT=SYSALLDA
//DD03 DD DISP=(,CATLG),SPACE=(TRK,(1,1)),
// DSN=MHLCAT9.BCAT.TEST03,UNIT=SYSALLDA
//DD04 DD DISP=(,CATLG),SPACE=(TRK,(1,1)),
// DSN=MHLCAT9.BCAT.TEST04,UNIT=SYSALLDA
//DD05 DD DISP=(,CATLG),SPACE=(TRK,(1,1)),
// DSN=MHLCAT9.BCAT.TEST05,UNIT=SYSALLDA
//DD06 DD DISP=(,CATLG),SPACE=(TRK,(1,1)),
// DSN=MHLCAT9.BCAT.TEST06,UNIT=SYSALLDA
//DD07 DD DISP=(,CATLG),SPACE=(TRK,(1,1)),
// DSN=MHLCAT9.BCAT.TEST07,UNIT=SYSALLDA
//DD08 DD DISP=(,CATLG),SPACE=(TRK,(1,1)),
// DSN=MHLCAT9.BCAT.TEST08,UNIT=SYSALLDA
//DD09 DD DISP=(,CATLG),SPACE=(TRK,(1,1)),
// DSN=MHLCAT9.BCAT.TEST09,UNIT=SYSALLDA
//DD10 DD DISP=(,CATLG),SPACE=(TRK,(1,1)),
// DSN=MHLCAT9.BCAT.TEST10,UNIT=SYSALLDA
//DD11 DD DISP=(,CATLG),SPACE=(TRK,(1,1)),
// DSN=MHLCAT9.BCAT.TEST11,UNIT=SYSALLDA
//DD12 DD DISP=(,CATLG),SPACE=(TRK,(1,1)),
// DSN=MHLCAT9.BCAT.TEST12,UNIT=SYSALLDA
//DD13 DD DISP=(,CATLG),SPACE=(TRK,(1,1)),
// DSN=MHLCAT9.BCAT.TEST13,UNIT=SYSALLDA
//DD14 DD DISP=(,CATLG),SPACE=(TRK,(1,1)),
// DSN=MHLCAT9.BCAT.TEST14,UNIT=SYSALLDA
//DD15 DD DISP=(,CATLG),SPACE=(TRK,(1,1)),
// DSN=MHLCAT9.BCAT.TEST15,UNIT=SYSALLDA
//DD16 DD DISP=(,CATLG),SPACE=(TRK,(1,1)),
// DSN=MHLCAT9.BCAT.TEST16,UNIT=SYSALLDA
//SYSIN DD *
LISTCAT LVL(MHLCAT9.BCAT)
/*

84
z/OS V1R8 DFSMS Technical Update
In Figure 3-69 we show the last part of the MHLRES1.BCAT.CATALOG.LISTING.G0006V00
data set, which shows the result of the EXPORT command.
Figure 3-69 ICFRU Readiness job last part of MHLRES1.BCAT.CATALOG.LISTING.G0002V00

For the purposes of simulating a catalog recovery, we noted the time and date that the
second EXPORT was created, as listed in the message:
IDC0594I PORTABLE DATA SET CREATED SUCCESSFULLY ON 03/16/07 AT 12:51:25
We then identified the SMF data sets that contained the SMF data from just before the first
export was created until just after the second export was created from 03/16/07 AT 12:39:15
to 03/16/07 AT 12:51:25.
The data sets were identified as SMFDATA.ALLRECS.G3350V00 through to
SMFDATA.ALLRECS.G3351V00.
We then used this information in the job to run program ICFRRSV.
In Figure 3-71 on page 85 we show the JCL to run the CRURRSV program. It has been set
up to use the SMF records that may contain updates to the catalog, and set up with the start
and stop date and times as taken from the IDCAMS EXPORT job and the SMF records.
The significant PARM values of catalog name, start_time, stop_time, and gap time are
highlighted in Figure 3-70.
Figure 3-70 CRURRSV program PARM values
EXPORT UCAT.FLETCHER OFILE(CATBACK) TEMP
IDC0005I NUMBER OF RECORDS PROCESSED WAS 22
IDC0594I PORTABLE DATA SET CREATED SUCCESSFULLY ON 03/16/07 AT 12:51:25
IDC1147I IT IS RECOMMENDED THAT DIAGNOSE AND EXAMINE BE RUN BEFORE
IDC1147I IMPORT OF CATALOG
IDC0001I FUNCTION COMPLETED, HIGHEST CONDITION CODE WAS 0

IDC0002I IDCAMS PROCESSING COMPLETE. MAXIMUM CONDITION CODE WAS 0
// PARM=('UCAT.FLETCHER',
// '03/16/07','12:39:15',
// '03/16/07','12:51:25',
// '0030',

Chapter 3. DFSMSdfp V1R8 enhancements
85
The appropriate SMF data set names and the output data set have been specified.
Figure 3-71 ICFRU Readiness check - CRURRSV job
CRURRSV produces several reports based on the analysis of the SMF data. For explanation
of the information you should refer to Appendix A of DFSMS Managing Catalogs, SC26-7409.
//MHLRES1I JOB (999,POK),'MHLRES2',CLASS=A,MSGCLASS=T,
// NOTIFY=&SYSUID,TIME=1440,REGION=6M
/*JOBPARM L=999,SYSAFF=*
//*****************************************************************
//* THIS JCL EXECUTES CRURRSV TO SELECT THE APPROPRIATE SMF
//* RECORDS FROM THE DATASETS COVERING THE PERIOD WHERE RECORDS
//* MAY BE LOST FROM THE CATALOG.
//* THE INPUT IS THE SMF DATA SET(S), OUTPUT IS A CONSOLIDATED
//* SET OF SMF RECORDS. SELECTION IS CONTROLED THROUGH THE PARM
//*
//RRSV EXEC PGM=CRURRSV,
// PARM=('UCAT.FLETCHER',
// '03/16/07','12:39:15',
// '03/16/07','12:51:25',
// '0030',
// '0000')
//*PARM=('CATALOG.NAME',
//* 'STARTDATE','STARTTIME', MM/DD/YY HH:MM:SS
//* 'STOPDATE,'STOPTIME',
//* 'GAPTIME', (MINUTES)
//* 'CLOCKDIFFERENCE') (SECONDS)
//SYSDUMP DD SYSOUT=*
//SYSPRINT DD SYSOUT=*
//SYSLOG DD SYSOUT=*
//SMFIN DD DISP=SHR,DCB=BUFNO=60,DSN=SMFDATA.ALLRECS.G3350V00
// DD DISP=SHR,DCB=BUFNO=60,DSN=SMFDATA.ALLRECS.G3351V00
//SMFOUT DD DISP=(,CATLG),DCB=BUFNO=60,
// UNIT=SYSDA,SPACE=(CYL,(10,2)),
// DSN=MHLRES1.BCAT.SMF.CAT.RECS.FLETCHER

86
z/OS V1R8 DFSMS Technical Update
In Figure 3-72 we show the output from running the job as shown in Figure 3-71 on page 85.
Figure 3-72 ICFRU Readiness check - CRURRSV job output
INTEGRATED CATALOG FORWARD RECOVERY UTILITY V2R1
CRURRSV SYSPRINT 03/16/07 (07.075) 17:42:17 PAGE 01
RECORD SELECTION AND VALIDATION REPORT
EXECUTION PARAMETERS
CATALOG NAME UCAT.FLETCHER
RECORD SELECTION START 03/16/07 (07.075) 12:39:15
RECORD SELECTION STOP 03/16/07 (07.075) 12:51:25
SIGNIFICANT GAP TIME 0030 MINUTES
MAXIMUM CLOCK DIFFERENCE 0000 SECONDS
REPORT FOR ALL SYSTEMS
RECORD SELECTION AND VALIDATION CONDITION CODE IS 00
0 ANOMALIES (LOST DATA, GAPS) DETECTED
16 RECORDS SELECTED FOR UCAT.FLETCHER
16 DEFINE (TYPE 61) RECORDS SELECTED
0 DELETE (TYPE 65) RECORDS SELECTED
0 ALTER (TYPE 66) RECORDS SELECTED
01 SYSTEM(S) RECORDED CHANGES TO THIS CATALOG
SC64
FOR CATALOG UCAT.FLETCHER
03/16/07 (07.075) 11:29:33.57 OLDEST SMF CATALOG RECORD FOUND
03/16/07 (07.075) 12:44:24.44 OLDEST SMF CATALOG RECORD SELECTED
03/16/07 (07.075) 12:44:24.59 NEWEST SMF CATALOG RECORD SELECTED
03/16/07 (07.075) 12:44:24.59 NEWEST SMF CATALOG RECORD FOUND
FOR ALL SMF RECORD TYPES
03/16/07 (07.075) 06:40:01.08 OLDEST SMF RECORD FOUND (ANY TYPE)
03/16/07 (07.075) 16:30:08.64 NEWEST SMF RECORD FOUND (ANY TYPE)
1 SYSTEM IDENTIFIERS WERE FOUND
26,478 TOTAL SMF RECORDS WERE READ
2 SMF SWITCH (TYPE 90, SUBTYPE 6) RECORDS WERE FOUND
0 SMF EOD (TYPE 90, SUBTYPE 7) RECORDS WERE FOUND
0 SMF IPL (TYPE 0) RECORDS WERE FOUND
0 SMF LOST DATA (TYPE 7) RECORDS WERE FOUND

0 FORWARD GAPS IN SINGLE-SYSTEM SMF RECORDS LONGER THAN
0030 MINUTES WERE FOUND
0 BACKWARD GAPS IN SINGLE-SYSTEM SMF RECORDS LONGER THAN
0030 MINUTES WERE FOUND
RECORD SELECTION AND VALIDATION CONDITION CODE IS 00
INTEGRATED CATALOG FORWARD RECOVERY UTILITY V2R1
CRURRSV SYSPRINT 03/16/07 (07.075) 17:42:17 PAGE 02
RECORD SELECTION AND VALIDATION REPORT
EXECUTION PARAMETERS
CATALOG NAME UCAT.FLETCHER
RECORD SELECTION START 03/16/07 (07.075) 12:39:15
RECORD SELECTION STOP 03/16/07 (07.075) 12:51:25
SIGNIFICANT GAP TIME 0030 MINUTES
MAXIMUM CLOCK DIFFERENCE 0000 SECONDS
REPORT FOR SYSTEM SC64
RECORD SELECTION AND VALIDATION CONDITION CODE IS 00
0 ANOMALIES (LOST DATA, GAPS) DETECTED
16 RECORDS SELECTED FOR UCAT.FLETCHER
16 DEFINE (TYPE 61) RECORDS SELECTED
0 DELETE (TYPE 65) RECORDS SELECTED
0 ALTER (TYPE 66) RECORDS SELECTED
FOR CATALOG UCAT.FLETCHER
03/16/07 (07.075) 11:29:33.57 OLDEST SMF CATALOG RECORD FOUND
03/16/07 (07.075) 12:44:24.44 OLDEST SMF CATALOG RECORD SELECTED
03/16/07 (07.075) 12:44:24.59 NEWEST SMF CATALOG RECORD SELECTED
03/16/07 (07.075) 12:44:24.59 NEWEST SMF CATALOG RECORD FOUND
FOR ALL SMF RECORD TYPES
03/16/07 (07.075) 06:40:01.08 OLDEST SMF RECORD FOUND (ANY TYPE)
03/16/07 (07.075) 16:30:01.08 NEWEST SMF RECORD FOUND (ANY TYPE)
26,478 TOTAL SMF RECORDS WERE READ
2 SMF SWITCH (TYPE 90, SUBTYPE 6) RECORDS WERE FOUND
0 SMF EOD (TYPE 90, SUBTYPE 7) RECORDS WERE FOUND
0 SMF IPL (TYPE 0) RECORDS WERE FOUND
0 SMF LOST DATA (TYPE 7) RECORDS WERE FOUND
0 FORWARD GAPS IN SINGLE-SYSTEM SMF RECORDS LONGER THAN
0030 MINUTES WERE FOUND
0 BACKWARD GAPS IN SINGLE-SYSTEM SMF RECORDS LONGER THAN
0030 MINUTES WERE FOUND
RECORD SELECTION AND VALIDATION CONDITION CODE IS 00

Chapter 3. DFSMSdfp V1R8 enhancements
87
In addition to the report output, a separate SYSLOG file is produced. This is useful for
determining the number of SMF switches that occurred and whether there might be gaps in
the SMF data.
In Figure 3-73 we show the output from the CRURRSV SYSLOG.
Figure 3-73 CFRU Readiness check - CRRRSV SYSLOG output
Once the SMF records have been selected, they must be sorted into the form required by
CRURRAP.
In Figure 3-74 we show the JCL to sort the SMF records. The input and output data sets are
bolded.
Figure 3-74 ICFRU Readiness check - SMF sort job
The output from the SORT should be checked to verify that the correct number of records as
reported by CRURRSV have been sorted.
INTEGRATED CATALOG FORWARD RECOVERY UTILITY V2R1
CRURRSV SYSLOG 03/16/07 (07.075) 17:42:17
CRU023I SWITCH SMF RECORD FOUND FOR SYSID SC64
03/16/07 (07.075) 06:40:01.17 RECORD BEING PROCESSED - DUMP FOLLOWS

+0000 00AEC8 00A40000 1E5A0024 9F750107 075FE2C3 F6F40000 00000024 000C0001 00000030 *.u...!.......¬SC64..............*
+0020 00AEE8 00740001 0006F0F1 E2D4C640 40404040 40404040 40404040 40404040 40404040 *......01SMF *
+0040 00AF08 40404040 0064FD95 0107056F E2E8E2F1 4BE2C3F6 F44BD4C1 D5F24040 40404040 * ...n...?SYS1.SC64.MAN2 *
+0060 00AF28 40404040 40404040 40404040 40404040 40404040 40404040 E2E8E2F1 4BE2C3F6 * SYS1.SC6*
+0080 00AF48 F44BD4C1 D5F14040 40404040 40404040 40404040 40404040 40404040 40404040 *4.MAN1 *
+00A0 00AF68 40404040 * *
CRU023I SWITCH SMF RECORD FOUND FOR SYSID SC64
03/16/07 (07.075) 11:40:00.57 RECORD BEING PROCESSED - DUMP FOLLOWS

+0000 00AEC8 00A40000 1E5A0040 16790107 075FE2C3 F6F40000 00000024 000C0001 00000030 *.u...!. .....¬SC64..............*
+0020 00AEE8 00740001 0006F0F1 E2D4C640 40404040 40404040 40404040 40404040 40404040 *......01SMF *
+0040 00AF08 40404040 0064FD95 0107056F E2E8E2F1 4BE2C3F6 F44BD4C1 D5F14040 40404040 * ...n...?SYS1.SC64.MAN1 *
+0060 00AF28 40404040 40404040 40404040 40404040 40404040 40404040 E2E8E2F1 4BE2C3F6 * SYS1.SC6*
+0080 00AF48 F44BD4C1 D5F24040 40404040 40404040 40404040 40404040 40404040 40404040 *4.MAN2 *
+00A0 00AF68 40404040 * *
//MHLRES1I JOB (999,POK),'MHLRES2',CLASS=A,MSGCLASS=T,
// NOTIFY=&SYSUID,TIME=1440,REGION=6M
/*JOBPARM L=999,SYSAFF=*
//*****************************************************************
//* THIS JCL EXECUTES SORT TO PREPARE FOR INPUT TO CRURRAP
//*
//RRSV EXEC PGM=ICEMAN
//SYSOUT DD SYSOUT=*
//SORTIN DD DISP=SHR,
// DSN=MHLRES1.BCAT.SMF.CAT.RECS.FLETCHER
//SORTOUT DD DISP=(,CATLG,DELETE),UNIT=SYSDA,
// SPACE=(CYL,(10,2),RLSE),
// DSN=MHLRES1.BCAT.SORTED.SMF.CAT.RECS.FLETCHER
//SYSIN DD *
OPTION DYNALLOC=SYSDA,FILSZ=E10000
SORT FIELDS=(218,44,CH,A,262,1,BI,A,11,4,PD,D,7,4,BI,D)
/*

88
z/OS V1R8 DFSMS Technical Update
In Figure 3-75 we show the message output resulting from running the JCL as shown in
Figure 3-74 on page 87.
Figure 3-75 ICFRU Readiness check - SORT job messages
The sorted SMF records must be processed by CRURRAP to produce a data set that is to be
used as a file for IDCAMS IMPORT. For the purposes of this ICFRU readiness check, it is not
used as import by IDCAMS, but it will be compared with the second IDCAMS EXPORT made
during this check.
BLOCKSET SORT TECHNIQUE SELECTED
VISIT http://www.ibm.com/storage/dfsort FOR DFSORT PAPERS, EXAMPLES AND MORE
- CONTROL STATEMENTS FOR 5694-A01, Z/OS DFSORT V1R5 - 17:42 ON FRI MAR 16, 2007 -
OPTION DYNALLOC=SYSDA,FILSZ=E10000
SORT FIELDS=(218,44,CH,A,262,1,BI,A,11,4,PD,D,7,4,BI,D)
RECORD TYPE IS V - DATA STARTS IN POSITION 5
C5-K21008 C6-K90007 C7-K90000 C8-K90007 E4-K90007 C9-BASE E5-K18181 E6-K18181 B0-Q96745 E7-K11698
ICEAM1 ENVIRONMENT IN EFFECT - ICEAM1 INSTALLATION MODULE SELECTED
MHLRES1I.RRSV . , INPUT LRECL = 32760, BLKSIZE = 4096, TYPE = VBS
MAIN STORAGE = (MAX,6291456,6291456)
MAIN STORAGE ABOVE 16MB = (6214096,6214096)
OPTIONS: OVFLO=RC0 ,PAD=RC0 ,TRUNC=RC0 ,SPANINC=RC16,VLSCMP=N,SZERO=Y,RESET=Y,VSAMEMT=Y,DYNSPC=256
OPTIONS: SIZE=6291456,MAXLIM=1048576,MINLIM=450560,EQUALS=Y,LIST=Y,ERET=RC16 ,MSGDDN=SYSOUT
OPTIONS: VIO=N,RESDNT=ALL ,SMF=NO ,WRKSEC=Y,OUTSEC=Y,VERIFY=N,CHALT=N,DYNALOC=(SYSDA
,004),ABCODE=MSG
OPTIONS: RESALL=4096,RESINV=0,SVC=109 ,CHECK=Y,WRKREL=Y,OUTREL=Y,CKPT=N,STIMER=Y,COBEXIT=COB2
OPTIONS: TMAXLIM=6291456,ARESALL=0,ARESINV=0,OVERRGN=65536,CINV=Y,CFW=Y,DSA=0
OPTIONS: VLSHRT=N,ZDPRINT=Y,IEXIT=N,TEXIT=N,LISTX=N,EFS=NONE ,EXITCK=S,PARMDDN=DFSPARM ,FSZEST=N
OPTIONS: HIPRMAX=OPTIMAL,DSPSIZE=MAX ,ODMAXBF=0,SOLRF=Y,VLLONG=N,VSAMIO=N,MOSIZE=MAX
OPTIONS: NULLOUT=RC0
EXCP ACCESS METHOD USED FOR SORTOUT
EXCP ACCESS METHOD USED FOR SORTIN
DC 8184 TC 0 CS DSV KSZ 58 VSZ 58
FSZ=8184 BC IGN=10000 E AVG=16380 0 WSP=10 C DYN=0 0
B1-K19866 B2-K17569 EC-K10929 B4-K17569 E8-K21008
OUTPUT LRECL = 32760, BLKSIZE = 27998, TYPE = VBS (SDB)
IN MAIN STORAGE SORT
INSERT 0, DELETE 0
RECORDS - IN: 16, OUT: 16
NUMBER OF BYTES SORTED: 4864
AVERAGE RECORD LENGTH = 304 BYTES
TOTAL WORK DATA SET TRACKS ALLOCATED: 0 , TRACKS USED: 0
MEMORY OBJECT STORAGE USED = 1M BYTES
HIPERSPACE STORAGE USED = 0K BYTES
DATA SPACE STORAGE USED = 0K BYTES
END OF DFSORT

Chapter 3. DFSMSdfp V1R8 enhancements
89
In Figure 3-76 we show the job set up to construct a pseudo IDCAMS EXPORT data set that
could be used to re-create a lost catalog.
Figure 3-76 ICFRU Readiness check - CRURRAP job
The PARM values used for CRURRAP must be the same as those used for CRURRSV.
For the purposes of this readiness check we compare it with the second IDCAMS EXPORT
data set. The two should be equal.
//MHLRES1I JOB (999,POK),'MHLRES2',CLASS=A,MSGCLASS=T,
// NOTIFY=&SYSUID,TIME=1440,REGION=6M
/*JOBPARM L=999,SYSAFF=*
//*****************************************************************
//* THIS JCL EXECUTES CRURRAP TO GENERATE THE EQUIVALENT OF
//* AND IDCAMS EXPORT.
//* THE INPUT IS THE SMF DATA SET(S) AS PRODUCED BY CRURRSV AND
//* SORTED BY SMF.
//* EXACTLY THE SAME SELECTION PARAMETERS MUST BE SET AS SPECIFIED
//* TO CRURRSV.
//*
//RRSV EXEC PGM=CRURRAP,
// PARM=('UCAT.FLETCHER',
// '03/16/07','12:39:15',
// '03/16/07','12:51:25',
// '0030',
// '0000')
//*PARM=('CATALOG.NAME',
//* 'STARTDATE','STARTTIME', MM/DD/YY HH:MM:SS
//* 'STOPDATE,'STOPTIME',
//* 'GAPTIME', (MINUTES)
//* 'CLOCKDIFFERENCE') (SECONDS)
//SYSDUMP DD SYSOUT=*
//SYSPRINT DD SYSOUT=*
//SYSLOG DD SYSOUT=*
//SMFIN DD DISP=SHR,DCB=BUFNO=60,
// DSN=MHLRES1.BCAT.SORTED.SMF.CAT.RECS.FLETCHER
//EXPIN DD DISP=SHR,DCB=BUFNO=60,
// DSN=MHLRES1.BCAT.CATALOG.BACKUP.G0006V00
//EXPOUT DD DISP=(,CATLG),DCB=BUFNO=60,
// UNIT=SYSDA,SPACE=(CYL,(10,2)),
// DSN=MHLRES1.BCAT.NEW.CATALOG.EXPORT.FLETCHER

90
z/OS V1R8 DFSMS Technical Update
In Figure 3-77 through to Figure 3-79 on page 91 we show the output from the CRURRAP job
as defined in Figure 3-76 on page 89.
Figure 3-77 ICFRU Readiness check - CRURRAP job output (part 1 of 3)
INTEGRATED CATALOG FORWARD RECOVERY UTILITY V2R1
CRURRAP SYSPRINT 03/16/07 (07.075) 20:10:32 PAGE 1
RECORD ANALYSIS AND PROCESSING REPORT
EXECUTION PARAMETERS
CATALOG NAME UCAT.FLETCHER
RECORD SELECTION START 03/16/07 (07.075) 12:39:15
RECORD SELECTION STOP 03/16/07 (07.075) 12:51:25
SIGNIFICANT GAP TIME 0030 MINUTES
MAXIMUM CLOCK DIFFERENCE 0000 SECONDS
RECORD ANALYSIS AND PROCESSING CONDITION CODE IS 08
ERROR REPORT
16 TOTAL ERRORS (CONDITION CODES 12 AND 8)
0 RECORDS REJECTED FROM EXPIN (LOGGED, DUMPED, CC=12)
0 RECORDS WITH INVALID LENGTHS (CRU302I)
0 RECORDS WITH INVALID CATALOG RECORD TYPES (CRU303I)
16 ERRORS IN EVENT SEQUENCE INVOLVING THE MOST CURRENT RECORD
(LOGGED, DUMPED, CC=8)
16 SEQUENCE ERRORS, BUT NO SYNCHRONIZATION CHECK
0 SMF UPDATE FOR A NON-EXISTENT RECORD (CRU203I)
0 SMF DELETE FOR A NON-EXISTENT RECORD (CRU204I)
0 SMF INSERT PRECEDED BY AN SMF INSERT (CRU205I)
0 SMF INSERT PRECEDED BY AN SMF UPDATE (CRU206I)
0 SMF UPDATE PRECEDED BY AN SMF DELETE (CRU207I)
0 SMF DELETE PRECEDED BY AN SMF DELETE (CRU208I)
16 SMF INSERT PRECEDED BY EXPORT RECORD (CRU209I)
0 SEQUENCE ERRORS, WITH A SYNCHRONIZATION CHECK
0 SMF INSERT PRECEDED BY AN SMF INSERT (CRU205I)
0 SMF INSERT PRECEDED BY AN SMF UPDATE (CRU206I)
0 SMF UPDATE PRECEDED BY AN SMF DELETE (CRU207I)
0 SMF DELETE PRECEDED BY AN SMF DELETE (CRU208I)

INTEGRATED CATALOG FORWARD RECOVERY UTILITY V2R1
CRURRAP SYSPRINT 03/16/07 (07.075) 20:10:32 PAGE 2
ANOMALY REPORT
0 ANOMALIES (CONDITION CODES 4 AND 0)
0 SYNCHRONIZATION CHECKS INVOLVING THE MOST CURRENT RECORD
BUT WITH NO EVENT SEQUENCE ERROR (CRU113I)
(LOGGED, DUMPED, CC=4)
0 SMF UPDATE PRECEDED BY AN SMF INSERT (CRU003I)
0 SMF DELETE PRECEDED BY AN SMF INSERT (CRU004I)
0 SMF UPDATE PRECEDED BY AN SMF UPDATE (CRU005I)
0 SMF DELETE PRECEDED BY AN SMF UPDATE (CRU006I)
0 SMF INSERT PRECEDED BY AN SMF DELETE (CRU007I)

0 ERRORS IN EVENT SEQUENCE INVOLVING A SUPERSEDED RECORD
(LOGGED, CC=4)
0 SEQUENCE ERRORS, BUT NO SYNCHRONIZATION CHECK
0 SMF UPDATE FOR A NON-EXISTENT RECORD (CRU106I)
0 SMF DELETE FOR A NON-EXISTENT RECORD (CRU107I)
0 SMF INSERT PRECEDED BY AN SMF INSERT (CRU108I)
0 SMF INSERT PRECEDED BY AN SMF UPDATE (CRU109I)
0 SMF UPDATE PRECEDED BY AN SMF DELETE (CRU110I)
0 SMF DELETE PRECEDED BY AN SMF DELETE (CRU111I)
0 SMF INSERT PRECEDED BY EXPORT RECORD (CRU112I)
0 SEQUENCE ERRORS, WITH A SYNCHRONIZATION CHECK
0 SMF INSERT PRECEDED BY AN SMF INSERT (CRU108I)
0 SMF INSERT PRECEDED BY AN SMF UPDATE (CRU109I)
0 SMF UPDATE PRECEDED BY AN SMF DELETE (CRU110I)
0 SMF DELETE PRECEDED BY AN SMF DELETE (CRU111I)
0 SYNCHRONIZATION CHECKS INVOLVING A SUPERSEDED RECORD
BUT WITH NO EVENT SEQUENCE ERROR (CRU020I)
(NOT LOGGED, CC=0)
0 SMF UPDATE PRECEDED BY AN SMF INSERT (CRU013I)
0 SMF DELETE PRECEDED BY AN SMF INSERT (CRU014I)
0 SMF UPDATE PRECEDED BY AN SMF UPDATE (CRU015I)
0 SMF DELETE PRECEDED BY AN SMF UPDATE (CRU016I)
0 SMF INSERT PRECEDED BY AN SMF DELETE (CRU017I)
INTEGRATED CATALOG FORWARD RECOVERY UTILITY V2R1

Chapter 3. DFSMSdfp V1R8 enhancements
91
Figure 3-78 ICFRU Readiness check - CRURRAP job output (part 2 of 3)
Figure 3-79 ICFRU Readiness check - CRURRAP job output (part 3 of 3)
In addition to the standard job output, as with CRURRSV, CRURRAP also produces a
SYSLOG.
CRURRAP SYSPRINT 03/16/07 (07.075) 20:10:32 PAGE 3
REPORT OF RECORDS PROCESSED WITHOUT ERROR OR ANOMALY
22 TOTAL RECORDS PROCESSED (NO ERROR/NO ANOMALY, CONDITION CODE 0)
6 MOST CURRENT RECORDS PROCESSED WITHOUT ERROR OR ANOMALY
0 SMF INSERT FOR A NEW RECORD (CRU002I)
0 SMF UPDATE PRECEDED BY AN SMF INSERT (CRU003I)
0 SMF DELETE PRECEDED BY AN SMF INSERT (CRU004I)
0 SMF UPDATE PRECEDED BY AN SMF UPDATE (CRU005I)
0 SMF DELETE PRECEDED BY AN SMF UPDATE (CRU006I)
0 SMF INSERT PRECEDED BY AN SMF DELETE (CRU007I)
0 SMF UPDATE PRECEDED BY EXPORT RECORD (CRU008I)
0 SMF DELETE PRECEDED BY EXPORT RECORD (CRU009I)
6 EXPORT RECORD CARRIED FORWARD (CRU001I)
16 SUPERSEDED RECORDS PROCESSED WITHOUT ERROR OR ANOMALY
0 SMF INSERT FOR A NEW RECORD (CRU012I)
0 SMF UPDATE PRECEDED BY AN SMF INSERT (CRU013I)
0 SMF DELETE PRECEDED BY AN SMF INSERT (CRU014I)
0 SMF UPDATE PRECEDED BY AN SMF UPDATE (CRU015I)
0 SMF DELETE PRECEDED BY AN SMF UPDATE (CRU016I)
0 SMF INSERT PRECEDED BY AN SMF DELETE (CRU017I)
0 SMF UPDATE PRECEDED BY EXPORT RECORD (CRU018I)
0 SMF DELETE PRECEDED BY EXPORT RECORD (CRU019I)
16 EXPORT RECORD SUPERSEDED (CRU011I)
INTEGRATED CATALOG FORWARD RECOVERY UTILITY V2R1
CRURRAP SYSPRINT 03/16/07 (07.075) 20:10:32 PAGE 4
REPORT OF RECORDS BY DATA SET
32 TOTAL RECORDS IN THE NEW EXPORT DATA SET (EXPOUT)
10 CONTROL RECORDS
22 CATALOG RECORDS
16 RECORDS FORWARDED FROM THE OLD EXPORT DATA SET (EXPIN)
10 CONTROL RECORDS
6 CATALOG RECORDS
16 CATALOG RECORDS SELECTED FROM THE SMF DATA SET (SMFIN)
32 TOTAL RECORDS FROM THE OLD EXPORT DATA SET (EXPIN)
16 RECORDS CARRIED FORWARD TO THE NEW EXPORT DATA SET
10 CONTROL RECORDS
6 CATALOG RECORDS
16 RECORDS SUPERSEDED OR DELETED (BASED ON SMF DATA)
0 RECORDS REJECTED BECAUSE OF ERRORS
0 INVALID LENGTH
0 UNRECOGNIZED CATALOG RECORD TYPE
16 TOTAL RECORDS FROM THE SMF DATA SET (SMFIN)
16 RECORDS CARRIED FORWARD TO THE NEW EXPORT DATA SET
0 RECORDS SUPERSEDED OR DELETED BY NEWER SMF RECORDS
0 RECORDS REJECTED
0 NOT AN MVS SMF RECORD
0 NOT AN SMF CATALOG RECORD
0 NOT AN SMF CATALOG RECORD FOR THIS CATALOG
0 DATE/TIME EARLIER THAN EFFECTIVE START TIME
0 DATE/TIME LATER THAN EFFECTIVE STOP TIME
32 TOTAL OF ALL OUTPUT RECORDS
16 TOTAL OF ALL RECORDS DISCARDED
48 TOTAL OF ALL INPUT RECORDS

92
z/OS V1R8 DFSMS Technical Update
In Figure 3-80 we show the SYLOG output. In this case the data reflected documents the
checking that occurred to determine whether a record found by the ICFRU analysis is more or
less up to date than that shown in the second IDCAMS EXPORT run.
Figure 3-80 ICFRU Readiness check - CRURRAP SYSLOG output
There are several instances of these checks. They are not all shown in the example because
they all reflect the same situation.
For the purpose of the ICFRU readiness check, the last step is to compare the second
IDCAMS EXPORT data set with the one constructed by running CRURRSV, DFSORT™, and
CRURRAP.
INTEGRATED CATALOG FORWARD RECOVERY UTILITY V2R1
CRURRAP SYSLOG 03/16/07 (07.075) 20:10:32
CRU104I SPECIFIED START PRECEDES EXPORT, ANOMALIES POSSIBLE

CRU209I SMF INSERT IS MOST CURRENT BUT IS PRECEDED BY EXPORT RECORD FOR
(A) MHLCAT9.BCAT.TEST01 /00
NEWER: DEFINE FROM SYS SC64 AT 12:44:24.44 ON 03/16/07 (07.075)
OLDER: EXPORT RECORD
SMF INSERT RECORD IS WRITTEN TO THE NEW EXPORT DATA SET

+0000 4F8000 01300000 1E3D0045 FBCC0107 075FE2C3 F6F44040 4040C9D5 00000028 000A0001 *.............¬SC64 IN........*
+0020 4F8020 00000032 009E0001 40F1C3C1 E3D4C7D4 E340D4C8 D3D9C5E2 F1C90045 FB950107 *........ 1CATMGMT MHLRES1I...n..*
+0040 4F8040 075FD4C8 D3D9C5E2 F14040E4 C3C1E34B C6D3C5E3 C3C8C5D9 40404040 40404040 *.¬MHLRES1 UCAT.FLETCHER *
+0060 4F8060 40404040 40404040 40404040 40404040 40404040 404040C1 D4C8D3C3 C1E3F94B * AMHLCAT9.*
+0080 4F8080 C2C3C1E3 4BE3C5E2 E3F0F140 40404040 40404040 40404040 40404040 40404040 *BCAT.TEST01 *
+00A0 4F80A0 40404040 40404040 40404040 40404040 40404040 40404040 40404040 40404040 * *
+00C0 4F80C0 40404040 40404040 40404040 00644040 00600034 C1000001 2DD4C8D3 C3C1E3F9 * .. .-..A....MHLCAT9*
+00E0 4F80E0 4BC2C3C1 E34BE3C5 E2E3F0F1 40404040 40404040 40404040 40404040 40404040 *.BCAT.TEST01 *
+0100 4F8100 40404040 40000014 01FFFFFF FFFFFFFF FF080707 5F00000F 01000016 0400E2C2 * ...............¬.........SB*
+0120 4F8120 D6E7C5F1 3010200F 08000000 00000000 *OXE1............ *

CRU011I EXPORT RECORD WAS SUPERSEDED AND WAS THE OLDEST RECORD FOR
(A) MHLCAT9.BCAT.TEST01 /00
RECORD IS BYPASSED, ACTION WAS TAKEN FOR A MORE CURRENT RECORD

CRU209I SMF INSERT IS MOST CURRENT BUT IS PRECEDED BY EXPORT RECORD FOR
(A) MHLCAT9.BCAT.TEST02 /00
NEWER: DEFINE FROM SYS SC64 AT 12:44:24.45 ON 03/16/07 (07.075)
OLDER: EXPORT RECORD
SMF INSERT RECORD IS WRITTEN TO THE NEW EXPORT DATA SET

+0000 4F8000 01300000 1E3D0045 FBCD0107 075FE2C3 F6F44040 4040C9D5 00000028 000A0001 *.............¬SC64 IN........*
+0020 4F8020 00000032 009E0001 40F1C3C1 E3D4C7D4 E340D4C8 D3D9C5E2 F1C90045 FB950107 *........ 1CATMGMT MHLRES1I...n..*
+0040 4F8040 075FD4C8 D3D9C5E2 F14040E4 C3C1E34B C6D3C5E3 C3C8C5D9 40404040 40404040 *.¬MHLRES1 UCAT.FLETCHER *
+0060 4F8060 40404040 40404040 40404040 40404040 40404040 404040C1 D4C8D3C3 C1E3F94B * AMHLCAT9.*
+0080 4F8080 C2C3C1E3 4BE3C5E2 E3F0F240 40404040 40404040 40404040 40404040 40404040 *BCAT.TEST02 *
+00A0 4F80A0 40404040 40404040 40404040 40404040 40404040 40404040 40404040 40404040 * *
+00C0 4F80C0 40404040 40404040 40404040 00644040 00600034 C1000001 2DD4C8D3 C3C1E3F9 * .. .-..A....MHLCAT9*
+00E0 4F80E0 4BC2C3C1 E34BE3C5 E2E3F0F2 40404040 40404040 40404040 40404040 40404040 *.BCAT.TEST02 *
+0100 4F8100 40404040 40000014 01FFFFFF FFFFFFFF FF080707 5F00000F 01000016 0400E2C2 * ...............¬.........SB*
+0120 4F8120 D6E7C1F7 3010200F 08000000 00000000 *OXA7............ *

CRU011I EXPORT RECORD WAS SUPERSEDED AND WAS THE OLDEST RECORD FOR
(A) MHLCAT9.BCAT.TEST02 /00
RECORD IS BYPASSED, ACTION WAS TAKEN FOR A MORE CURRENT RECORD
.
.
.

Chapter 3. DFSMSdfp V1R8 enhancements
93
In Figure 3-81 we show the job used to do the compare using the standard IEBCOMPR utility.
Figure 3-81 ICFRU Readiness check - IEBCOMPR compare job
//MHLRES1I JOB (999,POK),'MHLRES2',CLASS=A,MSGCLASS=T,
// NOTIFY=&SYSUID,TIME=1440,REGION=6M
/*JOBPARM L=999,SYSAFF=*
//*****************************************************************
//* THIS JCL EXECUTES IEBCOMPR TO COMPARE THE NEW EXPORT DATA SET
//* PRODUCED BY CRURRAP WITH THE SECOND EXPORT DATA SET CREATED
//* AS PART OF THE ICFRU READINESS CHECK
//*
//* THIS COMPARISON IS ONLY FOR TESTING THE CATALOG UPDATE PROCESS
//* SINCE THE SECOND EXPORT DATA SET WILL NOT EXIST IN A REAL
//* RECOVERY SITUATION
//*
//COMP EXEC PGM=IEBCOMPR
//SYSPRINT DD SYSOUT=*
//SYSUT1 DD DISP=SHR,DSN=MHLRES1.BCAT.CATALOG.BACKUP.G0006V00
//SYSUT2 DD DISP=SHR,DSN=MHLRES1.BCAT.NEW.CATALOG.EXPORT.FLETCHER
//SYSIN DD DUMMY

94
z/OS V1R8 DFSMS Technical Update
In Figure 3-82 we show the full output from the IEBCOMPR job because the utility does not
document the data sets in its output, so it is necessary to show the JES2 output showing the
SYSUT1 and SYSUT2 DDNAMEs with data sets.
Figure 3-82 ICFRU Readiness check - IEBCOMPR compare job output
3.10.5 ICFRU implementation final steps
One ICFRU has been validated as a means to reconstruct IDCAMS EXPORT data sets, the
processes necessary to regularly capture IDCAMS EXPORT copies of catalogs and to
capture SMF records need to be set up.

J E S 2 J O B L O G -- S Y S T E M S C 6 4 -- N O D E W T S C P L X 2

20.11.18 JOB22060 ---- FRIDAY, 16 MAR 2007 ----
20.11.18 JOB22060 IRR010I USERID MHLRES1 IS ASSIGNED TO THIS JOB.
20.11.18 JOB22060 ICH70001I MHLRES1 LAST ACCESS AT 20:10:32 ON FRIDAY, MARCH 16, 2007
20.11.18 JOB22060 $HASP373 MHLRES1I STARTED - INIT 1 - CLASS A - SYS SC64
20.11.18 JOB22060 IEF403I MHLRES1I - STARTED - TIME=20.11.18 - ASID=002A - SC64
20.11.18 JOB22060 - --TIMINGS (MINS.)-- ----PAGING COUNTS---
20.11.18 JOB22060 -JOBNAME STEPNAME PROCSTEP RC EXCP CPU SRB CLOCK SERV PG PAGE SWAP VIO SWAPS STEPNO
20.11.18 JOB22060 -MHLRES1I COMP 00 53 .00 .00 .00 209 0 0 0 0 0 1
20.11.18 JOB22060 IEF404I MHLRES1I - ENDED - TIME=20.11.18 - ASID=002A - SC64
20.11.18 JOB22060 -MHLRES1I ENDED. NAME-MHLRES2 TOTAL CPU TIME= .00 TOTAL ELAPSED TIME= .00
20.11.18 JOB22060 $HASP395 MHLRES1I ENDED
------ JES2 JOB STATISTICS ------
16 MAR 2007 JOB EXECUTION DATE
17 CARDS READ
54 SYSOUT PRINT RECORDS
0 SYSOUT PUNCH RECORDS
3 SYSOUT SPOOL KBYTES
0.00 MINUTES EXECUTION TIME
1 //MHLRES1I JOB (999,POK),'MHLRES2',CLASS=A,MSGCLASS=T, JOB22060
// NOTIFY=&SYSUID,TIME=1440,REGION=6M 00020000
/*JOBPARM L=999,SYSAFF=* 00030000
//*****************************************************************
//* THIS JCL EXECUTES IEBCOMPR TO COMPARE THE NEW EXPORT DATA SET
//* PRODUCED BY CRURRAP WITH THE SECOND EXPORT DATA SET CREATED
//* AS PART OF THE ICFRU READINESS CHECK

//*
//* THIS COMPARISON IS ONLY FOR TESTING THE CATALOG UPDATE PROCESS
//* SINCE THE SECOND EXPORT DATA SET WILL NOT EXIST IN A REAL
//* RECOVERY SITUATION
//*
IEFC653I SUBSTITUTION JCL - (999,POK),'MHLRES2',CLASS=A,MSGCLASS=T,NOTIFY=MHLRES1,TIME=1440,REGION=6M
2 //COMP EXEC PGM=IEBCOMPR
3 //SYSPRINT DD SYSOUT=*
4 //SYSUT1 DD DISP=SHR,DSN=MHLRES1.BCAT.CATALOG.BACKUP.G0006V00
5 //SYSUT2 DD DISP=SHR,DSN=MHLRES1.BCAT.NEW.CATALOG.EXPORT.FLETCHER
6 //SYSIN DD DUMMY
ICH70001I MHLRES1 LAST ACCESS AT 20:10:32 ON FRIDAY, MARCH 16, 2007

IEF236I ALLOC. FOR MHLRES1I COMP
IEF237I JES2 ALLOCATED TO SYSPRINT
IGD103I SMS ALLOCATED TO DDNAME SYSUT1
IGD103I SMS ALLOCATED TO DDNAME SYSUT2
IEF237I DMY ALLOCATED TO SYSIN
IEF142I MHLRES1I COMP - STEP WAS EXECUTED - COND CODE 0000
IEF285I MHLRES1.MHLRES1I.JOB22060.D0000101.? SYSOUT
IGD104I MHLRES1.BCAT.CATALOG.BACKUP.G0006V00 RETAINED, DDNAME=SYSUT1
IGD104I MHLRES1.BCAT.NEW.CATALOG.EXPORT.FLETCHER RETAINED, DDNAME=SYSUT2
IEF373I STEP/COMP /START 2007075.2011
IEF374I STEP/COMP /STOP 2007075.2011 CPU 0MIN 00.01SEC SRB 0MIN 00.00SEC VIRT 376K SYS 320K EXT 4K SYS
IEF375I JOB/MHLRES1I/START 2007075.2011
IEF376I JOB/MHLRES1I/STOP 2007075.2011 CPU 0MIN 00.01SEC SRB 0MIN 00.00SEC
COMPARE UTILITY PAGE 0001
END OF JOB-TOTAL NUMBER OF RECORDS COMPARED = 00000032
Note: It is important to take into account the necessity to capture the SMF records from all
systems that have access to the catalogs in question, and they must all participate in the
recovery process.

© Copyright IBM Corp. 2008. All rights reserved.
95
Chapter 4.
OAM enhancements
In this chapter we discuss new and changed functions in OAM. The following topics are
covered:
Binary large object support
Immediate backup copy
Automated selection of RECYCLE volumes
Global display keyword
Update from z/OS V1.7
4

96
z/OS V1R8 DFSMS Technical Update
4.1 Binary large object support
DB2 UDB for OS/390 Version 6 introduced the support for large objects (LOBs). They can
contain text documents, images, or movies, and can be stored directly in the DBMS with sizes
up to 2 gigabytes per object and 65,536 TB for a single LOB column in a 4,096 partition table.
The introduction of these new data types has implied some changes in the administration
processes and programming techniques. The IBM Redbook LOBs with DB2 for z/OS:
Stronger and Faster, SG24-7270, describes the usage of LOBs with DB2 for z/OS. In z/OS
V1R8, OAM has been enhanced for use with DB2’s LOB support.
If LOB support is not enabled, OAM stores objects larger than 32 K in multiple rows in the
32 K DB2 table. A 256 MB object for example would take approximately 8,000 rows to store.
When LOB support is enabled, objects up to 256 MB in size can be stored in a single row.
This has the effect of improving performance and reducing the frequency of lock escalations
and time outs.
4.1.1 Implementing
These are the steps we used in order to implement the LOB function:
1.Ensure that the values for LOBVALA and LOBVALS in DB2’s DSNZPARM are sufficient
for supporting large objects. They must allow for objects greater than 256 M in size in
order to use OAM’s new function. See the IBM Redbooks publication LOBs with DB2 for
z/OS: Stronger and Faster, SG24-7270, for more information about preparing DB2 for
LOBs.
2.Verify that the MOS=xxx value in your IEFSSNxx PARMLIB member is sufficient. The
default is 50 M, and future attempts to use OSREQ to store objects larger than 50 M fail if
this value is not specified. Figure 4-1 shows an example of an IEFSSNxx member with
MOS=xxx customized to support objects up to 256 M in size.
Figure 4-1 Setting the MOS=xxx value in IEFSSNxx
EDIT SYS1.PARMLIB(IEFSSNR7) - 01.76
Command ===>
000213 SUBSYS SUBNAME(OAM1)
000214 INITRTN(CBRINIT)
000215 INITPARM('MSG=EM,MOS=256')

Chapter 4. OAM enhancements
97
3.Add the LOB keyword in your IEFSSNxx PARMLIB member. The allowed values are A, P,
and N. LOB=N is the default. Specifying A puts objects for all storage groups into a LOB
storage structure, specifying P places objects for a partial list of storage groups into a LOB
storage structure, and specifying N does not put any objects into a LOB storage structure.
Figure 4-2 is an excerpt from the DFSMS OAM Planning, Installation, and Storage
Administration Guide for Object Support, SC26-0426, publication explaining each value in
detail.
Figure 4-2 Definition of the LOB parameter in PARMLIB
Figure 4-3 shows an example of an IEFSSNxx member customized to use LOB support
for a partial list of the object storage groups defined to SMS.
Figure 4-3 Setting the LOB=x value in IEFSSNxx
4.Modify and run the CBRSMR18 migration job. This job performs the migration from the
z/OS V1R7 version of the Object Storage Database to the z/OS V1R8 version that
supports DB2 large objects. Running this job is required even if you do not intend on using
the LOB function. The latest version of CBRSMR18 is found in SAMPLIB, and it has been
updated by recent APARs.
LOB=x
Specifies whether or not OAM exploits DB2 LOB support for large
objects that exceed 32 KB (32640 bytes). LOB has the following
options:
LOB=A specifies that, for all storage groups, objects that
exceed 32 KB are to be stored in a LOB storage structure
when stored to DB2. LOB=A indicates to OAM that the
installation has created LOB storage structures and
associated V_OSM_LOB_BASE_TBL views for ALL object storage
groups defined in the ACDS. This results in optimal
performance when you want to store large objects(greater
than 32 KB) to DB2, because OAM does not query DB2 to see if
the LOB base table view exists. If the LOB base table view
does not exist, the large object store fails.
LOB=P indicates to OAM that the installation has created LOB
storage structures and associated V_OSM_LOB_BASE_TBL views
for a PARTIAL list of object storage groups defined in the
ACDS. This requires OAM to query DB2 to see if the LOB base
table view exists for a given object storage group for each
large object stored. If the LOB base table view does exist
for a given object storage group, large objects are stored
in the associated LOB storage structure. If the LOB base
table view does not exist, large objects are stored in the
32 KB data table.
LOB=N specifies that objects that exceed 32 KB are to be
stored in a 32 KB data table when stored to DB2. This is the
default option.
EDIT SYS1.PARMLIB(IEFSSNR7) - 01.76
Command ===>
000213 SUBSYS SUBNAME(OAM1)
000214 INITRTN(CBRINIT)
000215 INITPARM('MSG=EM,MOS=256,LOB=P')

98
z/OS V1R8 DFSMS Technical Update
5.Modify and run the CBRILOB sample job found in SAMPLIB. This job defines the VSAM
ESDS that is used by DB2 to create the LOB storage structure and creates the LOB base
table, base table view, auxiliary table, and index that comprise the LOB storage structure
within the object storage table hierarchy. It should be customized to create the LOB
structure for every storage group that you intend to have exploit large objects. If you
specify LOB=A in IEFSSNxx, all groups need to have the LOB structure defined using this
job. Figure 4-4 shows some tips for calculating primary and secondary space needs within
this job if you do it manually.
Figure 4-4 Calculating primary and secondary space needs in CBRILOB
6.Modify and run the CBRPBIND sample job found in SAMPLIB. This job performs a DB2
bind for the packages needed to access the OAM storage object group, OAM
administration, and OAM configuration tables. It has been updated by recent APARs and it
now uses VALIDATE(RUN) instead of VALIDATE(BIND). The DFSMS Using the New
Functions, SC26-7473, publication contains more information about when you might want
to change this default to VALIDATE(BIND).
7.Verify the state of your OAM/DB2 environment and the functionality of the LOB support,
as discussed in “Validating” on page 98.
4.1.2 Validating
After implementing the LOB function, you should verify the state of your OAM/DB2
environment. The OSREQ command and SMF analysis are two good methods to accomplish
this task.
OSREQ
OSREQ is a TSO/E command processing tool that can be used to manipulate objects in your
OAM/DB2 environment. Refer to DFSMS OAM Planning, Installation, and Storage
Administration Guide for Object Support, SC26-0426, for more information about this tool.
OSREQ requires that the DSNALI module, or DSNCLI module when OAM is running under
CICS, be APF authorized and in the LINKLST concatenation or in the STEPLIBs of any jobs
using OSREQ to store and access objects. If you attempt to use OSREQ and these
prerequisites are not met, message CBR0401I is issued with a return code 16 and reason
code D8010000. CBRSAMIV is a sample member in SAMPLIB that runs OSREQ in a batch
job.
Calculating the primary and secondary space needs for OSMLBTS, OTLOBX1, OSMLATS
and OTLOBAX1 in a customized CBRILOB job.
Each formula results in number of cylinders.
The result could be calculated as number of tracks if you divide by 49152
instead of 720K.
OSMLBTS:
(Number of objects stored in this table * 78 ) / 720K
OTLOBX1:
(Number of objects stored in this table * 50) / 720K
OSMLATS:
(Number of objects stored in this table * average object size) / 720K
OTLOBAX1:
(Number of objects stored in this table * 21 ) / 720K

Chapter 4. OAM enhancements
99
Figure 4-5 is an example of a customized CBRSAMIV job that stores a new 200 million byte
object, lists it, queries it, retrieves it, and deletes it.
Figure 4-5 Customized JCL for the CBRSAMIV sample job
//STEP1 EXEC PGM=IKJEFT01,REGION=4096K
//SYSPRINT DD SYSOUT=*
//STEPLIB DD DSN=DB2M8.SDSNLOAD,DISP=SHR
//SYSTSPRT DD SYSOUT=*
//SYSTSIN DD *
OSREQ STORE OAMTEST.LOB200A OAMTEST.LOB200A LENGTH(200000000)
LISTCAT ENTRIES('OAMTEST.LOB200A') ALL
OSREQ QUERY OAMTEST.LOB200A OAMTEST.LOB200A
OSREQ RETRIEVE OAMTEST.LOB200A OAMTEST.LOB200A COMPARE VIEW(PRIMARY)
OSREQ DELETE OAMTEST.LOB200A OAMTEST.LOB200A
/*

100
z/OS V1R8 DFSMS Technical Update
Running the customized CVRSAMIV sample job results in the output seen in Figure 4-6.
Figure 4-6 Output from running the CBRSAMIV sample job
The return and reason code from the STORE operation above indicates that in this case the
catalog entry was created for this collection and that the storage class specified for the
collection was overridden. The DFSMSdfp Diagnosis, GY27-7618, publication is very useful
in determining what the OSREQ return codes mean.
SMF
SMF data should be analyzed in order to validate that the new LOB support and associated
performance improvements are being used when running the CBRSAMIV sample job.
OAM writes SMF Record type 85 subtype 2/3/6 to document the OSREQ macros use of the
LOB support.
IEF375I JOB/CBROAMIV/START 2007057.1654
IEF376I JOB/CBROAMIV/STOP 2007057.1654 CPU 0MIN 01.89SEC SRB 0MIN
00.00S
READY
OSREQ STORE OAMTEST.LOB200A OAMTEST.LOB200A LENGTH(200000000)
OSREQ STORE successful. Return code = 00000004, reason code = 04020480.
OSREQ STORE response time is 6845 milliseconds.
OSREQ STORE data rate is 28533 kilobytes/second.
READY
LISTCAT ENTRIES('OAMTEST.LOB200A') ALL
NONVSAM ------- OAMTEST.LOB200A
IN-CAT --- UCAT.VSBOX01
HISTORY
DATASET-OWNER-----(NULL) CREATION--------2007.057
RELEASE----------------2 EXPIRATION------9999.999
ACCOUNT-INFO-----------------------------------(NULL)
OAMDATA
DIRECTORYTOKEN----GROUP00
SMSDATA
STORAGECLASS ----OBJDASD MANAGEMENTCLASS--OBJDASD
DATACLASS --------(NULL) LBACKUP ---XXXX.XXX.XXXX
ASSOCIATIONS--------(NULL)
ATTRIBUTES
READY
OSREQ RETRIEVE OAMTEST.LOB200A OAMTEST.LOB200A COMPARE VIEW(PRIMARY)
OSREQ QUERY successful. Return code = 00000000, reason code = 00000000.
OSREQ QUERY response time is 12 milliseconds.
OSREQ RETRIEVE successful. Return code = 00000000, reason code = 00000000.
OSREQ RETRIEVE response time is 1611 milliseconds.
OSREQ RETRIEVE data rate is 121236 kilobytes/second.
Data comparison for object OAMTEST.LOB200A OAMTEST.LOB200A successful.
READY
OSREQ DELETE OAMTEST.LOB200A OAMTEST.LOB200A
OSREQ DELETE successful. Return code = 00000000, reason code = 00000000.
OSREQ DELETE response time is 13 milliseconds.
READY
END

Chapter 4. OAM enhancements
101
We have written a simple program called SMF85TA to scan the SMF records and summarize
activity. The program itself and how to construct it is documented in “SMF record type 85
subtype 1-7 data display program” on page 513
In Figure 4-7 we show the JCL to extract the SMF records and run the program.
If you do not want output from all the types that the program can process, change the
SMFselection statement to only include those subtypes that you do want. For example,
change OUTDD(OUTDD,TYPE(85(1,2,3,4,5,6,7))) to OUTDD(OUTDD,TYPE(85(2,3,4,5,6)))
to only exclude subtypes 1 and 7.
Figure 4-7 SMF85TA program execution JCL
In Figure 4-8 on page 102 and Figure 4-9 on page 103 we show an example of output from
the execution of program SMF85TA with all records selected in the SMFEXTR step. At the
bottom in bold, you can see an OSREQ DELETE with a value of 81000000. Looking at the
CBRSMF member of MACLIB, you can see that the 8 indicates “PRIMARY COPY OF
OBJECT DELETED FROM DASD” and the 1 indicates “PRIMARY COPY OF THE OBJECT
WAS DELETED FROM A LOB STORAGE STRUCTURE”.
Looking further up in the output you can see an OSREQ RETRIEVE with a value of
80040000. If we look at CBRSMF again, we can see that the 8 indicates “PRIMARY COPY
OF OBJECT RETRIEVED FROM DASD” and the 4 indicates “PRIMARY COPY OF THE
OBJECT WAS RETRIEVED FROM A LOB STORAGE STRUCTURE”. This same method is
also used to translate values for output from OSREQ STORE. Another important thing about
STORE is that ST2FLGS8 equals X'80' when an immediate backup copy is scheduled, as
seen in the CBRSMF data pasted in Figure 4-12 on page 106. The SMF Type 85 Subtype 39
record is preferred, however, for analyzing Immediate Backup Copy results, as seen in “SMF”
on page 112.
Note: This JCL assumes that an extract is being taken from the active SMF data set, which
is then passed to the SMFT85I program. The JCL could be changed to make a permanent
extract of the SMF data, or to read an already created SMF data extract.
//MHLRES1O JOB (999,POK),MSGLEVEL=1,NOTIFY=MHLRES1
//SMFEXTR EXEC PGM=IFASMFDP
//SYSPRINT DD SYSOUT=*
//DUMPIN DD DISP=SHR,DSN=SYS1.SC64.MAN1
//OUTDD DD DSN=&SMFT85,
// SPACE=(CYL,(10,5)),
// RECFM=VB,LRECL=5096,
// DISP=(,PASS,DELETE),
// UNIT=SYSDA
//SYSIN DD *
INDD(DUMPIN,OPTIONS(DUMP))
OUTDD(OUTDD,TYPE(85(1,2,3,4,5,6,7)))
/*
// EXEC PGM=SMF85TA
//STEPLIB DD DISP=SHR,DSN=MHLRES1.SMF85TA.LOAD
//SYSUDUMP DD SYSOUT=A
//SMFIN DD DISP=SHR,DCB=BFTEK=A,
// DSN=&SMFT85
//PRINT DD SYSOUT=A,RECFM=UA

102
z/OS V1R8 DFSMS Technical Update
Figure 4-8 SMF85TA output - all subtypes part (1 of 2)
SMF TYPE 85 SUBTYPE 1-7 RECORDS
SMFDTE/TME: 2007061 14:14:01.013
STYPE: 1 OSREQ ACCESS
COLN/OBJN:
SGN/SCN/MCN/LEN/TTOK/TOK: 00000000000 00000000000000000000000000000000 D6E2D4C97F667A08
VSN/MT/RC/RS/FLGS: 00000000 00000000 00000000

SMFDTE/TME: 2007061 14:14:01.020
STYPE: 2 OSREQ STORE
COLN/OBJN: OAMTEST.MAR020A OAMTEST.OBJ0020A
SGN/SCN/MCN/LEN/TTOK/TOK: GROUP00 OBJDASD OBJDASD 00000500000 00000000000000000000000000000000 D6E2D4C97F667A08
VSN/MT/RC/RS/FLGS: 00000004 07241088 80C00000

SMFDTE/TME: 2007061 14:14:01.021
STYPE: 7 OSREQ UNACCESS
COLN/OBJN:
SGN/SCN/MCN/LEN/TTOK/TOK: 00000000000 00000000000000000000000000000000 D6E2D4C97F667A08
VSN/MT/RC/RS/FLGS: 00000000 00000000 00000000

SMFDTE/TME: 2007061 14:14:02.022
STYPE: 1 OSREQ ACCESS
COLN/OBJN:
SGN/SCN/MCN/LEN/TTOK/TOK: 00000000000 00000000000000000000000000000000 D6E2D4C97F667A08
VSN/MT/RC/RS/FLGS: 00000000 00000000 00000000

SMFDTE/TME: 2007061 14:14:02.023
STYPE: 4 OSREQ QUERY
COLN/OBJN: OAMTEST.MAR020A OAMTEST.OBJ0020A
SGN/SCN/MCN/LEN/TTOK/TOK: GROUP00 OBJDASD OBJDASD 00000000001 00000000000000000000000000000000 D6E2D4C97F667A08
VSN/MT/RC/RS/FLGS: 00000000 00000000 00000000

SMFDTE/TME: 2007061 14:14:02.026
STYPE: 7 OSREQ UNACCESS
COLN/OBJN:
SGN/SCN/MCN/LEN/TTOK/TOK: 00000000000 00000000000000000000000000000000 D6E2D4C97F667A08
VSN/MT/RC/RS/FLGS: 00000000 00000000 00000000

SMFDTE/TME: 2007061 14:14:04.031
STYPE: 1 OSREQ ACCESS
COLN/OBJN:
SGN/SCN/MCN/LEN/TTOK/TOK: 00000000000 00000000000000000000000000000000 D6E2D4C97F667A08
VSN/MT/RC/RS/FLGS: 00000000 00000000 00000000

SMFDTE/TME: 2007061 14:14:04.032
STYPE: 5 OSREQ CHANGE
COLN/OBJN: OAMTEST.MAR020A OAMTEST.OBJ0020A
SGN/SCN/MCN/LEN/TTOK/TOK: GROUP00 00000000000 00000000000000000000000000000000 D6E2D4C97F667A08
VSN/MT/RC/RS/FLGS: 00000000 00000000 20000000
SMFDTE/TME: 2007061 14:14:04.033
STYPE: 7 OSREQ UNACCESS
COLN/OBJN:
SGN/SCN/MCN/LEN/TTOK/TOK: 00000000000 00000000000000000000000000000000 D6E2D4C97F667A08
VSN/MT/RC/RS/FLGS: 00000000 00000000 00000000

SMFDTE/TME: 2007061 14:14:05.034
STYPE: 1 OSREQ ACCESS
COLN/OBJN:
SGN/SCN/MCN/LEN/TTOK/TOK: 00000000000 00000000000000000000000000000000 D6E2D4C97F667A08
VSN/MT/RC/RS/FLGS: 00000000 00000000 00000000

SMFDTE/TME: 2007061 14:14:05.035
STYPE: 4 OSREQ QUERY
COLN/OBJN: OAMTEST.MAR020A OAMTEST.OBJ0020A
SGN/SCN/MCN/LEN/TTOK/TOK: GROUP00 OBJDASD OBJDASD 00000000001 00000000000000000000000000000000 D6E2D4C97F667A08
VSN/MT/RC/RS/FLGS: 00000000 00000000 00000000
SMFDTE/TME: 2007061 14:14:05.038
STYPE: 7 OSREQ UNACCESS
COLN/OBJN:
SGN/SCN/MCN/LEN/TTOK/TOK: 00000000000 00000000000000000000000000000000 D6E2D4C97F667A08
VSN/MT/RC/RS/FLGS: 00000000 00000000 00000000

Chapter 4. OAM enhancements
103
Figure 4-9 SMF85TA output - all subtypes part (2 of 2)

SMFDTE/TME: 2007061 14:14:06.039
STYPE: 1 OSREQ ACCESS
COLN/OBJN:
SGN/SCN/MCN/LEN/TTOK/TOK: 00000000000 00000000000000000000000000000000 D6E2D4C97F667A08
VSN/MT/RC/RS/FLGS: 00000000 00000000 00000000

SMFDTE/TME: 2007061 14:14:06.040
STYPE: 4 OSREQ QUERY
COLN/OBJN: OAMTEST.MAR020A OAMTEST.OBJ0020A
SGN/SCN/MCN/LEN/TTOK/TOK: GROUP00 OBJDASD OBJDASD 00000000001 00000000000000000000000000000000 D6E2D4C97F667A08
VSN/MT/RC/RS/FLGS: 00000000 00000000 00000000

SMFDTE/TME: 2007061 14:14:06.041
STYPE: 3 OSREQ RETRIEVE
COLN/OBJN: OAMTEST.MAR020A OAMTEST.OBJ0020A
SGN/SCN/MCN/LEN/TTOK/TOK: GROUP00 00000500000 00000000000000000000000000000000 D6E2D4C97F667A08
VSN/MT/RC/RS/FLGS: 00000000 00000000 80040000

SMFDTE/TME: 2007061 14:14:06.042
STYPE: 7 OSREQ UNACCESS
COLN/OBJN:
SGN/SCN/MCN/LEN/TTOK/TOK: 00000000000 00000000000000000000000000000000 D6E2D4C97F667A08
VSN/MT/RC/RS/FLGS: 00000000 00000000 00000000
SMFDTE/TME: 2007061 14:14:07.043
STYPE: 1 OSREQ ACCESS
COLN/OBJN:
SGN/SCN/MCN/LEN/TTOK/TOK: 00000000000 00000000000000000000000000000000 D6E2D4C97F667A08
VSN/MT/RC/RS/FLGS: 00000000 00000000 00000000

SMFDTE/TME: 2007061 14:14:07.044
STYPE: 6 OSREQ DELETE
COLN/OBJN: OAMTEST.MAR020A OAMTEST.OBJ0020A
SGN/SCN/MCN/LEN/TTOK/TOK: GROUP00 00000500000 00000000000000000000000000000000 D6E2D4C97F667A08
VSN/MT/RC/RS/FLGS: 00000000 00000000 81000000

SMFDTE/TME: 2007061 14:14:07.045
STYPE: 7 OSREQ UNACCESS
COLN/OBJN:
SGN/SCN/MCN/LEN/TTOK/TOK: 00000000000 00000000000000000000000000000000 D6E2D4C97F667A08

104
z/OS V1R8 DFSMS Technical Update
In Figure 4-10 we show example of output from the execution of program SMF85TA with
records 1 and 7 not selected in the SMFEXTR step. Using this method to leave out the
OSREQ ACCESS and UNACCESS data serves to reduce the amount of data that appears in
your output, thereby simplifying things.
Figure 4-10 SMF85TA output - excluding subtypes 1 and 7
SMF TYPE 85 SUBTYPE 1-7 RECORDS
SMFDTE/TME: 2007061 14:14:01.020
STYPE: 2 OSREQ STORE
COLN/OBJN: OAMTEST.MAR020A OAMTEST.OBJ0020A
SGN/SCN/MCN/LEN/TTOK/TOK: GROUP00 OBJDASD OBJDASD 00000500000 00000000000000000000000000000000 D6E2D4C97F667A08
VSN/MT/RC/RS/FLGS: 00000004 07241088 80C00000

SMFDTE/TME: 2007061 14:14:02.023
STYPE: 4 OSREQ QUERY
COLN/OBJN: OAMTEST.MAR020A OAMTEST.OBJ0020A
SGN/SCN/MCN/LEN/TTOK/TOK: GROUP00 OBJDASD OBJDASD 00000000001 00000000000000000000000000000000 D6E2D4C97F667A08
VSN/MT/RC/RS/FLGS: 00000000 00000000 00000000

SMFDTE/TME: 2007061 14:14:04.032
STYPE: 5 OSREQ CHANGE
COLN/OBJN: OAMTEST.MAR020A OAMTEST.OBJ0020A
SGN/SCN/MCN/LEN/TTOK/TOK: GROUP00 00000000000 00000000000000000000000000000000 D6E2D4C97F667A08
VSN/MT/RC/RS/FLGS: 00000000 00000000 20000000

SMFDTE/TME: 2007061 14:14:05.035
STYPE: 4 OSREQ QUERY
COLN/OBJN: OAMTEST.MAR020A OAMTEST.OBJ0020A
SGN/SCN/MCN/LEN/TTOK/TOK: GROUP00 OBJDASD OBJDASD 00000000001 00000000000000000000000000000000 D6E2D4C97F667A08
VSN/MT/RC/RS/FLGS: 00000000 00000000 00000000

SMFDTE/TME: 2007061 14:14:06.040
STYPE: 4 OSREQ QUERY
COLN/OBJN: OAMTEST.MAR020A OAMTEST.OBJ0020A
SGN/SCN/MCN/LEN/TTOK/TOK: GROUP00 OBJDASD OBJDASD 00000000001 00000000000000000000000000000000 D6E2D4C97F667A08
VSN/MT/RC/RS/FLGS: 00000000 00000000 00000000

SMFDTE/TME: 2007061 14:14:06.041
STYPE: 3 OSREQ RETRIEVE
COLN/OBJN: OAMTEST.MAR020A OAMTEST.OBJ0020A
SGN/SCN/MCN/LEN/TTOK/TOK: GROUP00 00000500000 00000000000000000000000000000000 D6E2D4C97F667A08
VSN/MT/RC/RS/FLGS: 00000000 00000000 80040000

SMFDTE/TME: 2007061 14:14:07.044
STYPE: 6 OSREQ DELETE
COLN/OBJN: OAMTEST.MAR020A OAMTEST.OBJ0020A
SGN/SCN/MCN/LEN/TTOK/TOK: GROUP00 00000500000 00000000000000000000000000000000 D6E2D4C97F667A08
VSN/MT/RC/RS/FLGS: 00000000 00000000 81000000

Chapter 4. OAM enhancements
105
In Figure 4-11 to Figure 4-15 on page 109 we show the SMF records that relate to this output.
The Flags field, as shown in the output, reflects the flag bits as mapped by the flag fields in
the SMF records corresponding to the particular subtype.
SMF record type 85 subtype 1 is used to map subtype records 1, 2, 3, 4, 5, 6, and 7.
Figure 4-11 SMF record type 85 subtype 1 significant fields (extract from CBRSMF macro) (1 of 5)
Note: You should not use the contents of the CBRSMF macro as presented here. When
you assemble the program, or wish to refer to the macro, it will be found in SYS1.MACLIB.
ST1 DSECT SUBTYPES 1 - 7
ST1COLN DS CL44' ' COLLECTION NAME
ST1OBJN DS CL44' ' OBJECT NAME
ST1SGN DS CL8' ' STORAGE GROUP NAME
ST1SCN DS CL8' ' STORAGE CLASS NAME
ST1MCN DS CL8' ' MANAGEMENT CLASS NAME
ST1OFF DS BL4'0' OFFSET FOR PARTIAL OBJECT
* RETRIEVE (SUBTYPE 3), ZERO FOR
* ALL OTHERS.
ST1LEN DS BL4'0' LENGTH,
* SUBTYPE 1 - UNUSED
* SUBTYPE 2 - LENGTH OF OBJECT STORED
* SUBTYPE 3 - NUMBER OF BYTES RETRIEVED
* SUBTYPE 4 - NUMBER OF QEL ELEMENTS
* RETURNED.
* SUBTYPE 5 - UNUSED
* SUBTYPE 6 - LENGTH OF OBJECT DELETED
* SUBTYPE 7 - UNUSED
ST1TTOK DS CL16' ' OSREQ TRACKING TOKEN, SUPPLIED
* WITH TTOKEN KEYWORD ON OSREQ
* MACRO
ST1TOK DS CL8' ' OSREQ ACCESS TOKEN
ST1VSN DS CL6' ' VOLUME SERIAL NUMBER
ST1VMT DS CL2' ' VOLUME MEDIA TYPE
ST1RC DS BL4'0' OSREQ RETURN CODE, IN REGISTER 15
* FOLLOWING OSREQ MACRO
ST1RS DS BL4'0' OSREQ REASON CODE, IN REGISTER 15
* FOLLOWING OSREQ MACRO
ST1FLGS DS BL4'0' PROCESSING FLAGS. MEANING
* DEPENDENT ON RECORD SUBTYPE.

106
z/OS V1R8 DFSMS Technical Update
Figure 4-12 SMF record type 85 subtype 1 significant fields (extract from CBRSMF macro) (2 of 5)
***********************************************************************
* *
* SUBTYPE 2 - OSREQ STORE FLAGS *
* *
***********************************************************************
ST2FLGS0 EQU X'80' OBJECT STORE TO DASD
ST2FLGS1 EQU X'40' OBJECT STORE TO OPTICAL
ST2FLGS2 EQU X'20' OBJECT STORE TO TAPE
ST2FLGS3 EQU X'10' UNUSED
ST2FLGS4 EQU X'08' UNUSED
ST2FLGS5 EQU X'04' WHEN ON, THE OSREQ STORE
* REQUEST RESULTED IN THE MOUNTING
* OF A SHELF-RESIDENT REMOVABLE
* MEDIA VOLUME (TAPE OR OPTICAL)
* BY A HUMAN OPERATOR. ONLY VALID
* IF BIT 1 OR 2 IS ON.
ST2FLGS6 EQU X'02' WHEN ON, THE OSREQ STORE
* REQUEST RESULTED IN THE MOUNTING
* OF A LIBRARY-RESIDENT REMOVABLE
* MEDIA VOLUME (TAPE OR OPTICAL)
* BY A HUMAN OPERATOR. ONLY VALID
* IF BIT 1 OR 2 IS ON.
ST2FLGS7 EQU X'01' WHEN ON, THE OSREQ STORE
* REQUEST WAS SATISIFIED USING
* AN ALREADY MOUNTED REMOVEABLE
* MEDIA VOLUME (TAPE OR OPTICAL).
* ONLY VALID
* IF BIT 1 OR 2 IS ON.
ST2FLGS8 EQU X'80' WHEN ON, AN IMMEDIATE BACKUP COPY
* WAS SCHEDULED FOR THIS OBJECT. @L5A
ST2FLGS9 EQU X'40' WHEN ON, THE OBJECT IS STORED
* TO LOB STORAGE STRUCTURE @L5A

Chapter 4. OAM enhancements
107
Figure 4-13 SMF record type 85 subtype 1 significant fields (extract from CBRSMF macro) (3 of 5)
***********************************************************************
* *
* SUBTYPE 3 - OSREQ RETRIEVE FLAGS *
* *
***********************************************************************
ST3FLGS0 EQU X'80' WHEN ON, PRIMARY COPY OF OBJECT
* RETRIEVED FROM DASD.
ST3FLGS1 EQU X'40' WHEN ON, PRIMARY COPY OF OBJECT
* RETRIEVED FROM OPTICAL.
ST3FLGS2 EQU X'20' WHEN ON, PRIMARY COPY OF OBJECT
* RETRIEVED FROM TAPE.
ST3FLGS3 EQU X'10' WHEN ON, EITHER THE FIRST OR THE
* SECOND BACKUP COPY OF THE OBJECT WAS
* RETRIEVED FROM OPTICAL AS RESULT OF
* VIEW=BACKUP OR VIEW=BACKUP2 BEING
* SPECIFIED ON THE OSREQ MACRO.
* REFER TO BIT 10 TO INDICATE WHICH
* BACKUP COPY WAS RETRIEVED. @L2C
ST3FLGS4 EQU X'08' WHEN ON, EITHER THE FIRST OR THE
* SECOND BACKUP COPY OF THE OBJECT WAS
* RETRIEVED FROM TAPE AS RESULT OF
* VIEW=BACKUP OR VIEW=BACKUP2 BEING
* SPECIFIED ON THE OSREQ MACRO.
* REFER TO BIT 10 TO INDICATE WHICH
* BACKUP COPY WAS RETRIEVED. @P1C
ST3FLGS5 EQU X'04' WHEN ON, EITHER THE FIRST OR THE
* SECOND BACKUP COPY OF THE OBJECT WAS
* RETRIEVED FROM OPTICAL AS A RESULT
* OF THE PRIMARY COPY OF THE OBJECT
* RESIDING ON AN UNREADABLE OPTICAL
* DISK VOLUME AND THE AUTOMATIC ACCESS
* TO BACKUP WAS ACTIVE.
* REFER TO BIT 10 TO INDICATE WHICH
* BACKUP COPY WAS RETRIEVED. @P1C
ST3FLGS6 EQU X'02' WHEN ON, EITHER THE FIRST OR THE
* SECOND BACKUP COPY OF THE OBJECT WAS
* RETRIEVED FROM TAPE AS A RESULT
* OF THE PRIMARY COPY OF THE OBJECT
* RESIDING ON AN UNREADABLE OPTICAL
* DISK VOLUME AND THE AUTOMATIC ACCESS
* TO BACKUP WAS ACTIVE.
* REFER TO BIT 10 TO INDICATE WHICH
* BACKUP COPY WAS RETRIEVED. @P1C
ST3FLGS7 EQU X'01' WHEN ON, THE OSREQ RETRIEVE
* REQUEST RESULTED IN THE MOUNTING
* OF A SHELF-RESIDENT REMOVABLE
* MEDIA VOLUME (TAPE OR OPTICAL)
* BY A HUMAN OPERATOR. ONLY VALID
* IF BIT 1, 2, 3, 5 OR 6 IS ON.

108
z/OS V1R8 DFSMS Technical Update
Figure 4-14 SMF record type 85 subtype 1 significant fields (extract from CBRSMF macro) (4 of 5)
ST3FLGS8 EQU X'80' WHEN ON, THE OSREQ RETRIEVE
* REQUEST RESULTED IN THE MOUNTING
* OF A LIBRARY-RESIDENT REMOVABLE
* MEDIA VOLUME (TAPE OR OPTICAL)
* BY A HUMAN OPERATOR. ONLY VALID
* IF BIT 1, 2, 3, 5 OR 6 IS ON.
ST3FLGS9 EQU X'40' WHEN ON, THE OSREQ RETRIEVE
* REQUEST WAS SATISIFIED USING
* AN ALREADY MOUNTED REMOVEABLE
* MEDIA VOLUME (TAPE OR OPTICAL).
* ONLY VALID
* IF BIT 1, 2, 3, 5 OR 6 IS ON.
ST3FLGS10 EQU X'20' WHEN ON, THE SECOND BACKUP COPY OF
* THE OBJECT WAS RETRIEVED @L4C
ST3FLGS11 EQU X'10' WHEN ON, A RECALL WAS SCHEDULED
* FOR THIS OBJECT @L4A
ST3FLGS12 EQU X'08' WHEN ON, A RECALL WAS EXPLICITYLY
* SPECIFIED ON THE OSREQ RETRIEVE
* REQUEST @L4A
ST3FLGS13 EQU X'04' WHEN ON, THE PRIMARY COPY OF
* THE OBJECT WAS RETRIEVED FROM
* A LOB STORAGE STRUCTURE @L5A
*
***********************************************************************
* *
* SUBTYPE 4 - OSREQ QUERY FLAGS *
* *
***********************************************************************
***********************************************************************
* *
* SUBTYPE 5 - OSREQ CHANGE FLAGS *
* *
***********************************************************************
ST5FLG0 EQU X'80' WHEN ON, MANAGEMENT CLASS
* SPECIFIED ON OSREQ CHANGE.
ST5FLG1 EQU X'40' WHEN ON, STORAGE CLASS
* SPECIFIED ON OSREQ CHANGE.
ST5FLG2 EQU X'20' WHEN ON, RETENTION PERIOD
* SPECIFIED ON OSREQ CHANGE.

Chapter 4. OAM enhancements
109
Figure 4-15 SMF record type 85 subtype 1 significant fields (extract from CBRSMF macro) (5 of 5)
Migration and coexistence
Migration considerations include running the CBRILOB, CBRSMR18, and CBRPBIND jobs
as described in 4.1.1, “Implementing” on page 96. Running CBRSMR18 is required even if
you do not intend to exploit large object support. Pre-V1R8 systems in the same OAMplex as
a V1R8 system must have the fix for APAR OA12683 applied, regardless of whether the
V1R8 system is exploiting large object support. DB2 Version 6.1 or later is required before
OAM can use DB2’s LOB support.
Maintenance
The fix for APAR OA16562 must be applied prior to running the CBRILOB and CBRSMR18
jobs associated with LOB implementation. This APAR updates those sample jobs.
4.2 Immediate backup copy
You now have the ability to immediately back up your objects when they first get stored. Prior
to this support, objects would not get backed up until the object storage management cycle
(OSMC) was run, which would often times be several hours later.
***********************************************************************
* *
* SUBTYPE 6 - OSREQ DELETE FLAGS *
* *
***********************************************************************
ST6FLG0 EQU X'80' WHEN ON, PRIMARY COPY OF OBJECT
* DELETED FROM DASD.
ST6FLG1 EQU X'40' WHEN ON, PRIMARY COPY OF OBJECT
* DELETED FROM OPTICAL.
ST6FLG2 EQU X'20' WHEN ON, PRIMARY COPY OF OBJECT
* DELETED FROM TAPE.
ST6FLG3 EQU X'10' WHEN ON, BACKUP COPY OF OBJECT
* DELETED FROM OPTICAL.
ST6FLG4 EQU X'08' WHEN ON, BACKUP COPY OF OBJECT
* DELETED FROM TAPE.
ST6FLG5 EQU X'04' WHEN ON, 2ND BACKUP COPY OF OBJECT
* DELETED FROM OPTICAL. @L2A
ST6FLG6 EQU X'02' WHEN ON, 2ND BACKUP COPY OF OBJECT
* DELETED FROM TAPE. @L2A
ST6FLG7 EQU X'01' WHEN ON, THE PRIMARY OCPY OF
* THE OBJECT WAS DELETED FROM
* A LOB STORAGE STRUCTURE @L5A
***********************************************************************
* *
* SUBTYPE 7 - OSREQ UNACCESS FLAGS *
* *
***********************************************************************

110
z/OS V1R8 DFSMS Technical Update
4.2.1 Implementing
Two changes must be made in the management class definition for the objects that you want
to exploit this function. The BACKUP FREQUENCY field must be set to 0 and the AUTO
BACKUP field must be set to Y, as seen in Figure 4-16. When these two fields are updated,
an immediate backup copy will automatically be used.
Figure 4-16 Altering an existing management class in ISMF for immediate backup
4.2.2 Validating
After you have implemented immediate backup copy, there are several methods that you can
use to validate its function.
RETCODE2
As described in the binary large object support section, OSREQ can be used to store objects.
A new RETCODE2 parameter can be coded when using the OSREQ store function, and
indicates whether the store request successfully scheduled an immediate backup copy for the
object. The RETCODE2 output is only valid when the STORE is successful, in which case
one of the following return codes is provided:
0: Immediate backup copy request successfully scheduled.
4: Immediate backup copy request not required.
8: An attempt to schedule an immediate backup for this object was not successful
because OSMC is not up and running.
14: An attempt to schedule an immediate backup for this object was not successful due to
an unexpected scheduling error.
MANAGEMENT CLASS ALTER Page 3 of 5
Command ===>

SCDS Name . . . . . . : SYS1.SMS.SCDS
Management Class Name : OBJDASD

To ALTER Management Class, Specify:
Backup Attributes
Backup Frequency . . . . . . . . 0 (0 to 9999 or blank)
Number of Backup Vers . . . . . . 1 (1 to 100 or blank)
(Data Set Exists)
Number of Backup Vers . . . . . . 1 (0 to 100 or blank)
(Data Set Deleted)
Retain days only Backup Ver . . . 60 (1 to 9999, NOLIMIT or blank)
(Data Set Deleted)
Retain days extra Backup Vers . . 30 (1 to 9999, NOLIMIT or blank)
Admin or User command Backup . . BOTH (BOTH, ADMIN or NONE)
Auto Backup . . . . . . . . . . . Y (Y or N)
Backup Copy Technique . . . . . . P (P=Conc Preferred, R=Conc
Required or S=Standard)
Use ENTER to Perform Verification; Use UP/DOWN Command to View other Panels;

Chapter 4. OAM enhancements
111
Figure 4-17 provides an example JCL showing an OSREQ STORE with the RETCODE2
parameter.
Figure 4-17 Example JCL for OSREQ STORE with RETCODE2 parameter
JOBLOG output from running this sample JCL can be seen in Figure 4-18. This output
indicates that the immediate backup was scheduled successfully.
Figure 4-18 Example of an OSREQ STORE with successful immediate backup
//CBROAMIV JOB CLASS=A,MSGCLASS=A,MSGLEVEL=(1,1),REGION=0M
/*JOBPARM SYSAFF=SC64
//STEP1 EXEC PGM=IKJEFT01,REGION=4096K
//SYSPRINT DD SYSOUT=*
//STEPLIB DD DSN=DB2M8.SDSNLOAD,DISP=SHR
//SYSTSPRT DD SYSOUT=*
//SYSTSIN DD *
OSREQ STORE OAMTEST.LOB200C OAMTEST.LOB200C LENGTH(2000) RETCODE2
IEF375I JOB/CBROAMIV/START 2007057.1950
IEF376I JOB/CBROAMIV/STOP 2007057.1950 CPU 0MIN 00.05SEC SRB 0MIN
00.00
1READY
OSREQ STORE OAMTEST.LOB200C OAMTEST.LOB200C LENGTH(2000) RETCODE2
OSREQ STORE successful. Return code = 00000004, reason code = 04020480,
retcode2 = 00000000.
OSREQ STORE response time is 181 milliseconds.
OSREQ STORE data rate is 10 kilobytes/second.
READY

112
z/OS V1R8 DFSMS Technical Update
SYSLOG
You can look at SYSLOG immediately after an OSREQ STORE is run in order to see if the
immediate backup copy is performed. Figure 4-19 shows an example of the mount seen when
backing up to tape.
Figure 4-19 SYSLOG indicating OAM is backing up to tape after OSREQ STORE
SMF
OAM writes SMF record type 85 subtype 39 to document the use of the immediate backup
facility. It also uses subtype 2 to document when an immediate backup copy has been
scheduled as a result of running the OSREQ macro. See the SMF section for large object
support for more information about immediate backup and subtype 2.
We have written a simple program called SMF85TI to scan the SMF records and summarize
activity. The program itself and how to construct it is documented in “SMF record type 85
subtype 39 data display program” on page 490.
$HASP373 CBROAMIV STARTED - INIT 1 - CLASS A - SYS SC64
IEF403I CBROAMIV - STARTED - TIME=17.13.54 - ASID=002A - SC64
+IGD01009I MC ACS GETS CONTROL &ACSENVIR=STORE
+IGD01010I SG ACS GETS CONTROL &ACSENVIR=STORE
IEC501A M 0B91,TST006,SL,COMP,OAM,OAM,OAM.BACKUP.DATA
- --TIMINGS (MINS.)--
----PAGING COUNTS---
-JOBNAME STEPNAME PROCSTEP RC EXCP CPU SRB CLOCK SERV
PG PAGE SWAP VIO SWAPS STEPNO
-CBROAMIV STEP1 00 379 .03 .00 .19 537K
0 0 0 0 0 1
IEF404I CBROAMIV - ENDED - TIME=17.14.06 - ASID=002A - SC64
-CBROAMIV ENDED. NAME- TOTAL CPU TIME= .03
TOTAL ELAPSED TIME= .19
$HASP395 CBROAMIV ENDED
$HASP309 INIT 1 INACTIVE ******** C=ABCDE
IEC205I CBRRT001,OAM,OAM,FILESEQ=1, COMPLETE VOLUME LIST, 241
DSN=OAM.BACKUP.DATA,VOLS=TST006,TOTALBLOCKS=1228
IEF234E K 0B91,TST006,PVT,OAM,OAM

Chapter 4. OAM enhancements
113
In Figure 4-40 on page 130 we show the JCL to extract the SMF records and run the
program.
Figure 4-20 SMF85TI program execution JCL
Note: This JCL assumes that an extract is being taken from the active SMF data set, which
is then passed to the SMFT85I program. The JCL could be changed to make a permanent
extract of the SMF data, or to read of an already created SMF data extract.
//MHLRES1O JOB (999,POK),MSGLEVEL=1,NOTIFY=MHLRES1
// EXEC PGM=IFASMFDP
//SYSPRINT DD SYSOUT=*
//DUMPIN DD DISP=SHR,DSN=SYS1.SC64.MAN1
//OUTDD DD DSN=&SMFT85,
// SPACE=(CYL,(10,5)),
// RECFM=VB,LRECL=5096,
// DISP=(,PASS,DELETE),
// UNIT=SYSDA
//SYSIN DD *
INDD(DUMPIN,OPTIONS(DUMP))
OUTDD(OUTDD,TYPE(85(39)))
/*
// EXEC PGM=SMF85TO
//STEPLIB DD DISP=SHR,DSN=MHLRES1.SMF85TI.LOAD
//SYSUDUMP DD SYSOUT=A
//SMFIN DD DISP=SHR,DCB=BFTEK=A,
// DSN=&SMFT85
//PRINT DD SYSOUT=A,RECFM=UA

114
z/OS V1R8 DFSMS Technical Update
In Figure 4-21 we show an example of output from the execution of program SMF85TI.
Figure 4-21 SMF85TI output
The left-hand side of the report contains abbreviations from the SMF records. Record type 85
subtype 39 contains the fields of interest to verify that the function of OAM immediate backup
is occurring. In the SMF records, as shown in Figure 4-22 on page 115, the fields are prefixed
with ST39. In the report the fields are shown without the ST39, and have been compressed to
have more than one entry per line.
The fourth example shows a successful operation involving object OAMTEST.LOB200R and
tape volume TST004. The Flgs field shows 8280000, which (as per the SMF record extract
below) indicates that the primary copy is stored to DASD and the backup is on tape, and that
the write to tape was successful. The VSN field is also filled in with TST004.
The fifth example shows an unsuccessful operation involving object OAMTEST.LOB200S.
The flgs field shows 8000000, which indicates that the primary copy is on DASD, but there is
no indication of a successful backup.
In Figure 4-22 on page 115 we show the SMF records that relate to this output.
SMF TYPE 85 SUBTYPE 39 RECORDS
COLN/CNID: OAMTEST.LOB200O 00000000035
OBJN/SGN/OLEN: OAMTEST.LOB200O GROUP00 00000002000
VSN/MT/TKN/FLGS: TST008 00000000004 82800000

COLN/CNID: OAMTEST.LOB200P 00000000036
OBJN/SGN/OLEN: OAMTEST.LOB200P GROUP00 00020000000
VSN/MT/TKN/FLGS: TST003 00000000008 82800000

COLN/CNID: OAMTEST.LOB200Q 00000000037
OBJN/SGN/OLEN: OAMTEST.LOB200Q GROUP00 00020000000
VSN/MT/TKN/FLGS: TST003 00000000622 82800000

COLN/CNID: OAMTEST.LOB200R 00000000038
OBJN/SGN/OLEN: OAMTEST.LOB200R GROUP00 00020000000
VSN/MT/TKN/FLGS: TST004 00000000004 82800000

COLN/CNID: OAMTEST.LOB200S 00000000039
OBJN/SGN/OLEN: OAMTEST.LOB200S GROUP00 00020000000
VSN/MT/TKN/FLGS: 01077952576 80000000
Note: You should not use the contents of the CBRSMF macro as presented here. When
you assemble the program, or wish to refer to the macro, it will be found in SYS1.MACLIB.

Chapter 4. OAM enhancements
115
Figure 4-22 SMF record type 85 subtype 39 significant fields (extract from CBRSMF macro)
ST39 DSECT SUBTYPE 39 @L5A
ST39COLN DS CL44' ' COLLECTION NAME @L5A
ST39CNID DS BL4'0' COLLECTION ID @L5A
ST39OBJN DS CL44' ' OBJECT NAME @L5A
ST39SGN DS CL8' ' STORAGE GROUP NAME @L5A
ST39MCN DS CL8' ' MANAGEMENT CLASS NAME @L5A
ST39OLEN DS BL4'0' OBJECT LENGTH @L5A
ST39SVSN DS CL6' ' SOURCE VOLUME SERIAL NUMBER OF
* OPTICAL VOLUME OR TAPE VOLUME
* FROM WHICH THE PRIMARY OBJECT
* WAS READ.
* ONLY VALID IF THE BIT 1 OR 2
* IS ON IN FIELD ST39FLGS @L5A
ST39SMT DS CL2' ' SOURCE MEDIA TYPE OF THE VOLUME
* FROM WHICH THE PRIMARY OBJECT
* WAS READ.
* ONLY VALID IF THE BIT 1 OR 2
* IS ON IN FIELD ST39FLGS @L5A
ST39TVSN DS CL6' ' TARGET VOLUME SERIAL NUMBER OF
* OPTICAL VOLUME OR TAPE VOLUME
* ON WHICH THE BACKUP COPY OF
* THE OBJECT WAS WRITTEN.
* ONLY VALID IF THE BIT 1 OR 2
* IS ON IN FIELD ST39FLGS @L5A
ST39TMT DS CL2' ' TARGET MEDIA TYPE OF THE VOLUME
* ON WHICH THE BACKUP COPY OF
* THE OBJECT WAS WRITTEN.
* ONLY VALID IF THE BIT 1 OR 2
* IS ON IN FIELD ST39FLGS @L5A
ST39BTKN DS BL4'0' VOLUME LOCATION TOKEN ASSOCIATED
* WITH THE COPY OF THE OBJECT
* ON THE VOLUME SPECIFIED IN THE
* ST39TVSN FIELD. @L5A
ST39FLGS DS BL4'0' PROCESSING FLAGS @L5A
ST39FLG0 EQU X'80' WHEN ON, THE PRIMARY COPY
* IS STORED TO DASD. @L5A
ST39FLG1 EQU X'40' WHEN ON, THE PRIMARY COPY
* IS STORED TO OPTICAL. @L5A
ST39FLG2 EQU X'20' WHEN ON, THE PRIMARY COPY
* IS STORED TO TAPE. @L5A
ST39FLG3 EQU X'10' RESERVED @L5A
ST39FLG4 EQU X'08' RESERVED @L5A
ST39FLG5 EQU X'04' WHEN ON, THE BACKUP COPY
* IS STORED TO OPTICAL. @L5A
ST39FLG6 EQU X'02' WHEN ON, THE BACKUP COPY
* IS STORED TO TAPE. @L5A
ST39FLG7 EQU X'01' RESERVED @L5A
ST39FLG8 EQU X'80' WHEN ON, WRITE TO BACKUP COPY
* WAS SUCCESSFUL. @L5A

116
z/OS V1R8 DFSMS Technical Update
Migration and coexistence
Pre-V1R8 systems in the same OAMplex as a V1R8 system must have the fix for APAR
OA12683 applied.
Maintenance
The fix for APAR OA18519 must be applied or immediate backup copy could fail if OSREQ
STORE is consuming all of the available DB2 threads.
4.3 Automated selection of RECYCLE volumes
OAM now provides a method to automatically recycle volumes based on user-defined criteria.
It is intended for use in recycling full tape volumes associated with object or object backup
storage groups. Optical volumes do not use this function and continue to be recycled
individually with the MODIFY,OAM,START,MOVEVOL command.
4.3.1 Implementing
The MODIFY OAM,START command has been enhanced with a RECYCLE parameter. This
command is used to start the RECYCLE based on criteria specified in the command and in
PARMLIB. As seen in the DFSMS OAM Planning, Installation, and Storage Administration
Guide for Object Support, SC26-0426, publication, the syntax of the command is as shown in
Example 4-1.
Example 4-1 Syntax of the MODIFY OAM,START,RECYCLE command
MODIFY OAM,START,RECYCLE,scope{,PV=xxx}[,LIM=yy|,DISPLAY]
Displays a list of candidate volumes that meet user-defined criteria
to the hardcopy log, and in turn, automatically selects recycle
candidate volumes and initiates the MOVEVOL with RECYCLE process on
those candidates until either the user-specified limit is reached or
no more volumes meeting criteria are available.

scope
Indicates one of the following:

Specified name of an object or object backup storage group,
indicates that only tape volumes marked full, that belong to
the specified object or object backup storage group, are
considered candidates for this RECYCLE command.

ALLGRP all full tape volumes that belong to all primary object
storage groups defined in the ACTIVE SCDS are considered candidates
for this RECYCLE command.

ALLBK1 all full tape volumes that belong to all first backup
storage groups defined in the ACTIVE SCDS are considered
candidates for this RECYCLE command.

ALLBK2 all full tape volumes that belong to all second backup
storage groups defined in the ACTIVE SCDS are considered
candidates for this RECYCLE command.
PV

Chapter 4. OAM enhancements
117
=nnn An optional keyword indicating the valid data threshold to be
used in determining whether a volume is a candidate for RECYCLE.
Full tape volumes that have a percentage of valid data less than
or equal to nnn are candidates for RECYCLE. If PV=nnn is not
specified, the percent valid to be used to determine RECYCLE
candidates is derived from the PERCENTVALID default value as
defined through the SETOAM command in the CBROAMxx PARMLIB member.
Valid values for nnn are 0 to 100.

DISPLAY
An optional parameter that produces a list of volumes that meet
criteria to be recycle candidates. This list is sorted by the
percentage of valid data on each volume and is written to hardcopy
system log through the CBR9875I message. This option does not
initiate Recycle processing, and can be issued at anytime, whether
a RECYCLE command is actively processing or not. The list of
candidate volumes might be large as it shows all volumes that meet
the user-specified criteria for RECYCLE.
If DISPLAY is not specified then LIM=yy must be specified.

LIM
=yy If the DISPLAY parameter is not specified, this keyword is
required to indicate the maximum number of volumes to be selected
for RECYCLE processing. Valid values for yy are 1 to 40.
If LIM=yy is not specified, then DISPLAY must be specified.
New keywords have been added for the SETOAM statement in the CBROAMxx member of
PARMLIB. They are:
MAXRECYCLETASKS
SGMAXRECYCLETASKS
PERCENTVALID

118
z/OS V1R8 DFSMS Technical Update
Figure 4-23 shows the definitions for these new keywords from the DFSMS OAM Planning,
Installation, and Storage Administration Guide for Object Support, SC26-0426, publication.
Figure 4-23 The new SETOAM statements for CBROAMxx in PARMLIB
MAXRECYCLETASKS(nn)
Can be specified at the global level. The nn is the maximum number of
MOVEVOL tasks that can be run concurrently by the RECYCLE function.
Valid values for nn are 0 - 15. The default is 1 if no value is
specified. A value of 0 indicates that no RECYCLE operations can be
run at the storage group or global level.
SGMAXRECYCLETASKS(nn)
An optional parameter that you can specify at the storage group
level. The nn is the maximum number of MOVEVOL tasks that can be
run concurrently by the RECYCLE function for a storage group. The
value for SGMAXRECYCLETASKS cannot exceed the value for
MAXRECYCLETASKS. Valid values for nn are 0 - 15. The default is 1
if no value is specified. A value of 0 indicates that no RECYCLE
operations can be run at the storage group level specified.

If you only want to recycle volumes from one group, the setting
for all other groups would be 0 to ensure that the group with a
non-zero value receives all the recycling processing.

If you specify a value for a group that is higher than the value
for another group, the system selects more of the volumes to
recycle from the group with the higher value. However, RECYCLE
processing might be working with the original order of volumes
that are sorted by the amount of valid data for each volume, and
might select volumes from other groups to satisfy the limit before
it processes the higher-value group.
PERCENTVALID(nnn)
You can only specify this keyword at the global (all storage groups)
level. nnn represents the global default percentage of valid data
threshold that is used to determine whether a full tape volume is a
candidate for RECYCLE processing. This SETOAM value is used only if
the optional PV= keyword is not specified on the RECYCLE command. The
PERCENTVALID value that is specified on the RECYCLE command takes
precedence over the PERCENTVALID value in the SETOAM statement. Valid
values for nnn are 0 - 100. The default is 0 if no value is
specified.

Chapter 4. OAM enhancements
119
An example of an updated CBROAMxx PARMLIB member can be seen in Figure 4-24.
Figure 4-24 Example CBROAMxx PARMLIB member updated for RECYCLE
4.3.2 Validating
In this section we discuss methods to validate and use the new function.
START RECYCLE
The DISPLAY option can be used to display RECYCLE candidates without actually
performing the MOVEVOL with RECYCLE function. Figure 4-25 is an example of a
RECYCLE command with the DISPLAY parameter for all first backup storage groups. You can
also see that we are overriding the PERCENTVALID value that we have in PARMLIB.
Figure 4-25 Showing RECYCLE candidates with the DISPLAY option
SYS1.PARMLIB(CBROAM00) - 01.44
===>
***************************** Top of Data ********
SETOAM TAPEDISPATCHERDELAY(45)
TAPERECYCLEMODE(MVSSCRATCH)
STORAGEGROUP(GROUP00 TAPEUNITNAME(3590-1))
STORAGEGROUP(OBJBKP TAPEUNITNAME(3590-1))
MAXRECYCLETASKS(2)
STORAGEGROUP(GROUP00 SGMAXRECYCLETASKS(2))
PERCENTVALID(10)
SETOSMC MAXRECALLTASKS(2)
RECALLALL(15)
CLEAROLDLOC(TAPE)
RECALLTAPE(0)
FIRSTBACKUPGROUP(OBJBKP)
STORAGEGROUP(GROUP00 RECALLOFF(ON)
FIRSTBACKUPGROUP(OBJBKP))
MODIFY OAM,START,RECYCLE,(ALLBK1),DISPLAY,PV=50
CBR9880I OAM START RECYCLE command starting.
CBR9875I Recycle Candidates: 720
The following volumes are candidates for OAM RECYCLE command
processing using pv=50, lim=N/A, scope=(ALLBK1), maxrecycletasks=2.
VOLSER %VAL SGNAME STAT VOLSER %VAL SGNAME STAT
TST016 16 OBJBKP
CBR9879I OAM Recycle: End of OAM Recycle candidate volumes.
CBR9881I OAM START RECYCLE command ending successfully. Reason is
display specified.

120
z/OS V1R8 DFSMS Technical Update
Figure 4-26 is an example of an actual RECYCLE taking place. In this case we again override
the PERCENTVALID that we previously specified in PARMLIB because the PERCENTVALID
value of the candidate volume is greater. We also setting LIM=2, although we could have set
LIM=1 because we have only one candidate volume. The LIM= keyword is required when the
DISPLAY parameter is not specified.
Figure 4-26 Issuing the RECYCLE for first backup storage group tape volumes
SMF
OAM writes SMF record type 85 subtype 40 to document the use of the RECYCLE facility.
We have written a simple program called SMF85TJ to scan the SMF records and summarize
activity. The program itself and how to construct it is documented in “SMF record type 85
subtype 40 data display program” on page 496.
MODIFY OAM,START,RECYCLE,(ALLBK1),LIM=2,PV=50
CBR9880I OAM START RECYCLE command starting.
CBR9875I Recycle Candidates: 725
The following volumes are candidates for OAM RECYCLE command
processing using pv=50, lim=2, scope=(ALLBK1), maxrecycletasks=2.
VOLSER %VAL SGNAME STAT VOLSER %VAL SGNAME STAT
TST016 16 OBJBKP
CBR9879I OAM Recycle: End of OAM Recycle candidate volumes.
CBR9800I OAM Move Volume Recycle starting for volumes TST016 and N/A.
CBR9852I Move Volume Utility processing objects in storage group
GROUP00 for volume TST016.
IEC501A M 0B91,TST016,SL,COMP,OAM,OAM,OAM.BACKUP.DATA
IEC205I CBRRT001,OAM,OAM,FILESEQ=1, COMPLETE VOLUME LIST, 730
DSN=OAM.BACKUP.DATA,VOLS=TST017,TOTALBLOCKS=1
IEF234E K 0B90,TST017,PVT,OAM,OAM
IEC501A M 0B90,TST017,SL,COMP,OAM,OAM,OAM.BACKUP.DATA
CBR9858I Move Volume Utility status for volume TST016. Total: 6,
Attempted: 6, Successful: 6, Unsuccessful: 0.
CBR9859I Move Volume Utility ending for volumes TST016 and N/A.
CBR9881I OAM START RECYCLE command ending successfully. Reason is no
more volumes available.
IEC205I CBRRT001,OAM,OAM,FILESEQ=1, COMPLETE VOLUME LIST, 760
DSN=OAM.BACKUP.DATA,VOLS=TST017,TOTALBLOCKS=1233
IEF234E K 0B91,TST016,PVT,OAM,OAM
CBR2164I Tape volume TST016 has had all objects expired or deleted
and has been returned to OAM scratch status.
IEF234E K 0B90,TST017,PVT,OAM,OAM

Chapter 4. OAM enhancements
121
In Figure 4-27 we show the JCL to extract the SMF records and run the program.
Figure 4-27 SMF85TJ program execution JCL
In Figure 4-28 we show an example of output from the execution of program SMF85TI. This
example shows that volumes TST003 and TST004 were successfully processed by the
OSMC command migration facility.
Figure 4-28 SMF85TJ output
Note: This JCL assumes that an extract is being taken from the active SMF data set, which
is then passed to the SMFT85I program. The JCL could be changed to make a permanent
extract of the SMF data, or to read an already created SMF data extract.
//MHLRES1O JOB (999,POK),MSGLEVEL=1,NOTIFY=MHLRES1
// EXEC PGM=IFASMFDP
//SYSPRINT DD SYSOUT=*
//DUMPIN DD DISP=SHR,DSN=SYS1.SC64.MAN2
//*DUMPIN DD DISP=SHR,DSN=SMFDATA.ALLRECS.G3196V00
//OUTDD DD DSN=&SMFT85,
// SPACE=(CYL,(10,5)),
// RECFM=VB,LRECL=5096,
// DISP=(,PASS,DELETE),
// UNIT=SYSDA
//SYSIN DD *
INDD(DUMPIN,OPTIONS(DUMP))
OUTDD(OUTDD,TYPE(85(40)))
/*
// EXEC PGM=SMF85TJ
//STEPLIB DD DISP=SHR,DSN=MHLRES1.SMF85TJ.LOAD
//SYSUDUMP DD SYSOUT=A
//SMFIN DD DISP=SHR,DCB=BFTEK=A,
// DSN=&SMFT85
//PRINT DD SYSOUT=A,RECFM=UA
SMF TYPE 85 SUBTYPE 40 RECORDS
STRD/ENDD/VOLN/PCTV/LIM: 2007-02-28 2007-02-28 00000000002 00000000010 00000000003
VSN:TST003
VSN:TST004

122
z/OS V1R8 DFSMS Technical Update
In Figure 4-29 we show the SMF records that relate to this output.
Figure 4-29 SMF record type 85 subtype 40 significant fields (extract from CBRSMF macro)
More SMF
OAM also writes SMF record type 85 subtype 32/33/34/35 to document the use of this facility.
For MOVEVOL (subtype 35), bit 0 in ST32FLGS is set to ON to indicate that the MOVEVOL
was invoked automatically under software control as the result of a RECYCLE.
We have written a simple program called SMF85TH to scan the SMF records and summarize
activity. The program itself and how to construct it are documented in “SMF record type 85
subtype 32-35 data display program” on page 502.
In Figure 4-30 we show the JCL to extract the SMF records and run the program.
Figure 4-30 SMF85TH program execution JCL
Note: You should not use the contents of the CBRSMF macro as presented here. When
you assemble the program, or wish to refer to the macro, it will be found in SYS1.MACLIB.
ST40 DSECT SUBTYPE 40
ST40STRD DS CL10' ' DATE RECYCLE COMMAND STARTED @L5A
ST40ENDD DS CL10' ' DATE RECYCLE COMMAND ENDED @L5A
ST40VOLN DS BL2'0' NUMBER OF VOLSERS COMPLETED @L5A
ST40PCTV DS BL2'0' PERCENT VALID USED FOR COMMAND @L5A
ST40LIM DS BL2'0' LIMIT USED FOR COMMAND @L5A
DS BL2'0' RESERVED @L5A
ST40END DS 0C END OF BASE SECTION @L5A
***********************************************************************
* SUBTYPE 40 VOLUME ARRAY SECTION *
***********************************************************************
ST40VOLD DSECT ARRAY OF VOLS COMPLETED RECYCLE @L5A
ST40VSN DS 40CL6' ' VOLSER @L5A
//MHLRES1O JOB (999,POK),MSGLEVEL=1,NOTIFY=MHLRES1
//SMFEXTR EXEC PGM=IFASMFDP
//SYSPRINT DD SYSOUT=*
//DUMPIN DD DISP=SHR,DSN=SYS1.SC64.MAN1
//OUTDD DD DSN=&SMFT85,
// SPACE=(CYL,(10,5)),
// RECFM=VB,LRECL=5096,
// DISP=(,PASS,DELETE),
// UNIT=SYSDA
//SYSIN DD *
INDD(DUMPIN,OPTIONS(DUMP))
OUTDD(OUTDD,TYPE(85(32,33,34,35)))
/*
// EXEC PGM=SMF85TH
//STEPLIB DD DISP=SHR,DSN=MHLRES1.SMF85TH.LOAD
//SYSUDUMP DD SYSOUT=A
//SMFIN DD DISP=SHR,DCB=BFTEK=A,
// DSN=&SMFT85
//PRINT DD SYSOUT=A,RECFM=UA

Chapter 4. OAM enhancements
123
If you do not want output from all the types that the program can process, change the SMF
selection statement to only include those subtypes that you do want. For example, change
OUTDD(OUTDD,TYPE(85(32,33,34,35))) to OUTDD(OUTDD,TYPE(85(35))) to only select
subtype 35.
In Figure 4-31 we show an example of output from the execution of program SMF85TH. You
can see that a RECYCLE command was processed for volume TST003 at 11:44 on day
2007059.
Figure 4-31 SMF85TH output
In Figure 4-32 on page 124 to Figure 4-36 on page 128 we show the SMF records that relate
to this output.
Note: This JCL assumes that an extract is being taken from the active SMF data set, which
is then passed to the sSMFT85I program. The JCL could be changed to make a
permanent extract of the SMF data, or to read of an already created SMF data extract.
SMF TYPE 85 SUBTYPE 32-35 RECORDS
SMFDTE/TME: 2007059 09:01:52.000
STYPE/SGN/VSN0/VSN1/MT: 32 GROUP00 (STORAGE GROUP PROCESSING)
PDWO/PDWK/PDRO/PDRK/PDDO/PDDK: 00000000000 00000000000 00000000004 00000078128 00000000004 00000078128
POWO/POWK/PORO/PORD/PODO/PODK: 00000000000 00000000000 00000000000 00000000000 00000000000 00000000000
PTWO/PTWK/PTRO/PTRD/PTDO/PTDK: 00000000004 00000078128 00000000000 00000000000 00000000000 00000000000
BOWO/BOWK/BORO/BORK/BODO/BODK: 00000000000 00000000000 00000000000 00000000000 00000000000 00000000000
BTWO/BTWK/BTRO/BTRK/BTDO/BTDK: 00000000001 00000019532 00000000000 00000000000 00000000000 00000000000
B2OWO/B2OWK/B2ORO/B2ORK/B2ODO/B2ODK: 00000000000 00000000000 00000000000 00000000000 00000000000 00000000000
B2TWO/B2TWK/B2TRO/B2TRK/B2TDO/B2TDK: 00000000000 00000000000 00000000000 00000000000 00000000000 00000000000
DTUP/DTDE/4KIN/4KDE/32KI/32KD/NCE: 00000000017 00000000000 00000000000 00000000000 00000000000 00000000000 00000000000
FLGS/NTE/RCLD/RCLK/LOBI/LOBD: 81000000 00000000000 00000000000 00000000000 00000000000 00000000004

SMFDTE/TME: 2007059 10:00:00.047
STYPE/SGN/VSN0/VSN1/MT: 32 OBJBKP (STORAGE GROUP PROCESSING)
PDWO/PDWK/PDRO/PDRK/PDDO/PDDK: 00000000000 00000000000 00000000000 00000000000 00000000000 00000000000
POWO/POWK/PORO/PORD/PODO/PODK: 00000000000 00000000000 00000000000 00000000000 00000000000 00000000000
PTWO/PTWK/PTRO/PTRD/PTDO/PTDK: 00000000000 00000000000 00000000000 00000000000 00000000000 00000000000
BOWO/BOWK/BORO/BORK/BODO/BODK: 00000000000 00000000000 00000000000 00000000000 00000000000 00000000000
BTWO/BTWK/BTRO/BTRK/BTDO/BTDK: 00000000000 00000000000 00000000000 00000000000 00000000000 00000000000
B2OWO/B2OWK/B2ORO/B2ORK/B2ODO/B2ODK: 00000000000 00000000000 00000000000 00000000000 00000000000 00000000000
B2TWO/B2TWK/B2TRO/B2TRK/B2TDO/B2TDK: 00000000000 00000000000 00000000000 00000000000 00000000000 00000000000
DTUP/DTDE/4KIN/4KDE/32KI/32KD/NCE: 00000000000 00000000000 00000000000 00000000000 00000000000 00000000000 00000000000
FLGS/NTE/RCLD/RCLK/LOBI/LOBD: 80800000 00000000000 00000000000 00000000000 00000000000 00000000000

SMFDTE/TME: 2007059 11:44:00.048
STYPE/SGN/VSN0/VSN1/MT: 35 OBJBKP TST003 N/A 06 (MOVE VOLUME (MOVEVOL) UTILITY)
PDWO/PDWK/PDRO/PDRK/PDDO/PDDK: 00000000000 00000000000 00000000000 00000000000 00000000000 00000000000
POWO/POWK/PORO/PORD/PODO/PODK: 00000000000 00000000000 00000000000 00000000000 00000000000 00000000000
PTWO/PTWK/PTRO/PTRD/PTDO/PTDK: 00000000000 00000000000 00000000000 00000000000 00000000000 00000000000
BOWO/BOWK/BORO/BORK/BODO/BODK: 00000000000 00000000000 00000000000 00000000000 00000000000 00000000000
BTWO/BTWK/BTRO/BTRK/BTDO/BTDK: 00000000006 00000039072 00000000006 00000039072 00000000000 00000000000
B2OWO/B2OWK/B2ORO/B2ORK/B2ODO/B2ODK: 00000000000 00000000000 00000000000 00000000000 00000000000 00000000000
B2TWO/B2TWK/B2TRO/B2TRK/B2TDO/B2TDK: 00000000000 00000000000 00000000000 00000000000 00000000000 00000000000
DTUP/DTDE/4KIN/4KDE/32KI/32KD/NCE: 00000000006 00000000000 00000000000 00000000000 00000000000 00000000000 00000000000
FLGS/NTE/RCLD/RCLK/LOBI/LOBD: 82000000 00000000000 00000000000 00000000000 00000000000 00000000000

124
z/OS V1R8 DFSMS Technical Update
SMF record type 85 subtype 32 is used to map subtype records 32, 33, 34, and 35.
Figure 4-32 SMF record type 85 subtype 32 significant fields (extract from CBRSMF macro) (1 of 5)
Note: You should not use the contents of the CBRSMF macro as presented here. When
you assemble the program, or wish to refer to the macro, it will be found in SYS1.MACLIB.
ST32 DSECT SUBTYPES 32 - 35
ST32SGN DS CL8' ' STORAGE GROUP NAME
ST32VSN0 DS CL6' ' VOLUME SERIAL NUMBER OF OPTICAL
* VOLUME. ONLY VALID FOR SUBTYPES
* 34 AND 35, CONTAINS BLANKS FOR
* OTHER SUBTYPES.
ST32VSN1 DS CL6' ' VOLUME SERIAL NUMBER OF OPPOSITE
* SIDE OF OPTICAL DISK
* VOLUME. ONLY VALID FOR SUBTYPES
* 34 AND 35, CONTAINS BLANKS FOR
* OTHER SUBTYPES.
ST32OMT DS CL2' ' OPTICAL MEDIA TYPE. ONLY VALID
* FOR SUBTYPES 34 AND 35, CONTAINS
* BLANKS FOR OTHER SUBTYPES.
DS CL2' ' RESERVED
***********************************************************************
* COUNTS OF PRIMARY OBJECTS (AND KILOBYTES) WRITTEN, READ AND
* DELETED FROM DASD.
***********************************************************************
ST32PDWO DS BL4'0' NUMBER OF PRIMARY OBJECTS
* WRITTEN TO DASD.
ST32PDWK DS BL4'0' NUMBER OF KILOBYTES PRIMARY
* OBJECTS WRITTEN TO DASD.
ST32PDRO DS BL4'0' NUMBER OF PRIMARY OBJECTS
* READ FROM DASD.
ST32PDRK DS BL4'0' NUMBER OF KILOBYTES PRIMARY
* OBJECTS READ FROM DASD.
ST32PDDO DS BL4'0' NUMBER OF PRIMARY OBJECTS
* DELETED FROM DASD.
ST32PDDK DS BL4'0' NUMBER OF KILOBYTES OF PRIMARY
* OBJECTS DELETED FROM DASD.
***********************************************************************
* COUNTS OF PRIMARY OBJECTS (AND KILOBYTES) WRITTEN, READ AND
* DELETED FROM OPTICAL.
***********************************************************************
ST32POWO DS BL4'0' NUMBER OF PRIMARY OBJECTS
* WRITTEN TO OPTICAL.
ST32POWK DS BL4'0' NUMBER OF KILOBYTES OF PRIMARY
* OBJECTS WRITTEN TO OPTICAL.
ST32PORO DS BL4'0' NUMBER OF PRIMARY OBJECTS
* READ FROM OPTICAL.
ST32PORK DS BL4'0' NUMBER OF KILOBYTES OF PRIMARY
* OBJECTS READ FROM OPTICAL.
ST32PODO DS BL4'0' NUMBER OF PRIMARY OBJECTS
* DELETED FROM OPTICAL.
ST32PODK DS BL4'0' NUMBER OF KILOBYTES OF PRIMARY
* OBJECTS DELETED FROM OPTICAL.

Chapter 4. OAM enhancements
125
Figure 4-33 SMF record type 85 subtype 32 significant fields (extract from CBRSMF macro) (2 of 5)
***********************************************************************
* COUNTS OF PRIMARY OBJECTS (AND KILOBYTES) WRITTEN, READ AND
* DELETED FROM TAPE.
***********************************************************************
ST32PTWO DS BL4'0' NUMBER OF PRIMARY OBJECTS
* WRITTEN TO TAPE.
ST32PTWK DS BL4'0' NUMBER OF KILOBYTES OF PRIMARY
* OBJECTS WRITTEN TO TAPE.
ST32PTRO DS BL4'0' NUMBER OF PRIMARY OBJECTS
* READ FROM TAPE.
ST32PTRK DS BL4'0' NUMBER OF KILOBYTES OF PRIMARY
* OBJECTS READ FROM TAPE.
ST32PTDO DS BL4'0' NUMBER OF PRIMARY OBJECTS
* LOGICALLY DELETED FROM TAPE.
ST32PTDK DS BL4'0' NUMBER OF KILOBYTES OF PRIMARY
* OBJECTS LOGICALLY DELETED FROM
* TAPE.
***********************************************************************
* COUNTS OF BACKUP OBJECTS (AND KILOBYTES) WRITTEN, READ AND
* DELETED FROM OPTICAL.
***********************************************************************
ST32BOWO DS BL4'0' NUMBER OF BACKUP OBJECTS
* WRITTEN TO OPTICAL.
ST32BOWK DS BL4'0' NUMBER OF KILOBYTES OF BACKUP
* OBJECTS WRITTEN TO OPTICAL.
ST32BORO DS BL4'0' NUMBER OF BACKUP OBJECTS
* READ FROM OPTICAL.
ST32BORK DS BL4'0' NUMBER OF KILOBYTES OF BACKUP
* OBJECTS READ FROM OPTICAL.
ST32BODO DS BL4'0' NUMBER OF BACKUP OBJECTS
* DELETED FROM OPTICAL.
ST32BODK DS BL4'0' NUMBER OF KILOBYTES OF BACKUP
* OBJECTS DELETED FROM OPTICAL.
***********************************************************************
* COUNTS OF BACKUP OBJECTS (AND KILOBYTES) WRITTEN, READ AND
* DELETED FROM TAPE.
***********************************************************************
ST32BTWO DS BL4'0' NUMBER OF BACKUP OBJECTS
* WRITTEN TO TAPE.
ST32BTWK DS BL4'0' NUMBER OF KILOBYTES OF BACKUP
* OBJECTS WRITTEN TO TAPE.
ST32BTRO DS BL4'0' NUMBER OF BACKUP OBJECTS
* READ FROM TAPE.
ST32BTRK DS BL4'0' NUMBER OF KILOBYTES OF BACKUP
* OBJECTS READ FROM TAPE.
ST32BTDO DS BL4'0' NUMBER OF BACKUP OBJECTS
* LOGICALLY DELETED FROM TAPE.
ST32BTDK DS BL4'0' NUMBER OF KILOBYTES OF BACKUP
* OBJECTS LOGICALLY DELETED FROM
* TAPE.

126
z/OS V1R8 DFSMS Technical Update
Figure 4-34 SMF record type 85 subtype 32 significant fields (extract from CBRSMF macro) (3 of 5)
***********************************************************************
* COUNTS OF BACKUP2 OBJECTS (AND KILOBYTES) WRITTEN, READ AND @L2A
* DELETED FROM OPTICAL. @L2A
***********************************************************************
ST32B2OWO DS BL4'0' NUMBER OF BACKUP2 OBJECTS @L2A
* WRITTEN TO OPTICAL. @L2A
ST32B2OWK DS BL4'0' NUMBER OF KILOBYTES OF BACKUP2 @L2A
* OBJECTS WRITTEN TO OPTICAL. @L2A
ST32B2ORO DS BL4'0' NUMBER OF BACKUP2 OBJECTS @L2A
* READ FROM OPTICAL. @L2A
ST32B2ORK DS BL4'0' NUMBER OF KILOBYTES OF BACKUP2 @L2A
* OBJECTS READ FROM OPTICAL. @L2A
ST32B2ODO DS BL4'0' NUMBER OF BACKUP2 OBJECTS @L2A
* DELETED FROM OPTICAL. @L2A
ST32B2ODK DS BL4'0' NUMBER OF KILOBYTES OF BACKUP2 @L2A
* OBJECTS DELETED FROM OPTICAL. @L2A
***********************************************************************
* COUNTS OF BACKUP2 OBJECTS (AND KILOBYTES) WRITTEN, READ AND @L2A
* DELETED FROM TAPE. @L2A
***********************************************************************
ST32B2TWO DS BL4'0' NUMBER OF BACKUP2 OBJECTS @L2A
* WRITTEN TO TAPE. @L2A
ST32B2TWK DS BL4'0' NUMBER OF KILOBYTES OF BACKUP2 @L2A
* OBJECTS WRITTEN TO TAPE. @L2A
ST32B2TRO DS BL4'0' NUMBER OF BACKUP2 OBJECTS @L2A
* READ FROM TAPE. @L2A
ST32B2TRK DS BL4'0' NUMBER OF KILOBYTES OF BACKUP2 @L2A
* OBJECTS READ FROM TAPE. @L2A
ST32B2TDO DS BL4'0' NUMBER OF BACKUP2 OBJECTS @L2A
* LOGICALLY DELETED FROM TAPE. @L2A
ST32B2TDK DS BL4'0' NUMBER OF KILOBYTES OF BACKUP2 @L2A
* OBJECTS LOGICALLY DELETED FROM @L2A
* TAPE. @L2A

Chapter 4. OAM enhancements
127
Figure 4-35 SMF record type 85 subtype 32 significant fields (extract from CBRSMF macro) (4 of 5)
***********************************************************************
* COUNTS OF ACTIVITY AGAINST THE OBJECT STORAGE DATABASE
* (OBJECT DIRECTORY TABLE, 4K OBJECT STORAGE TABLE AND 32K
* OBJECT STORAGE TABLE).
***********************************************************************
ST32DTUP DS BL4'0' NUMBER OF ROWS UPDATED IN THE
* OBJECT DIRECTORY TABLE.
ST32DTDE DS BL4'0' NUMBER OF ROWS DELETED FROM THE
* OBJECT DIRECTORY TABLE.
ST324KIN DS BL4'0' NUMBER OF ROWS INSERTED INTO THE
* 4K OBJECT STORAGE TABLE.
ST324KDE DS BL4'0' NUMBER OF ROWS DELETED FROM THE
* 4K OBJECT STORAGE TABLE.
ST3232KI DS BL4'0' NUMBER OF ROWS INSERTED INTO THE
* 32K OBJECT STORAGE TABLE.
ST3232KD DS BL4'0' NUMBER OF ROWS DELETED FROM THE
* 32K OBJECT STORAGE TABLE.
ST32NCE DS BL4'0' NUMBER OF OPTICAL CARTRIDGES
* EXPIRED. VALID ONLY FOR
* SUBTYPE 32.
ST32FLGS DS BL4'0' PROCESSING FLAGS
ST32FLG0 EQU X'80' WHEN ON, THIS PROCESS WAS
* INVOKED AUTOMATICALLY UNDER
* SOFTWARE CONTROL.
ST32FLG1 EQU X'40' WHEN ON, THIS PROCESS WAS
* INVOKED BY A MODIFY OAM,START
* COMMAND.
ST32FLG2 EQU X'20' WHEN ON, THIS PROCESS WAS
* INVOKED USING AN ISMF LINE
* OPERATOR.
ST32FLG3 EQU X'10' WHEN ON, VOL RECOVERY WAS
* INVOKED W/ BACKUP1 KEYWORD OR
* DEFAULTED TO BACKUP1 @L2A
ST32FLG4 EQU X'08' WHEN ON, VOL RECOVERY WAS
* INVOKED W/ BACKUP2 KEYWORD @L2A
ST32FLG5 EQU X'04' WHEN ON, VOL RECOVERY OR MOVEVOL WAS
* SPECIFIED WITH DELETE OPTION @L3A
ST32FLG6 EQU X'02' WHEN ON, VOL RECOVERY OR MOVEVOL WAS
* SPECIFIED WITH RECYCLE OPTION @L3A
ST32FLG7 EQU X'01' WHEN ON, INDICATED PROCESSING OBJECT
* STORAGE GROUP @L3A
ST32FLG8 EQU X'80' WHEN ON, INDICATED PROCESSING BACKUP
* OBJECT STORAGE GROUP @L3A
ST32FLG9 EQU X'40' WHEN ON, CYCLE ENDTIME EXCEEDED @L3A
*

128
z/OS V1R8 DFSMS Technical Update
Figure 4-36 SMF record type 85 subtype 32 significant fields (extract from CBRSMF macro) (5 of 5)
Migration and coexistence
Pre-V1R8 systems in the same OAMplex as a V1R8 system must have the fix for APAR
OA12683 applied.
Maintenance
The fix for APAR OA17310 must be applied prior to using the OAM RECYCLE function. It
fixes an ABEND0C4, a message problem when there are no more volumes to be recycled,
and a problem with SMF record type 85 missing new fields.
4.4 GLOBAL display keyword
A GLOBAL keyword has been added to the MODIFY OAM,DISPLAY command.
4.4.1 Usage
You can use the MODIFY OAM,DISPLAY command to display the current settings of the
SETOAM, SETOPT, or SETOSMC statements for the OAM address space.
The syntax of this command, from the DFSMS OAM Planning, Installation, and Storage
Administration Guide for Object Support, SC26-0426, publication, is as follows:
MODIFY OAM,DISPLAY,keyword1,keyword2
keyword1 must be one of the options shown in Figure 4-37.
Figure 4-37 The options for keyword1
ST32NTE DS BL4'0' NUMBER OF TAPE VOLUMES EXPIRED
* VALID ONLY FOR SUBTYPE 32 @L3A
ST32RCLD DS BL4'0' NUMBER OF RECALLED OBJECTS
* VALID ONLY FOR SUBTYPE 32 @00A
ST32RCLK DS BL4'0' NUMBER OF KILOBYTES OF
* RECALLED OBJECTS
* VALID ONLY FOR SUBTYPE 32 @00A
ST32LOBI DS BL4'0' NUMBER OF LOB ROWS INSERTED
* VALID ONLY FOR SUBTYPE 32 @00A
ST32LOBD DS BL4'0' NUMBER OF LOB ROWS DELETED
* VALID ONLY FOR SUBTYPE 32 @00A
SETOPT | SETOAM | SETOSMC
Specifies the command parameter being displayed. Use SETOAM to display
values of settings in an object tape environment. SETOPT is used to
display values of settings in an optical environment. Use SETOSMC to
display the values of settings for OSMC processing.

Chapter 4. OAM enhancements
129
keyword2 must be one of the options shown in Figure 4-38.
Figure 4-38 The new GLOBAL keyword2
Figure 4-39 is an example of the new GLOBAL keyword in action when displaying SETOSMC
settings.
Figure 4-39 Displaying SETOSMC settings with the GLOBAL keyword
Migration and coexistence
There are no migration or coexistence considerations for use of this new keyword.
4.5 Update from z/OS V1.7
z/OS V1R7 DFSMS Technical Update, SG24-7225, contains a description of various OAM
enhancements. One in particular relates to OSMC SINGLE OBJECT RECALL UTILITY.
There has been no change to that in z/OS V1R8, but the function was retested because an
SMF MACRO error was found in z/OS V1R7.
The SMF error is still present in z/OS V1R8, but it now has an APAR number OA20170
(“OA20170” on page 464).
ALL|GLOBAL|storgrp
Specifies the kind of information that the system is to display. For
the SETOAM, SETOSMC, and SETOPT parameters, the valid values are as
follows:

ALL
Displays the settings for each valid storage group as well as the
global default settings. If ALL is specified, the global default,
if applicable, is displayed as well as the settings for each
valid storage group in the active SMS configuration.

GLOBAL
Displays only the OAM global keywords.

storgrp
Displays only the settings for the specified storage group
name. There can be up to fourteen storage group names
indicated on a single DISPLAY command.

Displays the settings for the FIRSTBACKUPGROUP and
SECONDBACKUPGROUP for the specified storage group name.
F OAM,DISPLAY,SETOSMC,GLOBAL
CBR1075I GLOBAL value for BACKUP1 is OBJBKP
CBR1075I GLOBAL value for BACKUP2 is -NONE-
CBR1075I GLOBAL value for CYCLEW is STRTONLY
CBR1075I GLOBAL value for MAXRECAL is 2
CBR1075I GLOBAL value for RECALLO is 15, ON
CBR1075I GLOBAL value for RECALLT is 0, ON
CBR1075I GLOBAL value for CLEAROLD is TAPE

130
z/OS V1R8 DFSMS Technical Update
In z/OS V1R7 DFSMS OAM introduced the ability to recall objects from tape and store them
in DB2 to improve performance for subsequent retrievals.
OAM writes SMF record type 85 subtype 38 to document the use of this facility.
We have written a simple program called SMF85TO to scan the SMF records and summarize
activity. This was originally made available in z/OS V1R7 DFSMS Technical Update,
SG24-7225. The example provided here is smaller but has the same function.
The program itself and how to construct it is documented in “SMF record type 85 subtype 38
data display program” on page 484.
In Figure 4-40 we show the JCL to extract the SMF records and run the program.
Figure 4-40 SMF85TO program execution JCL
In Figure 4-41 we show an example of output from the execution of program SMF85TO. In
Figure 4-42 on page 131 we show the SMF records that relate to this output.
Figure 4-41 SMF85TO output
Note: This JCL assumes that an extract is being taken from the active SMF data set, which
is then passed to the SMFT85O program. The JCL could be changed to make a
permanent extract of the SMF data, or to read an already created SMF data extract.
//MHLRES1O JOB (999,POK),MSGLEVEL=1,NOTIFY=MHLRES1
// EXEC PGM=IFASMFDP
//SYSPRINT DD SYSOUT=*
//DUMPIN DD DISP=SHR,DSN=SYS1.SC64.MAN1
//OUTDD DD DSN=&SMFT85,
// SPACE=(CYL,(10,5)),
// RECFM=VB,LRECL=5096,
// DISP=(,PASS,DELETE),
// UNIT=SYSDA
//SYSIN DD *
INDD(DUMPIN,OPTIONS(DUMP))
OUTDD(OUTDD,TYPE(85(38)))
/*
// EXEC PGM=SMF85TO
//STEPLIB DD DISP=SHR,DSN=MHLRES1.SMF85TO.LOAD
//SYSUDUMP DD SYSOUT=A
//SMFIN DD DISP=SHR,DCB=BFTEK=A,
// DSN=&SMFT85
//PRINT DD SYSOUT=A,RECFM=UA
SMF TYPE 85 SUBTYPE 38 RECORDS
COLN/CNID: OAMTEST.MAR210E 00000000046
OBJN/SGN/OLEN: OAMTEST.OBJ0210E GROUP00 00000050000
VSN/MT/TKN/VT/BT/FLGS:TST000 06 00000001242 FLG0 ON

Chapter 4. OAM enhancements
131
The left-hand side of the report contains abbreviations from the SMF records. Record type 85
subtype 38 contains the fields of interest to verify the function of OAM object recall to DB2. In
the SMF records, as shown in Figure 4-42, the fields are prefixed with ST38. In the report the
fields are shown without the ST38, and have been compressed to have more than one entry
per line.
In this example, Collection OAMTEST.MAR210E, which contains object
OMATEST.OBJ0210E, was recalled. Object OAMTEST.OBJ0210E is 50000 long. The flags
field shown in the fourth line as been interpreted from the SMF field ST38FLG0 as being ON,
which indicates that the recall was successful.
Figure 4-42 SMF record type 85 subtype 38 significant fields (extract from CBRSMF macro)
Note: You should not use the contents of the CBRSMF macro as presented here. When
you assemble the program, or wish to refer to the macro, it will be found in SYS1.MACLIB.
ST38 DSECT SUBTYPE 38 @P2A
ST38COLN DS CL44' ' COLLECTION NAME @P2A
ST38CNID DS BL4'0' COLLECTION ID @P2A
ST38OBJN DS CL44' ' OBJECT NAME @P2A
ST38SGN DS CL8' ' STORAGE GROUP NAME @P2A
ST38OLEN DS BL4'0' OBJECT LENGTH @P2A
ST38VSN DS CL6' ' VOLUME SERIAL NUMBER OF OPTICAL
* VOLUME OR TAPE VOLUME FROM WHICH
* THE COPY OF THE OBJECT
* WAS READ. @P2A
ST38MT DS CL2' ' MEDIA TYPE OF THE VOLUME FROM
* WHICH THE COPY OF THE OBJECT
* WAS READ. @P2A
ST38TKN DS BL4'0' VOLUME LOCATION TOKEN ASSOCIATED
* WITH THE COPY OF THE OBJECT
* ON THE VOLUME SPECIFIED IN THE
* ST39VSN FIELD. @P2A
ST38VT DS CL1' ' VOLUME TYPE @P2A
ST38BT DS CL1' ' BACKUP TYPE @P2A
DS CL2' ' RESERVED @P2A
ST38FLGS DS BL4'0' PROCESSING FLAGS @P2A
*
ST38FLG0 EQU X'80' WHEN ON, OBJECT RECALL
* WAS SUCCESSFUL. @P2A

132
z/OS V1R8 DFSMS Technical Update

© Copyright IBM Corp. 2008. All rights reserved.
133
Chapter 5.
DFSMSdss enhancements
In z/OZ V1R8 a number of enhancements have been incorporated into DFSMSdss. The
majority of the enhancements are used by DFSMShsm in support of Fast Replicate.
Users should refer to DFSMShsm for information concerning fast replication.
The interface to Physical Data Set COPY and Physical Data Set DUMP/RESTORE is
available for users of DFSMSdss without using DFSMShsm. This document concentrates on
this use of DFSMSdss.
DFSMSdss continues to support logical data set operations.
5

134
z/OS V1R8 DFSMS Technical Update
5.1 Defining logical and physical processing
The following information is based on Chapter 3 of DFSMSdss Storage Administration Guide,
SC26-0423, as it provides a background to the new physical processing in z/OS V1R8.
DFSMSdss can perform two kinds of processing when executing COPY, DUMP, and
RESTORE commands:

Logical processing
operates against data sets independently of physical device format.

Physical processing
moves data at the track-image level and operates against volumes,
tracks, and data sets.
Each type of processing offers different capabilities and advantages.
During a restore operation, the data is processed the same way in which it is dumped
because physical and logical dump tapes have different formats. If a data set is dumped
logically, it is restored logically. If it is dumped physically, it is restored physically. A data set
restore operation from a full-volume dump is a physical data set restore operation.
Logical processing
A logical copy, dump, or restore operation treats each data set and its associated information
as a logical entity, and processes an entire data set before beginning the next one.
Each data set is moved by tracks from the source device and is potentially written to the
target device as a set of data records, allowing data movement between devices with
different track and cylinder configurations. Checking of data record consistency is not
performed during dump operations.
DFSMSdss performs logical processing if:
You specify the DATASET keyword with the COPY command. A data set copy is always a
logical operation regardless of how or whether you specify input volumes.
You specify the DATASET keyword with the DUMP command, and either no input volume
is specified, or LOGINDDNAME, LOGINDYNAM, or STORGRP is used to specify input
volumes.
The RESTORE command is performed, and the input volume was created by a logical
dump.
DFSMSdss uses catalogs or VTOCs to select data sets for logical processing. If you do not
specify input volumes, DFSMSdss uses the catalogs to select data sets for copy and dump
operations. If you specify input volumes using the LOGINDDNAME, LOGINDYNAM, or
STORGRP keywords on the COPY or DUMP command, DFSMSdss uses VTOCs to select
data sets for processing.
When to use logical processing
Use logical processing for the following situations:
Data is copied to an unlike device.
Logical processing is the only way to move data between unlike device types.
Note: To copy or dump entire multivolume data sets, you do not need to specify all the
volumes in the LOGINDDNAME or LOGINDYNAM volume list. However, you must specify
the SELECTMULTI keyword with either the FIRST or ANY subkeywords.

Chapter 5. DFSMSdss enhancements
135
Data that may need to be restored to an unlike device is dumped.
Data must be restored the same way in which it is dumped. This is particularly important to
keep in mind when making backups that you plan to retain for a long period of time (such
as vital records backups). If a backup is retained for a long period of time, it is possible
that the device type it originally resided on will no longer be in use at your site when you
want to restore it. This means that you have to restore it to an unlike device, which can
only be done if the backup was made logically.
Aliases of VSAM user catalogs are to be preserved during the copy and restore functions.
Aliases are not preserved for physical processing.
Unmovable data sets or data sets with absolute track allocation are moved to different
locations.
Multivolume data sets are processed.
VSAM and multivolume data sets are cataloged as part of DFSMSdss processing.
Data sets are deleted from the source volume after a successful copy or restore operation.
Non-VSAM and VSAM data sets are renamed after a successful copy or restore
operation.
You want to control the percentage of space allocated on each of the output volumes for
copy or restore operations.
You want to copy and convert a PDS to PDSE or vice versa.
You want to copy or restore a data set with an undefined DSORG to an unlike device.
You want to keep together all parts of a VSAM sphere.
Physical processing
Physical processing moves data based on physical track images. Because data movement is
carried out at the track level, only target devices with track sizes equal to those of the source
device are supported. Physical processing operates on volumes, ranges of tracks, or data
sets. For data sets, it relies on volume information (in the VTOC and VVDS) for data set
selection, and processes only that part of a data set residing on the specified input volumes.
DFSMSdss performs physical processing when the following conditions exist:
You specify the FULL or TRACKS keyword with the COPY or DUMP command.
This results in a physical volume or physical tracks operation.
Notes: VSAM data sets are
not
cataloged during physical processing within SMS or
non-SMS environments. The CATALOG keyword is ignored for VSAM data set during
physical restore. Use IDCAMS DEFINE RECATALOG to catalog the data sets after the
restore.
The RENAME and RENAMEUNCONDITIONAL keywords are ignored for VSAM data sets
during physical restore.
Attention: Take care when invoking the TRACKS keyword with the COPY and
RESTORE keywords. The TRACKS keyword should only be used for a data set
recovery operation. For example, you can use it to repair a bad track in the VTOC or a
data set, or to retrieve data from a damaged data set. You cannot use it in place of a
full-volume or a logical data set operation. Doing so could destroy or impair data
integrity.

136
z/OS V1R8 DFSMS Technical Update
You specify the DATASET keyword on the COPY or DUMP command and input volumes
with the PHYSINDDNAME or PHYSINDYNAM keyword. This produces a physical data
set copy or physical data set dump.
The RESTORE command is executed and the input volume is created by a physical dump
operation.
When to use physical processing
Use physical processing when the following conditions exist:
Backing up system volumes that you might want to restore with a stand-alone DFSMSdss
restore operation.
Stand-Alone DFSMSdss restore only supports physical dump tapes.
Performance is an issue.
Generally, the fastest way (measured by elapsed time) to copy or dump an entire volume
is with a physical full-volume command. This is primarily because minimal catalog
searching is necessary for physical processing.
Substituting one physical volume for another or recovering an entire volume.
With a COPY or RESTORE (full-volume or track) command, the volume serial number of
the input DASD volume can be copied to the output DASD volume.
Dealing with I/O errors. Physical processing provides the capability to copy, dump, and
restore a specific track or range of tracks.
Dumping or copying between volumes of the same device type but different capacity.
5.2 Physical and logical data set difference summary
The form that data sets are used by application programs in is what DFSMSdss refers to as
the logical form. The data set physical structure and location are provided to the application
through the system catalog (usually) and supplementary information from the DASD VTOC
and VVDS entries for the data set. For single volume non-VSAM data sets there is not much
further information to be obtained, but when the data set becomes multi-volume and has a
more complicated structure such as VSAM and may be multi-volume it is necessary to use
these sources of data for effective use of the data set. There is a trade off, however, in that
obtaining all that information takes time.
For utility programs such as DFSMSdss, whether used in its own right or as support for
DFSMShsm, for example, there may be situations where a copy of a data set may be able to
be made more rapidly by using a physical process. This is where the DFSMSdss Physical
COPY can be used.
5.3 Physical and logical data set copy specification changes
z/OS V1R8 DFSMSdss command specifications have been updated to better emphasize the
difference between logical and physical processing.
The INDDNAME/INDYNAM keyword is replaced by LOGINDDNAME/LOGINDYNAM and
PHYSINDDNAME/PHYSINDDNAM. LOGINDDNAME/LOGINDYNAM was available prior to
z/OS V1R8.

Chapter 5. DFSMSdss enhancements
137
Use of the INDDNAME/INDYNAM keywords is accepted, but they are internally replaced by
PHYSINDDNAME/PHYSINDDNAM. Message ADR146I is issued to advise of this once
APAR OA20242 has been implemented (as is the case for Physical DUMP). See “OA20242”
on page 465 for details.
Physical DUMP operation changes
Although there are code changes in the Physical DUMP part of DFSMSdss to support
Physical COPY, to the user there is minimal change. New keywords PHYSINDDNAME and
PHYSINDYNAM, if used, make it more obvious that the operations relate to physical
operations.
The main keyword difference is the change from the use of the INDDNAME and INDYNAM
keywords, which in the past were used in conjunction with the DUMP command to specify
physical copies. The new keywords PHYSINDDNAME and PHYSINDYNAM to replace
INDDNAME and INDYNAM make it more obvious that the operations relate to physical
operations.
You can continue to use existing DUMP jobs using INDDNAME or INDYNAM, and will
receive a warning message ADR146I advising that the obsolete keyword has been replaced.
In Figure 5-1 we show use of the INDD keyword (short form of INDDNAME) to specify the
source volume.
Figure 5-1 DFSMSdss DUMP job using INDD
Note: There is no prior system toleration for the new PHYSINDDNAME or
PHYSINDYNAM keywords, so jobs should not be changed to use these new keywords
until all systems are converted to z/OS V1R8 or later.
//MHLRES1D JOB (999,POK),MSGLEVEL=1,NOTIFY=MHLRES1
//STEP01 EXEC PGM=ADRDSSU,TIME=1440,REGION=6000K,
// PARM='UTILMSG=YES'
//SYSPRINT DD SYSOUT=*
//SBOX38 DD DISP=SHR,UNIT=SYSDA,DSN=MHLRES1.JOBS.JCL
//TEMP DD DISP=(,PASS),UNIT=SYSDA,SPACE=(CYL,(1,1))
//SYSIN DD *
DUMP DATASET(INCLUDE(-
MHLRES1.JOBS.JCL -
)) -
SHARE -
INDD(SBOX38) OUTDD(TEMP)

138
z/OS V1R8 DFSMS Technical Update
In Figure 5-2 we show the output from the job shown in Figure 5-1 on page 137, where
DFSMSdss has substituted the keyword PHYSINDDNAME for INDD.
Figure 5-2 DFSMSdss DUMP job output showing message about substitution of PHYSINDDNAME for INDD
5.3.1 Logical copy operations
To make a logical copy of a data set, the COPY command and keywords DATASET and
OUTDD are required, and they can be used together with many optional modifiers.
In Figure 5-3 we show the JCL to copy a data set using the system catalog to locate the
volume that the data set is on. The example demonstrates the input data set lookup with as
few other operands as possible, so the data set is not renamed. As a result, because this is a
COPY operation, the copy fails. This situation differs from the case when Physical COPY is
used (see 5.3.3, “Physical copy operations” on page 140).
Figure 5-3 Logical COPY without LOGINDDNAME
PAGE 0001 5695-DF175 DFSMSDSS V1R08.0 DATA SET SERVICES 2007.065 11:40
DUMP DATASET(INCLUDE(-
MHLRES1.JOBS.JCL -
)) -
SHARE -
INDD(SBOX38) OUTDD(TEMP)
ADR101I (R/I)-RI01 (01), TASKID 001 HAS BEEN ASSIGNED TO COMMAND 'DUMP '
ADR109I (R/I)-RI01 (01), 2007.065 11:40:49 INITIAL SCAN OF USER CONTROL STATEMENTS COMPLETED.
ADR146I (R/I)-RI03 (13), OBSOLETE KEYWORD 'INDDNAME ' SPECIFIED. 'PHYSINDDNAME ' WILL BE USED.
ADR016I (001)-PRIME(01), RACF LOGGING OPTION IN EFFECT FOR THIS TASK
ADR006I (001)-STEND(01), 2007.065 11:40:49 EXECUTION BEGINS
ADR378I (001)-DTDS (01), THE FOLLOWING DATA SETS WERE SUCCESSFULLY PROCESSED FROM VOLUME SBOX38
MHLRES1.JOBS.JCL
ADR006I (001)-STEND(02), 2007.065 11:40:50 EXECUTION ENDS
ADR013I (001)-CLTSK(01), 2007.065 11:40:50 TASK COMPLETED WITH RETURN CODE 0000
ADR012I (SCH)-DSSU (01), 2007.065 11:40:50 DFSMSDSS PROCESSING COMPLETE. HIGHEST RETURN CODE IS 0000
//MHLRES1D JOB (999,POK),MSGLEVEL=1,NOTIFY=MHLRES1
//STEP01 EXEC PGM=ADRDSSU,TIME=1440,REGION=6000K,
// PARM='UTILMSG=YES'
//SYSPRINT DD SYSOUT=*
//TEMP DD DISP=(,PASS),UNIT=SYSDA,SPACE=(CYL,(1,1))
//SYSIN DD *
COPY DATASET(INCLUDE(-
MHLRES1.JOBS.JCL -
)) -
SHARE -
OUTDD(TEMP)

Chapter 5. DFSMSdss enhancements
139
In Figure 5-4 we show the output from the job selecting data set by just using the system
catalog. As predicted, the copy was not allowed to proceed because no rename keyword was
specified.
Figure 5-4 Logical COPY output without LOGINDDNAME
If there is a reason to locate a data set other than through the system catalog, for example,
because there is a copy on another volume, the LOGINDDNAME keyword can be specified. If
the DDNAME specified on LOGINDDNAME refers to the volume that the requested data set
is on, the copy is able to proceed.
If, however, the volume referred to was not where the requested data set was, the copy would
not be able to proceed.
In Figure 5-5 we show a sample job to make a copy of a data set, illustrating the use of the
LOGINDDNAME keyword, but the DDNAME referred to does not contain the data set to be
copied, so the job should fail.
Figure 5-5 Logical COPY with LOGINDDNAME referring to the wrong volume
PAGE 0001 5695-DF175 DFSMSDSS V1R08.0 DATA SET SERVICES 2007.066 15:08
COPY DATASET(INCLUDE(-
MHLRES1.JOBS.JCL -
)) -
SHARE -
OUTDD(TEMP)
ADR101I (R/I)-RI01 (01), TASKID 001 HAS BEEN ASSIGNED TO COMMAND 'COPY '
ADR109I (R/I)-RI01 (01), 2007.066 15:08:13 INITIAL SCAN OF USER CONTROL STATEMENTS COMPLETED.
ADR016I (001)-PRIME(01), RACF LOGGING OPTION IN EFFECT FOR THIS TASK
ADR006I (001)-STEND(01), 2007.066 15:08:13 EXECUTION BEGINS
ADR713E (001)-ALLOC(01), UNABLE TO ALLOCATE SMS MANAGED DATA SET MHLRES1.JOBS.JCL BECAUSE NEITHER DELETE NOR RENAMEU WAS SPE
ADR801I (001)-DDDS (01), DATA SET FILTERING IS COMPLETE. 1 OF 1 DATA SETS WERE SELECTED: 0 FAILED SERIALIZATION AND 0 FAILED
OTHER REASONS.
ADR455W (001)-DDDS (02), THE FOLLOWING DATA SETS WERE NOT SUCCESSFULLY PROCESSED
MHLRES1.JOBS.JCL
ADR006I (001)-STEND(02), 2007.066 15:08:14 EXECUTION ENDS
ADR013I (001)-CLTSK(01), 2007.066 15:08:14 TASK COMPLETED WITH RETURN CODE 0008
ADR012I (SCH)-DSSU (01), 2007.066 15:08:14 DFSMSDSS PROCESSING COMPLETE. HIGHEST RETURN CODE IS 0008 FROM:
TASK 001
//MHLRES1D JOB (999,POK),MSGLEVEL=1,NOTIFY=MHLRES1
//STEP01 EXEC PGM=ADRDSSU,TIME=1440,REGION=6000K,
// PARM='UTILMSG=YES'
//SYSPRINT DD SYSOUT=*
//SBOXA8 DD DISP=SHR,UNIT=SYSDA,VOL=SER=SBOXA8
//TEMP DD DISP=(,PASS),UNIT=SYSDA,SPACE=(CYL,(1,1))
//SYSIN DD *
COPY DATASET(INCLUDE(-
MHLRES1.JOBS.JCL -
)) -
SHARE -
LOGINDDNAME(SBOXA8) OUTDD(TEMP)

140
z/OS V1R8 DFSMS Technical Update
In Figure 5-6 we show the output from running the example in Figure 5-5 on page 139.
The volume referred to by LOGINDDNAME(SBOXA8) in the JCL DDNAME(SBOXA8) does
not contain data set MHLRES1.JOBS.JCL, so the job fails.
Figure 5-6 Logical COPY output with LOGINDDNAME referring to the wrong volume
5.3.2 DSS LOGICAL COPY warning
A VSAM KSDS with data and index components on different volumes may lose its SMS
RLSDATA attribute during DSS COPY. Refer to “OA18319” on page 462, for further
information.
5.3.3 Physical copy operations
To make a physical copy of a data set, the COPY command and keywords DATASET,
OUTDDNAME, and PHYSINDDNAME or PHYSINDYNAM are required, which can be used
together with many optional modifiers. The significant difference over a logical COPY is the
presence of the PHYSINDDNAME or PHYSINDYNAM keyword. In this respect the syntax is
like the version using LOGINDDNAME or LOGINDYNAM.
When using a physical COPY, a data set that is SMS managed can be copied. This is not
allowed with a logical copy, as demonstrated in Figure 5-4 on page 139.
The user must then adjust the catalog to refer to the new data set if that is what is intended.
PAGE 0001 5695-DF175 DFSMSDSS V1R08.0 DATA SET SERVICES 2007.066 15:15
COPY DATASET(INCLUDE(-
MHLRES1.JOBS.JCL -
)) -
SHARE -
LOGINDDNAME(SBOXA8) OUTDD(TEMP)
ADR101I (R/I)-RI01 (01), TASKID 001 HAS BEEN ASSIGNED TO COMMAND 'COPY '
ADR109I (R/I)-RI01 (01), 2007.066 15:15:53 INITIAL SCAN OF USER CONTROL STATEMENTS COMPLETED.
ADR016I (001)-PRIME(01), RACF LOGGING OPTION IN EFFECT FOR THIS TASK
ADR006I (001)-STEND(01), 2007.066 15:15:53 EXECUTION BEGINS
ADR383W (001)-DDDS (01), DATA SET MHLRES1.JOBS.JCL NOT SELECTED
ADR455W (001)-DDDS (03), THE FOLLOWING DATA SETS WERE NOT SUCCESSFULLY PROCESSED
MHLRES1.JOBS.JCL
ADR470W (001)-DDDS (04), NO DATA SETS SELECTED FOR PROCESSING
ADR006I (001)-STEND(02), 2007.066 15:15:54 EXECUTION ENDS
ADR013I (001)-CLTSK(01), 2007.066 15:15:54 TASK COMPLETED WITH RETURN CODE 0004
ADR012I (SCH)-DSSU (01), 2007.066 15:15:54 DFSMSDSS PROCESSING COMPLETE. HIGHEST RETURN CODE IS 0004 FROM:
TASK 001

Chapter 5. DFSMSdss enhancements
141
In Figure 5-7 we show sample JCL to make a physical copy of a data set. The input data set
is SMS managed (and must therefore be catalogged), and the target volume as specified in
OUTDDNAME must also refer to an SMS managed volume.
Figure 5-7 Physical COPY JCL
In Figure 5-8 we show the output from the job shown in Figure 5-7. There is no specific
indication that this is the result of a Physical COPY, but that can be inferred from the fact that
the input data set MHLRES1.JOBS.JCL, which is SMS managed, has been copied, which
would not have been permitted with a Logical COPY.
Figure 5-8 Physical COPY job output
//MHLRES1D JOB (999,POK),MSGLEVEL=1,NOTIFY=MHLRES1
//STEP01 EXEC PGM=ADRDSSU,TIME=1440,REGION=6000K,
// PARM='UTILMSG=YES'
//SYSPRINT DD SYSOUT=*
//SBOX08 DD DISP=SHR,UNIT=SYSDA,DSN=MHLRES1.JOBS.JCL
//TEMP DD DISP=SHR,VOL=SER=MLD40C
//SYSIN DD *
COPY DATASET(INCLUDE(-
MHLRES1.JOBS.JCL -
)) -
SHARE -
PHYSINDDNAME(SBOX08) OUTDD(TEMP)
PAGE 0001 5695-DF175 DFSMSDSS V1R08.0 DATA SET SERVICES 2007.066 17:22
COPY DATASET(INCLUDE(-
MHLRES1.JOBS.JCL -
)) -
SHARE -
PHYSINDDNAME(SBOX08) OUTDD(TEMP)
ADR101I (R/I)-RI01 (01), TASKID 001 HAS BEEN ASSIGNED TO COMMAND 'COPY '
ADR109I (R/I)-RI01 (01), 2007.066 17:22:21 INITIAL SCAN OF USER CONTROL STATEMENTS COMPLETED.
ADR016I (001)-PRIME(01), RACF LOGGING OPTION IN EFFECT FOR THIS TASK
ADR006I (001)-STEND(01), 2007.066 17:22:21 EXECUTION BEGINS
ADR396I (001)-PCNVS(01), DATA SET MHLRES1.JOBS.JCL ALLOCATED, ON VOLUME(S): MLD40C
ADR801I (001)-DDDS (01), DATA SET FILTERING IS COMPLETE. 1 OF 1 DATA SETS WERE SELECTED: 0 FAILED
SERIALIZATION AND 0 FAILED
OTHER REASONS.
ADR454I (001)-DDDS (02), THE FOLLOWING DATA SETS WERE SUCCESSFULLY PROCESSED
MHLRES1.JOBS.JCL
ADR006I (001)-STEND(02), 2007.066 17:22:22 EXECUTION ENDS
ADR013I (001)-CLTSK(01), 2007.066 17:22:22 TASK COMPLETED WITH RETURN CODE 0000
ADR012I (SCH)-DSSU (01), 2007.066 17:22:22 DFSMSDSS PROCESSING COMPLETE. HIGHEST RETURN CODE IS 0000

142
z/OS V1R8 DFSMS Technical Update
In Figure 5-9 we show the results of attempting to display the characteristics of the
MHLRES1.JOBS.JCL data set on volume MLD40C. The warning duplicate data set name is
displayed in the top right-hand corner of the panel, as highlighted in bold.
Figure 5-9 ISPF 3.4 listing of data set MHLRES1.JOBS.JCL on volume MLD40C
In Figure 5-10 we show the help information about the situation that exists as a result of
making a physical copy of the data set and trying to locate it.
Figure 5-10 ISPF Help information for the duplicate data set name situation
ISPF prevented any attempt to process the copy of the data set while the original was still
catalogged.
The original was renamed so that there was no longer a duplicate catalog situation, but the
copied data set could not be cataloged using ISPF as it, being SMS managed, is in an error
state.
This copied data set was cataloged by means of the IDCAMS command
DEFINE NVSAM(NAME('MHLRES1.jobs.jcl') VOLUMES(Mld40c) DEVT(3390) recatalog)
issued under TSO.
Menu Options View Utilities Compilers Help
ssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssss
DSLIST - Data Sets on volume MLD40C Duplicate data set name
Command ===> Scroll ===> CSR

Command - Enter "/" to select action Message Volume
-------------------------------------------------------------------------------
I MHLRES1.JOBS.JCL MLD40C
***************************** End of Data Set list ****************************
TUTORIAL ----------------- DUPLICATE DATA SET NAMES ------------------ TUTORIAL
COMMAND ===>

The data set that you are accessing via ISPF:
MHLRES1.JOBS.JCL
is also cataloged on a volume other than MLD40C .

Since MLD40C is an SMS volume, the allocation of the data set
will use the catalog to find the data set and will allocate the
data set that is on another volume. Since this is not the volume
you specified, the command fails.

© Copyright IBM Corp. 2008. All rights reserved.
143
Chapter 6.
DFSMShsm enhancements
The DFSMShsm enhancements in DFSMS V1R8 provide improvements in the areas of error
handling for alternate duplex tapes, tape recycling, migration of non-VSAM data sets,
ARECOVER processing, and altering the priority of queued requests. For details on the
DFSMShsm fast replication function in DFSMS V1R8 refer to Chapter 7, “DFSMShsm fast
replication” on page 169.
The following topics are covered:
Error handling on alternate duplex tapes
Recycle SYNCDEV at intervals
Migration scratch queue for non-VSAM data set
Individual data set restore for ARECOVER processing
New command ALTERPRI
6

144
z/OS V1R8 DFSMS Technical Update
6.1 Error handling on alternate duplex tapes
Before z/OS V1R8.0, when an error was detected on the alternate tape, during the duplex
process, this alternate tape was de-mounted and returned to scratch, and DFSMShsm
continues to write to the original volume.
It has become problematic for users with high capacity tapes and using duplex tape function
to create alternate copies for disaster recovery, due the tape size and the total time the user
would be without a valid tape copy.
In z/OS V1R8.0, the duplex tape migration option is improved. A new keyword,
ERRORALTERNATE(CONTINUE | MARKFULL), is added to the existing SETSYS
DUPLEX(MIGRATION(Y)). If you choose the DUPLEX option, two tapes are created
concurrently, and the most common use of that is to keep one on site and the other on a
remote tape library.
The new option lets users choose to MARK FULL original and alternate tapes in case of a
problem writing on the alternate tape. In this case, both tapes are demounted and the failing
data set is retried on a new set of tapes.
The other option on this command is CONTINUE. If you specify CONTINUE, and an error
occurs on the alternate volume, then the process continues on the original tape. DFSMShsm
creates a TCN record (Tape Copy Need) and a TAPECOPY is immediately attempted after
completion of the original tape, or else it will be scheduled to the next auto function window.
The default to ERRORALTERNATE is CONTINUE.
This new option is valid when duplexing a tape migration or recycling a tape migration. The
syntax of the SETSYS DUPLEX command is shown in Figure 6-1.
Figure 6-1 New keyword ERRORALTERNATE syntax
Remember that if you want this option working all the time, do not forget to write the
command on the DFSMShsm PARMLIB member that you use, as a regular SETSYS
command. This causes the command to be issued at DFSMShsm start up. It can be issued
dynamically also.
SETSYS DUPLEX(MIGRATION(Y ERRORALTERNATE(CONTINUE | MARKFULL)))
The default for ERRORALTERNATE is CONTINUE.
-SETSYS-----------------------------------------------------------------------
| |-Y-| |
|DUPLEX-(-BACKUP(--N--)----------------------------------------------)|
| |
| |-MARKFULL-| |
| |-Y-(ERRORALTERNATE(--CONTINUE--))-| |
|-MIGRATION(--N---------------------------------)----------|
| |
| |MARKFULL| |
| |-Y-| |-Y-(ERRORALTERNATE(-CONTINUE-))-|
|-BACKUP(--N--)-MIGRATION(--N-----------------------------)|

Chapter 6. DFSMShsm enhancements
145
Figure 6-2 shows the result of a dynamic command issue.
Figure 6-2 Dynamic command output
Figure 6-2 shows how SETSYS was before
X
and after
Y
the command SETSYS
DUPLEX(MIGRATION(Y ERRORALTERNATE(MARKFULL)).
Figure 6-3 shows the effect of the command to DUPLEX migration tape. One data set is
migrated, but two mounts occur due to the DUPLEX(MIGRATION(Y) option.
Figure 6-3 Output of DUPLEX command
Message ARC0120I in Figure 6-3 tells you that, if the return code (RC=) was 0 (zero)
X,
an
ADDVOL command has been processed for the type of volume
volser
specified in the
message. If the volume record is successfully created or updated, the return code is zero.
When SETSYS DUPLEX(MIGRATION(Y ERRORALTERNATE(MARKFULL))) has been
specified, migration processing ignores the patch provided by APAR OA09928 (refer to
X
STC18988 ARC0442I TAPE OUTPUT PROMPT FOR TAPECOPY=NO, DUPLEX
ARC0442I (CONT.) BACKUP TAPES=NO, DUPLEX MIGRATION TAPES=NO
...
...
-F DFHSM64,SETSYS DUPLEX(MIGRATION(Y ERRORALTERNATE(MARKFULL))
STC18988 ARC0100I SETSYS COMMAND COMPLETED
...
...
Y
STC18988 ARC0442I TAPE OUTPUT PROMPT FOR TAPECOPY=NO, DUPLEX
ARC0442I (CONT.) BACKUP TAPES=NO, DUPLEX MIGRATION TAPES=(Y,
ARC0442I (CONT.) ERRORALTERNATE=MARKFULL)
STC18988 00000290 IEC501A M
0B90,PRIVAT,SL,COMP,DFHSM64,DFHSM64,HSM.HMIGTAPE.DATASET
STC18988 00000090 IEC705I TAPE ON
0B90,TST029,SL,COMP,DFHSM64,DFHSM64,HSM.HMIGTAPE.DATASET,MEDIA3
STC18988 00000090 ARC0120I MIGRATION VOLUME TST029 ADDED, RC=0000
X
, 863
863 00000090 ARC0120I (CONT.) REAS=0000
STC18988 00000290 IEC501A M
0B91,PRIVAT,SL,COMP,DFHSM64,DFHSM64,HSM.COPY.HMIGTAPE.DATASE T
STC18988 00000090 IEC705I TAPE ON
0B91,TST000,SL,COMP,DFHSM64,DFHSM64,HSM.COPY.HMIGTAPE.
DATASET,MEDIA3
STC18988 00000090 IEC205I SYS00303,DFHSM64,DFHSM64,FILESEQ=1, COMPLETE VOLUME
LIST, 866
866 00000090 DSN=HSM.HMIGTAPE.DATASET,VOLS=TST029,TOTALBLOCKS=133
STC18988 00000090 IEC205I SYS00304,DFHSM64,DFHSM64,FILESEQ=1, COMPLETE VOLUME
LIST, 867
867 00000090 DSN=HSM.COPY.HMIGTAPE.DATASET,VOLS=TST000,TOTALBLOCKS=133
00000290 IEA989I SLIP TRAP ID=X33E MATCHED. JOBNAME=DFHSM64 ,
ASID=0076.
STC18988 00000290 IEF234E K 0B90,TST029,PVT,DFHSM64,DFHSM64
STC18988 00000290 IEF234E K 0B91,TST000,PVT,DFHSM64,DFHSM64

146
z/OS V1R8 DFSMS Technical Update
“OA09928” on page 465) that turns off SYNCDEVs as each data set is written to the
migration alternate volume.
Figure 6-4 shows the PATCH provided by APAR OA09928.
Figure 6-4 PATCH from APAR OA09928
In this case, with the patch ON (see Figure 6-4) and SETSYS DUPLEX(MIGRATION(Y
ERRORALTERNATE(MARKFULL))), a new PDA trace is added, with text (‘APAR OW43224 is
ignored’), so it is easily identified during a problem determination. For more information about
APAR OW43224, see “OW43224” on page 466.
Note: PATCH OA0998, OW45264, and OW43224 deal with error handling on alternate
duplex tapes. The PATCH to turn off the SYNCDEV was suggested to decrease the
elapsed recycle processing time. With z/OS V1R8.0, it is no longer beneficial and no
longer recommended.
...
...
Customers can now suppress syncs on the alternate tape during duplex migration.
Syncs for the migration alternate tape can be turned off via:
PATCH .MCVT.+196 BITS(..1.....)
and back on via:
PATCH .MCVT.+196 BITS(..0.....)
The default is to do syncs on the alternate tapes.
This patch is not recommended for earlier technology tape drives such as the 3490 and
3590.
Prior APAR OW45264 describes a similar patch for turning off alternate tape syncs for
Recycle. This can also dramatically improve performance with the 3592 -J tape drive.
...
...

Chapter 6. DFSMShsm enhancements
147
Figure 6-5 shows the PATCH provided by APAR OW45264.

Figure 6-5 PATCH from APAR OW45264
To simulate a problem on the alternate tape, we issued a V unit,xxxx,FORCE on its unit tape
drive that caused an interruption. See Figure 6-6 for log details.
Figure 6-6 VARY OFFLINE FORCE to simulate the tape problem
...
...
Bypassing the SYNCDEV for the duplex alternate tape during Recycle is now
supported. This reduces the time Recycle takes by bypassing the SYNCDEV of the
duplex alternate tape after each dataset. To bypass the alternate tapes SYNCDEV
operation, enter the following PATCH command for your release:
For releases HDZ11D0 AND HDZ11E0:
PATCH .YGCB.+88 BITS(......1.)
.
For releases HDZ11F0, HDZ11G0, HDZ11H0, HDZ11J0 & HDZ11K0:
PATCH .YGCB.+C8 BITS(......1.)
.
To preserve the PATCH when HSM is restarted, place the PATCH command in the HSM
startup PARMLIB member ARCCMDxx.
.
To remove the patch and allow the SYNCDEV to be again occur on the alternate
volume, enter the following PATCH command for your release and remove the
previous PATCH from ARCCMDxx:
.
For releases HDZ11D0 AND HDZ11E0:
PATCH .YGCB.+88 BITS(......0.)
For releases HDZ11F0, HDZ11G0, HDZ11H0, HDZ11J0 & HDZ11K0:
PATCH .YGCB.+C8 BITS(......0.)
.
If you had the following PATCH installed from APAR OW43224:
PATCH .MCVT.+295 BITS(....1...)
it should be removed at this time.
...
...
00000090 *038 IEE800D CONFIRM VARY FORCE FOR B91 - REPLY NO OR YES
00000290 R 38,YES
00000090 IEE600I REPLY TO 038 IS;YES
00000090 IOS102I DEVICE 0B91 BOXED, OPERATOR REQUEST, WAS ASSIGNED
00000090 IEE793I 0B91 PENDING OFFLINE AND BOXED
00000090 IOS000I 0B91,**,SIM,**,**06,,**,TST000,DFHSM64
00000090 IEC020I 001-3,DFHSM64,DFHSM64,SYS00325,0B91,TST000,
00000090 IEC020I HSM.COPY.HMIGTAPE.DATASET
00000090 IEC020I NON-ACCEPTABLE ERROR
00000090 IEF524I 0B91 PENDING OFFLINE

148
z/OS V1R8 DFSMS Technical Update
Figure 6-7 shows the DFSMShsm behavior after the VARY OFFLINE FORCE command.
Figure 6-7 Output of DFSMShsm behavior after VARY OFFLINE FORCE
00000090 IEA794I SVC DUMP HAS CAPTURED: 983
00000090 DUMPID=001 REQUESTED BY JOB (DFHSM64 )
00000090 DUMP TITLE=DFSMSDSS-DUMP ABEND=001,COMPID=DF175,ADRFTARR,JOBNAM
00000090 E=DFHSM64
00000090 ARC0421I MIGRATION VOLUME TST029 IS NOW MARKED FULL
00000090 ARC0443I MIGRATION VOLUME TST029 WITH ALTERNATE TST000 985
00000090 ARC0443I (CONT.) IS NOW MARKED FULL
...
...
00000090 IEC205I SYS00324,DFHSM64,DFHSM64,FILESEQ=1, COMPLETE VOLUME LIST, 990
00000090 DSN=HSM.HMIGTAPE.DATASET,VOLS=TST029,TOTALBLOCKS=11126
00000090 IEC215I 714-0C,IFG0200Z,DFHSM64,DFHSM64,SYS00325,0B91,TST000,
HSM.COPY.HMIGTAPE.DATASET
00000090 IOS000I 0B91,**,SIM,**,**06,,**,TST000,DFHSM64
00000090 IOS000I 0B91,**,SIM,**,**06,,**,TST000,DFHSM64
00000090 IOS000I 0B91,**,SIM,**,**06,,**,TST000,DFHSM64
00000290 IEA989I SLIP TRAP ID=X33E MATCHED. JOBNAME=MHLRES2 , ASID=005B.
00000290 IEF234E K 0B90,TST029,PVT,DFHSM64,DFHSM64
00000290 IEF196I IGD104I DUMP.D070223.H23.SC64.DFHSM64.S00001
RETAINED,
00000290 IEF196I DDNAME=SYS00011
00000090 IEA611I COMPLETE DUMP ON DUMP.D070223.H23.SC64.DFHSM64.S00001 999
...
...
00000090 INCIDENT TOKEN: SANDBOX SC64 02/23/2007 23:56:48
00000090 ERROR ID = SEQ00744 CPU00 ASID0074 TIME18.56.48.1
00000290 IEF196I IEF237I 8074 ALLOCATED TO IPCSDDIR
00000290 IEF234E K 0B91,TST000,PVT,DFHSM64,DFHSM64
00000090 IEF244I DFHSM64 DFHSM64 - UNABLE TO ALLOCATE 1 UNIT(S) 002
00000090 AT LEAST 1 OFFLINE UNIT(S) NEEDED.
00000090 IEF877E DFHSM64 NEEDS 1 UNIT(S) 003
00000090 FOR DFHSM64 SYS00333
...
...
STC19100 00000290
X
IEF234E D 0B91,TST000,,DFHSM64,DFHSM64
STC19100 00000290 IEC501A M
0B90,PRIVAT,SL,COMP,DFHSM64,DFHSM64,HSM.HMIGTAPE.DATASET
...
...
STC19100 00000090 IEC705I TAPE ON
Y

0B90,TST001,SL,COMP,DFHSM64,DFHSM64,HSM.HMIGTAPE.DATAS
ET,MEDIA3
STC19100 00000090 ARC0120I MIGRATION VOLUME TST001 ADDED, RC=0000, 150
150 00000090 ARC0120I (CONT.) REAS=0000
STC19100 00000290 IEC501A M
0B91,PRIVAT,SL,COMP,DFHSM64,DFHSM64,HSM.COPY.HMIGTAPE.DATASE
T
STC19100 00000090 IEC705I TAPE ON
0B91,TST002,SL,COMP,DFHSM64,DFHSM64,HSM.COPY.HMIGTAPE.
DATASET,MEDIA3

Chapter 6. DFSMShsm enhancements
149
In the sequence in Figure 6-7 on page 148
X
, tape - TST000 - is demounted. What happens
with the tape depends on what TAPE manager you are using in your environment and the
respective policy you have adopted. After that, on
Y
, DFSMShsm requests two new tapes to
continue from where it had a problem.
Because of this new option on SETSYS DUPLEX, an additional keyword was added on the
LIST TTOC command so that you can produce a list with only tapes that were closed due to
the ERRORALTERNATE option. Figure 6-8 shows some optional parameters on LIST TTOC
command.
Figure 6-8 Optional parameters from TTOC
Figure 6-9 is a sample of the LIST TTOC command with the new option ERRORALTERNATE.
Figure 6-9 Output of LIST TTOC command
This enhancement only applies to DFSMShsm at the V1.8 or later level that have specified
the option SETSYS DUPLEX with the ERRORALTERNATE(MARKFULL) option. If
DFSMShsm is not at the V1.8 level, DFSMShsm continues to work as it did previously.
Message change
Message ARC0442I changed because of this improvement, including the new keywords. See
Figure 6-10 for details.
Figure 6-10 Output of changed message ARC0442I
For more information about DFSMShsm messages, refer to z/OS MVS System Messages
Vol 2 (ARC-ASA), SA22-7632.
Coexistence and migration
Table 6-1 details coexistence and migration for SESTYS DUPLEX with the new
ERRORALTERNATE option.
TTOC SELECT Optional Parameters:
_ ________________ __
|_EMPTY__________|
|_FULL___________|
|_NOTFULL________|
|_ASSOCIATED_____|
|_NOTASSOCIATED__|
|_ERRORALTERNATE_|
I
new option
LIST TTOC SELECT(ERRORALTERNATE) ODS('MHLRES4.HSM.OUTPUT')

- DFSMSHSM CONTROL DATASET - TAPE VOLUME TTOC - LISTING - AT 17:05:15 ON 07/03/02 FOR SYSTEM=SC64

VOLSER UNIT VOL REUSE VALID PCT VOL RACF PREV SUCC NUM ONE ALT LIB STORAGE
NAME TYPE CAPACITY BLKS VALID STATUS VOL VOL REC FILE VOL GROUP
TST001 3590-1 ML2 00197300 00197270 100 FULL NO *NONE* *NONE* 001 YES TST002 LIB1 SGLIB1
----- END OF - TAPE VOLUME TTOC - LISTING -----

ARC0442I TAPE OUTPUT PROMPT FOR TAPECOPY = x, DUPLEX BACKUP TAPES =x,
DUPLEX MIGRATION TAPES = x, ERRORALTERNATE= {CONTINUE | MARKFULL}

150
z/OS V1R8 DFSMS Technical Update
Table 6-1 Coexistence table for SESTYS DUPLEX with new option ERRORALTERNATE
.
6.2 Recycle SYNCDEV at intervals
Prior to z/OS V1.8, using the option SETSYS DUPLEX caused DFSMShsm to issue a
SYNCDEV after each successful processed data set for both primary and alternate tapes.
This is a time-consuming operation. To avoid the SYNCDEV operation and improve
performance, a patch was provided to turn it off.
With z/OS V1.8 the recycle process issues the SYNCDEV at intervals. It occurs at about each
20 data sets or 500 Mb, whichever comes first. So, even if you are duplexing, performance
should improve on your process. After the last data set on the recycle input volume has been
moved to the output tape volumes, a SYNCDEV is done regardless of any counts. This
guarantees that all data from the input tape has been written to the recycle output tape (both
original and alternate).
With this new function implemented, DFSMShsm continues to honor the existing patch to turn
off all SYNCDEV, except in the case where DFSMShsm recycles a migration tape and
SETSYS DUPLEX(MIGRATION(Y ERRORALTERNATE(MARKFULL))) is specified. In this
case, DFSMShsm issues a PDA trace entry with text (‘APAR OW43224 is ignored’) to identify
this condition.
The specifics of how Recycle SYNCDEV at intervals works are:
If you are working with NO duplexing:
– Standard: SYNCDEV executes after 20 data sets or 500 Mb, whichever comes first.
– Error: If an error occurs on the output volume, it is marked full, a new output volume is
selected, and the process continues starting on the first data set after the last
successful SYNCDEV on the previous tape.
If you are working with duplexing and recycling a BACKUP volume:
– Standard: SYNCDEV executes after each 20 data sets or 500 Mb, whichever comes
first.
Element or feature DFSMShsm
When change was introduced z/OS V1R8.0
Applies to migration from None
Timing After the first IPL of z/OS V1R8.0
Target system hardware requirements None
Target system software requirements None
Other system (coexistence or fallback)
requirements
None
Restrictions None
System impacts None
Note: The PATCH to turn off the SYNCDEV was suggested to decrease the elapsed
recycle processing time. On z/OS V1R8.0 it is no longer beneficial and no longer
recommended.

Chapter 6. DFSMShsm enhancements
151
– Original volume error: If an output error occurs on the original volume, the result is that
both original and alternate are marked full, a new set of tapes is selected, and recycle
continues after the first data set after the last successful SYNCDEV on the previous
tape.
– Alternate volume error: If an error occurs on the alternate volume, DFSMShsm
continues to write on the original volume tape and the alternate is demounted. When it
has completed writing the original tape, it marks the tape full and schedules a
TAPECOPY to create a new alternate tape.
If you are working with duplexing and recycling a MIGRATE volume:
– Standard: SYNCDEV executes after each 20 data sets or 500 Mb, whichever comes
first.
– Original volume error: If an output error occurs on the original volume, the result is that
both original and alternate are marked full, a new set of tapes is selected, and recycle
continues after the first data set after the last successful SYNCDEV on the previous
tape.
– Alternate volume error: If an error occurs on the alternate volume, DFSMShsm works
as defined on the ERRORALTERNATE parameter of the SETSYS
DUPLEX(MIGRATON(Y)) option.
• If ERRORALTERNATE(CONTINUE) is specified, the alternate tape is demounted
and DFSMShsm continues write on the original tape. When DFSMShsm finishes
writing the original tape, it schedules a TAPECOPY to create a new alternate
volume.
• If ERRORALTERNATE(MARKFULL) is specified, both the original and alternate
tape are marked FULL. DFSMShsm recycle requests two new volumes and
continues writing with the first data set after the last SYNCDEV on the previous
tape.
Coexistence and migration
The enhancement on SYNCDEV on z/OS V1.R8 made the patch to bypass it no longer
beneficial. See Table 6-2 for details.
Table 6-2 Coexistence table for SYNCDEV enhancemen
t
Important: If you currently do calculations to track recycle times, take into consideration
that now SYNCDEV works with a
block
of 20 data sets or 500 Mb, whichever comes first.
Element or feature DFSMShsm
When change was introduced z/OS V1R8.0
Applies to migration from None
Timing After the first IPL of z/OS V1R8.0
Is the migration action required? No, but recommended because the patch is no
longer beneficial
Target system hardware requirements None
Target system software requirements None
Other system (coexistence or fallback)
requirements
None
Restrictions None

152
z/OS V1R8 DFSMS Technical Update
6.3 Migration scratch queue for non-VSAM data set
Prior to z/OS V1.8, during the migration of non-VSAM data sets, DFSMShsm used
considerable time doing the scratch operation from the L0 volume.
With z/OS V1R8 DFSMShsm is able to create a new task under the migration queue when
migrating from L0 to tape that does the scratch activity (from scratch queue requests), so the
performance is improved, as the requests are now handled asynchronously. If you have
multiple migrate tasks, you have this new task to each one.
This new process is applied when migrating conditions are:
From L0 volume
On command volume migration
On primary space management
On interval migration
And are not applied when:
You are migrating a single data set.
During L1 to L2 migration - secondary space management.
6.3.1 Individual data set restore for ARECOVER processing
Aggregate Backup and Recovery Support (ABARS) facilitates a point-in-time backup of a
collection of related data in a consistent manner. This group of related data is defined to
ABARS as an aggregate. The backup copies are created in a device-independent format.
During backup processing, the data is packaged as a single entity in preparation for taking it
off-site. This enables the recovery of individual applications in user-priority sequence.
ABARS has three components:
Aggregate group definition: The aggregate group lists the selection data set names,
instruction data set name, and the management class for the aggregate. This component
is implemented through the Interactive Storage Management Facility (ISMF) and requires
the storage management subsystem (SMS) address space to be active.
System impacts None
Element or feature DFSMShsm

Chapter 6. DFSMShsm enhancements
153
Aggregate backup (ABACKUP): The ABACKUP command backs up a user-defined group
(aggregate group) of data sets at that moment in time. This DFSMShsm function also
requires the SMS address space to be active. ABACKUP is capable to get data sets from
user tapes, ML1, ML2, SMS, and Non-SMS volumes. See a simplified Aggregate backup
(ABACKUP) flow in Figure 6-11.
Figure 6-11 Simplified aggregate backup flow

154
z/OS V1R8 DFSMS Technical Update
Aggregate recovery (ARECOVER): The ARECOVER command recovers data sets that
were previously backed up by an aggregate backup. The SMS address space is not
required, but it is highly recommended to simplify the recovery process. This address
space is required to recover VSAM data sets in the ALLOCATE list and any extended
format data sets. See a simplified aggregate backup (ARECOVER) flow in Figure 6-12.
Figure 6-12 Simplified aggregate recover flow
Prior to z/OS V1R8 we were not able to select one single data set to recover. Now, we have
additional keywords ONLYDATASET and LISTOFNAMES that make it possible.
Figure 6-13 shows the keywords that were added to ARECOVER in DFSMShsm 1.8.0.
Figure 6-13 ARECOVER new keywords
The optional ONLYDATASET parameter allows you to recover individual data sets from an
aggregate backup version. You can specify a single fully qualified data set name using the
NAME(dsname) parameter of ONLYDATASET to recover a particular data set.
To recover multiple data sets, you can supply the name of a data set that contains a list of
data set names to be recovered on the LISTOFNAMES(listdsname) parameter of
ONLYDATASET. The LISTOFNAMES data set must be a sequential data set, fixed-block,
>>_ARECOVER...
______________________________________________________________... ><
| |
|___ONLYDATASET___(_____NAME(dsname)___________________)_|
|_ONLYDS______| | |
|___LISTOFNAMES(listdsname)___|
|_LON______________________|

Chapter 6. DFSMShsm enhancements
155
record size of 80, and be cataloged. The LISTOFNAMES data set should contain one fully
qualified data set name per record.
The abbreviation for ONLYDATASET is ONLYDS and the abbreviation for LISTOFNAMES is
LON.
For more details about these new keywords, refer to DFSMShsm Storage Administration
Guide, SC26-0421.
First you need a valid ABACKUP file. So, to accomplish that, you should have an ABARS
environment prepared to do backup and recovery. We do not show in this book how to set up
the ABARS function.
To reach ABARS panels under ISMF, go to the ISMF main panel and choose 9 Aggregate
Group. Check whether you have storage administrator privileges (on the main panel select
option 0, then option 0, and then option 2 for Storage Admin) or you will be able to list only.
After that, choose option 5, Abackup - Backup an Aggregate Group. Figure 6-14 and
Figure 6-15 on page 156 show the AGGREGATE GROUP BACKUP panels used to do this
preparation.
Figure 6-14 Output 1 of 2 about creating a AGGREGATE GROUP BACKUP
Panel Utilities Help
ssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssss
AGGREGATE GROUP BACKUP Page 1 of 2
Command ===>

CDS Name : ACTIVE

Aggregate Group Name . . MHLRES4

Unit Name . . . . . . . .
X
Processing Option . . . . 1 (1=Verify, 2=Execute)
Y
Wait for Completion . . . N (Y or N)
Stack / Nostack . . . . . (S=STACK, N=NOSTACK or blank)
Optimize . . . . . . . . 3 (1 to 4 or blank)
Delete Data Sets After Abackup . . N (Y or N)

Filter Output Data Set Name (1 to 44 Characters)
Z
===> 'MHLRES4.ABARS.OUTPUT'

Enter "/" to select option
Process only
[
/ L0 ML1 ML2 USERTAPE

Use ENTER to Continue;

156
z/OS V1R8 DFSMS Technical Update
To prepare our backup, we choose in the first panel (Figure 6-14 on page 155) the
parameters:

X
and
Y
, which are self-explanitory.

Z
, which will be a file that will receive our output results.

[
, which is a slash before L0, so this specific backup selects from the filter list only data
sets from L0.
At anytime, using the panel, you can choose Help (usually pre-set as PF1).
Figure 6-15 Output 2 of 2 about creating a AGGREGATE GROUP BACKUP
It is a good practice to use the VERIFY option, as shown in Figure 6-15. After the VERIFY,
check the results and do any required adjustments. Run the VERIFY option until you get a RC
00 as a result. After that, change the option to EXECUTE (2). You can use the EXECUTE
option directly.
Figure 6-16 on page 157 shows where you are going to find the ACTIVITY LOG from
ABACKUP function.
Attention: If you are a first-time TSO user, issue a PFSHOW ON/OFF or FKA ON/OFF to
check your predefined PF Keys.
Panel Utilities Help
ssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssss
AGGREGATE GROUP BACKUP Page 2 of 2
Command ===>


DFSMShsm Command and Processing Option:
NOWAIT ABACKUP MHLRES4 VERIFY
FILTEROUTPUTDATASET('MHLRES4.ABARS.OUTPUT')
OPTIMIZE(3) PONLY(L0)







Enter 1 to Submit DFSMShsm ABACKUP COMMAND
Enter 2 to Save Generated ABACKUP PARAMETERS

Select Option . . 1 (1=SUBMIT, 2=SAVE)

Use ENTER to Perform Selection;

Chapter 6. DFSMShsm enhancements
157
Figure 6-16 Support information where your dynamically allocated ACTIVITY LOG file is placed
Figure 6-17 shows you a sample of the ACTIVITY LOG from ABARS.
Figure 6-17 ACTIVITY LOG containing information about the current ABACKUP execution
ARC6054I AGGREGATE BACKUP STARTING FOR AGGREGATE GROUP
ARC6054I (CONT.) MHLRES4, AT 19:16:35,
ARC6054I (CONT.) STARTED TASK = DFHSMABR.ABAR0135
ARC6030I ACTIVITY LOG FOR AGGREGATE GROUP MHLRES4 WILL BE
ARC6030I (CONT.) ROUTED TO
ARC6030I (CONT.) HSMACT.H2.ABACKUP.MHLRES4.D07051.T191635
COMMAND REQUEST 00000018 SENT TO DFSMSHSM
PAGE 0001 Z/OS DFSMSHSM 1.8.0 DATA FACILITY HIERARCHICAL STORAGE MANAGER
07.051 19:16
ARC6000I ABACKUP MHLRES4 EXECUTE FILTEROUTPUTDATASET('MHLRES4.ABARS.OUTPUT')
OPTIMIZE(3) PONLY(L0 )
ARC6054I AGGREGATE BACKUP STARTING FOR AGGREGATE GROUP MHLRES4, AT 19:16:35,
STARTED TASK = DFHSMABR.ABAR0135
ARC6030I ACTIVITY LOG FOR AGGREGATE GROUP MHLRES4 WILL BE ROUTED TO
HSMACT.H2.ABACKUP.MHLRES4.D07051.T191635
ARC6379I THE MANAGEMENT CLASS CONSTRUCTS USED IN THE AGGREGATE GROUP, MHLRES4,
ARE:
CLASS NAME : MCDB22
DESCRIPTION:
MGMT CLASS FOR DB2 TS
EXPIRATION ATTRIBUTES
EXPIRE AFTER DAYS NON-USAGE: NOLIMIT
EXPIRE AFTER DATE/DAYS : NOLIMIT
...
...
ARC6004I 0079 ABACKUP DATA SET FILTERING IS COMPLETE. 18 OF 18 DATA SETS WERE
SELECTED: 0 FAILED SERIALIZATION AND 0
ARC6004I 0079 ABACKUP FAILED FOR
PAGE 0002 Z/OS DFSMSHSM 1.8.0 DATA FACILITY HIERARCHICAL STORAGE MANAGER
07.051 19:16
ARC6004I 0079 ABACKUP OTHER REASONS.
ARC6004I 0079 ABACKUP ADR454I (001)-DTDSC(01), THE FOLLOWING DATA SETS WERE
SUCCESSFULLY PROCESSED
X

ARC6004I 0079 ABACKUP MHLRES4.SC64.SPFLOG2.LIST
ARC6004I 0079 ABACKUP MHLRES4.JCL.CNTL
ARC6004I 0079 ABACKUP MHLRES4.SC63.ISPF42.ISPPROF
...
...
ARC6071I VOLUMES USED FOR INSTRUCTION/ACTIVITY LOG FILE
MHLRES4.ABARS.OUTPUT.I.C01V0001 DURING AGGREGATE BACKUP FOR
AGGREGATE GROUP MHLRES4 ARE:
TST022
Y

ARC6051I AN INSTRUCTION DATA SET WAS NOT SPECIFIED FOR AGGREGATE GROUP MHLRES4
ARC6055I AGGREGATE BACKUP HAS COMPLETED FOR AGGREGATE GROUP MHLRES4, AT
19:18:08, RETCODE = 000

158
z/OS V1R8 DFSMS Technical Update
In this ACTIVITY LOG (Figure 6-17 on page 157) you will find
X
files that you saved at this
time and the tape you used
Y
, among other information.
The output data set that you chose on the first panel on the ABACKUP function (see
Figure 6-14 on page 155,
Z
) is similar to the one shown in Figure 6-18.
Figure 6-18 Print panel from MHLRES4.ABARS.OUTPUT used on out test
Now, let us do the ARECOVER choosing just one data set. Note that the first panel has a new
line:
Recover Individual Data Sets. Y (Y or N)
See
[
in Figure 6-19 on page 159.
To get this panel, go to the ISMF main panel, then choose option 9. Aggregate Group and
after that option 6. Arecover - Recover an Aggregate Group.
000001 FILTEROUTPUTDATASET LIST FOR AGGREGATE GROUP MHLRES4,
000002 VERSION 0001, CREATED ON 2007/02/20 AT 19:16:35
000003
000004 THE FOLLOWING DATA SETS WERE SPECIFIED IN THE INCLUDE LIST:
000005 MHLRES4.ABARS.OUTPUT VOLSER= MLD30C
000006 MHLRES4.ABARS.SELECT VOLSER= MLD30C
000007 MHLRES4.BRODCAST VOLSER= SBOX88
000008 MHLRES4.JCL.CNTL VOLSER= MLD20C
000009 MHLRES4.LIST VOLSER= MLD50C
000010 MHLRES4.LOG.MISC VOLSER= SBOX88
000011 MHLRES4.SC63.ISPF42.ISPPROF VOLSER= MLD20C
000012 MHLRES4.SC64.SPFLOG1.LIST VOLSER= MLD20C
000013 MHLRES4.SC64.SPFLOG2.LIST VOLSER= MLD00C
000014 MHLRES4.SC64.SPFLOG3.LIST VOLSER= MLD30C
000015 MHLRES4.SC64.SPFTEMP2.CNTL VOLSER= SBOX88
000010 MHLRES4.SC63.ISPF42.ISPPROF VOLSER= MLD20C
...
...
000022 MHLRES4.TCPIP.PROFILE VOLSER= SBOX88
000023 A TOTAL OF 000018 DATA SETS SPECIFIED IN THE INCLUDE LIST WERE SELECTED.
Note: Not all panels were used from all available. On some panels, pressing DOWN, you
have more options.

Chapter 6. DFSMShsm enhancements
159
Figure 6-19 shows panel 1 of 8 from AGGREGATE GROUP RECOVER.
Figure 6-19 ARECOVER panel 01 of 08
In Figure 6-19, we used the ABACKUP CONTROL DATASET
X
field to specify the name of
the aggregate group backup control file to be used for aggregate recovery.
Other fields we choose on this test were:

Y
In a first step, we did a verify and after all checked, we choose Execute

Z
Usually we choose N here. If you choose Y for any reason, be advised that you will do
a synchronous operation, so your session will be locked up until the end of the process.

[
To demonstrate the new option, we choose Y.
Panel Utilities Scroll Help
ssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssss
AGGREGATE GROUP RECOVER Page 1 of 8
Command ===>

X
Abackup Control Dataset . . 'MHLRES4.ABARS.OUTPUT.C.C01V0001'
(1 to 44 Characters)
Xmit . . . . . . . . . . . N (Y or N)
Stack / Nostack . . . . . (S=STACK, N=NOSTACK or blank)
Aggregate Group Name . . . .
Date . . . . . . . . . . . (yyyy/mm/dd)
Version . . . . . . . . . (1 to 9999)

Y
Processing Option . . . . . 2 (1=Prepare, 2=Verify, 3=Execute)
Z
Wait for Completion . . . . N (Y or N)
Target GDG Data Set Status (A=ACTIVE, D=DEFERRED, R=ROLLEDOFF,
S=SOURCE or blank)
Volume Count . . . . . . . . (A=ANY or blank)
Recover Instruction Data Set . . N (Y or N)
Recover Activity Log . . . . . . N (Y or N)
[
Recover Individual Data Sets . . Y (Y or N)

Use ENTER to Continue; Use DOWN to View Additional Options;

160
z/OS V1R8 DFSMS Technical Update
Figure 6-20 shows panel 2 of 8 from AGGREGATE GROUP RECOVER.
Figure 6-20 ARECOVER panel 2 of 8
In Figure 6-20 we choose
X
Y for the Backup Migrated Data Sets field to indicate whether
you want to cause a backup of all data sets that were successfully recovered to ML1 DASD
volumes the next time automatic backup is run.
Y
For New Migration Level choose 3 to
indicate that migrated data sets are to be recovered to the same migration level as the
original data set from the aggregate backup site. Specify N for
Z
Rename Selected Data
Sets so on this test, your recovered data set is not renamed. Specify N
[
for the Specify
Conflict Resolution Option field to indicate whether you want to specify the action to be taken
when a like-named data set conflict occurs.
Panel Utilities Scroll Help
ssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssss
AGGREGATE GROUP RECOVER Page 2 of 8
Command ===>

ABACKUP CONTROL DATASET: 'MHLRES4.ABARS.OUTPUT.C.C01V0001'
Model Entity . . . .
(1 to 44 Characters)
Target Unit Name . .

Target Utilization Percentage . . (1 to 100)

X
Backup Migrated Data Sets . . . . Y (Y or N)
Y
New Migration Level . . . . . . . 3 (1=ML1, 2=ML2, 3=SOURCELEVEL)
Rename All to High Level Qualifier . . (1 to 8 characters)
Z
Rename Selected Data Sets . . . . . . . N (Y or N)

[
Specify Conflict Resolution Option . . N (Y or N)
Use ENTER to Continue; Use UP to View Previous Options;

Chapter 6. DFSMShsm enhancements
161
Figure 6-21 shows panel 3 of 8 from AGGREGATE GROUP RECOVER. It lets you choose
your unit name
X
and tape volume
Y
.
Figure 6-21 ARECOVER panel 03 of 08
In Figure 6-17 on page 157 you have information about the volume that was used to save
your data. This information is duplicated here
X
in Figure 6-22.
Figure 6-22 The tape volume that was used in our demonstration
We jump to panel 7 of 8, but you have more options on panels that were not shown. It is your
decision as to whether you need these extra options. To complete our test, we do not.
Panel Utilities Help
ssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssss
AGGREGATE GROUP RECOVER Page 3 of 8
Command ===>

Abackup Control Dataset: 'MHLRES4.ABARS.OUTPUT.C.C01V0001'

Unit Name . . 3590-1
X


Volumes ===> TST022
Y
===> ===> ===>
===> ===> ===> ===>
===> ===> ===> ===>
===> ===> ===>









Use ENTER to Continue;
...
...
ARC6071I VOLUMES USED FOR INSTRUCTION/ACTIVITY LOG FILE
MHLRES4.ABARS.OUTPUT.I.C01V0001 DURING AGGREGATE BACKUP FOR
AGGREGATE GROUP MHLRES4 ARE:
X
TST022
...
...

162
z/OS V1R8 DFSMS Technical Update
Figure 6-23 shows you panel 7 of 8 from AGGREGATE GROUP RECOVER. Here you can
choose one specific data set to restore.
Figure 6-23 ARECOVER panel 7 of 8
In Figure 6-23, you can choose from
X
and specify one data set to recover (this is our test at
this time) or from
Y
and specify the name of the data set that consists of names in the
aggregate group to be used for recovery.
For the List Of Names Data Set you can specify at
Y
(Figure 6-23), there are the following
rules:
Must be a sequential file
Can be fixed format, block size=80 and length=80 or fixed blocked format, any multiple of
80, length=80
Must be cataloged
Only fully qualified data set names accepted
PDS member name not accepted
GDG data sets - only fully qualified accepted (no relative generation here)
Aliases not supported
Panel Utilities Help
ssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssss
AGGREGATE GROUP RECOVER Page 7 of 8
Command ===>

ABACKUP CONTROL DATASET:
'MHLRES4.ABARS.OUTPUT.C.C01V0001'

Specify Data Set Name for Individual Recovery:

X
Single Data Set Name . . . . 'MHLRES4.SC64.SPFTEMP2.CNTL'
(1 to 44 Characters)
Y
List Of Names Data Set . . .
(1 to 44 Characters)








Use ENTER to Continue;

Chapter 6. DFSMShsm enhancements
163
Panel 8 of 8 from AGGREGATE GROUP RECOVER on Figure 6-24 shows you the set of
commands that DFSMShsm executes through the ABARS task.
Figure 6-24 ARECOVER panel 8 of 8
After this recovery, you see a message like the one in Figure 6-25 telling you about your
success by RETCODE = 000. It will be shown on your current session and will be recorded in
the recover log. For details about how the recover log name is created, see Table 6-3 on
page 164.
Figure 6-25 Message indicating ARECOVER PROCESSING ENDED
You can receive other messages, indicating a failure or a non-ZERO return code. In this case,
you can search on your environment the high level qualifier HSMACT. This is the default used
by DFSMShsm DASD activity logs. It may be the case that in your installation, it had been
changed using a patch. Check with your DFSMShsm administrator about that. In our case,
the recover log was recorded in data set
HSMACT.H2.ARECOVER.MHLRES4.D07051.T202329. See the Table 6-3 on page 164
about the conventions used for activity log names. If you want more details regarding this
Panel Utilities Help
ssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssss
AGGREGATE GROUP RECOVER Page 8 of 8
Command ===>


DFSMShsm Command and Processing Option:
NOWAIT ARECOVER
EXECUTE
DATASETNAME('MHLRES4.ABARS.OUTPUT.C.C01V0001')
MIGRATEDDATA(SOURCELEVEL)
UNIT(3590-1) VOLUMES(...)
ONLYDATASET(NAME('MHLRES4.SC64.SPFTEMP2.CNTL'))




Enter 1 to Submit DFSMShsm ARECOVER COMMAND
Enter 2 to Save Generated ARECOVER PARAMETERS

Select Option . . 1 (1=SUBMIT, 2=SAVE)

Use ENTER to Perform Selection;
ARC6103I AGGREGATE RECOVERY HAS COMPLETED FOR AGGREGATE
ARC6103I (CONT.) GROUP MHLRES4,
ARC6103I (CONT.) USING CONTROL FILE DATA SET
ARC6103I (CONT.) MHLRES4.ABARS.OUTPUT.C.C01V0001,
ARC6103I (CONT.) AT 20:24:37, RETCODE = 000
ARC1000I MHLRES4.ABARS.OUTPUT.C.C01V0001 ARECOVER PROCESSING ENDED
Note: Only the main panels from the ARECOVER were shown here. You can navigate
there and see all other options.

164
z/OS V1R8 DFSMS Technical Update
point, see the manual DFSMShsm Implementation and Customization Guide, SC35-0418,
under the topic “Controlling the Device Type for the Activity Logs.”
Table 6-3 provides details about the activity log data sets names conventions.
Table 6-3 Type of activity log name
Messages
With z/OS DFSMS V1.8 there are updated messages and a new one, ARC6411E. See
details on Table 6-4.
Table 6-4 Changed messages
For more details about messages in Table 6-4, refer to z/OS MVS System Messages Vol 2
(ARC-ASA), SA22-7632.
Coexistence and migration
The new algorithm of ARECOVER on z/OS V1.8 made it possible to specify a single fully
qualified data set name to recover from an AGGREGATE GROUP. See Table 6-5 for details.
Table 6-5 Coexistence table for ARECOVER new option
ABARS mcvtactn.Hmcvthost.function.agname.Dyyddd.Thhmmss
All other types mcvtactn.Hmcvthost.function.Dyyddd.Thhmmss
mcvtactn Activity log high-level qualifier (in our test, HSMACT (This is the default.))
H, D, and T Constants
mcvthost Identifier for the DFSMShsm host that creates these activity logs
function ABACKUP, ARECOVER, CMDLOG, BAKLOG, DMPLOG, or MIGLOG
agname Aggregate group name
yyddd Year and day of allocation
hhmmss Hour, minute, and second of allocation
ARC6173E Added LOCATE (indicating the data set not cataloged) and OBTAIN (indicating the
data set received an OBTAIN error).
ARC6309E Added LISTOFNAMES keyword and explanation.
ARC6310E Added LISTOFNAMES keyword and explanation.
ARC6311I Added LISTOFNAMES keyword and explanation and change the type code from “E”
to “I”.
ARC6312I Added LISTOFNAMES keyword and explanation and change the type code from “E”
to “I”.
|
ARC6411E
Data set data-set-name1 specified on the ONLYDATASET with the NAME parameter
on ARECOVER command is not identified as being part of the aggregate group being
recovered.
Element or feature DFSMShsm
When change was introduced z/OS V1R8.0
Applies to migration from None

Chapter 6. DFSMShsm enhancements
165
6.4 New command ALTERPRI
There is a new DFSMShsm command, ALTERPRI, to alter the priority of queued requests.
You can alter the priority of the following request types:
ABACKUP
ARECOVER
BACKDS
BACKVOL
DELETE
FRBACKUP
FREEVOL
FRRECOV
MIGRATE
RECALL
RECOVER
You cannot alter the priority of BACKVOL CDS commands and requests that have already
been selected for processing. You should use the ALTERPRI command to alter the priority of
queued requests on an as-needed basis. You should not use this command as the primary
means of assigning priority values to new requests.
There are two options of priority here:
The HIGH parameter, which is the default, alters the specified request so that it has the
highest priority on its respective queue.
Conversely, the LOW parameter alters the request so that it has the lowest priority on its
respective queue.
The mutually exclusive REQUEST, USER, or DATASETNAME parameters indicate which
requests DFSMShsm should re-prioritize. Use the QUERY REQUEST command to
determine the request number to issue on the ALTERPRI command. DFSMShsm
re-prioritizes all queued requests that match the REQUEST, USERID, or DATASETNAME
criteria specified on the ALTERPRI command.
To re-prioritize a recall request on the Common Recall Queue, issue the ALTERPRI
command on the same host that originated the recall request.
Timing After the first IPL of z/OS V1R8.0
Is the migration action required? None
Target system hardware requirements None
Target system software requirements None
Other system (coexistence or fallback)
requirements
None
Restrictions None
System impacts None
Element or feature DFSMShsm

166
z/OS V1R8 DFSMS Technical Update
Figure 6-26 shows ALTERPRI syntax.
Figure 6-26 Output of ALTERPRI command keywords
The following command alters all requests for a particular data set so that they have the
highest priority on their respective queues:
ALTERPRI DATASETNAME(dsname)
The following command alters all requests with a particular request number so that they have
the highest priority on their respective queue:
ALTERPRI REQUEST(reqnum) HIGH
The following command alters all requests issued by a particular user so that they have the
lowest priority on their respective queues:
ALTERPRI USERID(userid) LOW
Figure 6-27 shows the result of a query command, See that MHLRES4.TCPIP.PROFILE
X
is
in the last position in the recall queue.
Figure 6-27 Output of command F DFHSM64,Q REQUEST before ALTERPRI
>>__ALTERPRI__ _REQUEST(reqnum)___________ __ __________ ___________><
|_USERID(userid)____________| | _HIGH_ |
|__ _DATASETNAME_ _(dsname)_| |_|_LOW__|_|
|_DSNAME______|
-F DFHSM64,Q REQUEST
STC20554 ARC0101I QUERY REQUEST COMMAND STARTING ON HOST=2
STC20554 ARC1543I RECALL MWE FOR DATASET MHLRES4.DUMP.OUT.#1,
ARC1543I (CONT.) FOR USER MHLRES4, REQUEST 00000211, WAITING TO BE
ARC1543I (CONT.) PROCESSED ON A COMMON QUEUE,00000000 MWES AHEAD OF
ARC1543I (CONT.) THIS ONE
STC20554 ARC1543I RECALL MWE FOR DATASET MHLRES4.DUMP.OUT.#2,
ARC1543I (CONT.) FOR USER MHLRES4, REQUEST 00000212, WAITING TO BE
ARC1543I (CONT.) PROCESSED ON A COMMON QUEUE,00000001 MWES AHEAD OF
ARC1543I (CONT.) THIS ONE
STC20554 ARC1543I RECALL MWE FOR DATASET MHLRES4.DUMP.OUT.#3,
ARC1543I (CONT.) FOR USER MHLRES4, REQUEST 00000213, WAITING TO BE
ARC1543I (CONT.) PROCESSED ON A COMMON QUEUE,00000002 MWES AHEAD OF
ARC1543I (CONT.) THIS ONE
STC20554 ARC1543I RECALL MWE FOR DATASET MHLRES4.TCPIP.PROFILE
X
,
ARC1543I (CONT.) FOR USER MHLRES4, REQUEST 00000214, WAITING TO BE
ARC1543I (CONT.) PROCESSED ON A COMMON QUEUE,00000003 MWES AHEAD OF
ARC1543I (CONT.) THIS ONE

Chapter 6. DFSMShsm enhancements
167
To demonstrate our test, we issue an ALTERPRI command and move
MHLRES4.TCPIP.PROFILE
X
(see Figure 6-27 on page 166) to the highest position.
Figure 6-28 shows the results of the ALTERPRI command.
Figure 6-28 Output of F DFHSM64,ALTERPRI command
After issuing the ALTERPRI command the file MHLRES4.TCPIP.PROFILE
X
is now on the
top.
Figure 6-29 shows the result of a QUERY command. Note that MHLRES4.TCPIP.PROFILE
X
is the first one to recall.
Figure 6-29 Output of F DFHSM64,Q REQUEST after the ALTERPRI command
The first recall results after changing the queue priority are shown in Figure 6-30.
Figure 6-30 Output of first results after change the recall priority
For more details about this command, you can refer to z/OS V1R8.0 DFSMS Storage
Administration Reference (for DFSMShsm, DFSMSdss, DFSMSdfp).
00000290 F DFHSM64,ALTERPRI REQUEST(00000214) HIGH
00000090 ARC0980I ALTERPRI REQUEST COMMAND STARTING
00000090 ARC0982I RECALL MWE FOR DATA SET MHLRES4.TCPIP.PROFILE 546
00000090 ARC0982I (CONT.) FOR USER MHLRES4, REQUEST 00000214, REPRIORITIZED TO
00000090 ARC0982I (CONT.) HIGH
00000090 ARC0981I ALTERPRI REQUEST COMMAND COMPLETED, RC=0000
-F DFHSM64,Q REQUEST
STC20554 ARC0101I QUERY REQUEST COMMAND STARTING ON HOST=2
STC20554 ARC1543I RECALL MWE FOR DATASET MHLRES4.TCPIP.PROFILE
X
,
ARC1543I (CONT.) FOR USER MHLRES4, REQUEST 00000214, WAITING TO BE
ARC1543I (CONT.) PROCESSED ON A COMMON QUEUE,00000000 MWES AHEAD OF
ARC1543I (CONT.) THIS ONE
STC20554 ARC1543I RECALL MWE FOR DATASET MHLRES4.DUMP.OUT.#1,
ARC1543I (CONT.) FOR USER MHLRES4, REQUEST 00000211, WAITING TO BE
ARC1543I (CONT.) PROCESSED ON A COMMON QUEUE,00000001 MWES AHEAD OF
ARC1543I (CONT.) THIS ONE
STC20554 ARC1543I RECALL MWE FOR DATASET MHLRES4.DUMP.OUT.#2,
ARC1543I (CONT.) FOR USER MHLRES4, REQUEST 00000212, WAITING TO BE
ARC1543I (CONT.) PROCESSED ON A COMMON QUEUE,00000002 MWES AHEAD OF
ARC1543I (CONT.) THIS ONE
STC20554 ARC1543I RECALL MWE FOR DATASET MHLRES4.DUMP.OUT.#3,
ARC1543I (CONT.) FOR USER MHLRES4, REQUEST 00000213, WAITING TO BE
ARC1543I (CONT.) PROCESSED ON A COMMON QUEUE,00000003 MWES AHEAD OF
ARC1543I (CONT.) THIS ONE
STC20554 ARC0101I QUERY REQUEST COMMAND COMPLETED ON HOST=2
ARC1000I MHLRES4.TCPIP.PROFILE RECALL PROCESSING ENDED
ARC0612I VOLUME MOUNT ISSUED FOR RECALL OR RECOVER OF MHLRES4.DUMP.OUT.#1

168
z/OS V1R8 DFSMS Technical Update
Protecting ALTERPRI command
Each storage administrator command can be protected through the following fully qualified
discrete FACILITY class profile:
STGADMIN.ARC.command
In this case, a security administrator can create the fully qualified, discrete profile to authorize
this command to storage administrators:
STGADMIN.ARC.ALTERPRI
You can find more details like the entire command list and specific RACF profiles in the
manual DFSMShsm Implementation and Customization Guide, SC35-0418.
Coexistence and migration
See Table 6-6 for details.
Table 6-6 Coexistence table for ALTERPRI new command
Tip: Search for ALTERPRI in the DFSMShsm Implementation and Customization Guide,
SC35-0418. You will find an item about protecting the DFSMShsm FACILITY.
Element or feature DFSMShsm
When change was introduced z/OS V1R8.0
Applies to migration from None
Timing After the first IPL of z/OS V1R8.0
Is the migration action required? None
Target system hardware requirements None
Target system software requirements None
Other system (coexistence or fallback)
requirements
None
Restrictions None
System impacts None

© Copyright IBM Corp. 2008. All rights reserved.
169
Chapter 7.
DFSMShsm fast replication
The DFSMShsm fast replication function in DFSMS V1R8 allows you to dump fast replication
backup copies to tape (through operator commands or automatic dump processing), and
allows you to restore individual data sets from copy pool backup copies.
The following topics are covered:
Fast replication review
Backup and recovery of copy pools
Tape support
Data set recovery
Reporting on the DFSMShsm fast replication environment
Security for DFSMShsm fast replication
Audit and error recovery
7

170
z/OS V1R8 DFSMS Technical Update
7.1 Fast replication overview
Point in time is the appearance of an almost instantaneous volume copy.
Fast Data Replication occurs so fast because it builds a map, with pointers, to the source
volume tracks or extents. There is no longer a need to wait for the physical copy to complete
before applications can resume their access to the data. Both the source and target data are
available for read/write access almost immediately, while the copy process continues in the
background. This process guarantees that the contents of the target volume are an exact
duplicate of the source volume at that point in time. You can back up, recover to, and resume
processing from that point in time.
Point-in-time copy is a mirroring technology that provides an instant copy of the data. It is a
duplication of source files by means of mirrored disks. In other words, all data residing on disk
is mirrored to another disk residing on the same storage subsystem. Most 24x7 database
operations require a mirror solution. It has no impact on application performance, and
provides immediate access to a copy of the mirrored data. This access to a copy of mirrored
data, which can be obtained by various techniques with a minimal impact on mirroring, is one
of the main returns on investment that many companies practice.
Point-in-time copy is also known as a T0 (time-zero) copy. After the copy is finished, the
connection between source and mirror is split off. The mirror can be mounted to another host
and can be used for backup and recovery purposes.
A Split of a Continuous Mirror is caused by the creation and maintenance of a mirror
relationship between the source and the target volumes. This relationship occurs on-site, as
opposed to a remote environment. The target volume is not accessible until a split or detach
of the mirror relationship is initiated. A Split of a Continuous Mirror guarantees that the
contents of the target volume are an exact duplicate of the source volume at the point in time
where the split occurs. After the split is complete, you can back up, recover to, and resume
processing from that point in time.
In the past, volume level point-in-time copies have been used almost exclusively as disaster
recovery backups. With new software services that are available, these copies, clones, or
mirrors can be used for job restart, data mining, and application testing.
Point-in-time hardware support
The DFSMShsm fast replication line item supports the volume-level FlashCopy® function of
the IBM System Storage™ DS8000™, IBM TotalStorage Enterprise Storage Server® (ESS)
DASD, SnapShot feature of the IBM RAMAC® Virtual Array (RVA) and STK Shared Virtual
Array™ (SVA), and any other DASD that support FlashCopy APIs.
The process of creating a point-in-time copy has two phases. The first phase completes in a
few seconds by constructing a map to describe the source volume.
At the completion of this first phase, the data has been cloned and both the source and target
volumes are available for read and write access. From the user’s perspective, the source and
target volumes’ contents are an exact duplicate at this point in time, even though actual
copying of data has not actually been initiated.
The second phase consists of copying the physical source volume data to the target volume.
The point-in-time copy is maintained and preserved by an on-demand copy of the data to the
Note: Appropriate microcode levels may be required to activate these features on the
storage device.

Chapter 7. DFSMShsm fast replication
171
target volume. When an update request has been issued for either a source or target volume
data set that has not yet been copied to the target volume, the Fast Data Replication tool
immediately copies the data from the source volume to the target volume before the update is
applied. The possible situations are:
Read request to the source volume
The data is read from the source volume.
Read request to the target volume
If the data has already been copied from the source volume to the target volume, the data
is read from the target volume. If the data has not yet been copied to the target volume,
the data is read from the source volume.
Write request to the source volume
If the data has already been copied from the source volume to the target volume, this
results in a normal write to source volume. However, if the data has not yet been copied
from the source volume to the target volume, the data will be copied from the source
volume to the target volume before the source volume update occurs.
Write request to the target volume
If the data has already been copied from the source volume to the target volume, the write
request results in a normal write to the target volume. However, if the data has not yet
been copied from the source volume to the target volume, the data will be copied to the
target volume before the target volume update occurs.
7.1.1 FlashCopy
FlashCopy provides a point-in-time copy of a logical volume, with almost instant availability
for the application of both the source and target volumes. Only a minimal interruption is
required for the FlashCopy relationship to be established, so the copy operation can be
initiated. The copy is then created in the background by the IBM TotalStorage ESS, with
minimal impact on other ESS activities.
FlashCopy: background copy
By default, FlashCopy performs a background copy. The background copy task makes a
physical copy of all tracks from the source volume to the target volume. De-staging
algorithms are used to efficiently manage the background ESS copy process. The
background copy task runs at a lower priority than normal I/O on the ESS, so as not to affect
the normal application I/O processing.
The ESS, using the metadata structure that was created during establish, keeps track of
which data has been copied from the source to the target and manages the integrity of both
copies. If an application wants to read some data from the target that has not yet been
copied, the data is read from the source. Otherwise, the read can be satisfied from the target
volume.
Before updating a not-yet-copied track on the source volume, the ESS performs an on
demand copy of the track to the target volume. Subsequent reads to this track on the target
volume are satisfied from the target volume.
Before updating a not-yet-copied track on the target volume, the ESS performs an on
demand copy of this track to the target volume. This on demand activity is done
Note: FlashCopy is an optional feature that must be enabled in the DS8000.

172
z/OS V1R8 DFSMS Technical Update
asynchronously after the host has received I/O completion (after the data is written in cache
and NVS), so host I/O is not delayed.
After some time, when all tracks have been copied to the target volume, the FlashCopy
relationship automatically ends unless the persistent FlashCopy option was specified.
As Figure 7-1 illustrates, a FlashCopy relationship goes through three stages:
Establishing the relationship
Copying the data
Terminating the relationship
Figure 7-1 FlashCopy with background copy
FlashCopy: no background copy
When selecting not to perform the background copy, the relationship is established but the
background copy task — of all source volume tracks — is not initiated. Only the source tracks
that receive application updates are copied to the target. Before updating a track on the
source volume, the ESS performs an on demand copy of the track to the target volume, thus
preserving the T0 copy. Similarly, before updating a track on the target volume, the ESS
performs an on demand copy of this track to the target volume.
A FlashCopy relationship that was established selecting no-background remains active until
one of the following occurs:
An explicit FlashCopy withdraw is done to terminate the relationship.
All source device tracks have been copied on the target device because they were all
updated.
All target device tracks have been updated by user applications.
Both source and target volumes
immediately available
Read and write to both source and
target volumes possible
Write
Read
When copy is complete,
relationship between
source and target ends
Time
Source
Target
Bitmap
FlashCopy requested
FlashCopy relationship is
established

Chapter 7. DFSMShsm fast replication
173
When a no-background copy FlashCopy relationship is terminated, the target volume is left in
an indeterminate state. Some of the tracks on the volume may contain data from the source
volume, while other tracks may contain residual data that was on the target volume before the
copy. The volume should not be used in these conditions unless it is reformatted or used for
another copy operation.
FlashCopy Version 1
FlashCopy V1 is invoked at volume level. The following considerations apply:
The source and target volumes must have the same track format.
The target volume must be as large as the source volume.
The source and target volumes must be within the same ESS logical subsystem (LSS).
A source and a target volume can only be involved in one FlashCopy relationship at a
time.
As soon as a FlashCopy establish command is issued (either invoked by a TSO command,
the DFSMSdss utility, the ANTRQST macro, or by means of the ESS Copy Services Web
user interface (WUI)), the ESS establishes a FlashCopy relationship between the target
volume and the source volume.
FlashCopy Version 2
FlashCopy Version 2 supports all of the FlashCopy V1 functions plus these enhancements:
FlashCopy V2 can be used for data set copies, as well as volume copies.
The source and target of a FlashCopy can be on different LSSs within a DS8000 or an
ESS.
Multiple FlashCopy relationships are allowed.
Incremental copies are possible.
Inband commands can be sent over PPRC links to a remote site.
FlashCopy consistency groups can be created.

174
z/OS V1R8 DFSMS Technical Update
In addition, there has been a reduction in the FlashCopy establish times. Figure 7-2
compares the features of FlashCopy Version 1 and FlashCopy Version 2.
Figure 7-2 FlashCopy V1 and FlashCopy V2 comparison
7.1.2 DFSMShsm fast replication
DFSMShsm was enhanced in z/OS DFSMS V1.5 to manage full volume fast replication
backup versions. DFSMShsm fast replication supports both FlashCopy Version 1 and
FlashCopy Version 2.
Copy pool constructs
Copy pools are defined through a new SMS construct named copy pool. This construct
enables customers to define which storage group should be processed collectively for fast
replication functions and specifies how many backup versions of the pool DFSMShsm should
maintain. When the maximum number of versions has been reached, the oldest version is
rolled off.
SMS also introduced a copy pool backup storage group type that is used to define which
volumes DFSMShsm may use as the target volumes of the fast replication backup versions.
Restriction: When you are using FlashCopy Version 1, any subsequent FRBACKUP
COPYPOOL command that is issued against the same copy pool or against another copy
pool with a common storage group before all of the background copies are complete
causes the command to fail. This restriction does not exist for FlashCopy Version 2
because a source volume can be in multiple concurrent relationships.
FlashCopy V1
Original
FlashCopy V1
Updated
FlashCopy V2
Lic 2.2.0
COPY or
NOCOPY
COPY, NOCOPY
N
OCOPY->COPY
COPY, NOCOPY
NOCOPY->COPY
Ended
Ended / Persistent
Ended / Persistent
Full
Full
Full or Incremental
Single
Single
Multiple
Volumes, Tracks
Volumes, Tracks
Volumes, Tracks,
Data Set
Single
Single
Multiple
Same LSS
Same LSS
Cross LSS
Cross Cluster
Applies to DS8000 and ESS models F10, F20, and 800
Background
Copy Mode
Disposition
Data Movement
Relationship(s)
Granularity
Data Set
Source/Target
Boundries

Chapter 7. DFSMShsm fast replication
175
DFSMShsm invokes the DFSMSdss COPY FULL function to perform a full volume copy of
the data from the source devices in a copy pool to the target devices in a copy pool backup
storage group.
DFSMShsm invokes the DFSMSdss physical data set COPY function to perform a fast
replication recovery at the data set level from the target devices in a copy pool backup
storage group to the source devices in a copy pool.
Related DFSMShsm commands
Three new DFSMShsm commands were being introduced in z/OS DFSMS V1.5 to support
this new function:
FRBACKUP
Create a fast replication backup version for each volume in a specified copy pool. This
command is not supported as part of automatic backup or dump. It is invoked by entering
the command at the operator console, HSEND TSO command, and batch or by a program
through the ARCHSEND macro interface.
FRDELETE
Delete one or more unneeded fast replication backup versions.
FRRECOV
Use fast replication to recover a data set, single volume, or a pool of volumes from the
managed backup versions.
ISMF is updated to support these enhancements to SMS. The DFSMShsm LIST and QUERY
commands and the ARCXTRCT macro are also modified to aid you in the use and monitoring
of the fast replication backup versions.
DFSMSdss support
DFSMShsm invokes DFSMSdss to invoke fast replication support. The control is returned to
DFHSMhsm after the fast replication has been initiated for all of the volumes (within minutes).
If a volume failed, DFSMShsm messages are issued.
DFSMSdss was enhanced to support the following functions:
Individual data set restore for extended format VSAM data sets, including extended format
linear striped data sets, which are often used by DB2 customers.
Provides a new FASTREPLICATION(REQUIRED) keyword.
Callers of DFSMShsm may specify the new DFSMShsm keyword NOVTOCENQ to
indicate no volume serialization (no reserve on the VTOC) to DFSMSdss.
Bypass security verification options during full volume copy.
What is a copy pool
A copy pool is an SMS construct that defines which storage groups should be processed
collectively by the DFSMShsm function. A copy pool can contain the names of up to 256
Note: Recovery can be performed at the volume or copy pool level, and since z/OS
DFSMS V1.8 at the data set level.
Note: The NOVTOCENQ must only be used when another utility, such as DB2, is
providing the serialization. If it is used improperly, the results can be disastrous.

176
z/OS V1R8 DFSMS Technical Update
storage groups that should be processed collectively for fast replication. An optional
parameter can be used to specify how many backup versions of the pool DFSMShsm should
maintain.
The name of this new construct may be up to 30 characters in length. The first character of
the name can be an uppercase alphabetic or national character. Any of the remaining
characters can be uppercase alphabetic, national, or numeric.
When defining a copy pool, the individual source pool storage group names are recorded.
The actual volumes that are associated with each storage group are retrieved during
subsequent processing. Consequently, the volumes that are processed during a function,
such as fast replication backup, may or may not be the same volumes that are associated
with the storage group at the time the copy pool was established. For example, if one or more
volumes are added to a storage group within a copy pool, then those volumes are processed
as part of the next fast replication backup automatically. There is no need to update the copy
pool definition unless storage groups are to be added or removed from the pool.
An individual storage group may be associated with more than one copy pool. Due to
implementation limitations, DFSMShsm must limit to 50 the number of copy pools with which
a storage group may be associated.
As many as 85 backup versions may be maintained for each copy pool. Each version that is
maintained requires a unique target volume for each source volume. If specifying five backup
versions, then five target volumes must be available for each source volume that is
associated with the copy pool. All target volumes must be available when the backup
command is issued.
With ESS FlashCopy Version 1, the source and target volumes must be in the same Logical
Subsystem (LSS), which is limited to a maximum of 256 total volumes.
With ESS FlashCopy Version 2, the source and target volumes may be in separate LSSs.
Restriction: An individual storage group can be associated with more than one copy pool,
but with no more than 50. Although you can include a storage group in multiple copy pools,
IBM recommends against doing so when FlashCopy Version 1 is used as the fast
replication utility. Each individual copy pool with its associated background copies for all of
the volumes in the common storage group must process completely before the next copy
pool can be processed. This processing can take several hours and prevents the volumes
in the non-common storage groups in those copy pools from having a timely backup
created. (FlashCopy Version 2 and SnapShot do not have this restriction.)
Restriction: You can maintain as many as 85 backup versions for each copy pool. Each
maintained version requires a unique target volume for each source volume. If you specify
five dedicated versions, there must be five target volumes that are available for each
source volume that is associated with the copy pool. All target volumes must be available
when you issue the FRBACKUP COPYPOOL(cpname) PREPARE command. If the
PREPARE function is not performed, the target volumes need not be available until the
actual backup version is created.
Restriction: With ESS FlashCopy Version 1, the source and target volumes must be in the
same LSS, which is limited to 256 total volumes. Thus, for three unique versions, you
could have up to 64 source volumes, leaving 192 volumes available as target volumes.
The source volumes within a storage group can span logical and physical subsystems.

Chapter 7. DFSMShsm fast replication
177
What is a copy pool backup storage group
SMS storage group types consist of Pool, VIO, Dummy, Tape, Object, and Object Backup. A
new storage group type, copy pool backup, is introduced to specify candidate target volumes
where the copies of the pool storage groups are kept.
Volumes that are associated with the new copy pool backup storage group are for
DFSMShsm use. SMS SCDS validation fails if a copy pool backup storage group is assigned
for allocation by the SMS ACS storage group selection routines.
The SMS status of storage group and volume are used during data set allocation and
end-of-volume extend. The SMS status of a copy pool backup storage group and volume are
not used by DFSMShsm during a fast replication request. (For example, a status of Disable
New or Disable All does not prevent DFSMShsm from processing the volume.) You may
restrict DFSMShsm access to a volume in a copy pool backup storage group by varying the
volume off-line to MVS.
A new field is provided in the storage group definition for POOL storage groups to specify the
name of the associated copy pool backup storage group. Only pool storage groups may
define associated copy pool backup storage groups. A pool storage group may specify its
own copy pool backup storage group or specify a copy pool backup storage group that is
shared by multiple pool storage groups.
For each source volume in a storage group to be copied, there must exist enough eligible
target volumes in the copy pool backup storage group to satisfy the needs of the number of
specified backup versions.
An eligible target volume must:
Have the same track format as the source volume.
Be the exact size of the source volume.
For FlashCopy:
Not also be a primary or secondary volume in an XRC or PPRC volume pair.
– For Version 1, reside in the same LSS as the source volume.
– For Version 1, at the time of the backup, not be in a FlashCopy relationship.
For SnapShot:
Reside in the same RVA/SVA as the source volume.

178
z/OS V1R8 DFSMS Technical Update
ISMF panels
The Interactive Storage Management Facility (ISMF) provides a series of applications for
storage administrators to define and manage the SMS configuration. Figure 7-3 shows the
ISMF primary option menu for storage administrators.
This primary option menu differs from the one that users see. In addition to the options found
on the ISMF Primary Option Menu for users, the primary option menu for storage
administrators includes:
Storage Group
Automatic Class Selection
Control Data Set
Library Management
Data Collection
Figure 7-3 ISMF Primary Option Menu
7.1.3 Preparing for fast replication
Your storage administrator must define copy pools and associated source and backup
storage groups in order to use the DFSMShsm fast replication function.
The following steps should be used to define the necessary copy pool construct, copy pool
backup storage group, and associate existing pool storage groups for fast replication support.
Panel Help
------------------------------------------------------------------------------
ISMF PRIMARY OPTION MENU - z/OS DFSMS V1 R8
Enter Selection or Command ===>

Select one of the following options and press Enter:
0 ISMF Profile - Specify ISMF User Profile
1 Data Set - Perform Functions Against Data Sets
2 Volume - Perform Functions Against Volumes
3 Management Class - Specify Data Set Backup and Migration Criteria
4 Data Class - Specify Data Set Allocation Parameters
5 Storage Class - Specify Data Set Performance and Availability
6 Storage Group - Specify Volume Names and Free Space Thresholds
7 Automatic Class Selection - Specify ACS Routines and Test Criteria
8 Control Data Set - Specify System Names and Default Criteria
9 Aggregate Group - Specify Data Set Recovery Parameters
10 Library Management - Specify Library and Drive Configurations
11 Enhanced ACS Management - Perform Enhanced Test/Configuration Management
C Data Collection - Process Data Collection Function
L List - Perform Functions Against Saved ISMF Lists
P Copy Pool - Specify Pool Storage Groups for Copies
R Removable Media Manager - Perform Functions Against Removable Media
X Exit - Terminate ISMF
Use HELP Command for Help; Use END Command or X to Exit.

Chapter 7. DFSMShsm fast replication
179
Defining copy pools
Define your copy pool definitions by selecting option P Copy Pool from the ISMF Primary
Option Menu shown in Figure 7-3 on page 178.
The Copy Pool Application Selection Panel appears as shown in Figure 7-4. You can work
with existing copy pools or define a new one by entering the copy pool name and selecting
option 3 Define a Copy Pool.
Figure 7-4 Copy Pool Application Selection panel
Panel Utilities Help
------------------------------------------------------------------------------
COPY POOL APPLICATION SELECTION
Command ===>

To perform Copy Pool Operations, Specify:
CDS Name . . . . SYS1.SMS.SCDS
(1 to 44 character data set name or 'Active' )
Copy Pool Name CP1 (For Copy Pool List, fully
or partially specified or * for all)

Select one of the following options :
3 1. List - Generate a list of Copy Pools
2. Display - Display a Copy Pool
3. Define - Define a Copy Pool
4. Alter - Alter a Copy Pool

If List Option is chosen,
Enter "/" to select option Respecify View Criteria
Respecify Sort Criteria


Use ENTER to Perform Selection;
Use HELP Command for Help; Use END Command to Exit.

180
z/OS V1R8 DFSMS Technical Update
The Copy Pool Define panel appears, as shown in Figure 7-5. Specify the dump classes to be
used with this copy pool, and you can specify the number of backup versions that you want to
keep of this copy pool. The default is two copies.
Figure 7-5 Copy Pool Define panel (page 1 of 2)
Note: DFSMShsm can keep up to 85 backup versions for each copy pool. We recommend
keeping at least two versions. Before a new version is created, the oldest version is
invalidated and the target volumes from that version are used as targets for the new
version. If the new version fails for some reason, then only n-1 valid versions are available,
where n is the number of requested versions. Maintaining two versions ensures that at
least one valid version is always available.
We recommend that if n backups are required, n+1 should be kept.
Panel Utilities Scroll Help
------------------------------------------------------------------------------
COPY POOL DEFINE Page 1 of 4
Command ===>

SCDS Name . . : SYS1.SMS.SCDS
Copy Pool Name : CP1

To DEFINE Copy Pool, Specify:
Description ==> COPY POOL 1
==>

Auto Dump . . . N (Y or N) Dump Sys/Sys Group Name . . .
Dump Class . . Dump Class . .
Dump Class . . Dump Class . .
Dump Class . .

Number of Recoverable DASD Fast
Replicate Backup Versions . . . . 2 (0 to 85 or blank)



Use ENTER to Perform Verification; Use DOWN Command to View next Panel;
Use HELP Command for Help; Use END Command to Save and Exit; CANCEL to Exit.

Chapter 7. DFSMShsm fast replication
181
Scroll down to reach the next page (shown in Figure 7-6) in order to specify the Storage
Groups to include in the copy pool. You can specify up to 256 Storage Groups in a copy pool.
Scroll down to pages 3 and 4 if needed.
Figure 7-6 Copy Pool Define panel (page 2 of 2)
Panel Utilities Scroll Help
------------------------------------------------------------------------------
COPY POOL DEFINE Page 2 of 4
Command ===>

SCDS Name . . : SYS1.SMS.SCDS
Copy Pool Name : CP1

To DEFINE Copy Pool, Specify:
Storage Group Names: (specify 1 to 256 names)
==> SG1
==>
==>
==>
==>
==>
==>
==>
==>
==>
==>

Use ENTER to Perform Verification; Use UP/DOWN Command to View other Panels;
Use HELP Command for Help; Use END Command to Save and Exit; CANCEL to Exit.
Note: You can specify Storage Groups not yet defined in the SCDS, but this makes the
SCDS invalid for activation. The Storage Groups specified must be defined in the SCDS
before activating the configuration.

182
z/OS V1R8 DFSMS Technical Update
Defining copy pool backup storage groups
You now have to define a copy pool backup storage group. This storage group is designated
as the target where the copies of the Pool Storage Groups will be kept. Start to define your
copy pool backup definitions by selecting option 6 Storage Group from the ISMF Primary
Option Menu shown in Figure 7-3 on page 178. Figure 7-7 shows the Storage Group
Application Selection panel.
Figure 7-7 Storage Group Application Selection panel
The Storage Group Application Selection panel enables you to:
Specify the storage group name for the copy pool backup.
Specify the storage group type of copy pool backup.
Associate your existing source pool storage groups with the copy pool backup storage
group where you want to keep your copies.
After entering the chosen storage group name and storage group type of copy pool backup,
select option 2 Define - Define a Storage Group.
When the copy pool backup storage group is defined you should also define volumes in it.
Here you need to know which pool storage group should be associated with this copy pool
backup storage group and the number of versions that are to be kept for the pool storage
group. The number of versions is defined in the related copy pool definition. For example, if
your pool storage group contains 10 volumes and you want to keep two backup versions,
then the copy pool backup storage group must contain 20 target volumes.
If the pool storage group has extended or overflow storage groups defined, you must also
define target volumes for those. For example, if the pool storage group with 10 volumes has
an overflow storage group with 5 volumes, you must define 30 target volumes in the copy
pool backup storage group for two backup versions.
Panel Utilities Help
------------------------------------------------------------------------------
STORAGE GROUP APPLICATION SELECTION
Command ===>

To perform Storage Group Operations, Specify:
CDS Name . . . . . . SYS1.SMS.SCDS
(1 to 44 character data set name or 'Active' )
Storage Group Name CPBSG1 (For Storage Group List, fully or
partially specified or * for all)
Storage Group Type COPY POOL BACKUP (VIO, POOL, DUMMY, COPY POOL BACKUP,
OBJECT, OBJECT BACKUP, or TAPE)

Select one of the following options :
2 1. List - Generate a list of Storage Groups
2. Define - Define a Storage Group
3. Alter - Alter a Storage Group
4. Volume - Display, Define, Alter or Delete Volume Information

If List Option is chosen,
Enter "/" to select option Respecify View Criteria
Respecify Sort Criteria
Use ENTER to Perform Selection;
Use HELP Command for Help; Use END Command to Exit.

Chapter 7. DFSMShsm fast replication
183
After defining the copy pool backup storage group, you must alter your existing pool storage
groups to define the copy pool backup storage group where you want to keep their copies.
Start by selecting option 6 Storage Group from the ISMF Primary Option Menu, as shown in
Figure 7-3 on page 178. The Storage Group Application Selection Panel appears. Specify the
desired storage group name and storage group type. Then select option 3 Alter (Figure 7-8).
Figure 7-8 Storage Group Application Selection panel: Alter a Storage Group
Panel Utilities Help
------------------------------------------------------------------------------
STORAGE GROUP APPLICATION SELECTION
Command ===>

To perform Storage Group Operations, Specify:
CDS Name . . . . . . SYS1.SMS.SCDS
(1 to 44 character data set name or 'Active' )
Storage Group Name SG1 (For Storage Group List, fully or
partially specified or * for all)
Storage Group Type POOL (VIO, POOL, DUMMY, COPY POOL BACKUP,
OBJECT, OBJECT BACKUP, or TAPE)

Select one of the following options :
3 1. List - Generate a list of Storage Groups
2. Define - Define a Storage Group
3. Alter - Alter a Storage Group
4. Volume - Display, Define, Alter or Delete Volume Information

If List Option is chosen,
Enter "/" to select option Respecify View Criteria
Respecify Sort Criteria
Use ENTER to Perform Selection;
Use HELP Command for Help; Use END Command to Exit.

184
z/OS V1R8 DFSMS Technical Update
The Pool Storage Group Alter panel appears. Specify the desired copy pool backup in the
copy pool backup SG Name field (Figure 7-9).
Figure 7-9 Pool Storage Group Alter panel
A pool storage group can have only one copy pool backup storage group associated with it.
Many pool storage groups can be associated with the same copy pool backup storage group,
so we can have different versions of different pool storage groups in a copy pool backup
storage group. DFSMShsm keeps control of the copies that you have.
From the Pool Storage Group Alter panel, you can also alter the SMS storage group status by
typing Y in the ALTER SMS Storage Group Status field. There is also a change related to
copy pool backup storage group SMS status — it can only have an SMS status of ENABLE or
NOTCON.
The modifications in the source CDS must be made in a z/OS V1.5 or later system.
Panel Utilities Help
------------------------------------------------------------------------------
POOL STORAGE GROUP ALTER
Command ===>

SCDS Name . . . . . : SYS1.SMS.SCDS
Storage Group Name : SG1
To ALTER Storage Group, Specify:
Description ==>
==>
Auto Migrate . . N (Y, N, I or P) Migrate Sys/Sys Group Name . .
Auto Backup . . N (Y or N) Backup Sys/Sys Group Name . .
Auto Dump . . . N (Y or N) Dump Sys/Sys Group Name . . .
Overflow . . . . N (Y or N) Extend SG Name . . . . . . . .
Copy Pool Backup SG Name . . . CPBSG1
Dump Class . . . (1 to 8 characters)
Dump Class . . . Dump Class . . .
Dump Class . . . Dump Class . . .
Allocation/migration Threshold: High . . 85 (1-99) Low . . (0-99)
Guaranteed Backup Frequency . . . . . . (1 to 9999 or NOLIMIT)

ALTER SMS Storage Group Status . . . N (Y or N)
Use ENTER to Perform Verification and Selection;
Use HELP Command for Help; Use END Command to Save and Exit; CANCEL to Exit.

Chapter 7. DFSMShsm fast replication
185
7.2 Backup and recovery of copy pools
We used for our tests an environment with one storage group per copy pool. The copy pool
CP1 was made up of storage group SG1, which was linked with copy pool backup storage
group CPBSG1, as shown in Figure 7-10.
Figure 7-10 Lab environment
The copy pool was defined (see Figure 7-15 on page 192) to hold a maximum of two fast
replication backup versions and to use dump class DCREDB18 for full volume dumps of
target volumes.
In order to show all DFSMShsm messages we used a setting of SETSYS MSGLEVEL(FULL).
7.2.1 Creating a fast replication backup copy
Now we created a fast replication backup version of copy pool CP1 by using the command
shown in Example 7-1.
Example 7-1 Requesting a new version of a fast replication backup
FRBACKUP COPYPOOL(CP1) EXECUTE TOKEN(EX1)
CP1
(COPY POOL)
SG1
(Storgrp)
MHL0A0
MHL0A1
MHL0AC
MHL0AD
MHL0AE
MHL0AF
MHL1AC
MHL1AD
MHL1AE
MHL1AF
MHL2AC
MHL2AD
MHL2AE
MHL2AF
CPBSG1
(Copy Pool Backup Storgrp)

186
z/OS V1R8 DFSMS Technical Update
The DFSMShsm backup log contained the messages shown in Example 7-2. There were no
messages in the dump log because full volume dump was not involved at this point.
Example 7-2 Backup log for fast replication of copy pool CP1 without copy to tape
ARC1801I FAST REPLICATION BACKUP IS STARTING FOR COPY POOL CP1, AT 14:49:20 ON 2007/02/19, TOKEN='EX1'
ARC0640I ARCFRTM - PAGE 0001 5695-DF175 DFSMSDSS V1R08.0 DATA SET SERVICES 2007.050 14:49
ARC0640I ARCFRTM - ADR035I (SCH)-PRIME(06), INSTALLATION EXIT ALTERED BYPASS FAC CLASS CHK DEFAULT TO YES
ARC0640I ARCFRTM - PARALLEL
ARC0640I ARCFRTM - ADR101I (R/I)-RI01 (01), TASKID 001 HAS BEEN ASSIGNED TO COMMAND 'PARALLEL'
ARC0640I ARCFRTM - COPY IDY(MHL0A0) ODY(MHL0AF) DUMPCOND FR(REQ) PUR ALLX ALLD(*)
ARC0640I ARCFRTM - ADR101I (R/I)-RI01 (01), TASKID 002 HAS BEEN ASSIGNED TO COMMAND 'COPY '
ARC0640I ARCFRTM - COPY IDY(MHL0A1) ODY(MHL1AC) DUMPCOND FR(REQ) PUR ALLX ALLD(*)
ARC0640I ARCFRTM - ADR101I (R/I)-RI01 (01), TASKID 003 HAS BEEN ASSIGNED TO COMMAND 'COPY '
ARC0640I ARCFRTM - ADR109I (R/I)-RI01 (01), 2007.050 14:49:20 INITIAL SCAN OF USER CONTROL STATEMENTS COMPLETED.
ARC0640I ARCFRTM - ADR014I (SCH)-DSSU (02),
2007.050 14:49:20 ALL PREVIOUSLY SCHEDULED TASKS COMPLETED. PARALLEL MODE NOW IN EFFECT
ARC0640I ARCFRTM - ADR050I (002)-PRIME(02), DFSMSDSS INVOKED VIA CROSS MEMORY APPLICATION INTERFACE
ARC0640I ARCFRTM - ADR016I (002)-PRIME(01), RACF LOGGING OPTION IN EFFECT FOR THIS TASK
ARC0640I ARCFRTM - ADR050I (003)-PRIME(02), DFSMSDSS INVOKED VIA CROSS MEMORY APPLICATION INTERFACE
ARC0640I ARCFRTM - ADR016I (003)-PRIME(01), RACF LOGGING OPTION IN EFFECT FOR THIS TASK
ARC0640I ARCFRTM - ADR006I (003)-STEND(01), 2007.050 14:49:20 EXECUTION BEGINS
ARC0640I ARCFRTM - ADR241I (003)-DDTFP(01), TARGET VTOC BEGINNING AT 000003:0000 AND ENDING AT 000008:0014 IS OVERLAID
ARC0640I ARCFRTM - ADR806I (003)-T0MI (02), VOLUME MHL0A1 WAS COPIED USING A FAST REPLICATION FUNCTION
ARC0640I ARCFRTM - ADR006I (003)-STEND(02), 2007.050 14:49:20 EXECUTION ENDS
ARC0640I ARCFRTM - ADR013I (003)-CLTSK(01), 2007.050 14:49:20 TASK COMPLETED WITH RETURN CODE 0000
ARC0640I ARCFRTM - ADR006I (002)-STEND(01), 2007.050 14:49:20 EXECUTION BEGINS
ARC0640I ARCFRTM - ADR241I (002)-DDTFP(01), TARGET VTOC BEGINNING AT 000003:0000 AND ENDING AT 000008:0014 IS OVERLAID
ARC0640I ARCFRTM - ADR806I (002)-T0MI (02), VOLUME MHL0A0 WAS COPIED USING A FAST REPLICATION FUNCTION
ARC0640I ARCFRTM - ADR006I (002)-STEND(02), 2007.050 14:49:21 EXECUTION ENDS
ARC0640I ARCFRTM - ADR013I (002)-CLTSK(01), 2007.050 14:49:21 TASK COMPLETED WITH RETURN CODE 0000
ARC0640I ARCFRTM - ADR012I (SCH)-DSSU (01), 2007.050 14:49:21 DFSMSDSS PROCESSING COMPLETE. HIGHEST RETURN CODE IS 0000
ARC1805I THE FOLLOWING 00002 VOLUME(S) WERE SUCCESSFULLY PROCESSED BY FAST REPLICATION BACKUP OF COPY POOL CP1
ARC1805I (CONT.) MHL0A0
ARC1805I (CONT.) MHL0A1
ARC1802I FAST REPLICATION BACKUP HAS COMPLETED FOR COPY POOL CP1, AT 14:49:21 ON 2007/02/19, FUNCTION RC=0000,
MAXIMUM VOLUME RC=0000
The resulting fast replication backup version can be shown by using the DFSMShsm LIST
command (see Example 7-3).
Example 7-3 LIST command to show copy pool backup versions
HSEND LIST COPYPOOL(CP1)

Chapter 7. DFSMShsm fast replication
187
The output of the LIST command shows a DUMPSTATE of NONE, which means that there is
at this point neither a full volume dump in creation, nor available for this fast replication
backup version (see Figure ). For a detailed description of the possible contents of the new
DUMPSTATE field see the explanations related to Figure on page 211.
Figure 7-11 Output of the LIST COPYPOOL(CP1) command
There is a new column heading FASTREPLICATIONSTATE that has replaced the former
VALID column to provide more detailed information about whether this version can be used
for recovery by the FRRECOV command.
Possible contents of the field FASTREPLICATIONSTATE are:
RECOVERABLE The copy pool has a valid DASD copy version that can be used for
recovery.
DUMPONLY The copy pool was defined with a number of 0 replicate backup
versions, so the DASD copies that are currently available are only kept
until the dump copies of a particular version have completed
successfully.
FAILED The copy pool cannot be recovered by using the DASD copy of this
version because the FRBACKUP did not complete successfully due to
a failure or a WITHDRAW being done before all volumes in the copy
pool were successfully copied.
NONE A DASD copy does not exist. For example, this status applies if only a
dump copy exists for a particular backup version of a copy pool.
1-- DFSMShsm CONTROL DATASET --COPY POOL--LISTING --------- AT 16:34:37 ON
07/02/22 FOR SYSTEM=SC64
0COPYPOOL=CP1
VERSION VTOCENQ DATE TIME FASTREPLICATIONSTATE DUMPSTATE
003 Y 2007/02/22 16:31:22 RECOVERABLE NONE
TOKEN(C)=C'EX3'
TOKEN(H)=X'C5E7F3'
TOTAL NUM OF VOLUMES=00002,INCREMENTAL=N
SGNAME SOURCE - TARGET SOURCE - TARGET SOURCE - TARGET SOURCE - TARGET
SG1 MHL0A0 - MHL0AF MHL0A1 - MHL1AC
Note: The output of the LIST COPYPOOL command has changed significantly since
DFSMShsm V1.7 and there are new keywords available in order to request more or less
information for display (see Figure 7-17 on page 196 and Figure 7-18 on page 197 for
more details).

188
z/OS V1R8 DFSMS Technical Update
So far, we have created a fast replication backup version of copy pool CP1, which copied the
contents of volumes MHL0A0 and MHL0A1 to volumes MHL0AF and MHL1AC (see
Figure 7-12).
Figure 7-12 FRBACKUP without dump processing
7.2.2 Using fast replication backups for recovery of copy pools and volumes
You can use the FRRECOV command to recover copy pools or volumes from the fast
replication backup versions.
Recovery of an entire copy pool is very easy to request and runs very quickly (at a similar
speed as fast replication backup). Usually, you start with a command (as shown in
Example 7-4) to make sure that no source volumes in the copy pool are in an existing
FlashCopy relationship. If one or more volumes are in an existing FlashCopy relationship, the
recovery fails.
Example 7-4 FRRECOV command that works at the copy pool level
HSEND FRRECOV CP(CP1) VERIFY(Y) GENERATION(0) FROMDASD
If one of the volumes cannot be serialized by DFSMShsm, the recovery of that particular
volume fails. A very common reason for this kind of failure is an open ICF catalog. You can
close the catalog by using the F CATALOG command, as shown in Example 7-5.
Example 7-5 MODIFY CATALOG,UNALLOCATE to close an open ICF catalog
F CATALOG,UNALLOCATE(UCAT.TESTFR)
CP1
(COPY POOL)
SG1
(Storgrp)
MHL0A0
MHL0A1
MHL0AC
MHL0AD
MHL0AE
MHL0AF
MHL1AC
MHL1AD
MHL1AE
MHL1AF
MHL2AC
MHL2AD
MHL2AE
MHL2AF
CPBSG1
(Copy Pool Backup Storgrp)
FRBACKUP CP(CP1) -
EXECUTE -
TOKEN(EX3)

Chapter 7. DFSMShsm fast replication
189
Figure 7-13 FRRECOV command at the copy pool level
The process during execution of command FRRECOV CP(CP1) FROMDASD is exactly the
same as during execution of command FRBACKUP CP(CP1), but source and target volumes
have swapped places with each other. An example for accompanying messages in the
JESMSGLG output of DFSMShsm is shown in Figure 7-14.
Figure 7-14 Messages during execution of FRRECOV CP(CP1) FROMDASD
If there is an open ICF catalog on one of the volumes, the resulting messages are as shown
in Example 7-6.
Example 7-6 Excerpt from the backup log during execution of a copy pool recovery
A
RC1801I FAST REPLICATION RECOVERY IS STARTING FOR COPY POOL CP1, AT 18:41:29 ON 2007/03/16
ARC0640I ARCFRTM - PAGE 0001 5695-DF175 DFSMSDSS V1R08.0 DATA SET SERVICES 2007.075 18:41
ARC0640I ARCFRTM - ADR035I (SCH)-PRIME(06), INSTALLATION EXIT ALTERED BYPASS FAC CLASS CHK DEFAULT TO YES
ARC0640I ARCFRTM - PARALLEL
ARC1801I FAST REPLICATION RECOVERY IS STARTING FOR 350
ARC1801I (CONT.) COPY POOL CP1, AT 19:05:10 ON 2007/03/16
ARC1805I THE FOLLOWING 00002 VOLUME(S) WERE 351
ARC1805I (CONT.) SUCCESSFULLY PROCESSED BY FAST REPLICATION RECOVERY
ARC1805I (CONT.) OF COPY POOL CP1
ARC1805I (CONT.) MHL0A0
ARC1805I (CONT.) MHL0A1
ARC1802I FAST REPLICATION RECOVERY HAS COMPLETED FOR 354
ARC1802I (CONT.) COPY POOL CP1, AT 19:05:11 ON 2007/03/16, FUNCTION
ARC1802I (CONT.) RC=0000, MAXIMUM VOLUME RC=0000
CP1
(COPY POOL)
SG1
(Storgrp)
MHL0A0
MHL0A1
MHL0AC
MHL0AD
MHL0AE
MHL0AF
MHL1AC
MHL1AD
MHL1AE
MHL1AF
MHL2AC
MHL2AD
MHL2AE
MHL2AF
CPBSG1
(Copy Pool Backup Storgrp)
FRRECOV CP(CP1) -
VERIFY(Y) -
FROMDASD -
GENERATION(0)
Flashcopy

190
z/OS V1R8 DFSMS Technical Update
ARC0640I ARCFRTM - ADR101I (R/I)-RI01 (01), TASKID 001 HAS BEEN ASSIGNED TO COMMAND 'PARALLEL'
ARC0640I ARCFRTM - COPY IDY(MHL0AF) ODY(MHL0A0) DUMPCOND FR(REQ) PUR ALLX ALLD(*)
ARC0640I ARCFRTM - ADR101I (R/I)-RI01 (01), TASKID 002 HAS BEEN ASSIGNED TO COMMAND 'COPY '
ARC0640I ARCFRTM - COPY IDY(MHL1AC) ODY(MHL0A1) DUMPCOND FR(REQ) PUR ALLX ALLD(*)
ARC0640I ARCFRTM - ADR101I (R/I)-RI01 (01), TASKID 003 HAS BEEN ASSIGNED TO COMMAND 'COPY '
ARC0640I ARCFRTM - ADR109I (R/I)-RI01 (01), 2007.075 18:41:30 INITIAL SCAN OF USER CONTROL STATEMENTS COMPLETED.
ARC0640I ARCFRTM - ADR014I (SCH)-DSSU (02),
2007.075 18:41:30 ALL PREVIOUSLY SCHEDULED TASKS COMPLETED. PARALLEL MODE NOW IN EFFECT
ARC0640I ARCFRTM - ADR050I (002)-PRIME(02), DFSMSDSS INVOKED VIA CROSS MEMORY APPLICATION INTERFACE
ARC0640I ARCFRTM - ADR016I (002)-PRIME(01), RACF LOGGING OPTION IN EFFECT FOR THIS TASK
ARC0640I ARCFRTM - ADR050I (003)-PRIME(02), DFSMSDSS INVOKED VIA CROSS MEMORY APPLICATION INTERFACE
ARC0640I ARCFRTM - ADR016I (003)-PRIME(01), RACF LOGGING OPTION IN EFFECT FOR THIS TASK
ARC0640I ARCFRTM - ADR006I (002)-STEND(01), 2007.075 18:41:30 EXECUTION BEGINS
ARC0640I ARCFRTM - ADR306E (002)-SBRTN(01),
UNABLE TO COPY THE VOLUME BECAUSE OUTPUT VOLUME MHL0A0 IS IN USE. TASK IS TERMINATED
ARC0640I ARCFRTM - ADR006I (002)-STEND(02), 2007.075 18:41:30 EXECUTION ENDS
ARC0640I ARCFRTM - ADR013I (002)-CLTSK(01), 2007.075 18:41:30 TASK COMPLETED WITH RETURN CODE 0008
ARC0640I ARCFRTM - ADR006I (003)-STEND(01), 2007.075 18:41:30 EXECUTION BEGINS
ARC0640I ARCFRTM - ADR241I (003)-DDTFP(01), TARGET VTOC BEGINNING AT 000003:0000 AND ENDING AT 000008:0014 IS OVERLAID
ARC0640I ARCFRTM - ADR806I (003)-T0MI (02), VOLUME MHL1AC WAS COPIED USING A FAST REPLICATION FUNCTION
ARC0640I ARCFRTM - ADR006I (003)-STEND(02), 2007.075 18:41:30 EXECUTION ENDS
ARC0640I ARCFRTM - ADR013I (003)-CLTSK(01), 2007.075 18:41:30 TASK COMPLETED WITH RETURN CODE 0000
ARC0640I ARCFRTM - ADR012I (SCH)-DSSU (01),
2007.075 18:41:30 DFSMSDSS PROCESSING COMPLETE. HIGHEST RETURN CODE IS 0008 FROM:
ARC0640I ARCFRTM - TASK 002
ARC0400I VOLUME MHL0A0 IS 00% FREE, 00000031 FREE TRACK(S), 000008 FREE CYLINDER(S), FRAG .198
ARC0401I LARGEST EXTENTS FOR MHL0A0 ARE CYLINDERS 8, TRACKS 120
ARC0402I VTOC FOR MHL0A0 IS 0090 TRACKS(04500 DSCBS), 04487 FREE DSCBS(99% OF TOTAL)
ARC0400I VOLUME MHL0A1 IS 00% FREE, 00000000 FREE TRACK(S), 000000 FREE CYLINDER(S), FRAG .000
ARC0401I LARGEST EXTENTS FOR MHL0A1 ARE CYLINDERS 0, TRACKS 0
ARC0402I VTOC FOR MHL0A1 IS 0090 TRACKS(04500 DSCBS), 04488 FREE DSCBS(99% OF TOTAL)
ARC1803E THE FOLLOWING 00001 VOLUME(S) FAILED DURING FAST REPLICATION RECOVERY OF COPY POOL CP1
ARC1803E (CONT.) MHL0A0
ARC1805I THE FOLLOWING 00001 VOLUME(S) WERE SUCCESSFULLY PROCESSED BY FAST REPLICATION RECOVERY OF COPY POOL CP1
ARC1805I (CONT.) MHL0A1
ARC1802I FAST REPLICATION RECOVERY HAS COMPLETED FOR COPY POOL CP1, AT 18:41:30 ON 2007/03/16, FUNCTION RC=0008,
MAXIMUM VOLUME RC=0008
After close of the catalog by using the command shown in Example 7-5 on page 188, the
recovery process is resumed by a request at the volume level (see Example 7-7). We
recommend this method because a new FRECOVER command at the copy pool level either
fails due to existing FlashCopy relationships of some volumes or causes excessive and
unnecessary copy activities for many volumes that were already successfully recovered (see
the explanation of message ARC1803E in z/OS MVS System Messages Vol 2 (ARC-ASA),
SA22-7632).
Example 7-7 FRRECOV command at the volume level
HSEND FRRECOV TOVOLUME(MHL0A0) FROMCOPYPOOL(CP1)
If the volume needed for recovery is currently associated with more than one copy pool, the
FRRECOV TOVOLUME command must specify the FROMCOPYPOOL keyword in order to
reference the appropriate copy pool’s fast replication backup version. Example 7-8 shows the
messages issued during volume level fast replication recovery in the DFSMShsm backup log.
Example 7-8 Excerpt from the backup log during execution of a fast replication recovery at the volume level
ARC1801I FAST REPLICATION RECOVERY IS STARTING FOR VOLUME MHL0A0, AT 18:52:47 ON 2007/03/16
ARC0640I ARCFRTM - PAGE 0001 5695-DF175 DFSMSDSS V1R08.0 DATA SET SERVICES 2007.075 18:52
ARC0640I ARCFRTM - ADR035I (SCH)-PRIME(06), INSTALLATION EXIT ALTERED BYPASS FAC CLASS CHK DEFAULT TO YES
ARC0640I ARCFRTM - COPY IDY(MHL0AF) ODY(MHL0A0) DUMPCOND FR(REQ) PUR ALLX ALLD(*)
ARC0640I ARCFRTM - ADR101I (R/I)-RI01 (01), TASKID 001 HAS BEEN ASSIGNED TO COMMAND 'COPY '
ARC0640I ARCFRTM - ADR109I (R/I)-RI01 (01), 2007.075 18:52:47 INITIAL SCAN OF USER CONTROL STATEMENTS COMPLETED.

Chapter 7. DFSMShsm fast replication
191
ARC0640I ARCFRTM - ADR050I (001)-PRIME(02), DFSMSDSS INVOKED VIA CROSS MEMORY APPLICATION INTERFACE
ARC0640I ARCFRTM - ADR016I (001)-PRIME(01), RACF LOGGING OPTION IN EFFECT FOR THIS TASK
ARC0640I ARCFRTM - ADR006I (001)-STEND(01), 2007.075 18:52:47 EXECUTION BEGINS
ARC0640I ARCFRTM - ADR241I (001)-DDTFP(01), TARGET VTOC BEGINNING AT 000003:0000 AND ENDING AT 000008:0014 IS OVERLAID
ARC0640I ARCFRTM - ADR806I (001)-T0MI (02), VOLUME MHL0AF WAS COPIED USING A FAST REPLICATION FUNCTION
ARC0640I ARCFRTM - ADR006I (001)-STEND(02), 2007.075 18:52:48 EXECUTION ENDS
ARC0640I ARCFRTM - ADR013I (001)-CLTSK(01), 2007.075 18:52:48 TASK COMPLETED WITH RETURN CODE 0000
ARC0640I ARCFRTM - ADR012I (SCH)-DSSU (01), 2007.075 18:52:48 DFSMSDSS PROCESSING COMPLETE. HIGHEST RETURN CODE IS 0000
ARC0400I VOLUME MHL0A0 IS 00% FREE, 00000018 FREE TRACK(S), 000000 FREE CYLINDER(S), FRAG .823
ARC0401I LARGEST EXTENTS FOR MHL0A0 ARE CYLINDERS 0, TRACKS 6
ARC0402I VTOC FOR MHL0A0 IS 0090 TRACKS(04500 DSCBS), 04487 FREE DSCBS(99% OF TOTAL)
ARC1802I FAST REPLICATION RECOVERY HAS COMPLETED FOR VOLUME MHL0A0, AT 18:52:48 ON 2007/03/16, FUNCTION RC=0000,

MAXIMUM VOLUME RC=0000
7.3 Tape support
Tape support of fast replication enhances DFSMShsm to manage full volume dumps of the
DASD volumes that are holding the replication backup data. The full volume dumps can be
created:
Automatically as part of the automatic dump process
By command as part of the fast replication backup process
By command in a separate process at any time after a replication backup version of a
copy pool was created
The resulting dump copies provide the same look and feel as dump generations that are
being created by dumping the volumes of the storage groups of a copy pool directly.
There is also available a request for fast replication backups that are processed by using the
NOCOPY option of FlashCopy. A DASD backup version created by this option cannot be
used for recovery of the copy pool and is only maintained until the target volumes have been
dumped to tape. By using this option you might need fewer volumes in the backup copy pool
storage groups.
7.3.1 Dump of target volumes during autodump
Automatic dump was enhanced with z/OS DFSMShsm 1.8 to process storage groups defined
to automatic dump enabled copy pools.

192
z/OS V1R8 DFSMS Technical Update
Prerequisites
In order to enable a copy pool for automatic dump processing, Auto Dump must be set to Y
and at least one dump class must be specified (see Figure 7-15).
Figure 7-15 Definition of copy pool CP1
You can request system affinity as with storage groups by entering the name of a system or a
system group specified in the System or Sysgrp fields in the base configuration of the SMS
configuration into the field Dump Sys/Sys Group Name.
You can use any dump class definition for copy pools. For our tests we have defined a dump
class by using the command shown in Example 7-9.
Example 7-9 DEFINE DUMPCLASS for use with fast replication tape support
HSEND DEFINE DUMPCLASS(DCREDB18 AUTOREUSE +
DATASETRESTORE FREQUENCY(7) +
FRDUMP(REQUIRED) +
FRRECOV(AFM(YES)) +
RETPD(356) +
UNIT(3590-1) +
STACK(30) +
VTOCCOPIES(3) +
DISPOSITION('FASTR TARGET TO TAPE'))
VTOC copies are not required to recover a data set from a dump tape generated from a fast
replication backup copy. Also, DATASETRESTORE is not required to be specified in the
dump class. If a dump class is only used for dump copies created from fast replication backup
copies, and if you plan to use the FRRECOV for individual data sets, then you should
Panel Utilities Scroll Help
------------------------------------------------------------------------------
COPY POOL ALTER Page 1 of 4
Command ===>

SCDS Name . . : SYS1.SMS.SCDS
Copy Pool Name : CP1

To ALTER Copy Pool, Specify:
Description ==> COPY POOL 1
==>

Auto Dump . . . Y (Y or N) Dump Sys/Sys Group Name . . .
Dump Class . . DCREDB18 Dump Class . .
Dump Class . . Dump Class . .
Dump Class . .

Number of Recoverable DASD Fast
Replicate Backup Versions . . . . 2 (0 to 85 or blank)



Use ENTER to Perform Verification; Use DOWN Command to View next Panel;
Use HELP Command for Help; Use END Command to Save and Exit; CANCEL to Exit.

Chapter 7. DFSMShsm fast replication
193
consider creating VTOCCOPIES to enhance your options for recovery (see “Recovery of data
sets from a fast replication dump copy” on page 224).
While not being required for any technical reason, it might be a good idea to create new dump
classes for fast replication dump processing. Unique dump classes for copy pools ensure that
copy pool volumes are dumped as a set, as each dump volume will always contain only dump
copies that were created using the same dump class during dump volume stacking.
There are new keywords available for definition of a dump class that is intended for use with
copy pools:
FRDUMP(REQUIRED) The default setting of FRDUMP(REQUIRED) indicates that every
volume in a copy pool must have been successfully dumped before
the DASD backup version that is being dumped is replaced with a
more recent backup copy.
FRDUMP(OPTIONAL) The alternative setting allows you to replace the DASD backup
version of a copy pool even if one or more volumes were not
successfully dumped.
FRRECOV(AFM(YES)) The default setting of FASTREPLICATIONRECOVER specifies
that a dump class copy is available for recovery without needing to
be specified on the FRRECOV command.
FRRECOV(AFM(NO)) The alternative setting specifies that a dump class copy must be
explicitly referenced by a FRRECOV command in order to be used.
Automatic dump (phase 2)
This phase is often referred to as volume function in contrast to phases 1 (deletion of expired
dump copies) and 3 (deletion of excess dump VTOC copies), which are so-called level
functions because they are executed on a primary host only. A complete description of what
happens in these phases is contained in the manual DFSMShsm Storage Administration
Guide, SC26-0421. In this section we emphasize only the changes that are related to fast
replication backup tape support.
1.At the beginning of this phase, DFSMShsm obtains volume lists from storage groups
defined to automatic dump enabled copy pools.
2.Volumes of copy pools are considered candidates if the frequency requirement for the
dump class to which the copy pool belongs has been met.
3.Volumes of copy pools are considered candidates
even
if they were processed by
automatic dump within the last 14 hours.
4.If a DASD fast replication backup copy generation 0 exists and has not previously been
successfully dumped, automatic dump uses the dump classes specified in the copy pool
to determine whether the frequency requirement is being met. If the generation 0 dump
copy is partially complete, automatic dump skips all volumes that were successfully
processed in the dump classes previously attempted.
5.The automatic dump settings (Y/N) of the source volumes’ storage groups are ignored by
automatic dump when it is processing a copy pool.
Note: Automatic dump never takes into account other fast replication backup copies
than generation 0. In case you need to create a dump copy of generation 1 to 84
consider using the FRBACKUP CP(..) DUMPONLY command.

194
z/OS V1R8 DFSMS Technical Update
6.Volumes associated with the same set of eligible dump classes, of which at least one
enables dump stacking (parameter STACK(n) in dump class definition requests n > 1) will
be always processed at first. Within those volumes, the order is:
a.Affinity copy pool volumes
b.Non-affinity copy pool volumes
c.Affinity SMS volumes
d.Non-affinity SMS volumes
e.Non-SMS volumes
7.Volumes associated with copy pools are not stacked with non-copy pool volumes.
8.After all volumes eligible for stacking have been dumped, the remaining volumes are
dumped, with DFSMShsm selecting the order of processing as listed above.
In Example 7-10 you can see the messages that were issued in the JESMSGLG data set of
DFSMShsm during automatic dump. The new messages ARC1841I and ARC1842I are
issued for each copy pool being processed during automatic dump.
Example 7-10 Excerpts from the JESMSGLG output of DFSMShsm during automatic dump
17.10.00 STC18988 ARC0620I AUTOMATIC DUMP STARTING
17.10.00 STC18988 ARC0570I AUTOMATIC DUMP FOR ALL SMS MANAGED VOLUME(S) 555
555 ARC0570I (CONT.) TERMINATED, RC=17 REASON=0
17.10.00 STC18988 ARC1841I AUTOMATIC DUMP STARTING FOR COPY POOL CP1
17.10.00 STC18988 ARC0622I FULL VOLUME DUMP STARTING ON VOLUME 557
557 ARC0622I (CONT.) MHL0A1(SMS) AT 17:10:00 ON 2007/02/22, SYSTEM SC64,
557 ARC0622I (CONT.) TASK ID=ARCDVOL1 , TO DUMP CLASS(ES)= DCREDB18
17.10.00 STC18988 ARC0728I VTOC FOR VOLUME MHL0A1 COPIED TO DATA SET 558
558 ARC0728I (CONT.) HSM.DUMPVTOC.T223116.VMHL0A1.D07053 ON VOLUME SBXHS4
17.10.00 STC18988 IEC501A M 0B90,PRIVAT,SL,COMP,DFHSM64,DFHSM64,HSM.DMP.DCREDB18.VMHL0A1.D07053.T223116
17.10.40 STC18988 IEC705I TAPE ON 0B90,TST025,SL,COMP,DFHSM64,DFHSM64,HSM.DMP.DCREDB18.VMHL0A1.D07053.T223116,MEDIA3
17.10.41 STC18988 ARC0120I DUMP VOLUME TST025 ADDED, RC= 0, REAS= 0
17.15.47 STC18988 IEC205I SYS00023,DFHSM64,DFHSM64,FILESEQ=1, COMPLETE VOLUME LIST, 599
599 DSN=HSM.DMP.DCREDB18.VMHL0A1.D07053.T223116,VOLS=TST025,
599 TOTALBLOCKS=50048
17.15.47 STC18988 ARC0637I DUMP COPY OF VOLUME MHL0A1 COMPLETE, 601
601 ARC0637I (CONT.) DCLASS=DCREDB18, EXPDT=2008/02/13, DISPOSITION= FASTR
601 ARC0637I (CONT.) TARGET TO TAPE
17.15.47 STC18988 ARC0623I FULL VOLUME DUMP OF VOLUME MHL0A1 ENDING AT 602
602 ARC0623I (CONT.) 17:15:47, PROCESSING SUCCESSFUL
17.15.47 STC18988 ARC0622I FULL VOLUME DUMP STARTING ON VOLUME 603
603 ARC0622I (CONT.) MHL0A0(SMS) AT 17:15:47 ON 2007/02/22, SYSTEM SC64,
603 ARC0622I (CONT.) TASK ID=ARCDVOL1 , TO DUMP CLASS(ES)= DCREDB18
17.15.48 STC18988 ARC0728I VTOC FOR VOLUME MHL0A0 COPIED TO DATA SET 604
604 ARC0728I (CONT.) HSM.DUMPVTOC.T223116.VMHL0A0.D07053 ON VOLUME SBXHS4
17.21.16 STC18988 IEC205I SYS00023,DFHSM64,DFHSM64,FILESEQ=2, COMPLETE VOLUME LIST, 624
624 DSN=HSM.DMP.DCREDB18.VMHL0A0.D07053.T223116,VOLS=TST025,
624 TOTALBLOCKS=49976
17.21.17 STC18988 ARC1842I AUTO DUMP HAS COMPLETED FOR COPY POOL CP1, AT 626
626 ARC1842I (CONT.) 17:21:17 ON 2007/02/22, MAXIMUM VOLUME RC=0000
17.21.17 STC18988 IEF234E K 0B90,TST025,PVT,DFHSM64,DFHSM64
17.21.18 STC18988 ARC0637I DUMP COPY OF VOLUME MHL0A0 COMPLETE, 628
628 ARC0637I (CONT.) DCLASS=DCREDB18, EXPDT=2008/02/13, DISPOSITION= FASTR
628 ARC0637I (CONT.) TARGET TO TAPE
17.21.18 STC18988 ARC0623I FULL VOLUME DUMP OF VOLUME MHL0A0 ENDING AT 629
629 ARC0623I (CONT.) 17:21:18, PROCESSING SUCCESSFUL
17.21.18 STC18988 ARC0621I AUTOMATIC DUMP ENDING
Note: If you want to make sure that only volumes of one copy pool will be stacked on
the same tape volume you should use different dump classes for each copy pool.

Chapter 7. DFSMShsm fast replication
195
Refer to Figure 7-16 to see what happens with copy pool CP1 in our test environment during
autmatic dump. As DFSMShsm always uses the DUMPCONDITIONING keyword when
calling DFSMSdss for the full volume copies during fast replication backup processing, the
dump generations appear to be made directly from the source volumes.
Figure 7-16 Automatic dump of copy pool CP1
The DFSMShsm dump log contained the messages shown in Example 7-11.
Example 7-11 Dump log messages issued during automatic dump
DFSMSHSM DUMP LOG, TIME 16:45:45, DATE 07/02/22
ARC0620I AUTOMATIC DUMP STARTING
ARC0570I AUTOMATIC DUMP FOR ALL SMS MANAGED VOLUME(S) TERMINATED, RC=17 REASON=0
ARC1841I AUTOMATIC DUMP STARTING FOR COPY POOL CP1
ARC0622I FULL VOLUME DUMP STARTING ON VOLUME MHL0A1(SMS) AT 17:10:00 ON 2007/02/22, SYSTEM SC64, TASK ID=ARCDVOL1 ,
TO DUMP CLASS(ES)= DCREDB18
ARC0728I VTOC FOR VOLUME MHL0A1 COPIED TO DATA SET HSM.DUMPVTOC.T223116.VMHL0A1.D07053 ON VOLUME SBXHS4
ARC0640I ARCDVOL1 - PAGE 0001 5695-DF175 DFSMSDSS V1R08.0 DATA SET SERVICES 2007.053 17:10
ARC0640I ARCDVOL1 - ADR035I (SCH)-PRIME(06), INSTALLATION EXIT ALTERED BYPASS FAC CLASS CHK DEFAULT TO YES
ARC0640I ARCDVOL1 - ADR035I (SCH)-PRIME(03), INSTALLATION EXIT ALTERED WORKUNIT DEFAULT TO
ARC0640I ARCDVOL1 - DUMP FULL INDDNAME(SYS00022) -
ARC0640I ARCDVOL1 - OUTDDNAME(SYS00023) -
ARC0640I ARCDVOL1 - ALLEXCP ALLDATA(*) OPTIMIZE(3) TOLERATE(IOERROR)
ARC0640I ARCDVOL1 - ADR101I (R/I)-RI01 (01), TASKID 001 HAS BEEN ASSIGNED TO COMMAND 'DUMP '
ARC0640I ARCDVOL1 - ADR109I (R/I)-RI01 (01), 2007.053 17:10:00 INITIAL SCAN OF USER CONTROL STATEMENTS COMPLETED.
ARC0640I ARCDVOL1 - ADR050I (001)-PRIME(01), DFSMSDSS INVOKED VIA APPLICATION INTERFACE
ARC0640I ARCDVOL1 - ADR016I (001)-PRIME(01), RACF LOGGING OPTION IN EFFECT FOR THIS TASK
ARC0640I ARCDVOL1 - ADR006I (001)-STEND(01), 2007.053 17:10:00 EXECUTION BEGINS
ARC0120I DUMP VOLUME TST025 ADDED, RC= 0, REAS= 0
ARC0640I ARCDVOL1 - ADR006I (001)-STEND(02), 2007.053 17:15:47 EXECUTION ENDS
ARC0640I ARCDVOL1 - ADR013I (001)-CLTSK(01), 2007.053 17:15:47 TASK COMPLETED WITH RETURN CODE 0000
ARC0640I ARCDVOL1 - ADR012I (SCH)-DSSU (01), 2007.053 17:15:47 DFSMSDSS PROCESSING COMPLETE. HIGHEST RETURN CODE IS 0000
ARC0637I DUMP COPY OF VOLUME MHL0A1 COMPLETE, DCLASS=DCREDB18, EXPDT=2008/02/13, DISPOSITION= FASTR TARGET TO TAPE
ARC0623I FULL VOLUME DUMP OF VOLUME MHL0A1 ENDING AT 17:15:47, PROCESSING SUCCESSFUL
ARC0622I FULL VOLUME DUMP STARTING ON VOLUME MHL0A0(SMS) AT 17:15:47 ON 2007/02/22, SYSTEM SC64, TASK ID=ARCDVOL1 ,
TO DUMP CLASS(ES)= DCREDB18
ARC0728I VTOC FOR VOLUME MHL0A0 COPIED TO DATA SET HSM.DUMPVTOC.T223116.VMHL0A0.D07053 ON VOLUME SBXHS4
CP1
(COPY POOL)
SG1
(Storgrp)
MHL0A0
MHL0A1
MHL0AC
MHL0AD
MHL0AE
MHL0AF
MHL1AC
MHL1AD
MHL1AE
MHL1AF
MHL2AC
MHL2AD
MHL2AE
MHL2AF
CPBSG1
(Copy Pool Backup Storgrp)
Automatic dump
TST025
HSM.DMP.DCREDB18.VMHL0A0.D07053.T223116
HSM.DMP.DCREDB18.VMHL0A1.D07053.T223116
Full volume
dump

196
z/OS V1R8 DFSMS Technical Update
ARC0640I ARCDVOL1 - PAGE 0001 5695-DF175 DFSMSDSS V1R08.0 DATA SET SERVICES 2007.053 17:15
ARC0640I ARCDVOL1 - ADR035I (SCH)-PRIME(06), INSTALLATION EXIT ALTERED BYPASS FAC CLASS CHK DEFAULT TO YES
ARC0640I ARCDVOL1 - ADR035I (SCH)-PRIME(03), INSTALLATION EXIT ALTERED WORKUNIT DEFAULT TO
ARC0640I ARCDVOL1 - DUMP FULL INDDNAME(SYS00025) -
ARC0640I ARCDVOL1 - OUTDDNAME(SYS00023) -
ARC0640I ARCDVOL1 - ALLEXCP ALLDATA(*) OPTIMIZE(3) TOLERATE(IOERROR)
ARC0640I ARCDVOL1 - ADR101I (R/I)-RI01 (01), TASKID 001 HAS BEEN ASSIGNED TO COMMAND 'DUMP '
ARC0640I ARCDVOL1 - ADR109I (R/I)-RI01 (01), 2007.053 17:15:48 INITIAL SCAN OF USER CONTROL STATEMENTS COMPLETED.
ARC0640I ARCDVOL1 - ADR050I (001)-PRIME(01), DFSMSDSS INVOKED VIA APPLICATION INTERFACE
ARC0640I ARCDVOL1 - ADR016I (001)-PRIME(01), RACF LOGGING OPTION IN EFFECT FOR THIS TASK
ARC0640I ARCDVOL1 - ADR006I (001)-STEND(01), 2007.053 17:15:48 EXECUTION BEGINS
ARC0640I ARCDVOL1 - ADR006I (001)-STEND(02), 2007.053 17:21:16 EXECUTION ENDS
ARC0640I ARCDVOL1 - ADR013I (001)-CLTSK(01), 2007.053 17:21:16 TASK COMPLETED WITH RETURN CODE 0000
ARC0640I ARCDVOL1 - ADR012I (SCH)-DSSU (01), 2007.053 17:21:16 DFSMSDSS PROCESSING COMPLETE. HIGHEST RETURN CODE IS 0000
ARC1842I AUTO DUMP HAS COMPLETED FOR COPY POOL CP1, AT 17:21:17 ON 2007/02/22, MAXIMUM VOLUME RC=0000
ARC0637I DUMP COPY OF VOLUME MHL0A0 COMPLETE, DCLASS=DCREDB18, EXPDT=2008/02/13, DISPOSITION= FASTR TARGET TO TAPE
ARC0623I FULL VOLUME DUMP OF VOLUME MHL0A0 ENDING AT 17:21:18, PROCESSING SUCCESSFUL
ARC0621I AUTOMATIC DUMP ENDING
When the new dump copy of copy pool CP1 in dump class DCREDB18 is created
successfully you can request the execution of a LIST COPYPOOL command, as shown in
Example 7-12, in order to see the results of the automatic dump as being stored in the BCDS.
Example 7-12 LIST COPYPOOL command to show all information about generation 0
HSEND LI CP(CP1) ALLVOLS
The output of the command is shown in Figure 7-19 on page 198.
Note the changed contents in field DUMPSTATE, which switched from NONE to
ALLCOMPLETE.
Since there is more information related with the fast replication backups of a copy pool if full
volume dumps of the target volumes are being requested, the LIST COPYPOOL command
was enhanced with z/OS DFSMShsm V1.8 by supplying additional keywords.
The new required parameters of the LIST COPYPOOL command are shown in Figure 7-17.
Figure 7-17 New required parameters of the LIST COPYPOOL command
The parameters are:
FRVOLS
Requests a list of source and target pairs (default).
NOVOLS
Requests a list of dump class information, if a dump copy version exists.
DUMPVOLS
Requests a list of source and dump volumes for all dump versions.
|-FRVOLS--------------|
|-FASTREPLICATIONVOLS-|
>>------------------------------------------------>
|-NOVOLS------|
|-DUMPVOLS----| |-GENERATION(0)----------|
|-ALLVOLS ---------(|------------------------|)->
|-GENERATION(--gennum--)-|
|-TOKEN(token)-----------|
|--ALLVERS---------------|

Chapter 7. DFSMShsm fast replication
197
ALLVOLS
Shows all available information including dump classes as well as DASD and tape
volumes in use for backups and dumps.
ALLVOLS(GENERATION(gennum))
A particular generation (generation 0 is the default).
ALLVOLS(TOKEN(token))
A backup version that was created by using this particular token.
ALLVOLS(ALLVERS)
All versions.
The new optional parameters of the LIST COPYPOOL command are shown in Figure 7-18.
FASTREPLICATIONSTATE allows you to limit the display to fast replication backup versions
based on the following parameters:
RECOVERABLE Available on DASD (copy pool requests #versions > 0)
DUMPONLY Still available on DASD (copy pool requests #versions = 0)
FAILED Not available for recovery of the copy pool due to a failure during the
FRBACKUP process or a request for WITHDRAW
NONE Only available on tape
Figure 7-18 New optional parameters of the LIST COPYPOOL command
DUMPSTATE allows you to limit the display to fast replication backup versions based on the
following parameters being specified:
ALLCOMPLETE Have all volumes dumped to all dump classes specified.
REQUIREDCOMPLETE Have all volumes dumped to at least the required dump
classes.
Note: While being introduced as required parameters, you need not really specify one of
these. If you do not specify any of the keywords above, FRVOLS is used as the default
setting.
-- SELECT (------------------------------------------------------)
|-FASTREPLICATIONSTATE (|-RECOVERABLE----|---)-|
|-|-FRSTATE----------| |-DUMPONLY-------| |
| |-FAILED---------| |
| |-NONE-----------| |

|----DUMPSTATE (------|-ALLCOMPLETE------|-)---|
| ---|-DSTATE-| -|-REQUIREDCOMPLETE-|
|-PARTIAL----------|
|-NONE-------------|
Note: A fast replication backup version is only marked with a DUMPSTATE of
REQUIREDCOMPLETE if at least one dump class is specified that is not required (see
Example 7-9 on page 192) and if not all volumes have been dumped to one of these
non-required dump classes. If all the specified dump classes are required, DUMPSTATE is
either PARTIAL or ALLCOMPLETE.

198
z/OS V1R8 DFSMS Technical Update
PARTIAL Have at least one volume dumped to a dump class specified
FAILED Have no volume dumped to any dump class specified
NONE Not required for dump because there is no relationship
between the copy pool and any dump class
Figure 7-19 Output of LIST CP(CP1) ALLVOLS
The possible contents of the field DUMPSTATE at the version level are:
ALLCOMPLETE All DASD volumes have been successfully dumped to the
dump classes specified.
REQUIRED COMPLETE All DASD volumes have been successfully dumped to at
least the required dump classes.
PARTIAL For at least one required dump class the DUMPSTATE at
the dump class level is PARTIAL.
NONE No dumps are associated with the copy pool.
FAILED The creation of a dump copy is being requested but so far
not a single volume has been successfully dumped.
The possible contents of the field DUMPSTATE at the dump class level are:
COMPLETE All DASD volumes have been successfully dumped to this
particular dump class.
PARTIAL At least one volume of the copy pool has not yet been successfully
dumped to this particular dump class.
FAILED A dump copy to this particular dump class is being requested but
so far not a single volume has been successfully dumped.
-- DFSMShsm CONTROL DATASET --COPY POOL--LISTING --------- AT 17:25:03 ON
07/02/22 FOR SYSTEM=SC64
COPYPOOL=CP1
VERSION VTOCENQ DATE TIME FASTREPLICATIONSTATE DUMPSTATE
003 Y 2007/02/22 16:31:22 RECOVERABLE ALLCOMPLETE
TOKEN(C)=C'EX3'
TOKEN(H)=X'C5E7F3'
TOTAL NUM OF VOLUMES=00002,INCREMENTAL=N
SGNAME SOURCE - TARGET SOURCE - TARGET SOURCE - TARGET SOURCE - TARGET
SG1 MHL0A0 - MHL0AF MHL0A1 - MHL1AC
0
DUMPCLASS REQUIRED DUMPSTATE VOLSSUC EXPDATE AVAILABLE
DCREDB18 Y COMPLETE 00002 2008/02/13 Y
HWCOMP ENCRYPT ENCTYPE RSAKEY/KPWD
NO NONE ********* **************************************
SOURCE DUMPVOLS DEVICE TYPE
MHL0A0 TST025 3590-1
FILE SEQ=02, DSNAME=HSM.DMP.DCREDB18.VMHL0A0.D07053.T223116
MHL0A1 TST025 3590-1
FILE SEQ=01, DSNAME=HSM.DMP.DCREDB18.VMHL0A1.D07053.T223116

Chapter 7. DFSMShsm fast replication
199
Automatic dump (phases 1 and 3)
These phases are executed in a primary host only.
During deletion of expired dump copies the expiration of dump copies of copy pool volumes is
determined at the dump class level. As with dump copies of non-copy pool volumes,
DFSMShsm does not automatically delete the last and only remaining copy of a source
volume. If you want to delete such copies you must do this with one of the following
commands:
You can use the DELVOL ... DUMP command, which must include the keywords
LASTCOPY and COPYPOOLCOPY for dumps of copy pools.
You can use the FRDELETE command.
Basically, there is no difference in processing of phase 1 between copy pool and no-copy
pool volumes.
Deletion of excess dump VTOC copy data sets (phase 3) works exactly the same way for all
kind of volumes.
7.3.2 FRBACKUP DUMP
In case you want the dump copies to be created as soon as possible after a successful
creation of a new fast replication backup version you can request the dump copy with the
FRBACKUP command by including the DUMP parameter (see Example 7-13).
Example 7-13 FRBACKUP command together with the DUMP parameter
HSEND FRBACKUP COPYPOOL(CP1) EXECUTE DUMP
The whole process is also called a fast replication dump. During execution of this command
there are two phases:
1.Fast replication backup - See Figure 7-20 on page 200.
2.Copy pool dump - See Figure 7-23 on page 201.

200
z/OS V1R8 DFSMS Technical Update
Copy pool fast replication backup
As soon as the fast replication backup completes successfully for the copy pool, the next
phase is started.
Figure 7-20 Phase 1 of FRBACKUP with imbedded dump
Figure 7-21 Output of a LIST CP(CP1) SELECT(DUMPSTATE(FAILED))
Successful completion means that a FASTREPLICATIONSTATE of RECOVERABLE or
DUMPONLY is displayed by the LIST COPYPOOL command (see Figure 7-21).
-- DFSMShsm CONTROL DATASET --COPY POOL--LISTING --------- AT 19:00:29 ON
07/02/19 FOR SYSTEM=SC64
COPYPOOL=CP1
VERSION VTOCENQ DATE TIME FASTREPLICATIONSTATE DUMPSTATE
002 Y 2007/02/19 18:56:14 RECOVERABLE FAILED
TOKEN(C)=C'EX2'
TOKEN(H)=X'C5E7F2'
TOTAL NUM OF VOLUMES=00002,INCREMENTAL=N
SGNAME SOURCE - TARGET SOURCE - TARGET SOURCE - TARGET SOURCE - TARGET
SG1 MHL0A0 - MHL1AD MHL0A1 - MHL1AE
0----- END OF -- COPY POOL -- LISTING -----
CP1
(COPY POOL)
SG1
(Storgrp)
MHL0A0
MHL0A1
MHL0AC
MHL0AD
MHL0AE
MHL0AF
MHL1AC
MHL1AD
MHL1AE
MHL1AF
MHL2AC
MHL2AD
MHL2AE
MHL2AF
CPBSG1
(Copy Pool Backup Storgrp)
FRBACKUP CP(CP1) -
EXECUTE DUMP
Flashcopy

Chapter 7. DFSMShsm fast replication
201
This does not necessarily imply that the internal FlashCopy process came to an end before
the start of phase 2, as we can see in Figure 7-22, which is the output of a QUERY command
that was requested at 18:57 (Example 7-14).
Example 7-14 QERY COPYPOOL
HSEND QUERY COPYPOOL(CP1)
Figure 7-22 Output of QUERY COPYPOOL command
This shows clearly that the background FlashCopy process is still going on while the copy
pool dump process started at 18:56:15, as we can see in Example 7-15 on page 202, which
is an excerpt of the JESMSGLG output of the HSM address space.
Copy pool dump
Figure 7-23 Phase 2 of FRBACKUP with imbedded dump
The full volume dump is performed to dump class DCREDB18, as we did not specify any
dump class with the FRBACKUP command (see Example 7-13 on page 199). Alternatively,
you can request a dump class with the command by using the DUMPCLASS parameter. In
ARC1820I THE FOLLOWING VOLUMES IN COPY POOL CP1, VERSION 002, HAVE AN ACTIVE
ARC1820I (CONT.) FLASHCOPY BACKGROUND COPY
ARC1820I (CONT.) SGNAME FR-PRIMARY FR-BACKUP
ARC1820I (CONT.) SG1 MHL0A0 MHL1AD
ARC1820I (CONT.) SG1 MHL0A1 MHL1AE
ARC1821I NONE OF THE VOLUMES IN COPY POOL CP1, VERSION 001, HAVE AN ACTIVE
ARC1821I (CONT.) FLASHCOPY BACKGROUND COPY
CP1
(COPY POOL)
SG1
(Storgrp)
MHL0A0
MHL0A1
MHL0AC
MHL0AD
MHL0AE
MHL0AF
MHL1AC
MHL1AD
MHL1AE
MHL1AF
MHL2AC
MHL2AD
MHL2AE
MHL2AF
CPBSG1
(Copy Pool Backup Storgrp)
FRBACKUP CP(CP1) -
EXECUTE DUMP
TST020
HSM.DMP.DCREDB18.VMHL0A0.D07050.T145618
HSM.DMP.DCREDB18.VMHL0A1.D07050.T145618
Full volume
dump

202
z/OS V1R8 DFSMS Technical Update
this case all other dump classes associated with the copy pool are ignored during execution
of this FRBACKUP command.
Example 7-15 JESMSGLG of HSM address space (excerpt)
18.56.14 STC18151 ARC1801I FAST REPLICATION DUMP IS STARTING FOR COPY 519
519 ARC1801I (CONT.) POOL CP1, AT 18:56:14 ON 2007/02/19, TOKEN='EX2'
18.56.15 STC18151 ARC1805I THE FOLLOWING 00002 VOLUME(S) WERE 527
527 ARC1805I (CONT.) SUCCESSFULLY PROCESSED BY FAST REPLICATION BACKUP OF
527 ARC1805I (CONT.) COPY POOL CP1
18.56.15 STC18151 ARC1805I (CONT.) MHL0A0
18.56.15 STC18151 ARC1805I (CONT.) MHL0A1
18.56.15 STC18151 ARC0622I FULL VOLUME DUMP STARTING ON VOLUME 530
530 ARC0622I (CONT.) MHL0A0(SMS) AT 18:56:15 ON 2007/02/19, SYSTEM SC64,
530 ARC0622I (CONT.) TASK ID=ARCDVOL1 , TO DUMP CLASS(ES)= DCREDB18
18.56.16 STC18151 IEC501A M 0B91,PRIVAT,SL,COMP,DFHSM64,DFHSM64,HSM.DMP.DCREDB18.VMHL0A0.D07050.T145618
18.57.17 STC18151 IEC705I TAPE ON 0B91,TST020,SL,COMP,DFHSM64,DFHSM64,HSM.DMP.DCREDB18.VMHL0A0.D07050.T145618,MEDIA3
18.57.17 STC18151 ARC0120I DUMP VOLUME TST020 ADDED, RC= 0, REAS= 0
19.02.25 STC18151 IEC205I SYS00105,DFHSM64,DFHSM64,FILESEQ=1, COMPLETE VOLUME LIST, 539
539 DSN=HSM.DMP.DCREDB18.VMHL0A0.D07050.T145618,VOLS=TST020,
539 TOTALBLOCKS=47476
19.02.25 STC18151 ARC0637I DUMP COPY OF VOLUME MHL0A0 COMPLETE, 541
541 ARC0637I (CONT.) DCLASS=DCREDB18, EXPDT=2008/02/10, DISPOSITION= FASTR
541 ARC0637I (CONT.) TARGET TO TAPE
19.02.25 STC18151 ARC0623I FULL VOLUME DUMP OF VOLUME MHL0A0 ENDING AT 542
542 ARC0623I (CONT.) 19:02:25, PROCESSING SUCCESSFUL
19.02.25 STC18151 ARC0622I FULL VOLUME DUMP STARTING ON VOLUME 543
543 ARC0622I (CONT.) MHL0A1(SMS) AT 19:02:25 ON 2007/02/19, SYSTEM SC64,
543 ARC0622I (CONT.) TASK ID=ARCDVOL1 , TO DUMP CLASS(ES)= DCREDB18
19.05.00 STC18151 ARC0405I HOST 2 UPDATING SPACE INFORMATION ON ALL 546
546 ARC0405I (CONT.) VOLUMES
19.06.07 STC18151 IEC205I SYS00105,DFHSM64,DFHSM64,FILESEQ=2, COMPLETE VOLUME LIST, 547
547 DSN=HSM.DMP.DCREDB18.VMHL0A1.D07050.T145618,VOLS=TST020,
547 TOTALBLOCKS=47447
19.06.07 STC18151 ARC1802I FAST REPLICATION BACKUP DUMP HAS COMPLETED 549
549 ARC1802I (CONT.) FOR COPY POOL CP1, AT 19:06:07 ON 2007/02/19,
549 ARC1802I (CONT.) FUNCTION RC=0000, MAXIMUM VOLUME RC=0000
19.06.08 STC18151 IEF234E K 0B91,TST020,PVT,DFHSM64,DFHSM64
19.06.08 STC18151 ARC0637I DUMP COPY OF VOLUME MHL0A1 COMPLETE, 551
551 ARC0637I (CONT.) DCLASS=DCREDB18, EXPDT=2008/02/10, DISPOSITION= FASTR
551 ARC0637I (CONT.) TARGET TO TAPE
19.06.08 STC18151 ARC0623I FULL VOLUME DUMP OF VOLUME MHL0A1 ENDING AT 552
552 ARC0623I (CONT.) 19:06:08, PROCESSING SUCCESSFUL
Another execution of the QUERY COPYPOOL command (as shown in Example 7-14 on
page 201) was requested at 19:01:00. As Figure 7-24 shows, the background copy is now
complete and the FlashCopy relationship of the volume pairs has ended.
Figure 7-24 Output of QUERY COPYPOOL command
ARC1821I NONE OF THE VOLUMES IN COPY POOL CP1, VERSION 002, HAVE AN ACTIVE
ARC1821I (CONT.) FLASHCOPY BACKGROUND COPY
ARC1821I NONE OF THE VOLUMES IN COPY POOL CP1, VERSION 001, HAVE AN ACTIVE
ARC1821I (CONT.) FLASHCOPY BACKGROUND COPY

Chapter 7. DFSMShsm fast replication
203
Figure 7-25 DFSMShsm backup log messages with fast replication dump
ARC1801I FAST REPLICATION DUMP IS STARTING FOR COPY POOL CP1, AT 18:56:14 ON 2007/02/19, TOKEN='EX2'
ARC0640I ARCFRTM - PAGE 0001 5695-DF175 DFSMSDSS V1R08.0 DATA SET SERVICES 2007.050 18:56
ARC0640I ARCFRTM - ADR035I (SCH)-PRIME(06), INSTALLATION EXIT ALTERED BYPASS FAC CLASS CHK DEFAULT TO YES
ARC0640I ARCFRTM - PARALLEL
ARC0640I ARCFRTM - ADR101I (R/I)-RI01 (01), TASKID 001 HAS BEEN ASSIGNED TO COMMAND 'PARALLEL'
ARC0640I ARCFRTM - COPY IDY(MHL0A0) ODY(MHL1AD) DUMPCOND FR(REQ) PUR ALLX ALLD(*)
ARC0640I ARCFRTM - ADR101I (R/I)-RI01 (01), TASKID 002 HAS BEEN ASSIGNED TO COMMAND 'COPY '
ARC0640I ARCFRTM - COPY IDY(MHL0A1) ODY(MHL1AE) DUMPCOND FR(REQ) PUR ALLX ALLD(*)
ARC0640I ARCFRTM - ADR101I (R/I)-RI01 (01), TASKID 003 HAS BEEN ASSIGNED TO COMMAND 'COPY '
ARC0640I ARCFRTM - ADR109I (R/I)-RI01 (01), 2007.050 18:56:15 INITIAL SCAN OF USER CONTROL STATEMENTS COMPLETED.
ARC0640I ARCFRTM - ADR014I (SCH)-DSSU (02),
2007.050 18:56:15 ALL PREVIOUSLY SCHEDULED TASKS COMPLETED. PARALLEL MODE NOW IN EFFECT
ARC0640I ARCFRTM - ADR050I (002)-PRIME(02), DFSMSDSS INVOKED VIA CROSS MEMORY APPLICATION INTERFACE
ARC0640I ARCFRTM - ADR016I (002)-PRIME(01), RACF LOGGING OPTION IN EFFECT FOR THIS TASK
ARC0640I ARCFRTM - ADR050I (003)-PRIME(02), DFSMSDSS INVOKED VIA CROSS MEMORY APPLICATION INTERFACE
ARC0640I ARCFRTM - ADR016I (003)-PRIME(01), RACF LOGGING OPTION IN EFFECT FOR THIS TASK
ARC0640I ARCFRTM - ADR006I (002)-STEND(01), 2007.050 18:56:15 EXECUTION BEGINS
ARC0640I ARCFRTM - ADR241I (002)-DDTFP(01), TARGET VTOC BEGINNING AT 000003:0000 AND ENDING AT 000008:0014 IS
OVERLAID
ARC0640I ARCFRTM - ADR806I (002)-T0MI (02), VOLUME MHL0A0 WAS COPIED USING A FAST REPLICATION FUNCTION
ARC0640I ARCFRTM - ADR006I (002)-STEND(02), 2007.050 18:56:15 EXECUTION ENDS
ARC0640I ARCFRTM - ADR013I (002)-CLTSK(01), 2007.050 18:56:15 TASK COMPLETED WITH RETURN CODE 0000
ARC0640I ARCFRTM - ADR006I (003)-STEND(01), 2007.050 18:56:15 EXECUTION BEGINS
ARC0640I ARCFRTM - ADR241I (003)-DDTFP(01), TARGET VTOC BEGINNING AT 000003:0000 AND ENDING AT 000008:0014 IS
OVERLAID
ARC0640I ARCFRTM - ADR806I (003)-T0MI (02), VOLUME MHL0A1 WAS COPIED USING A FAST REPLICATION FUNCTION
ARC0640I ARCFRTM - ADR006I (003)-STEND(02), 2007.050 18:56:15 EXECUTION ENDS
ARC0640I ARCFRTM - ADR013I (003)-CLTSK(01), 2007.050 18:56:15 TASK COMPLETED WITH RETURN CODE 0000
ARC0640I ARCFRTM - ADR012I (SCH)-DSSU (01), 2007.050 18:56:15 DFSMSDSS PROCESSING COMPLETE. HIGHEST RETURN CODE
IS 0000
ARC1805I THE FOLLOWING 00002 VOLUME(S) WERE SUCCESSFULLY PROCESSED BY FAST REPLICATION BACKUP OF COPY POOL CP1
ARC1805I (CONT.) MHL0A0
ARC1805I (CONT.) MHL0A1
ARC1802I FAST REPLICATION BACKUP DUMP HAS COMPLETED FOR COPY POOL CP1, AT 19:06:07 ON 2007/02/19, FUNCTION
RC=0000,
MAXIMUM VOLUME RC=0000

204
z/OS V1R8 DFSMS Technical Update
The processing of the fast replication dump is reported in the backup log of DFSMShsm, as
shown in Figure 7-25 on page 203. Note that the process is documented in messages
ARC1801I and ARC1802I as FAST REPLICATION DUMP. The end of the process is reported
after the last full volume dump has ended successfully.
Figure 7-26 Output of QUERY ACTIVE command during processing of a fast replication dump
ARC0101I QUERY ACTIVE COMMAND STARTING ON HOST=2
ARC0144I AUDIT=NOT HELD AND INACTIVE, LIST=NOT HELD AND INACTIVE, RECYCLE=NOT
ARC0144I (CONT.) HELD AND INACTIVE, REPORT=NOT HELD AND INACTIVE
ARC0160I MIGRATION=NOT HELD, AUTOMIGRATION=HELD, RECALL=NOT HELD,
ARC0160I (CONT.) TAPERECALL=NOT HELD, DATA SET MIGRATION=INACTIVE, VOLUME
ARC0160I (CONT.) MIGRATION=INACTIVE, DATA SET RECALL=INACTIVE
ARC0163I BACKUP=NOT HELD, AUTOBACKUP=HELD, RECOVERY=NOT HELD,
ARC0163I (CONT.) TAPEDATASETRECOVERY=NOT HELD, DATA SET BACKUP=NOT HELD, VOLUME
ARC0163I (CONT.) BACKUP=INACTIVE, DATA SET RECOVERY=INACTIVE, VOLUME
ARC0163I (CONT.) RECOVERY=INACTIVE
ARC0276I DATA SET BACKUP=INACTIVE, DATA SET BACKUP ACTUAL IDLETASKS=(ALLOC=00,
ARC0276I (CONT.) MAX=00)
ARC1826I FRBACKUP=NOT HELD AND INACTIVE,FRRECOV=NOT HELD AND INACTIVE,FRBACKUP
ARC1826I (CONT.) DUMP=NOT HELD AND ACTIVE,FRRECOV(TAPE)=NOT HELD AND INACTIVE,
ARC1826I (CONT.) FRRECOV(DATASET)=NOT HELD AND INACTIVE
ARC0642I DUMP=NOT HELD, AUTODUMP=NOT HELD, VOLUME DUMP=INACTIVE, VOLUME
ARC0642I (CONT.) RESTORE=INACTIVE, DATA SET RESTORE=INACTIVE
ARC1822I FRBACKUP DUMP OR DUMPONLY OF COPY POOL CP1 FOR USER MHLRES2, REQUEST
ARC1822I (CONT.) 59 ON HOST 2 IS IN PROGRESS: NOT PROCESSED = 2, TOTAL = 2
ARC0161I FRBACKUP DUMP OF VOLUME MHL0A0,COPY POOL=CP1 FOR USER MHLRES2,
ARC0161I (CONT.) REQUEST 00000059
ARC0437I - TAPECOPY NOT HELD AND INACTIVE
ARC0437I - TAPEREPL NOT HELD AND INACTIVE
ARC0415I EXPIREBV=NOT HELD AND INACTIVE, LAST STORED BACKUP VERSION KEY=, LAST
ARC0415I (CONT.) STORED ABARS VERSION KEY=, LAST PLANNED END KEY=
ARC0460I PRIVATE AREA LIMIT=8168K, UNALLOCATED=5560K, LARGEST FREE AREAS=5516K,
ARC0460I (CONT.) 40K
ARC0460I EXTENDED PRIVATE AREA LIMIT=1466M, UNALLOCATED=1436M, LARGEST FREE
ARC0460I (CONT.) AREAS=1436M, 56K
ARC6018I AGGREGATE BACKUP/RECOVERY = INACTIVE
ARC6019I AGGREGATE BACKUP = NOT HELD, AGGREGATE RECOVERY = NOT HELD
ARC1540I COMMON RECALL QUEUE PLACEMENT FACTORS: CONNECTION STATUS=CONNECTED,
ARC1540I (CONT.) CRQPLEX HOLD STATUS=NONE,HOST COMMONQUEUE HOLD STATUS=NONE,
ARC1540I (CONT.) STRUCTURE ENTRIES=000% FULL,STRUCTURE ELEMENTS=000% FULL
ARC1541I COMMON RECALL QUEUE SELECTION FACTORS: CONNECTION STATUS=CONNECTED,
ARC1541I (CONT.) HOST RECALL HOLD STATUS=NONE,HOST COMMONQUEUE HOLD STATUS=NONE
ARC0101I QUERY ACTIVE COMMAND COMPLETED ON HOST=2

Chapter 7. DFSMShsm fast replication
205
When you look at the output of a QUERY ACTIVE command during execution of a fast
replication dump process you will find the FRBACKUP DUMP to be reported as active (see
Figure 7-26 on page 204).
Figure 7-27 DFSMShsm dump log messages with fast replication dump
ARC0622I FULL VOLUME DUMP STARTING ON VOLUME MHL0A0(SMS) AT 18:56:15 ON 2007/02/19, SYSTEM SC64, TASK ID=ARCDVOL1
,
TO DUMP CLASS(ES)= DCREDB18
ARC0728I VTOC FOR VOLUME MHL0A0 COPIED TO DATA SET HSM.DUMPVTOC.T145618.VMHL0A0.D07050 ON VOLUME SBXHS6
ARC0640I ARCDVOL1 - PAGE 0001 5695-DF175 DFSMSDSS V1R08.0 DATA SET SERVICES 2007.050 18:56
ARC0640I ARCDVOL1 - ADR035I (SCH)-PRIME(06), INSTALLATION EXIT ALTERED BYPASS FAC CLASS CHK DEFAULT TO YES
ARC0640I ARCDVOL1 - ADR035I (SCH)-PRIME(03), INSTALLATION EXIT ALTERED WORKUNIT DEFAULT TO
ARC0640I ARCDVOL1 - DUMP FULL INDDNAME(SYS00104) -
ARC0640I ARCDVOL1 - OUTDDNAME(SYS00105) -
ARC0640I ARCDVOL1 - ALLEXCP ALLDATA(*) OPTIMIZE(3) TOLERATE(IOERROR)
ARC0640I ARCDVOL1 - ADR101I (R/I)-RI01 (01), TASKID 001 HAS BEEN ASSIGNED TO COMMAND 'DUMP '
ARC0640I ARCDVOL1 - ADR109I (R/I)-RI01 (01), 2007.050 18:56:16 INITIAL SCAN OF USER CONTROL STATEMENTS COMPLETED.
ARC0640I ARCDVOL1 - ADR050I (001)-PRIME(01), DFSMSDSS INVOKED VIA APPLICATION INTERFACE
ARC0640I ARCDVOL1 - ADR016I (001)-PRIME(01), RACF LOGGING OPTION IN EFFECT FOR THIS TASK
ARC0640I ARCDVOL1 - ADR006I (001)-STEND(01), 2007.050 18:56:16 EXECUTION BEGINS
ARC0120I DUMP VOLUME TST020 ADDED, RC= 0, REAS= 0
ARC0640I ARCDVOL1 - ADR006I (001)-STEND(02), 2007.050 19:02:25 EXECUTION ENDS
ARC0640I ARCDVOL1 - ADR013I (001)-CLTSK(01), 2007.050 19:02:25 TASK COMPLETED WITH RETURN CODE 0000
ARC0640I ARCDVOL1 - ADR012I (SCH)-DSSU (01), 2007.050 19:02:25 DFSMSDSS PROCESSING COMPLETE. HIGHEST RETURN CODE
IS 0000
ARC0637I DUMP COPY OF VOLUME MHL0A0 COMPLETE, DCLASS=DCREDB18, EXPDT=2008/02/10, DISPOSITION= FASTR TARGET TO
TAPE
ARC0623I FULL VOLUME DUMP OF VOLUME MHL0A0 ENDING AT 19:02:25, PROCESSING SUCCESSFUL
ARC0622I FULL VOLUME DUMP STARTING ON VOLUME MHL0A1(SMS) AT 19:02:25 ON 2007/02/19, SYSTEM SC64, TASK ID=ARCDVOL1
,
TO DUMP CLASS(ES)= DCREDB18
ARC0728I VTOC FOR VOLUME MHL0A1 COPIED TO DATA SET HSM.DUMPVTOC.T145618.VMHL0A1.D07050 ON VOLUME SBXHS6
ARC0640I ARCDVOL1 - PAGE 0001 5695-DF175 DFSMSDSS V1R08.0 DATA SET SERVICES 2007.050 19:02
ARC0640I ARCDVOL1 - ADR035I (SCH)-PRIME(06), INSTALLATION EXIT ALTERED BYPASS FAC CLASS CHK DEFAULT TO YES
ARC0640I ARCDVOL1 - ADR035I (SCH)-PRIME(03), INSTALLATION EXIT ALTERED WORKUNIT DEFAULT TO
ARC0640I ARCDVOL1 - DUMP FULL INDDNAME(SYS00108) -
ARC0640I ARCDVOL1 - OUTDDNAME(SYS00105) -
ARC0640I ARCDVOL1 - ALLEXCP ALLDATA(*) OPTIMIZE(3) TOLERATE(IOERROR)
ARC0640I ARCDVOL1 - ADR101I (R/I)-RI01 (01), TASKID 001 HAS BEEN ASSIGNED TO COMMAND 'DUMP '
ARC0640I ARCDVOL1 - ADR109I (R/I)-RI01 (01), 2007.050 19:02:25 INITIAL SCAN OF USER CONTROL STATEMENTS COMPLETED.
ARC0640I ARCDVOL1 - ADR050I (001)-PRIME(01), DFSMSDSS INVOKED VIA APPLICATION INTERFACE
ARC0640I ARCDVOL1 - ADR016I (001)-PRIME(01), RACF LOGGING OPTION IN EFFECT FOR THIS TASK
ARC0640I ARCDVOL1 - ADR006I (001)-STEND(01), 2007.050 19:02:25 EXECUTION BEGINS
ARC0640I ARCDVOL1 - ADR006I (001)-STEND(02), 2007.050 19:06:07 EXECUTION ENDS
ARC0640I ARCDVOL1 - ADR013I (001)-CLTSK(01), 2007.050 19:06:07 TASK COMPLETED WITH RETURN CODE 0000
ARC0640I ARCDVOL1 - ADR012I (SCH)-DSSU (01), 2007.050 19:06:07 DFSMSDSS PROCESSING COMPLETE. HIGHEST RETURN CODE
IS 0000
ARC1802I FAST REPLICATION BACKUP DUMP HAS COMPLETED FOR COPY POOL CP1, AT 19:06:07 ON 2007/02/19, FUNCTION
RC=0000,
MAXIMUM VOLUME RC=0000
ARC0637I DUMP COPY OF VOLUME MHL0A1 COMPLETE, DCLASS=DCREDB18, EXPDT=2008/02/10, DISPOSITION= FASTR TARGET TO
TAPE
ARC0623I FULL VOLUME DUMP OF VOLUME MHL0A1 ENDING AT 19:06:08, PROCESSING SUCCESSFUL
DFSMSHSM DUMP LOG, TIME 19:06:08, DATE 07/02/19

206
z/OS V1R8 DFSMS Technical Update
The processing of the fast replication dump is reported in the dump log of DFSMShsm, as
shown in Figure 7-27 on page 205. We can see similar messages as during full volume
dumps of level0 volumes. The end of the process is reported by message ARC1802I as FAST
REPLICATION DUMP after the last full volume dump has ended successfully.
Figure 7-28 Output of the LI COPYPOOL command after the end of the fast replication dump
After the end of the fast replication dump process you can see all the details by executing a
LIST command, as shown in Example 7-16.
Example 7-16 LIST COPYPOOL(CP1) ALLVOLS
HSEND LI CP(CP1) ALLVOLS(TOKEN(EX2))
The output (see Figure 7-28) shows a DUMPSTATE of ALLCOMPLETE at the version level,
which means that all outstanding dump copies for this version ended successfully and are
reported at the dump class level with a DUMPSTATE of COMPLETE.
The volume serial numbers of all the dump volumes that contain dump copies and the names
of all dump files are included as well.
-- DFSMShsm CONTROL DATASET --COPY POOL--LISTING --------- AT 19:13:20 ON
07/02/19 FOR SYSTEM=SC64
COPYPOOL=CP1
VERSION VTOCENQ DATE TIME FASTREPLICATIONSTATE DUMPSTATE
002 Y 2007/02/19 18:56:14 RECOVERABLE ALLCOMPLETE
TOKEN(C)=C'EX2'
TOKEN(H)=X'C5E7F2'
TOTAL NUM OF VOLUMES=00002,INCREMENTAL=N
SGNAME SOURCE - TARGET SOURCE - TARGET SOURCE - TARGET SOURCE - TARGET
SG1 MHL0A0 - MHL1AD MHL0A1 - MHL1AE
DUMPCLASS REQUIRED DUMPSTATE VOLSSUC EXPDATE AVAILABLE
DCREDB18 Y COMPLETE 00002 2008/02/10 Y

HWCOMP ENCRYPT ENCTYPE RSAKEY/KPWD
NO NONE ********* **************************************

SOURCE DUMPVOLS DEVICE TYPE
MHL0A0 TST020 3590-1
FILE SEQ=01, DSNAME=HSM.DMP.DCREDB18.VMHL0A0.D07050.T145618
MHL0A1 TST020 3590-1
FILE SEQ=02, DSNAME=HSM.DMP.DCREDB18.VMHL0A1.D07050.T145618
0----- END OF -- COPY POOL -- LISTING -----

Chapter 7. DFSMShsm fast replication
207
To determine the contents of such a dump copy we can use the LIST command, as shown in
Example 7-22 on page 211.
Figure 7-29 Output of LIST DUMPVOLUME DCONTENTS(MHL0A1)
The output (see Figure 7-29) shows the same look and feel as with other dump volumes with
the exception of column CP, which contains Y to indicate that the contents of this dump
volume were created while processing at least one target volume of a fast replication backup.
7.3.3 FRBACKUP DUMPONLY
If you want to create another dump copy of a copy pool or complete a dump copy for that
version (DUMPSTATE is PARTIAL) at any time (as long as the DASD copy you want to dump
exists) you can use the DUMPONLY parameter of the FRBACKUP command.
1-- DFSMSHSM CONTROL DATASET -DUMP VOLUME-BCDS-- LISTING --- AT 19:21:17 ON 07/02/19 FOR SYSTEM=SC64
0DUMP VOL UNIT FILE SOURCE DUMPED DUMPED PCT HW ENC C SET OF
DUMP
VOLSER STATUS TYPE SEQ VOLSER SMS CLASS DATE TIME EXP DATE IDRC LIBRARY FULL P VOLSERS
0TST020 UNEXP 3590-1 DCREDB18 2008/02/10 Y LIB1 04 N *** Y
01 MHL0A0 Y 2007/02/19 18:56:14 TST020
ENCTYPE RSAKEY/KPWD
********* ****************************************************************
DUMP COPY DATA SET NAME = HSM.DMP.DCREDB18.VMHL0A0.D07050.T145618
1CONTENTS OF VTOC COPY FOR SOURCE VOLUME MHL0A0
0DATASET NAME ORG MULTI CREATED REFERENCED EXP DATE RACF PSWD CHANGED
0MHLRES2.DCOLLECT.D997 PS NO 07/02/16 07/02/16 00/00/00 NO NO YES
MHLRES2.DCOLLECT.D998 PS NO 07/02/16 07/02/16 00/00/00 NO NO NO
MHLRES2.DCOLLECT.D999 PS NO 07/02/16 07/02/16 00/00/00 NO NO YES
SYS1.VTOCIX.MHL0A0 PS *** 03/10/20 00/00/00 00/00/00 NO NO NO
SYS1.VVDS.VMHL0A0 VS *** 03/11/07 00/00/00 00/00/00 *** *** NO
YYY.CNTL.JCL PO NO 07/02/19 07/02/19 00/00/00 NO NO YES
02 MHL0A1 Y 2007/02/19 18:56:14 TST020
ENCTYPE RSAKEY/KPWD
********* ****************************************************************
DUMP COPY DATA SET NAME = HSM.DMP.DCREDB18.VMHL0A1.D07050.T145618
0----- END OF - DUMP VOLUME - LISTING -----
1-- DFSMSHSM CONTROL DATASET -DUMP VOLUME-BCDS-- LISTING --- AT 19:21:41 ON 07/02/19 FOR SYSTEM=SC64
0DUMP VOL UNIT FILE SOURCE DUMPED DUMPED PCT HW ENC C SET OF
DUMP
VOLSER STATUS TYPE SEQ VOLSER SMS CLASS DATE TIME EXP DATE IDRC LIBRARY FULL P VOLSERS
0TST020 UNEXP 3590-1 DCREDB18 2008/02/10 Y LIB1 04 N *** Y
01 MHL0A0 Y 2007/02/19 18:56:14 TST020
ENCTYPE RSAKEY/KPWD
********* ****************************************************************
DUMP COPY DATA SET NAME = HSM.DMP.DCREDB18.VMHL0A0.D07050.T145618
02 MHL0A1 Y 2007/02/19 18:56:14 TST020
ENCTYPE RSAKEY/KPWD
********* ****************************************************************
DUMP COPY DATA SET NAME = HSM.DMP.DCREDB18.VMHL0A1.D07050.T145618
1CONTENTS OF VTOC COPY FOR SOURCE VOLUME MHL0A1
0DATASET NAME ORG MULTI CREATED REFERENCED EXP DATE RACF PSWD CHANGED
0MHLRES2.DCOLLECT.D987 PS NO 07/02/16 07/02/16 00/00/00 NO NO YES
MHLRES2.DCOLLECT.D988 PS NO 07/02/16 07/02/16 00/00/00 NO NO NO
MHLRES2.DCOLLECT.D989 PS NO 07/02/16 07/02/16 00/00/00 NO NO YES
SYS1.VTOCIX.MHL0A1 PS *** 03/10/20 00/00/00 00/00/00 NO NO NO
SYS1.VVDS.VMHL0A1 VS *** 03/11/07 00/00/00 00/00/00 *** *** NO
YYY.CMD.CLIST PE NO 07/02/19 07/02/19 00/00/00 NO NO YES
0----- END OF - DUMP VOLUME - LISTING -----

208
z/OS V1R8 DFSMS Technical Update
Suppose that there was a Version 1 DASD copy (token EX1) of copy pool CP1 created some
time ago (see Figure 7-30).
Figure 7-30 Creation of Version 1 DASD copy of copy pool CP1
If it is still available, we can display its properties by using the LIST COPYPOOL command,
as shown in Example 7-17.
Example 7-17 LIST COPYPOOL command to verify the existence of a DASD copy
HSEND LIST CP(CP1) ALLVOLS(TOKEN(EX1))
The output of the LIST command is shown in Figure 7-31 on page 209. The
FASTREPLICATIONSTATE of RECOVERABLE proves that the DASD copy is available and
complete. The DUMPSTATE of NONE shows that there is no dump copy available so far.
When we want to dump this particular fast replication backup version we can use the
command shown in Example 7-18.
Example 7-18 FRBACKUP COPYPOOL(CP1) DUMPONLY
HSEND FRBACKUP COPYPOOL(CP1) DUMPONLY(TOKEN(EX1))
CP1
(COPY POOL)
SG1
(Storgrp)
MHL0A0
MHL0A1
MHL0AC
MHL0AD
MHL0AE
MHL0AF
MHL1AC
MHL1AD
MHL1AE
MHL1AF
MHL2AC
MHL2AD
MHL2AE
MHL2AF
CPBSG1
(Copy Pool Backup Storgrp)
FRBACKUP CP(CP1) -
EXECUTE -
TOKEN(EX1)
Flashcopy

Chapter 7. DFSMShsm fast replication
209
If you do not specify a dump class with the FRBACKUP command DFSMShsm looks up the
copy pool definition and uses the dump classes specified with the copy pool.
Figure 7-31 Output of the LIST COPYPOOL command to verify the existence of a DASD copy
The resulting messages in the JESMSGLG data set of the HSM address space are shown in
Example 7-19. Note the time stamp of the dump data set. While the full volume dump process
actually starts at 17:42:50, the low-level qualifier of the dump data set name (T204914, format
Tssmmhh) reflects the time stamp of the fast replication backup as reported by the LIST
COPYPOOL command (see Figure on page 187).
Example 7-19 Messages in JESMSGLG for FRBACKUP COPYPOOL(SG1) DUMPONLY
17.42.49 STC18151 ARC1801I FAST REPLICATION DUMPONLY IS STARTING FOR 261
261 ARC1801I (CONT.) COPY POOL CP1, AT 17:42:49 ON 2007/02/19
17.42.50 STC18151 ARC0622I FULL VOLUME DUMP STARTING ON VOLUME 262
262 ARC0622I (CONT.) MHL0A0(SMS) AT 17:42:50 ON 2007/02/19, SYSTEM SC64,
262 ARC0622I (CONT.) TASK ID=ARCDVOL1 , TO DUMP CLASS(ES)= DCREDB18
17.42.50 STC18151 IEC501A M 0B90,PRIVAT,SL,COMP,DFHSM64,DFHSM64,HSM.DMP.DCREDB18.VMHL0A0.D07050.T204914
17.43.48 STC18151 IEC705I TAPE ON 0B90,TST019,SL,COMP,DFHSM64,DFHSM64,HSM.DMP.DCREDB18.VMHL0A0.D07050.T204914,MEDIA3
17.43.49 STC18151 ARC0120I DUMP VOLUME TST019 ADDED, RC= 0, REAS= 0
17.47.28 STC18151 IEC205I SYS00073,DFHSM64,DFHSM64,FILESEQ=1, COMPLETE VOLUME LIST, 267
267 DSN=HSM.DMP.DCREDB18.VMHL0A0.D07050.T204914,VOLS=TST019,
267 TOTALBLOCKS=47437
17.47.28 STC18151 ARC0637I DUMP COPY OF VOLUME MHL0A0 COMPLETE, 269
269 ARC0637I (CONT.) DCLASS=DCREDB18, EXPDT=2008/02/10, DISPOSITION= FASTR
269 ARC0637I (CONT.) TARGET TO TAPE
17.47.28 STC18151 ARC0623I FULL VOLUME DUMP OF VOLUME MHL0A0 ENDING AT 270
270 ARC0623I (CONT.) 17:47:28, PROCESSING SUCCESSFUL
17.47.28 STC18151 ARC0622I FULL VOLUME DUMP STARTING ON VOLUME 271
271 ARC0622I (CONT.) MHL0A1(SMS) AT 17:47:28 ON 2007/02/19, SYSTEM SC64,
271 ARC0622I (CONT.) TASK ID=ARCDVOL1 , TO DUMP CLASS(ES)= DCREDB18
17.51.09 STC18151 IEC205I SYS00073,DFHSM64,DFHSM64,FILESEQ=2, COMPLETE VOLUME LIST, 280
280 DSN=HSM.DMP.DCREDB18.VMHL0A1.D07050.T204914,VOLS=TST019,
280 TOTALBLOCKS=47437
17.51.10 STC18151 ARC1802I FAST REPLICATION BACKUP DUMPONLY HAS 282
282 ARC1802I (CONT.) COMPLETED FOR COPY POOL CP1, AT 17:51:10 ON
282 ARC1802I (CONT.) 2007/02/19, FUNCTION RC=0000, MAXIMUM VOLUME RC=0000
17.51.10 STC18151 IEF234E K 0B90,TST019,PVT,DFHSM64,DFHSM64
17.51.11 STC18151 ARC0637I DUMP COPY OF VOLUME MHL0A1 COMPLETE, 284
284 ARC0637I (CONT.) DCLASS=DCREDB18, EXPDT=2008/02/10, DISPOSITION= FASTR
284 ARC0637I (CONT.) TARGET TO TAPE
17.51.11 STC18151 ARC0623I FULL VOLUME DUMP OF VOLUME MHL0A1 ENDING AT 285
285 ARC0623I (CONT.) 17:51:11, PROCESSING SUCCESSFUL
-- DFSMShsm CONTROL DATASET --COPY POOL--LISTING --------- AT 15:05:04 ON
07/02/19 FOR SYSTEM=SC64
COPYPOOL=CP1
VERSION VTOCENQ DATE TIME FASTREPLICATIONSTATE DUMPSTATE
001 Y 2007/02/19 14:49:20 RECOVERABLE NONE
TOKEN(C)=C'EX1'
TOKEN(H)=X'C5E7F1'
TOTAL NUM OF VOLUMES=00002,INCREMENTAL=N
SGNAME SOURCE - TARGET SOURCE - TARGET SOURCE - TARGET SOURCE - TARGET
SG1 MHL0A0 - MHL0AF MHL0A1 - MHL1AC
0----- END OF -- COPY POOL -- LISTING -----

210
z/OS V1R8 DFSMS Technical Update
The fast replication DUMPONLY processing is documented by messages in the backup log
(see Figure 7-32), as well as in the dump log (see Example 7-20).
Figure 7-32 Messages in the DFSMShsm backup log for FRBACKUP COPYPOOL(SG1) DUMPONLY
The messages in the DFSMShsm dump log look very similar to what we can see at any time
when a full volume dump takes place under control of DFSMShsm. What we cannot see as
part of the dump log is which volumes stand behind the references that are used with the
INDDNAME statements.
There is one hint available that this was not a full volume dump of MHL0A0 as usual. Look at
the time stamp of the DUMPVTOC data set and you can once more realize that this does not
represent the current time of full volume dump processing (which started at 17:42:50), but
instead T204914 (format Tssmmhh) represents exactly the time (14:49:20) when fast
replication backup of volume MHL0A0 was started (see Figure on page 211).
Example 7-20 Messages in the DFSMShsm dump log for FRBACKUP COPYPOOL(SG1) DUMPONLY
DFSMSHSM DUMP LOG, TIME 15:26:04, DATE 07/02/19
ARC0622I FULL VOLUME DUMP STARTING ON VOLUME MHL0A0(SMS) AT 17:42:50 ON 2007/02/19, SYSTEM SC64, TASK ID=ARCDVOL1 ,
TO DUMP CLASS(ES)= DCREDB18
ARC0728I VTOC FOR VOLUME MHL0A0 COPIED TO DATA SET HSM.DUMPVTOC.T204914.VMHL0A0.D07050 ON VOLUME SBXHS6
ARC0640I ARCDVOL1 - PAGE 0001 5695-DF175 DFSMSDSS V1R08.0 DATA SET SERVICES 2007.050 17:42
ARC0640I ARCDVOL1 - ADR035I (SCH)-PRIME(06), INSTALLATION EXIT ALTERED BYPASS FAC CLASS CHK DEFAULT TO YES
ARC0640I ARCDVOL1 - ADR035I (SCH)-PRIME(03), INSTALLATION EXIT ALTERED WORKUNIT DEFAULT TO
ARC0640I ARCDVOL1 - DUMP FULL INDDNAME(SYS00072) -
ARC0640I ARCDVOL1 - OUTDDNAME(SYS00073) -
ARC0640I ARCDVOL1 - ALLEXCP ALLDATA(*) OPTIMIZE(3) TOLERATE(IOERROR)
ARC0640I ARCDVOL1 - ADR101I (R/I)-RI01 (01), TASKID 001 HAS BEEN ASSIGNED TO COMMAND 'DUMP '
ARC0640I ARCDVOL1 - ADR109I (R/I)-RI01 (01), 2007.050 17:42:50 INITIAL SCAN OF USER CONTROL STATEMENTS COMPLETED.
ARC0640I ARCDVOL1 - ADR050I (001)-PRIME(01), DFSMSDSS INVOKED VIA APPLICATION INTERFACE
ARC0640I ARCDVOL1 - ADR016I (001)-PRIME(01), RACF LOGGING OPTION IN EFFECT FOR THIS TASK
ARC0640I ARCDVOL1 - ADR006I (001)-STEND(01), 2007.050 17:42:50 EXECUTION BEGINS
ARC0120I DUMP VOLUME TST019 ADDED, RC= 0, REAS= 0
ARC0640I ARCDVOL1 - ADR006I (001)-STEND(02), 2007.050 17:47:28 EXECUTION ENDS
ARC0640I ARCDVOL1 - ADR013I (001)-CLTSK(01), 2007.050 17:47:28 TASK COMPLETED WITH RETURN CODE 0000
ARC0640I ARCDVOL1 - ADR012I (SCH)-DSSU (01), 2007.050 17:47:28 DFSMSDSS PROCESSING COMPLETE. HIGHEST RETURN CODE IS
0000
ARC0637I DUMP COPY OF VOLUME MHL0A0 COMPLETE, DCLASS=DCREDB18, EXPDT=2008/02/10, DISPOSITION= FASTR TARGET TO TAPE
ARC0623I FULL VOLUME DUMP OF VOLUME MHL0A0 ENDING AT 17:47:28, PROCESSING SUCCESSFUL
ARC0622I FULL VOLUME DUMP STARTING ON VOLUME MHL0A1(SMS) AT 17:47:28 ON 2007/02/19, SYSTEM SC64, TASK ID=ARCDVOL1 ,
TO DUMP CLASS(ES)= DCREDB18
ARC0728I VTOC FOR VOLUME MHL0A1 COPIED TO DATA SET HSM.DUMPVTOC.T204914.VMHL0A1.D07050 ON VOLUME SBXHS6
ARC0640I ARCDVOL1 - PAGE 0001 5695-DF175 DFSMSDSS V1R08.0 DATA SET SERVICES 2007.050 17:47
ARC0640I ARCDVOL1 - ADR035I (SCH)-PRIME(06), INSTALLATION EXIT ALTERED BYPASS FAC CLASS CHK DEFAULT TO YES
ARC0640I ARCDVOL1 - ADR035I (SCH)-PRIME(03), INSTALLATION EXIT ALTERED WORKUNIT DEFAULT TO
ARC0640I ARCDVOL1 - DUMP FULL INDDNAME(SYS00075) -
ARC0640I ARCDVOL1 - OUTDDNAME(SYS00073) -
ARC0640I ARCDVOL1 - ALLEXCP ALLDATA(*) OPTIMIZE(3) TOLERATE(IOERROR)
ARC0640I ARCDVOL1 - ADR101I (R/I)-RI01 (01), TASKID 001 HAS BEEN ASSIGNED TO COMMAND 'DUMP '
ARC0640I ARCDVOL1 - ADR109I (R/I)-RI01 (01), 2007.050 17:47:28 INITIAL SCAN OF USER CONTROL STATEMENTS COMPLETED.
ARC0640I ARCDVOL1 - ADR050I (001)-PRIME(01), DFSMSDSS INVOKED VIA APPLICATION INTERFACE
ARC0640I ARCDVOL1 - ADR016I (001)-PRIME(01), RACF LOGGING OPTION IN EFFECT FOR THIS TASK
ARC0640I ARCDVOL1 - ADR006I (001)-STEND(01), 2007.050 17:47:28 EXECUTION BEGINS
ARC0640I ARCDVOL1 - ADR006I (001)-STEND(02), 2007.050 17:51:09 EXECUTION ENDS
ARC0640I ARCDVOL1 - ADR013I (001)-CLTSK(01), 2007.050 17:51:09 TASK COMPLETED WITH RETURN CODE 0000
ARC0640I ARCDVOL1 - ADR012I (SCH)-DSSU (01), 2007.050 17:51:09 DFSMSDSS PROCESSING COMPLETE. HIGHEST RETURN CODE IS
0000
ARC1802I FAST REPLICATION BACKUP DUMPONLY HAS COMPLETED FOR COPY POOL CP1, AT 17:51:10 ON 2007/02/19, FUNCTION RC=0000,
ARC1801I FAST REPLICATION DUMPONLY IS STARTING FOR COPY POOL CP1, AT 17:42:49 ON 2007/02/19
ARC1802I FAST REPLICATION BACKUP DUMPONLY HAS COMPLETED FOR COPY POOL CP1, AT 17:51:10 ON 2007/02/19, FUNCTION
RC=0000,
MAXIMUM VOLUME RC=0000

Chapter 7. DFSMShsm fast replication
211
MAXIMUM VOLUME RC=0000
ARC0637I DUMP COPY OF VOLUME MHL0A1 COMPLETE, DCLASS=DCREDB18, EXPDT=2008/02/10, DISPOSITION= FASTR TARGET TO TAPE
ARC0623I FULL VOLUME DUMP OF VOLUME MHL0A1 ENDING AT 17:51:11, PROCESSING SUCCESSFUL
DFSMSHSM DUMP LOG, TIME 17:51:11, DATE 07/02/19
When a LIST COPYPOOL(CP1) command (as shown in Example 7-21) is performed after all
related full volume dumps have completed successfully, the DUMPSTATE indicates
ALLCOMPLETE (see Figure ).
Example 7-21 Syntax of LIST COPYPOOL command
LIST COPYPOOL(CP1) ALLVOLS(TOKEN(EX1))
Use the ALLVOLS parameter to request the display of all available information (volume pairs,
dump classes, dump volumes, names of dump copy data sets), as shown in Figure 7-33.
Figure 7-33 Output of the LIST COPYPOOL(CP1) ALLVOLS
To determine the contents of such a dump copy we can use the LIST command, as shown in
Example 7-22.
Example 7-22 LIST DUMPVOLUME command including DCONTENTS keyword
LIST DUMPVOLUME TST019 DCONTENTS(MHL0A1)
-- DFSMShsm CONTROL DATASET --COPY POOL--LISTING --------- AT 17:52:36 ON
07/02/19 FOR SYSTEM=SC64
COPYPOOL=CP1
VERSION VTOCENQ DATE TIME FASTREPLICATIONSTATE DUMPSTATE
001 Y 2007/02/19 14:49:20 RECOVERABLE ALLCOMPLETE
TOKEN(C)=C'EX1'
TOKEN(H)=X'C5E7F1'
TOTAL NUM OF VOLUMES=00002,INCREMENTAL=N
SGNAME SOURCE - TARGET SOURCE - TARGET SOURCE - TARGET SOURCE - TARGET
SG1 MHL0A0 - MHL0AF MHL0A1 - MHL1AC
0
DUMPCLASS REQUIRED DUMPSTATE VOLSSUC EXPDATE AVAILABLE
DCREDB18 Y COMPLETE 00002 2008/02/10 Y

HWCOMP ENCRYPT ENCTYPE RSAKEY/KPWD
NO NONE ********* ***************************************

SOURCE DUMPVOLS DEVICE TYPE
MHL0A0 TST019 3590-1
FILE SEQ=01, DSNAME=HSM.DMP.DCREDB18.VMHL0A0.D07050.T204914
MHL0A1 TST019 3590-1
FILE SEQ=02, DSNAME=HSM.DMP.DCREDB18.VMHL0A1.D07050.T204914
0----- END OF -- COPY POOL -- LISTING -----

212
z/OS V1R8 DFSMS Technical Update
The output (Example 7-23) shows the same look and feel as other dump volumes with the
exception of column CP, which contains Y to indicate that the contents of this dump volume
were created while processing at least one target volume of a fast replication backup.
Example 7-23 Output of LIST DUMPVOLUME DCONTENTS(MHL0A1)
1-- DFSMSHSM CONTROL DATASET -DUMP VOLUME-BCDS-- LISTING --- AT 18:08:58 ON 07/02/19 FOR SYSTEM=SC64
0DUMP VOL UNIT FILE SOURCE DUMPED DUMPED PCT HW ENC C SET OF DUMP
VOLSER STATUS TYPE SEQ VOLSER SMS CLASS DATE TIME EXP DATE IDRC LIBRARY FULL P VOLSERS
0TST019 UNEXP 3590-1 DCREDB18 2008/02/10 Y LIB1 04 N *** Y
01 MHL0A0 Y 2007/02/19 14:49:20 TST019
ENCTYPE RSAKEY/KPWD
********* ****************************************************************
DUMP COPY DATA SET NAME = HSM.DMP.DCREDB18.VMHL0A0.D07050.T204914
02 MHL0A1 Y 2007/02/19 14:49:20 TST019
ENCTYPE RSAKEY/KPWD
********* ****************************************************************
DUMP COPY DATA SET NAME = HSM.DMP.DCREDB18.VMHL0A1.D07050.T204914
1CONTENTS OF VTOC COPY FOR SOURCE VOLUME MHL0A1
0DATASET NAME ORG MULTI CREATED REFERENCED EXP DATE RACF PSWD CHANGED
0MHLRES2.DCOLLECT.D987 PS NO 07/02/16 07/02/16 00/00/00 NO NO YES
MHLRES2.DCOLLECT.D988 PS NO 07/02/16 07/02/16 00/00/00 NO NO YES
MHLRES2.DCOLLECT.D989 PS NO 07/02/16 07/02/16 00/00/00 NO NO YES
SYS1.VTOCIX.MHL0A1 PS *** 03/10/20 00/00/00 00/00/00 NO NO NO
SYS1.VVDS.VMHL0A1 VS *** 03/11/07 00/00/00 00/00/00 *** *** NO
0----- END OF - DUMP VOLUME - LISTING -----

Chapter 7. DFSMShsm fast replication
213
By specifying the DUMPONLY parameter of the FRBACKUP command, a full volume dump
is requested for all target volumes of a particular fast replication backup version of a copy
pool (see Figure 7-34).
Figure 7-34 FRBACKUP with DUMPONLY parameter
The DUMPONLY parameter of the FRBACKUP command can be used to request any dump
class even if it is not assigned at the copy pool level. By using up to five different dump
classes you can have up to five dump copies for each fast replication backup generation. This
allows for a maximum of 425 dump copies per copy pool.
7.3.4 Copy pools that request NOCOPY type FlashCopy processing
If you are interested in a full volume dump of a volume pool and if you can quiesce the
applications that are using these data only for a very short period of time, you should consider
using fast replication. When you are short on volumes to hold the fast replication backups,
you can request the FlashCopy process with NOCOPY and to maintain these DASD copies
only until a dump copy is successfully created.
CP1
(COPY POOL)
SG1
(Storgrp)
MHL0A0
MHL0A1
MHL0AC
MHL0AD
MHL0AE
MHL0AF
MHL1AC
MHL1AD
MHL1AE
MHL1AF
MHL2AC
MHL2AD
MHL2AE
MHL2AF
CPBSG1
(Copy Pool Backup Storgrp)
FRBACKUP CP(CP1) -
EXECUTE -
DUMPONLY-
(TOKEN(EX1))
TST019
HSM.DMP.DCREDB18.VMHL0A0.D07050.T204914
HSM.DMP.DCREDB18.VMHL0A1.D07050.T204914
Full volume
dump

214
z/OS V1R8 DFSMS Technical Update
Copy pool for NOCOPY processing
You must define a copy pool that contains 0 (zero) in the field number of recoverable DASD
fast replicate backup versions (see Figure 7-35).
Figure 7-35 Copy pool definition for NOCOPY processing during fast replication backup
Depending on your needs you can specify Y or N in the field auto dump. If you specify Y you
must specify at least one dump class, as shown in Example 7-24.
Example 7-24 FRBACKUP command for NOCOPY
HSEND FRBACKUP CP(CP0) DUMP(DCLASS(CRYPCOPY))
Panel Utilities Scroll Help
------------------------------------------------------------------------------
COPY POOL DEFINE Page 1 of 4
Command ===>

SCDS Name . . : SYS1.SMS.SCDS
Copy Pool Name : CP0

To DEFINE Copy Pool, Specify:
Description ==> DUMPONLY COPY POOL
==>

Auto Dump . . . Y (Y or N) Dump Sys/Sys Group Name . . .
Dump Class . . CRYPCOPY Dump Class . .
Dump Class . . Dump Class . .
Dump Class . .

Number of Recoverable DASD Fast
Replicate Backup Versions . . . . 0 (0 to 85 or blank)



Use ENTER to Perform Verification; Use DOWN Command to View next Panel;
Use HELP Command for Help; Use END Command to Save and Exit; CANCEL to Exit.

Chapter 7. DFSMShsm fast replication
215
When you request a command (as shown in Example 7-24 on page 214) DFSMShsm
requests a fast replication backup for CP0 followed by a full volume dump of all target
volumes that were paired with volumes of copy pool CP0 during fast replication backup (see
Figure 7-36).
Figure 7-36 Fast replication dump with NOCOPY
When you look at the backup log of DFSMShsm (see Example 7-25) while executing this
request, you can see that the copy commands for execution of the fast replication backup
phase of the process include FCNC keywords.
Example 7-25 DFSMShsm backup log while executing a fast replication dump with NOCOPY
DFSMSHSM BACKUP LOG, TIME 18:05:18, DATE 07/03/12
ARC1801I FAST REPLICATION DUMP IS STARTING FOR COPY POOL CP0, AT 18:13:14 ON 2007/03/13
ARC0640I ARCFRTM - PAGE 0001 5695-DF175 DFSMSDSS V1R08.0 DATA SET SERVICES 2007.072 18:13
ARC0640I ARCFRTM - ADR035I (SCH)-PRIME(06), INSTALLATION EXIT ALTERED BYPASS FAC CLASS CHK DEFAULT TO YES
ARC0640I ARCFRTM - PARALLEL
ARC0640I ARCFRTM - ADR101I (R/I)-RI01 (01), TASKID 001 HAS BEEN ASSIGNED TO COMMAND 'PARALLEL'
ARC0640I ARCFRTM - COPY IDY(MHL0A0) ODY(MHL0AF) DUMPCOND FR(REQ) PUR ALLX ALLD(*) FCNC
ARC0640I ARCFRTM - ADR101I (R/I)-RI01 (01), TASKID 002 HAS BEEN ASSIGNED TO COMMAND 'COPY '
ARC0640I ARCFRTM - COPY IDY(MHL0A1) ODY(MHL1AC) DUMPCOND FR(REQ) PUR ALLX ALLD(*) FCNC
ARC0640I ARCFRTM - ADR101I (R/I)-RI01 (01), TASKID 003 HAS BEEN ASSIGNED TO COMMAND 'COPY '
ARC0640I ARCFRTM - ADR109I (R/I)-RI01 (01), 2007.072 18:13:14 INITIAL SCAN OF USER CONTROL STATEMENTS COMPLETED.
ARC0640I ARCFRTM - ADR014I (SCH)-DSSU (02),
2007.072 18:13:14 ALL PREVIOUSLY SCHEDULED TASKS COMPLETED. PARALLEL MODE NOW IN EFFECT
ARC0640I ARCFRTM - ADR050I (002)-PRIME(02), DFSMSDSS INVOKED VIA CROSS MEMORY APPLICATION INTERFACE
ARC0640I ARCFRTM - ADR016I (002)-PRIME(01), RACF LOGGING OPTION IN EFFECT FOR THIS TASK
ARC0640I ARCFRTM - ADR050I (003)-PRIME(02), DFSMSDSS INVOKED VIA CROSS MEMORY APPLICATION INTERFACE
ARC0640I ARCFRTM - ADR016I (003)-PRIME(01), RACF LOGGING OPTION IN EFFECT FOR THIS TASK
ARC0640I ARCFRTM - ADR006I (002)-STEND(01), 2007.072 18:13:14 EXECUTION BEGINS
ARC0640I ARCFRTM - ADR241I (002)-DDTFP(01), TARGET VTOC BEGINNING AT 000003:0000 AND ENDING AT 000008:0014 IS OVERLAID
ARC0640I ARCFRTM - ADR806I (002)-T0MI (02), VOLUME MHL0A0 WAS COPIED USING A FAST REPLICATION FUNCTION
CP0
(COPY POOL)
#FRBUV = 0
SG1
(Storgrp)
MHL0A0
MHL0A1
MHL0AC
MHL0AD
MHL0AE
MHL0AF
MHL1AC
MHL1AD
MHL1AE
MHL1AF
MHL2AC
MHL2AD
MHL2AE
MHL2AF
CPBSG1
(Copy Pool Backup Storgrp)
FRBACKUP CP(CP0) -
EXECUTE -
DUMP(DCLASS( -
CRYPCOPY))
TST019
HSM.DMP.CRYPCOPY.VMHL0A0.D07072.T141318
HSM.DMP.CRYPCOPY.VMHL0A1.D07072.T141318
Full volume
dump
FCNOCOPY
(Flashcopy Nocopy)

216
z/OS V1R8 DFSMS Technical Update
ARC0640I ARCFRTM - ADR006I (002)-STEND(02), 2007.072 18:13:14 EXECUTION ENDS
ARC0640I ARCFRTM - ADR013I (002)-CLTSK(01), 2007.072 18:13:14 TASK COMPLETED WITH RETURN CODE 0000
ARC0640I ARCFRTM - ADR006I (003)-STEND(01), 2007.072 18:13:14 EXECUTION BEGINS
ARC0640I ARCFRTM - ADR241I (003)-DDTFP(01), TARGET VTOC BEGINNING AT 000003:0000 AND ENDING AT 000008:0014 IS OVERLAID
ARC0640I ARCFRTM - ADR806I (003)-T0MI (02), VOLUME MHL0A1 WAS COPIED USING A FAST REPLICATION FUNCTION
ARC0640I ARCFRTM - ADR006I (003)-STEND(02), 2007.072 18:13:14 EXECUTION ENDS
ARC0640I ARCFRTM - ADR013I (003)-CLTSK(01), 2007.072 18:13:14 TASK COMPLETED WITH RETURN CODE 0000
ARC0640I ARCFRTM - ADR012I (SCH)-DSSU (01), 2007.072 18:13:14 DFSMSDSS PROCESSING COMPLETE. HIGHEST RETURN CODE IS 0000
ARC1805I THE FOLLOWING 00002 VOLUME(S) WERE SUCCESSFULLY PROCESSED BY FAST REPLICATION BACKUP OF COPY POOL CP0
ARC1805I (CONT.) MHL0A0
ARC1805I (CONT.) MHL0A1
ARC1802I FAST REPLICATION BACKUP DUMP HAS COMPLETED FOR COPY POOL CP0, AT 18:43:45 ON 2007/03/13, FUNCTION RC=0000,
MAXIMUM VOLUME RC=0000
Execution of a LIST command (as shown in Example 7-26) at the beginning of the process of
a fast replication dump shows a FASTREPLICATIONSTATE of DUMPONLY (see
Figure 7-37), which means that you cannot use this fast replication backup for recovery of the
copy pool.
Example 7-26 LIST CP(CP0) command
HSEND LIST CP(CP0) ALLVOLS
The DUMPSTATE is FAILED on the version level as well as on the dump class level because
the full volume dumps of the target volumes are still in progress (see Figure 7-37).
Figure 7-37 Output of LIST CP(CP0) command while DUMPSTATE was FAILED
1-- DFSMShsm CONTROL DATASET --COPY POOL--LISTING --------- AT 18:14:40 ON 07/0

0COPYPOOL=CP0

VERSION VTOCENQ DATE TIME FASTREPLICATIONSTATE DUMPSTATE
001 Y 2007/03/13 18:13:14 DUMPONLY FAILED
TOKEN(C)=C''
TOKEN(H)=X''
TOTAL NUM OF VOLUMES=00002,INCREMENTAL=N

SGNAME SOURCE - TARGET SOURCE - TARGET SOURCE - TARGET SOURCE - TARGET
SG1 MHL0A0 - MHL0AF MHL0A1 - MHL1AC
0
DUMPCLASS REQUIRED DUMPSTATE VOLSSUC EXPDATE AVAILABLE
CRYPCOPY N FAILED ***** 2008/03/03 N


SOURCE DUMPVOLS DEVICE TYPE
MHL0A0 ****** ********
FILE SEQ=**, DSNAME=**
MHL0A1 ****** ********
FILE SEQ=**, DSNAME=**

0----- END OF -- COPY POOL -- LISTING -----

Chapter 7. DFSMShsm fast replication
217
When we repeated the same LIST command, after some minutes we saw messages as
shown in Figure 7-38.
Since at least one copy pool volume was successfully dumped to tape, the DUMPSTATE
switched from FAILED to PARTIAL.
The target volume that was successfully dumped so far is no longer reported as a member of
a volume pair.
Figure 7-38 Output of LIST CP(CP0) command while DUMPSTATE was PARTIAL
-- DFSMShsm CONTROL DATASET --COPY POOL--LISTING --------- AT 18:43:36 ON 07/03

COPYPOOL=CP0

VERSION VTOCENQ DATE TIME FASTREPLICATIONSTATE DUMPSTATE
001 Y 2007/03/13 18:13:14 DUMPONLY PARTIAL
TOKEN(C)=C''
TOKEN(H)=X''
TOTAL NUM OF VOLUMES=00002,INCREMENTAL=N

SGNAME SOURCE - TARGET SOURCE - TARGET SOURCE - TARGET SOURCE - TARGET
SG1 MHL0A0 - MHL0A1 - MHL1AC

DUMPCLASS REQUIRED DUMPSTATE VOLSSUC EXPDATE AVAILABLE
CRYPCOPY N PARTIAL 00001 2008/03/03 N

HWCOMP ENCRYPT ENCTYPE RSAKEY/KPWD
NO KEYPW CLRAES128 PASSW0RD

SOURCE DUMPVOLS DEVICE TYPE
MHL0A0 TST019 3590-1
FILE SEQ=01, DSNAME=HSM.DMP.CRYPCOPY.VMHL0A0.D07072.T141318
MHL0A1 ****** ********
FILE SEQ=**, DSNAME=**

----- END OF -- COPY POOL -- LISTING -----

218
z/OS V1R8 DFSMS Technical Update
When all copy pool volumes were successfully dumped to tape, the same LIST command
shows output as shown in Figure 7-39:
DUMPSTATE was changing to ALLCOMPLETE / COMPLETE.
FASTREPLICATIONSTATE changed from DUMPONLY to NONE.
The former shown target volumes of the volume pairs are no longer presented.
Volumes MHL0AF and MHL1AC can be reused by fast replication for any other FlashCopy
relationships.
Figure 7-39 Output of LIST CP(CP0) command when DUMPSTATE was ALLCOMPLETE
7.3.5 Two kinds of dump copies
DFSMShsm can create dump copies of Level0 volumes by using one of the following
methods:
Automatic dump
BACKVOL command with keyword DUMP
FRBACKUP command with keyword DUMP or keyword DUMPONLY
In all cases a dump generation record (DGN) in the BCDS describes each dump copy and a
dump volume record (DVL) in the BCDS describes each dump volume.
-- DFSMShsm CONTROL DATASET --COPY POOL--LISTING --------- AT 18:44:36 ON 07/03

COPYPOOL=CP0

VERSION VTOCENQ DATE TIME FASTREPLICATIONSTATE DUMPSTATE
001 Y 2007/03/13 18:13:14 NONE ALLCOMPLETE
TOKEN(C)=C''
TOKEN(H)=X''
TOTAL NUM OF VOLUMES=00002,INCREMENTAL=N

SGNAME SOURCE - TARGET SOURCE - TARGET SOURCE - TARGET SOURCE - TARGET
SG1 MHL0A0 - ****** MHL0A1 - ******

DUMPCLASS REQUIRED DUMPSTATE VOLSSUC EXPDATE AVAILABLE
CRYPCOPY N COMPLETE 00002 2008/03/03 N

HWCOMP ENCRYPT ENCTYPE RSAKEY/KPWD
NO KEYPW CLRAES128 PASSW0RD
SOURCE DUMPVOLS DEVICE TYPE
MHL0A0 TST019 3590-1
FILE SEQ=01, DSNAME=HSM.DMP.CRYPCOPY.VMHL0A0.D07072.T141318
MHL0A1 TST019 3590-1
FILE SEQ=02, DSNAME=HSM.DMP.CRYPCOPY.VMHL0A1.D07072.T141318

----- END OF -- COPY POOL -- LISTING -----

Chapter 7. DFSMShsm fast replication
219
When we use a list command, as shown in Example 7-27, we can list the associated dump
copies for each level 0 volume.
Example 7-27 Request for a list of associated dump copies of a level 0 volume
HSEND LI PVOL(MHL0A0) ALLDUMPS BCDS
An example of the output of the list command is shown in Example 7-28.
Example 7-28 Output of LI PVOL(MHL0A0) ALLDUMPS BCDS command
1- DFSMSHSM CONTROL DATASET - PRIMARY VOLUME-BCDS--- ALLDUMPS----- AT 19:56:38 ON 07/02/28 FOR SYSTEM=SC64
SOURCE SET OF DUMP
VOLSER GEN SMS DUMPED TIME CLASS EXP DATE VOLSERS
X
MHL0A0 00 YES 07/02/28 15:52:42 MHLRES 07/03/07 TST015 ****** ****** ****** ****** ****** ****** ****** ******
FAST REPLICATION ASSOCIATED DATA FOR VOLUME MHL0A0:
COPY POOL GEN DUMPED TIME CLASS EXP DATE SET OF DUMP VOLSERS
Y
CP1 00 2007/02/27 17:03:06 DCREDB18 2008/02/18 TST028 ****** ****** ****** ****** ******
PLUSCOPY 2008/02/18 TST024 ****** ****** ****** ****** ******
01 2007/02/26 14:00:35 DCREDB18 2008/02/17 TST005 ****** ****** ****** ****** ******
02 2007/02/22 16:31:22 DCREDB18 2008/02/13 TST025 ****** ****** ****** ****** ******
03 2007/02/19 18:56:14 DCREDB18 2008/02/10 TST020 ****** ****** ****** ****** ******
04 2007/02/19 14:49:20 DCREDB18 2008/02/10 TST019 ****** ****** ****** ****** ******
----- END OF - PRIMARY VOLUME - LISTING -----
As you can see, there are two lists of generations of dump copies presented in the list:
The first list to be seen (X) results from full volume dumps of volume MHL0A0.
The second list (Y) results from full volume dumps of target volumes that appeared to be
volume MHL0A0 (same contents but with different volume label in cylinder 0 track 0).

220
z/OS V1R8 DFSMS Technical Update
The basic difference between the traditional full volume dump copies and the fast replication
associated full volume dump copies is that the fast replication associated dump copies are
created from volumes that were the target of a FULL COPY DUMPCONDITIONING
operation. This results in an incorrect volume serial number (that of the target volume) being
dumped from record three of cylinder 0 track 0. You do not need to worry about that at any
time because during any reuse of a dump copy like this for full volume recovery purposes
DFSMSdss remembers that this is a dump copy of a conditioned volume and will take
appropriate action to restore the volume label as of the source volume’s (see Figure 7-40).
Figure 7-40 DUMPCONDITIONING
To learn more about DUMPCONDITIONING see the manuals:
DFSMSdss Storage Administration Guide, SC26-0423
DFSMSdfp Storage Administration Reference, SC26-7402
APAR OW48234.
The two types of full volume dump copies look very similar but must be treated in different
ways by DFSMShsm.
Dump generation records of dump copies that are not associated with fast replication backup
are referenced by the BCDS eligible volume record (MCP) of the related volumes and can
only be used by the commands:
HRECOVER dsname
HSEND RECOVER dsname FROMDUMP
HSEND RECOVER * TOVOLUME(volser) FROMDUMP
In case of a fast replication dump copy (that was created from a volume of a copy pool
backup storage group), the dump generation records are not referenced by the BCDS eligible
DUMPCONDITIONING - Phase 2
Dump
Tape
COPY FULL
DUMPCONDITIONING
OW48234
DUMP
FULL
RESTORE
COPYVOLID
volume
Vol. A
VTOC.A
VVDS.A
Dump Tape appears to be
made directly from source
No longer required to CLIP
volume label after restore
Enables physical restore of
volume, or individual physical
data set restores
or
RESTORE
DATASET
Vol. A
VTOC.A
VVDS.A
Vol. B
VTOC.A
VVDS.A

Chapter 7. DFSMShsm fast replication
221
volume record (MCP). Such dump generations are referenced by the new fast replication
records (FRx®) and are exclusively available for recovery by the FRREVCOV command.
See Table 7-1 for a detailed comparison between traditional dump copies that DFSMShsm
has been providing for the last 20 years and the new dump copies created from fast
replication target volumes that are available since z/OS DFSMShsm V1.8.
Table 7-1 Comparison of fast replication dump copies and other dump copies
Dump of pool storage group
volumes
Dump of copy pool backup
storage group volumes
Created by command BACKVOL volser DUMP FRBACKUP CP(cpname)
{DUMP / DUMPONLY}
Created during Autodump Yes Yes
Described by DGN record Yes Yes
Described by DVL record Yes Yes
Referenced by MCP record Yes No
Available for use with
RECOVER dsn command
Yes No
Available for use with
RECOVER * command
Yes No
Available for use with command
HRECOVER dsn
Yes No
Available for use with
FRRECOV command
No Yes
Reported with LIST DVOL
command
Yes Yes
Reported with command LIST
PVOL(volser) ALLDUMPS
BCDS
Yes Yes
Reported with command LIST
COPYPOOL DUMPVOLS
No Yes
Reported with command LIST
COPYPOOL ALLVOLS
No Yes
Can be used by DFSMSdss in
batch directly for recovery of
volumes and data sets
Yes Yes
Warning in case of one of the
volumes of a storage group was
not successfully dumped
No Yes
Maximum number of dump
copies
100 generations
up to 5 copies per generation
85 generations
up to 5 copies per generation
Maximum number of parallel
processes during creation
SETSYS
MAXDUMPTASKS(32)
SETSYS
MAXDUMPTASKS(32)
Maximum number of parallel
processes during recovery of
volumes by DFSMShsm
1 1

222
z/OS V1R8 DFSMS Technical Update
7.3.6 Recovery from fast replication dumps
Fast replication recovery processing was enhanced with z/OS DFSMS 1.8 to support
recovery from dumps of copy pool volumes.
The FRRECOV command allows request for recovery of a single volume or data sets when
you refer to a replication dump copy.
Recovery of a single volume from a fast replication dump copy
Using the command shown in Example 7-29 selects the latest version (GEN(0)) of the set of
replication dump copies of copy pool CP1 for the recovery of volume MHL0A0. The
FROMCOPYPOOL is an optional keyword that must be used if the volume that is being
recovered resides within a storage group that is shared by multiple copy pools.
Example 7-29 FRRECOV command to recover a single volume from a dump copy
FRRECOV TOVOLUME(MHL0A1) FROMCOPYPOOL(CP1) FROMDUMP
If a version other than the current version is to be recovered, then use the GENERATION,
VERSION, DATE, or TOKEN keyword. There is no option to perform a recovery at the
storage group level.
As with FRBACKUP, DFSMShsm can process up to 64 concurrent invocations of
DFSMSdss, with 15 being the default.
Messages that were issued during execution of the FRRECOV command are in the dump
activity log (see Example 7-30).
Example 7-30 Excerpt from the dump log while processing a fast replication recovery at the volume level from dump
DFSMSHSM DUMP LOG, TIME 17:57:52, DATE 07/03/12
ARC0622I FULL VOLUME RESTORE STARTING ON VOLUME MHL0A0 AT 17:59:16 ON 2007/03/12, SYSTEM SC64, TASK ID=ARCGRVOL
ARC0640I ARCGRVOL - PAGE 0001 5695-DF175 DFSMSDSS V1R08.0 DATA SET SERVICES 2007.071 17:59
ARC0640I ARCGRVOL - ADR035I (SCH)-PRIME(06), INSTALLATION EXIT ALTERED BYPASS FAC CLASS CHK DEFAULT TO YES
ARC0640I ARCGRVOL - RESTORE FULL INDDNAME(SYS00332) OUTDDNAME(SYS00331) -
ARC0640I ARCGRVOL - PURGE COPYVOLID CANCELERROR
ARC0640I ARCGRVOL - ADR101I (R/I)-RI01 (01), TASKID 001 HAS BEEN ASSIGNED TO COMMAND 'RESTORE '
ARC0640I ARCGRVOL - ADR109I (R/I)-RI01 (01), 2007.071 17:59:16 INITIAL SCAN OF USER CONTROL STATEMENTS COMPLETED.
ARC0640I ARCGRVOL - ADR050I (001)-PRIME(01), DFSMSDSS INVOKED VIA APPLICATION INTERFACE
ARC0640I ARCGRVOL - ADR016I (001)-PRIME(01), RACF LOGGING OPTION IN EFFECT FOR THIS TASK
ARC0640I ARCGRVOL - ADR006I (001)-STEND(01), 2007.071 17:59:16 EXECUTION BEGINS
ARC0640I ARCGRVOL - ADR780I (001)-TDFP (01),
THE INPUT DUMP DATA SET BEING PROCESSED IS IN FULL VOLUME FORMAT AND WAS CREATED BY DFSMSDSS VERSION 1
ARC0640I ARCGRVOL - RELEASE 8 MODIFICATION LEVEL 0
ARC0640I ARCGRVOL - ADR808I (001)-TDFP (01),
THE INPUT DUMP DATA SET BEING PROCESSED WAS CREATED FROM A CONDITIONED VOLUME
ARC0640I ARCGRVOL - ADR460I (001)-UTMSG(01), UTILITY GENERATED MESSAGES FOLLOW FOR VOLUME MHL0A0
ARC0640I ARCGRVOL - ICKDSF - MVS/ESA DEVICE SUPPORT FACILITIES 17.0 TIME: 18:04:57 03/12/07
Maximum number of parallel
processes during recovery of
data sets
SETSYS
MAXDSTAPERECOVERTASK
S(64)
SETSYS
MAXDSTAPERECOVERTASK
S(64)
Dump of pool storage group
volumes
Dump of copy pool backup
storage group volumes
Note: Recovery of an entire copy pool by using just one command can only be performed
from the DASD backup copies.

Chapter 7. DFSMShsm fast replication
223
PAGE 1
ARC0640I ARCGRVOL -
ARC0640I ARCGRVOL - BUILDIX DDNAME(SYS00331) IXVTOC
ARC0640I ARCGRVOL - ICK01502I BUILDIX FUNCTION STARTED
ARC0640I ARCGRVOL - ICK00700I DEVICE INFORMATION FOR 8005 IS CURRENTLY AS FOLLOWS:
ARC0640I ARCGRVOL - PHYSICAL DEVICE = 3390
ARC0640I ARCGRVOL - STORAGE CONTROLLER = 2105
ARC0640I ARCGRVOL - STORAGE CONTROL DESCRIPTOR = E8
ARC0640I ARCGRVOL - DEVICE DESCRIPTOR = 0A
ARC0640I ARCGRVOL - ADDITIONAL DEVICE INFORMATION = 4A000035
ARC0640I ARCGRVOL - TRKS/CYL = 15, # PRIMARY CYLS = 3339
ARC0640I ARCGRVOL - ICK04000I DEVICE IS IN SIMPLEX STATE
ARC0640I ARCGRVOL - ICK01503I 8005 REQUEST RECEIVED TO CONVERT VTOC TO IXFORMAT
ARC0640I ARCGRVOL - ICK01504I 8005 VTOC FORMAT IS CURRENTLY OSFORMAT, REQUEST ACCEPTED
ARC0640I ARCGRVOL - ICK01513I 8005 BUILDIX PROCESSING COMPLETED: VTOC IS NOW IN IXFORMAT
ARC0640I ARCGRVOL - ICK01317I VTOC-INDEX IS LOCATED AT CCHH=X'0001 0000' AND IS 30 TRACKS.
ARC0640I ARCGRVOL - 18:05:05 03/12/07
ARC0640I ARCGRVOL -
ARC0640I ARCGRVOL - ICK00002I ICKDSF PROCESSING COMPLETE. MAXIMUM CONDITION CODE WAS 0
ARC0640I ARCGRVOL - ADR006I (001)-STEND(02), 2007.071 18:05:16 EXECUTION ENDS
ARC0640I ARCGRVOL - PAGE 0002 5695-DF175 DFSMSDSS V1R08.0 DATA SET SERVICES 2007.071 17:59
ARC0640I ARCGRVOL - ADR013I (001)-CLTSK(01), 2007.071 18:05:16 TASK COMPLETED WITH RETURN CODE 0000
ARC0640I ARCGRVOL - ADR012I (SCH)-DSSU (01), 2007.071 18:05:16 DFSMSDSS PROCESSING COMPLETE. HIGHEST RETURN CODE IS
0000
ARC0400I VOLUME MHL0A0 IS 26% FREE, 00000018 FREE TRACK(S), 000900 FREE CYLINDER(S), FRAG .008
ARC0401I LARGEST EXTENTS FOR MHL0A0 ARE CYLINDERS 890, TRACKS 13350
ARC0402I VTOC FOR MHL0A0 IS 0090 TRACKS(04500 DSCBS), 04489 FREE DSCBS(99% OF TOTAL)
ARC0623I FULL VOLUME RESTORE OF VOLUME MHL0A0 ENDING AT 18:05:18, DCLASS=DCREDB18, DGEN=000, DATE=2007/03/12,
PROCESSING SUCCESSFUL
ARC1802I FAST REPLICATION RECOVERY FROM DUMP HAS COMPLETED FOR VOLUME MHL0A0, AT 18:05:18 ON 2007/03/12,
FUNCTION RC=0000, MAXIMUM VOLUME RC=0000
DFSMSHSM DUMP LOG, TIME 18:05:18, DATE 07/03/12
Volume restore FROMDUMP(APPLYINCREMENTAL)
You can request a recovery from a fast replication backup copy on tape that includes
APPLYINCREMENTAL processing by using a command, as shown in Example 7-31.
Example 7-31 FRRECOV command that includes FROMDUMP(APPLYINCREMENTAL)
HSEND FRRECOV TOVOLUME(MHL0A0) FROMCP(CP1) FROMDUMP(APPLYINCREMENTAL) GEN(0)
This process is well-known from the command:
RECOVER * TOVOLUME(...) FROMDUMP(APPLYINCREMENTAL)
It works exactly the same way. After successful restore of a DASD volume from a full volume
dump copy, an incremental volume recovery process follows on. A data set will be selected
for processing during the incremental recovery phase if the data set was backed up by
command or during volume backup after the creation of the fast replication backup.

224
z/OS V1R8 DFSMS Technical Update
In Figure 7-41 you can see the messages from the backup log when we executed the
command as shown in Example 7-31 on page 223. Note that there was an ICF catalog on
volume MHL0A0.
Figure 7-41 Backup log messages from fast replication recovery including APPLYINCREMENTAL
Additional messages can be found in the DFSMShsm dump log, as shown in Example 7-30
on page 222.
Recovery of data sets from a fast replication dump copy
This is covered in 7.4, “Data set recovery” on page 224.
7.4 Data set recovery
Starting with z/OS DFSMShsm V1.8, you can use the FRROCOV command to recover an
individual data set from either DASD or tape fast replication backup copies. Since
DFSMShsm does not record any information at the data set level in the BCDS during fast
replication backup, the current catalog entry of the data set is looked up by DFSMShsm in
order to see which volumes are involved in the recovery process. DFSMShsm always riles on
the catalog entry, and the FRRECOV command fails either any of the following is true:
The catalog entry is not available.
The catalog entry points to different volumes as when the fast replication backup was
created.
ARC1801I FAST REPLICATION RECOVERY FROM DUMP IS STARTING FOR VOLUME MHL0A0, AT 17:59:16 ON 2007/03/12
IDC0001I FUNCTION COMPLETED, HIGHEST CONDITION CODE WAS 0
IDC0604I DATA SET BEING IMPORTED WAS EXPORTED ON 03/12/07 AT 17:53:45
IGD01010I SG ACS GETS CONTROL &ACSENVIR=RECOVER
IGD01010I SG ACS GETS CONTROL &ACSENVIR=RECOVER
IDC0181I MANAGEMENTCLASS USED IS MCDB22
IDC0181I STORAGECLASS USED IS HSMFR
IDC0508I DATA ALLOCATION STATUS FOR VOLUME MHL0A0 IS 0
IDC0509I INDEX ALLOCATION STATUS FOR VOLUME MHL0A0 IS 0
IDC01654I ALIASES FROM THE PORTABLE DATA SET WERE NOT DEFINED
TESTFR
IDC0001I FUNCTION COMPLETED, HIGHEST CONDITION CODE WAS 0
IDC0002I IDCAMS PROCESSING COMPLETE. MAXIMUM CONDITION CODE WAS 0
ARC0778I DATA SET UCAT.TESTFR WAS RECOVERED FROM A BACKUP MADE AT 17:53:45 ON 2007/03/12 WITHOUT SERIALIZATION
ARC0734I ACTION=RECOVER FRVOL=SBXHS5 TOVOL=MHL0A0 TRACKS= 16 RC= 0, REASON= 0, AGE= ***, DSN=UCAT.TESTFR
ARC0734I ACTION=RECOVER FRVOL=****** TOVOL=MHL0A0 TRACKS= *** RC= 43, REASON= 18, AGE= ***,
DSN=SYS1.VVDS.VMHL0A0
ARC0734I ACTION=RECOVER FRVOL=****** TOVOL=MHL0A0 TRACKS= *** RC= 58, REASON= 36, AGE= ***,
DSN=TESTFR.DCOLLECT.ESDS01
ARC0734I ACTION=RECOVER FRVOL=****** TOVOL=MHL0A0 TRACKS= *** RC= 58, REASON= 36, AGE= ***,
DSN=TESTFR.MHLRES2.MVOL.DATA
ARC0734I ACTION=RCVSCHD FRVOL=****** TOVOL=MHL0A0 TRACKS= *** RC= 0, REASON= 0, AGE= ***,
DSN=TESTFR.CNTL.JCL
ARC0734I ACTION=RCVSCHD FRVOL=****** TOVOL=MHL0A0 TRACKS= *** RC= 0, REASON= 0, AGE= ***,
DSN=TESTFR.DCOLLECT.ESDS0
ARC0778I DATA SET TESTFR.CNTL.JCL WAS RECOVERED FROM A BACKUP MADE AT 17:53:19 ON 2007/03/12
ARC0734I ACTION=RECOVER FRVOL=SBXHS4 TOVOL=MHL0A0 TRACKS= 150 RC= 0, REASON= 0, AGE= ***,
DSN=TESTFR.CNTL.JCL
ARC0778I DATA SET TESTFR.DCOLLECT.ESDS0 WAS RECOVERED FROM A BACKUP MADE AT 17:55:21 ON 2007/03/12
ARC0734I ACTION=RECOVER FRVOL=SBXHS5 TOVOL=MHL0A0 TRACKS= 13350 RC= 0, REASON= 0, AGE= ***,
DSN=TESTFR.DCOLLECT.ESDS0
ARC0773I RECOVERY ENDED ON VOLUME MHL0A0 TIME 18:06:18
Note: To be eligible for recovery by using the FRRECOV command, a data set must be
cataloged and the catalog entry must point to the same volumes on which the data set
resided when the backup copy was created.

Chapter 7. DFSMShsm fast replication
225
The keywords that are available to request a data set recovery from a fast replication backup
are shown in Figure 7-42.
Figure 7-42 FRRECOV command keywords for data set recovery
For dsname you can substitute one or more fully or partially specified data set names. For
partially specified data set names you can use wild cards of the form:
% To represent a single character
* To represent a single qualifier or parts of a qualifier
** To represent anything (even more than one qualifier)
The standard search order for catalogs applies. You may not specify ** as the high-level
qualifier or as the only qualifier.
7.4.1 Data set filtering
You can request, for example, the recovery of two data sets by using the command as shown
in Example 7-32.
Example 7-32 FRRECOV DSNAME command to recover two data sets
HSEND FRRECOV DSNAME(TESTFR.CMD.CLIST TESTFR.CNTL.JCL) REPLACE FROMCP(CP1)
>>-FRRECOV------------------------------------------------------------------><
| | |-GENERATION(gennum)-| |-B-|
| |--,---| | |-VERSION(vernum)----|
|-DSNAME(-dsname-)----------| |-DATE(date)---------|
|-A-| |-TOKEN(token)-------|
A>>----------------------------------------------REPLACE---------------------->
|-FROMDASD-------------------------------|
|-FROMDUMP-------------------------------|
|-(----------------------------------)-|
|-DUMPCLASS(dclass)-|
|-DCLASS(dclass)----|
|-DUMPVOLUME(dvol)--|
|-DVOL(dvol)--------|

>----------------------------------------------------------------------------->
| | |-FROMCOPYPOOL(cpname)-|
|-FASTREPLICATION(-PREFERRED--)--|
|FR-| |-REQUIRED-|
|-NONE-----|
>---------------------------------------------------------------------------><
| |-RC8-| |
|-NOCOPYPOOLBACKUP(--RC4--)--|
|-NOCPB-|
B>>-----------------------------><
|-RSA(keylabel)-|

226
z/OS V1R8 DFSMS Technical Update
Note that the specification of FROMCP(CP1) is required because the volumes where the data
sets reside are assigned to more than one copy pool (CP0 and CP1). If you omit FROMCP
you end up with message ARC1866I, as shown in Figure 7-43.
Figure 7-43 ARC1866I RC=0020
Messages issued during the data set recovery process can be found in the:
JESMSGLG (Example 7-33)
Backup log (Example 7-34)
Dump log (Example 7-35)
Example 7-33 JESMSGLG during FRRECOV processing of two data sets
16.06.15 STC21322 ARC1861I THE FOLLOWING 0002 DATA SET(S) WERE 870
870 ARC1861I (CONT.) SUCCESSFULLY PROCESSED DURING FAST REPLICATION DATA
870 ARC1861I (CONT.) SET RECOVERY:
16.06.15 STC21322 ARC1861I (CONT.) TESTFR.CMD.CLIST, COPYPOOL=CP1, DEVTYPE=DASD
16.06.15 STC21322 ARC1861I (CONT.) TESTFR.CNTL.JCL, COPYPOOL=CP1, DEVTYPE=DASD
16.06.15 STC21322 ARC1802I FAST REPLICATION DATA SET RECOVERY HAS 873
873 ARC1802I (CONT.) COMPLETED FOR DATA SET TESTFR.CMD.CLIST, ***, AT
873 ARC1802I (CONT.) 16:06:15 ON 2007/03/19, FUNCTION RC=0000, MAXIMUM
873 ARC1802I (CONT.) DATA SET RC=0000
The catalog search found entries for both data sets, so both of them are selected for further
processing.
Example 7-34 Backup log during FRRECOV processing of two data sets
ARC1801I FAST REPLICATION DATA SET RECOVERY IS STARTING FOR DATA SET TESTFR.CMD.CLIST, ***, AT 16:06:14 ON 2007/03/19
ARC1861I THE FOLLOWING 0002 DATA SET(S) WERE SUCCESSFULLY PROCESSED DURING FAST REPLICATION DATA SET RECOVERY:
ARC1861I (CONT.) TESTFR.CMD.CLIST, COPYPOOL=CP1, DEVTYPE=DASD
ARC1861I (CONT.) TESTFR.CNTL.JCL, COPYPOOL=CP1, DEVTYPE=DASD
ARC1802I FAST REPLICATION DATA SET RECOVERY HAS COMPLETED FOR DATA SET TESTFR.CMD.CLIST, ***, AT 16:06:15 ON 2007/03/19,
FUNCTION RC=0000, MAXIMUM DATA SET RC=0000
The dump log shows the keywords that are generated for execution of the data set copy. It is
a physical data set operation (keyword PHYSINDY) that requests the same management
class, storage class, and volume as was found in the catalog entry of each data set. While
BYPASSACS is requested as well, this does not really change the catalog entry of the data
sets.
In this particular case, preallocated data sets were found to be in place, which fitted exactly
the needs of the data sets that were to be recovered.
Example 7-35 Dump log during FRRECOV processing of two data sets
DFSMSHSM DUMP LOG, TIME 18:43:46, DATE 07/03/13
ARC0640I GDSN01 - PAGE 0001 5695-DF175 DFSMSDSS V1R08.0 DATA SET SERVICES 2007.078 16:06
ARC0640I GDSN01 - ADR035I (SCH)-PRIME(06), INSTALLATION EXIT ALTERED BYPASS FAC CLASS CHK DEFAULT TO YES
ARC0640I GDSN02 - PAGE 0001 5695-DF175 DFSMSDSS V1R08.0 DATA SET SERVICES 2007.078 16:06
ARC0640I GDSN02 - ADR035I (SCH)-PRIME(06), INSTALLATION EXIT ALTERED BYPASS FAC CLASS CHK DEFAULT TO YES
ARC0640I GDSN01 - COPY DS(INC(TESTFR.CMD.CLIST )) -
ARC0640I GDSN02 - COPY DS(INC(TESTFR.CNTL.JCL )) -
ARC0640I GDSN01 - PHYSINDY(MHL1AC) OUTDYNAM(MHL0A1) -
ARC0640I GDSN02 - PHYSINDY(MHL0AF) OUTDYNAM(MHL0A0) -
ARC0640I GDSN01 - BYPASSACS(TESTFR.CMD.CLIST ) -
ARC0640I GDSN02 - BYPASSACS(TESTFR.CNTL.JCL ) -
ARC0640I GDSN01 - FASTREPLICATION(PREFERRED) -
ARC0640I GDSN02 - FASTREPLICATION(PREFERRED) -
ARC1866I FAST REPLICATION RECOVERY HAS FAILED FOR DATA SET TESTFR.CMD.CLIST,
ARC1866I (CONT.) RC=0020

Chapter 7. DFSMShsm fast replication
227

ARC0640I GDSN01 - ALLDATA(*) ALLEXCP REPLACEU CANCELERROR DEBUG(FRMSG(DTL)) -
ARC0640I GDSN02 - ALLDATA(*) ALLEXCP REPLACEU CANCELERROR DEBUG(FRMSG(DTL)) -
ARC0640I GDSN01 - TGTGDS( DEFERRED) -
ARC0640I GDSN02 - TGTGDS( DEFERRED) -
ARC0640I GDSN01 - FORCECP(0) PROCESS(SYS1) -
ARC0640I GDSN02 - FORCECP(0) PROCESS(SYS1) -
ARC0640I GDSN01 - STORCLAS(HSMFR ) -
ARC0640I GDSN02 - STORCLAS(HSMFR ) -
ARC0640I GDSN01 - MGMTCLAS(MCDB22 )
ARC0640I GDSN01 - ADR101I (R/I)-RI01 (01), TASKID 001 HAS BEEN ASSIGNED TO COMMAND 'COPY '
ARC0640I GDSN02 - MGMTCLAS(HFS )
ARC0640I GDSN01 - ADR109I (R/I)-RI01 (01), 2007.078 16:06:14 INITIAL SCAN OF USER CONTROL STATEMENTS COMPLETED.
ARC0640I GDSN02 - ADR101I (R/I)-RI01 (01), TASKID 001 HAS BEEN ASSIGNED TO COMMAND 'COPY '
ARC0640I GDSN02 - ADR109I (R/I)-RI01 (01), 2007.078 16:06:14 INITIAL SCAN OF USER CONTROL STATEMENTS COMPLETED.
ARC0640I GDSN01 - ADR050I (001)-PRIME(02), DFSMSDSS INVOKED VIA CROSS MEMORY APPLICATION INTERFACE
ARC0640I GDSN01 - ADR016I (001)-PRIME(01), RACF LOGGING OPTION IN EFFECT FOR THIS TASK
ARC0640I GDSN02 - ADR050I (001)-PRIME(02), DFSMSDSS INVOKED VIA CROSS MEMORY APPLICATION INTERFACE
ARC0640I GDSN02 - ADR016I (001)-PRIME(01), RACF LOGGING OPTION IN EFFECT FOR THIS TASK
ARC0640I GDSN01 - ADR006I (001)-STEND(01), 2007.078 16:06:14 EXECUTION BEGINS
ARC0640I GDSN02 - ADR006I (001)-STEND(01), 2007.078 16:06:14 EXECUTION BEGINS
ARC0640I GDSN01 - ADR442I (001)-PPRNV(01), DATA SET TESTFR.CMD.CLIST PREALLOCATED, ON VOLUME(S): MHL0A1
ARC0640I GDSN02 - ADR442I (001)-PPRNV(01), DATA SET TESTFR.CNTL.JCL PREALLOCATED, ON VOLUME(S): MHL0A0
ARC0640I GDSN01 - ADR806I (001)-T0MI (01), DATA SET TESTFR.CMD.CLIST COPIED USING A FAST REPLICATION FUNCTION
ARC0640I GDSN02 - ADR806I (001)-T0MI (01), DATA SET TESTFR.CNTL.JCL COPIED USING A FAST REPLICATION FUNCTION
ARC0640I GDSN01 - ADR454I (001)-DDDS (02), THE FOLLOWING DATA SETS WERE SUCCESSFULLY PROCESSED
ARC0640I GDSN01 - TESTFR.CMD.CLIST
ARC0640I GDSN01 - ADR006I (001)-STEND(02), 2007.078 16:06:14 EXECUTION ENDS
ARC0640I GDSN02 - ADR454I (001)-DDDS (02), THE FOLLOWING DATA SETS WERE SUCCESSFULLY PROCESSED
ARC0640I GDSN02 - TESTFR.CNTL.JCL
ARC0640I GDSN02 - ADR006I (001)-STEND(02), 2007.078 16:06:14 EXECUTION ENDS
ARC0640I GDSN01 - ADR013I (001)-CLTSK(01), 2007.078 16:06:14 TASK COMPLETED WITH RETURN CODE 0000
ARC0640I GDSN01 - ADR012I (SCH)-DSSU (01), 2007.078 16:06:14 DFSMSDSS PROCESSING COMPLETE. HIGHEST RETURN CODE IS 0000
ARC0640I GDSN02 - ADR013I (001)-CLTSK(01), 2007.078 16:06:14 TASK COMPLETED WITH RETURN CODE 0000
ARC0640I GDSN02 - ADR012I (SCH)-DSSU (01), 2007.078 16:06:14 DFSMSDSS PROCESSING COMPLETE. HIGHEST RETURN CODE IS 0000
Let us look at another example that requests data set filtering by some masking characters
(Example 7-36).
Example 7-36 FRRECOV DSNAME command to recover a group of data sets
HSEND FRRECOV DSNAME(TESTFR.C*.* TESTFR.N*.*) REPLACE FROMCP(CP1)
When we did a catalog search by using the PDF utility DSLIST (3.4) providing the same data
set name masks, we got a display as shown in Figure 7-44.
Figure 7-44 DSLIST display for data set name masks as used with FRRECOV
Menu Options View Utilities Compilers Help
-------------------------------------------------------------------------------
DSLIST - Data Sets Matching TESTFR.C* 2 new append(s)
Command ===> Scroll ===> CSR

Command - Enter "/" to select action Message Volume
-------------------------------------------------------------------------------
TESTFR.CMD.CLIST MHL0A1
TESTFR.CNTL.JCL MHL0A0
TESTFR.NONFRBA.TEST1 SBOX1E
TESTFR.NONFRBA.TEST2 SBOX1D
***************************** End of Data Set list ****************************

228
z/OS V1R8 DFSMS Technical Update
When you compare the results of data set filtering by using Figure 7-44 on page 227,
Figure 7-45, and Example 7-37, you can verify that during processing of the FRRECOV
command the same data sets are selected as when looking up the catalog by using DSLIST
(3.4).
Figure 7-45 Messages returned to TSO session in case of processing a group of data sets
If some of the matching data set names cannot be recovered because there are no fast
replication backups available, recovery fails for these data sets but command processing
continues (see Figure 7-45 and Example 7-37) and all data sets currently cataloged on a
recoverable copy pool volume will be recovered.
Example 7-37 Messages issued in backup log in case of processing a group of data sets
ARC1801I FAST REPLICATION DATA SET RECOVERY IS STARTING FOR DATA SET TESTFR.C*.*, ***, AT 18:12:53 ON 2007/03/19
ARC1861I THE FOLLOWING 0002 DATA SET(S) WERE SUCCESSFULLY PROCESSED DURING FAST REPLICATION DATA SET RECOVERY:
ARC1861I (CONT.) TESTFR.CMD.CLIST, COPYPOOL=CP1, DEVTYPE=DASD
ARC1861I (CONT.) TESTFR.CNTL.JCL, COPYPOOL=CP1, DEVTYPE=DASD
ARC1862I THE FOLLOWING 0002 DATA SET(S) WERE NOT SELECTED FOR FAST REPLICATION DATA SET RECOVERY PROCESSING:
ARC1862I (CONT.) TESTFR.NONFRBA.TEST1
ARC1862I (CONT.) TESTFR.NONFRBA.TEST2
ARC1802I FAST REPLICATION DATA SET RECOVERY HAS COMPLETED FOR DATA SET TESTFR.C*.*, ***, AT 18:12:53 ON 2007/03/19,
FUNCTION RC=0008, MAXIMUM DATA SET RC=0066
7.4.2 Multi-volume data sets
Multi-volume data sets are supported by FRRECOV DSNAME command processing. But you
should always keep in mind that recovering a multi-volume data set from full volume copies or
dumps may fail. Even if the recovery process ends without any error messages you might end
up in a corrupted data set because the data set’s catalog entry is from than at the time of fast
replication backup.
ARC1001I TESTFR.NONFRBA.TEST1 RECOVER FAILED, RC=0093, REAS=0066-0024
ARC1193I FAST REPLICATION RECOVERY FAILED
ARC1001I TESTFR.NONFRBA.TEST2 RECOVER FAILED, RC=0093, REAS=0066-0024
ARC1193I FAST REPLICATION RECOVERY FAILED
ARC1000I TESTFR.CMD.CLIST RECOVER PROCESSING ENDED
ARC1001I TESTFR.C*.*, *** FRRECOV FAILED, RC=0008, REAS=0000
ARC1808E ONE OR MORE FAILURES OCCURRED DURING FAST REPLICATION RECOVERY OF
ARC1808E (CONT.) DATA SET TESTFR.C*.*, ***
ARC1000I TESTFR.CNTL.JCL RECOVER PROCESSING ENDED
***

Chapter 7. DFSMShsm fast replication
229
Suppose that we want to recover data set TESTFR.DCOLLECT ESDS01, which is a
multi-volume VSAM data set. Its catalog entry is shown in Figure 7-46.
Figure 7-46 Output of LISTC command for TESTFR.DCOLLECT.ESDS01
Recovery when volume pointers in ICF are unchanged
The fast replication backup version we use for recovery is listed inFigure 7-47.
Figure 7-47 LIST CP of the fast replication backup version to be used for recovery
The catalog entry of the data set did not change since the fast replication backup was made.
CLUSTER ------- TESTFR.DCOLLECT.ESDS01
IN-CAT --- UCAT.TESTFR
HISTORY
DATASET-OWNER-----(NULL) CREATION--------2007.058
RELEASE----------------2 EXPIRATION------0000.000
SMSDATA
STORAGECLASS ------HSMFR MANAGEMENTCLASS---MCDB22
DATACLASS --------(NULL) LBACKUP ---0000.000.0000
BWO STATUS------00000000 BWO TIMESTAMP---00000 00:00:00.0
BWO---------------(NULL)
RLSDATA
LOG ----------------(NULL) RECOVERY REQUIRED --(NO) FRLOG -------
----(NULL)
VSAM QUIESCED -------(NO) RLS IN USE ---------(NO)
LOGSTREAMID-----------------------------(NULL)
RECOVERY TIMESTAMP LOCAL-----X'0000000000000000'
RECOVERY TIMESTAMP GMT-------X'0000000000000000'
DATA ------- TESTFR.DCOLLECT.ESDS01.DATA
IN-CAT --- UCAT.TESTFR
HISTORY
DATASET-OWNER-----(NULL) CREATION--------2007.058
RELEASE----------------2 EXPIRATION------0000.000
ACCOUNT-INFO-----------------------------------(NULL)
VOLUMES
VOLSER------------MHL0A0 DEVTYPE------X'3010200F'
VOLSER------------MHL0A1 DEVTYPE------X'3010200F'
-- DFSMShsm CONTROL DATASET --COPY POOL--LISTING --------- AT 13:16:07 ON 07/03

COPYPOOL=CP1
VERSION VTOCENQ DATE TIME FASTREPLICATIONSTATE DUMPSTATE
006 Y 2007/03/12 16:01:45 RECOVERABLE ALLCOMPLETE
TOKEN(C)=C'EX6'
TOKEN(H)=X'C5E7F6'
TOTAL NUM OF VOLUMES=00002,INCREMENTAL=N

SGNAME SOURCE - TARGET SOURCE - TARGET SOURCE - TARGET SOURCE - TARGET
SG1 MHL0A0 - MHL0AF MHL0A1 - MHL1AC

230
z/OS V1R8 DFSMS Technical Update
Execution of the command as shown in Example 7-38 leads to a perfectly recovered data set.
Example 7-38 FRRECOV DSNAME command
FRRECOV DSNAME(TESTFR.DCOLLECT.ESDS01) REPLACE FROMCP(CP1)
The messages in the dump log of DFSMShsm (see Example 7-39) show clearly that two copy
commands are being executed internally by DFSMShsm:
One for the data set’s components on volume MHL0A0
One for the data set’s components on volume MHL0A1
Example 7-39 DFSMShsm dump log while processing FRRECOV for TESTFR.DCOLLECT.ESDS01
ARC0640I GDSN01 - PAGE 0001 5695-DF175 DFSMSDSS V1R08.0 DATA SET SERVICES 2007.081
18:45
ARC0640I GDSN01 - ADR035I (SCH)-PRIME(06), INSTALLATION EXIT ALTERED BYPASS FAC CLASS CHK
DEFAULT TO YES
ARC0640I GDSN01 - COPY DS(INC(TESTFR.DCOLLECT.ESDS01 )) -
ARC0640I GDSN01 - PHYSINDY(MHL0AF) OUTDYNAM(MHL0A0) -
ARC0640I GDSN01 - BYPASSACS(TESTFR.DCOLLECT.ESDS01 ) -
ARC0640I GDSN02 - PAGE 0001 5695-DF175 DFSMSDSS V1R08.0 DATA SET SERVICES 2007.081
18:45
ARC0640I GDSN01 - FASTREPLICATION(PREFERRED) -
ARC0640I GDSN02 - ADR035I (SCH)-PRIME(06), INSTALLATION EXIT ALTERED BYPASS FAC CLASS CHK
DEFAULT TO YES
ARC0640I GDSN01 - ALLDATA(*) ALLEXCP REPLACEU CANCELERROR DEBUG(FRMSG(DTL)) -
ARC0640I GDSN01 - FORCECP(0) PROCESS(SYS1) -
ARC0640I GDSN02 - COPY DS(INC(TESTFR.DCOLLECT.ESDS01 )) -
ARC0640I GDSN01 - STORCLAS(HSMFR ) -
ARC0640I GDSN02 - PHYSINDY(MHL1AC) OUTDYNAM(MHL0A1) -
ARC0640I GDSN01 - MGMTCLAS(MCDB22 )
ARC0640I GDSN02 - BYPASSACS(TESTFR.DCOLLECT.ESDS01 ) -
ARC0640I GDSN01 - ADR101I (R/I)-RI01 (01), TASKID 001 HAS BEEN ASSIGNED TO COMMAND 'COPY '
ARC0640I GDSN02 - FASTREPLICATION(PREFERRED) -
ARC0640I GDSN01 - ADR109I (R/I)-RI01 (01), 2007.081 18:45:50 INITIAL SCAN OF USER CONTROL
STATEMENTS COMPLETED.
ARC0640I GDSN02 - ALLDATA(*) ALLEXCP REPLACEU CANCELERROR DEBUG(FRMSG(DTL)) -
ARC0640I GDSN02 - FORCECP(0) PROCESS(SYS1) -
ARC0640I GDSN02 - STORCLAS(HSMFR ) -
ARC0640I GDSN02 - MGMTCLAS(MCDB22 )
ARC0640I GDSN02 - ADR101I (R/I)-RI01 (01), TASKID 001 HAS BEEN ASSIGNED TO COMMAND 'COPY '
ARC0640I GDSN02 - ADR109I (R/I)-RI01 (01), 2007.081 18:45:50 INITIAL SCAN OF USER CONTROL
STATEMENTS COMPLETED.
ARC0640I GDSN01 - ADR050I (001)-PRIME(02), DFSMSDSS INVOKED VIA CROSS MEMORY APPLICATION
INTERFACE
ARC0640I GDSN01 - ADR016I (001)-PRIME(01), RACF LOGGING OPTION IN EFFECT FOR THIS TASK
ARC0640I GDSN02 - ADR050I (001)-PRIME(02), DFSMSDSS INVOKED VIA CROSS MEMORY APPLICATION
INTERFACE
ARC0640I GDSN02 - ADR016I (001)-PRIME(01), RACF LOGGING OPTION IN EFFECT FOR THIS TASK
ARC0640I GDSN01 - ADR006I (001)-STEND(01), 2007.081 18:45:50 EXECUTION BEGINS
ARC0640I GDSN02 - ADR006I (001)-STEND(01), 2007.081 18:45:50 EXECUTION BEGINS
ARC0640I GDSN01 - ADR442I (001)-PPRVS(01), DATA SET TESTFR.DCOLLECT.ESDS01 PREALLOCATED, IN
CATALOG UCAT.TESTFR,
ON VOLUME(S): MHL0A0
ARC0640I GDSN02 - ADR442I (001)-PPRVS(01), DATA SET TESTFR.DCOLLECT.ESDS01 PREALLOCATED, IN
CATALOG UCAT.TESTFR,

Chapter 7. DFSMShsm fast replication
231
ON VOLUME(S): MHL0A1
ARC0640I GDSN02 - ADR806I (001)-T0MI (03), DATA SET TESTFR.DCOLLECT.ESDS01 COPIED USING A FAST
REPLICATION FUNCTION
ARC0640I GDSN02 - ADR454I (001)-DDDS (01), THE FOLLOWING DATA SETS WERE SUCCESSFULLY PROCESSED
ARC0640I GDSN02 - CLUSTER NAME TESTFR.DCOLLECT.ESDS01
ARC0640I GDSN02 - COMPONENT NAME TESTFR.DCOLLECT.ESDS01.DATA
ARC0640I GDSN02 - ADR006I (001)-STEND(02), 2007.081 18:45:51 EXECUTION ENDS
ARC0640I GDSN02 - ADR013I (001)-CLTSK(01), 2007.081 18:45:51 TASK COMPLETED WITH RETURN CODE
0000
ARC0640I GDSN02 - ADR012I (SCH)-DSSU (01), 2007.081 18:45:51 DFSMSDSS PROCESSING COMPLETE.
HIGHEST RETURN CODE IS 0000
ARC0640I GDSN01 - ADR806I (001)-T0MI (03), DATA SET TESTFR.DCOLLECT.ESDS01 COPIED USING A FAST
REPLICATION FUNCTION
ARC0640I GDSN01 - ADR454I (001)-DDDS (01), THE FOLLOWING DATA SETS WERE SUCCESSFULLY PROCESSED
ARC0640I GDSN01 - CLUSTER NAME TESTFR.DCOLLECT.ESDS01
ARC0640I GDSN01 - COMPONENT NAME TESTFR.DCOLLECT.ESDS01.DATA
ARC0640I GDSN01 - ADR006I (001)-STEND(02), 2007.081 18:45:51 EXECUTION ENDS
ARC0640I GDSN01 - ADR013I (001)-CLTSK(01), 2007.081 18:45:51 TASK COMPLETED WITH RETURN CODE
0000
ARC0640I GDSN01 - ADR012I (SCH)-DSSU (01), 2007.081 18:45:51 DFSMSDSS PROCESSING COMPLETE.
HIGHEST RETURN CODE IS 0000
Recovery after volume count has been reduced
Now we have reorganized the data set and it is no longer made up of pieces spread among
several volumes (see the catalog entry in Figure 7-48).
Figure 7-48 Output of LISTC command for TESTFR.DCOLLECT.ESDS01
CLUSTER ------- TESTFR.DCOLLECT.ESDS01
IN-CAT --- UCAT.TESTFR
HISTORY
DATASET-OWNER-----(NULL) CREATION--------2007.058
RELEASE----------------2 EXPIRATION------0000.000
SMSDATA
STORAGECLASS ------HSMFR MANAGEMENTCLASS---MCDB22
DATACLASS --------(NULL) LBACKUP ---0000.000.0000
BWO STATUS------00000000 BWO TIMESTAMP---00000 00:00:00.0
BWO---------------(NULL)
RLSDATA
LOG ----------------(NULL) RECOVERY REQUIRED --(NO) FRLOG -------
----(NULL)
VSAM QUIESCED -------(NO) RLS IN USE ---------(NO)
LOGSTREAMID-----------------------------(NULL)
RECOVERY TIMESTAMP LOCAL-----X'0000000000000000'
RECOVERY TIMESTAMP GMT-------X'0000000000000000'
DATA ------- TESTFR.DCOLLECT.ESDS01.DATA
IN-CAT --- UCAT.TESTFR
HISTORY
DATASET-OWNER-----(NULL) CREATION--------2007.058
RELEASE----------------2 EXPIRATION------0000.000
ACCOUNT-INFO-----------------------------------(NULL)
VOLUMES
VOLSER------------MHL0A0 DEVTYPE------X'3010200F'
VOLSER------------* DEVTYPE------X'3010200F'

232
z/OS V1R8 DFSMS Technical Update
The backup copy we use for recovery is still the same that we last used (see Figure 7-47 on
page 229). Remember that during creation of this backup the data set was spread among the
two volumes.
The recovery of the data set seems to work fine, as the messages in Example 7-40 lead us to
believe.
Example 7-40 DFSMShsm dump log while processing FRRECOV for TESTFR.DCOLLECT.ESDS01
ARC0640I GDSN01 - PAGE 0001 5695-DF175 DFSMSDSS V1R08.0 DATA SET SERVICES 2007.081
19:36
ARC0640I GDSN01 - ADR035I (SCH)-PRIME(06), INSTALLATION EXIT ALTERED BYPASS FAC CLASS CHK
DEFAULT TO YES
ARC0640I GDSN01 - COPY DS(INC(TESTFR.DCOLLECT.ESDS01 )) -
ARC0640I GDSN01 - PHYSINDY(MHL0AF) OUTDYNAM(MHL0A0) -
ARC0640I GDSN01 - BYPASSACS(TESTFR.DCOLLECT.ESDS01 ) -
ARC0640I GDSN01 - FASTREPLICATION(PREFERRED) -
ARC0640I GDSN01 - ALLDATA(*) ALLEXCP REPLACEU CANCELERROR DEBUG(FRMSG(DTL)) -
ARC0640I GDSN01 - FORCECP(0) PROCESS(SYS1) -
ARC0640I GDSN01 - STORCLAS(HSMFR ) -
ARC0640I GDSN01 - MGMTCLAS(MCDB22 )
ARC0640I GDSN01 - ADR101I (R/I)-RI01 (01), TASKID 001 HAS BEEN ASSIGNED TO COMMAND 'COPY '
ARC0640I GDSN01 - ADR109I (R/I)-RI01 (01), 2007.081 19:36:17 INITIAL SCAN OF USER CONTROL
STATEMENTS COMPLETED.
ARC0640I GDSN01 - ADR050I (001)-PRIME(02), DFSMSDSS INVOKED VIA CROSS MEMORY APPLICATION
INTERFACE
ARC0640I GDSN01 - ADR016I (001)-PRIME(01), RACF LOGGING OPTION IN EFFECT FOR THIS TASK
ARC0640I GDSN01 - ADR006I (001)-STEND(01), 2007.081 19:36:17 EXECUTION BEGINS
ARC0640I GDSN01 - ADR442I (001)-PPRVS(01), DATA SET TESTFR.DCOLLECT.ESDS01 PREALLOCATED, IN
CATALOG UCAT.TESTFR,
ON VOLUME(S): MHL0A0
ARC0640I GDSN01 - ADR390I (001)-PPRVS(01),
DATA SET TESTFR.DCOLLECT.ESDS01 WILL BE SCRATCHED FROM MHL0A0 BECAUSE OF UNMATCHED SIZE. IT
WILL BE
ARC0640I GDSN01 - REALLOCATED
ARC0640I GDSN01 - ADR431I (001)-DYNA (02), DATA SET TESTFR.DCOLLECT.ESDS01.DATA HAS BEEN
DELETED
ARC0640I GDSN01 - ADR396I (001)-PCVSM(01),
DATA SET CLUSTER TESTFR.DCOLLECT.ESDS01 COMPONENT TESTFR.DCOLLECT.ESDS01.DATA ALLOCATED, ON
VOLUME(S):
ARC0640I GDSN01 - MHL0A0
ARC0640I GDSN01 - ADR806I (001)-T0MI (03), DATA SET TESTFR.DCOLLECT.ESDS01 COPIED USING A FAST
REPLICATION FUNCTION
ARC0640I GDSN01 - ADR454I (001)-DDDS (01), THE FOLLOWING DATA SETS WERE SUCCESSFULLY PROCESSED
ARC0640I GDSN01 - CLUSTER NAME TESTFR.DCOLLECT.ESDS01
ARC0640I GDSN01 - COMPONENT NAME TESTFR.DCOLLECT.ESDS01.DATA
ARC0640I GDSN01 - ADR006I (001)-STEND(02), 2007.081 19:36:17 EXECUTION ENDS
ARC0640I GDSN01 - ADR013I (001)-CLTSK(01), 2007.081 19:36:17 TASK COMPLETED WITH RETURN CODE
0000
ARC0640I GDSN01 - ADR012I (SCH)-DSSU (01), 2007.081 19:36:17 DFSMSDSS PROCESSING COMPLETE.
HIGHEST RETURN CODE IS 0000
But when we compare the contents of the data set after the FRRECOV process with the one
before we did the recover, we see that there is some missing data — the piece of data that
was backed up from MHL0A1.

Chapter 7. DFSMShsm fast replication
233
The likelihood for a decrease in volume count is not as high as an increase in volume count (it
depends on the installation’s behavior), but this is an issue that you should keep in mind.
Recovery after volume count has been increased
We tested the option where the volume count has been increased. The data set is spread
among two volumes at the time of recovery because of a volume overflow that took place
some time since the last fast replication backup.
Now we use a different fast replication backup version (see Figure 7-49).
Figure 7-49 LIST CP of the fast replication backup version to be used for recovery
Note: Be very careful when you recover a multi-volume data set from a fast replication
backup and when the volume count of the data set has decreased since the backup was
made.
-- DFSMShsm CONTROL DATASET --COPY POOL--LISTING --------- AT 13:16:07 ON 07/03

COPYPOOL=CP1

VERSION VTOCENQ DATE TIME FASTREPLICATIONSTATE DUMPSTATE
007 Y 2007/03/22 19:39:37 RECOVERABLE NONE
TOKEN(C)=C''
TOKEN(H)=X''
TOTAL NUM OF VOLUMES=00002,INCREMENTAL=N

SGNAME SOURCE - TARGET SOURCE - TARGET SOURCE - TARGET SOURCE - TARGET
SG1 MHL0A0 - MHL1AD MHL0A1 - MHL1AE

234
z/OS V1R8 DFSMS Technical Update
TESTFR.DCOLLECT.ESDS01 was, at time of backup, a single volume data set (see
Figure 7-50).
Figure 7-50 Output of LISTC command for TESTFR.DCOLLECT.ESDS01
The data set was successfully recovered by the process and the catalog entry was changed
accordingly to show just a candidate volume instead of the second volume pointer.
As you can see in the dump log of DFSMShsm (see Example 7-41), a physical data set copy
from MHL1AE to MHL0A1 was tried (GDSN02) but did not end successfully because the data
set was not found on volume MHL1AE (remember that at backup time the data set had no
extents on any volume other than MHL0A0).
Example 7-41 DFSMShsm dump log while processing FRRECOV for TESTFR.DCOLLECT.ESDS01
ARC0640I GDSN01 - PAGE 0001 5695-DF175 DFSMSDSS V1R08.0 DATA SET SERVICES 2007.082
12:41
ARC0640I GDSN01 - ADR035I (SCH)-PRIME(06), INSTALLATION EXIT ALTERED BYPASS FAC CLASS CHK
DEFAULT TO YES
ARC0640I GDSN01 - COPY DS(INC(TESTFR.DCOLLECT.ESDS01 )) -
ARC0640I GDSN01 - PHYSINDY(MHL1AD) OUTDYNAM(MHL0A0) -
ARC0640I GDSN01 - BYPASSACS(TESTFR.DCOLLECT.ESDS01 ) -
ARC0640I GDSN01 - FASTREPLICATION(PREFERRED) -
ARC0640I GDSN01 - ALLDATA(*) ALLEXCP REPLACEU CANCELERROR DEBUG(FRMSG(DTL)) -
ARC0640I GDSN01 - FORCECP(0) PROCESS(SYS1) -
ARC0640I GDSN01 - STORCLAS(HSMFR ) -
ARC0640I GDSN01 - MGMTCLAS(MCDB22 )
ARC0640I GDSN01 - ADR101I (R/I)-RI01 (01), TASKID 001 HAS BEEN ASSIGNED TO COMMAND 'COPY '
CLUSTER ------- TESTFR.DCOLLECT.ESDS01
IN-CAT --- UCAT.TESTFR
HISTORY
DATASET-OWNER-----(NULL) CREATION--------2007.058
RELEASE----------------2 EXPIRATION------0000.000
SMSDATA
STORAGECLASS ------HSMFR MANAGEMENTCLASS---MCDB22
DATACLASS --------(NULL) LBACKUP ---0000.000.0000
BWO STATUS------00000000 BWO TIMESTAMP---00000 00:00:00.0
BWO---------------(NULL)
RLSDATA
LOG ----------------(NULL) RECOVERY REQUIRED --(NO) FRLOG -------
----(NULL)
VSAM QUIESCED -------(NO) RLS IN USE ---------(NO)
LOGSTREAMID-----------------------------(NULL)
RECOVERY TIMESTAMP LOCAL-----X'0000000000000000'
RECOVERY TIMESTAMP GMT-------X'0000000000000000'
DATA ------- TESTFR.DCOLLECT.ESDS01.DATA
IN-CAT --- UCAT.TESTFR
HISTORY
DATASET-OWNER-----(NULL) CREATION--------2007.058
RELEASE----------------2 EXPIRATION------0000.000
ACCOUNT-INFO-----------------------------------(NULL)
VOLUMES
VOLSER------------MHL0A0 DEVTYPE------X'3010200F'
VOLSER------------* DEVTYPE------X'3010200F'

Chapter 7. DFSMShsm fast replication
235
ARC0640I GDSN01 - ADR109I (R/I)-RI01 (01), 2007.082 12:41:45 INITIAL SCAN OF USER CONTROL
STATEMENTS COMPLETED.
ARC0640I GDSN01 - ADR050I (001)-PRIME(02), DFSMSDSS INVOKED VIA CROSS MEMORY APPLICATION
INTERFACE
ARC0640I GDSN01 - ADR016I (001)-PRIME(01), RACF LOGGING OPTION IN EFFECT FOR THIS TASK
ARC0640I GDSN01 - ADR006I (001)-STEND(01), 2007.082 12:41:45 EXECUTION BEGINS
ARC0640I GDSN02 - PAGE 0001 5695-DF175 DFSMSDSS V1R08.0 DATA SET SERVICES 2007.082
12:41
ARC0640I GDSN02 - ADR035I (SCH)-PRIME(06), INSTALLATION EXIT ALTERED BYPASS FAC CLASS CHK
DEFAULT TO YES
ARC0640I GDSN02 - COPY DS(INC(TESTFR.DCOLLECT.ESDS01 )) -
ARC0640I GDSN02 - PHYSINDY(MHL1AE) OUTDYNAM(MHL0A1) -
ARC0640I GDSN02 - BYPASSACS(TESTFR.DCOLLECT.ESDS01 ) -
ARC0640I GDSN02 - FASTREPLICATION(PREFERRED) -
ARC0640I GDSN02 - ALLDATA(*) ALLEXCP REPLACEU CANCELERROR DEBUG(FRMSG(DTL)) -
ARC0640I GDSN02 - FORCECP(0) PROCESS(SYS1) -
ARC0640I GDSN02 - STORCLAS(HSMFR ) -
ARC0640I GDSN02 - MGMTCLAS(MCDB22 )
ARC0640I GDSN02 - ADR101I (R/I)-RI01 (01), TASKID 001 HAS BEEN ASSIGNED TO COMMAND 'COPY '
ARC0640I GDSN02 - ADR109I (R/I)-RI01 (01), 2007.082 12:41:45 INITIAL SCAN OF USER CONTROL
STATEMENTS COMPLETED.
ARC0640I GDSN02 - ADR050I (001)-PRIME(02), DFSMSDSS INVOKED VIA CROSS MEMORY APPLICATION
INTERFACE
ARC0640I GDSN02 - ADR016I (001)-PRIME(01), RACF LOGGING OPTION IN EFFECT FOR THIS TASK
ARC0640I GDSN02 - ADR006I (001)-STEND(01), 2007.082 12:41:45 EXECUTION BEGINS
ARC0640I GDSN02 - ADR801I (001)-DDDS (01),
DATA SET FILTERING IS COMPLETE. 0 OF 0 DATA SETS WERE SELECTED: 0 FAILED SERIALIZATION AND 0
FAILED FOR
ARC0640I GDSN02 - OTHER REASONS.
ARC0640I GDSN02 - ADR383W (001)-DDDS (01), DATA SET TESTFR.DCOLLECT.ESDS01 NOT SELECTED
ARC0640I GDSN02 - ADR470W (001)-DDDS (04), NO DATA SETS SELECTED FOR PROCESSING
ARC0640I GDSN02 - ADR006I (001)-STEND(02), 2007.082 12:41:45 EXECUTION ENDS
ARC0640I GDSN02 - ADR013I (001)-CLTSK(01), 2007.082 12:41:45 TASK COMPLETED WITH RETURN CODE
0004
ARC0640I GDSN02 - ADR012I (SCH)-DSSU (01),
2007.082 12:41:45 DFSMSDSS PROCESSING COMPLETE. HIGHEST RETURN CODE IS 0004 FROM:
ARC0640I GDSN02 - TASK 001
ARC0640I GDSN01 - ADR442I (001)-PPRVS(01), DATA SET TESTFR.DCOLLECT.ESDS01 PREALLOCATED, IN
CATALOG UCAT.TESTFR,
ON VOLUME(S): MHL0A0
ARC0640I GDSN01 - ADR390I (001)-PPRVS(01),
DATA SET TESTFR.DCOLLECT.ESDS01 WILL BE SCRATCHED FROM MHL0A0 BECAUSE OF UNMATCHED SIZE. IT
WILL BE
ARC0640I GDSN01 - REALLOCATED
ARC0640I GDSN01 - ADR431I (001)-DYNA (02), DATA SET TESTFR.DCOLLECT.ESDS01.DATA HAS BEEN
DELETED
ARC0640I GDSN01 - ADR396I (001)-PCVSM(01),
DATA SET CLUSTER TESTFR.DCOLLECT.ESDS01 COMPONENT TESTFR.DCOLLECT.ESDS01.DATA ALLOCATED, ON
VOLUME(S):
ARC0640I GDSN01 - MHL0A0
ARC0640I GDSN01 - ADR806I (001)-T0MI (03), DATA SET TESTFR.DCOLLECT.ESDS01 COPIED USING A FAST
REPLICATION FUNCTION
ARC0640I GDSN01 - ADR454I (001)-DDDS (01), THE FOLLOWING DATA SETS WERE SUCCESSFULLY PROCESSED
ARC0640I GDSN01 - CLUSTER NAME TESTFR.DCOLLECT.ESDS01
ARC0640I GDSN01 - COMPONENT NAME TESTFR.DCOLLECT.ESDS01.DATA

236
z/OS V1R8 DFSMS Technical Update
ARC0640I GDSN01 - ADR006I (001)-STEND(02), 2007.082 12:41:46 EXECUTION ENDS
ARC0640I GDSN01 - ADR013I (001)-CLTSK(01), 2007.082 12:41:46 TASK COMPLETED WITH RETURN CODE
0000
ARC0640I GDSN01 - ADR012I (SCH)-DSSU (01), 2007.082 12:41:46 DFSMSDSS PROCESSING COMPLETE.
HIGHEST RETURN CODE IS 0000
During further processing the current data set was deleted, reallocated, and restored from the
copy that was found on MHL1AD, which is the target volume of MHL0A0.
Conclusion
If you recover multi-volume data sets you should check the results of FRRECOV.
But how can you remember the catalog entry of a data set at backup time? Basically, you
need some kind of VTOC copies that you can provide easily as part of the fast replication
dump process. Another option is to create your own meta data by using IDCAMS
DCOLLECT. Go to 7.4.4, “Hints and tips for the recovery of data sets that no longer exist” on
page 239, for more information.
7.4.3 Data set recovery from fast replication dumps
Fast replication recovery at the data set level is initiated by a command, as shown in
Example 7-42.
Example 7-42 FRRECOV DSNAME command to request recovery from fast replication dump
HSEND FRRECOV DSNAME(TESTFR.DCOLLECT.ESDS01) REPLACE FROMCP(CP1) GEN(1) FROMDUMP

Chapter 7. DFSMShsm fast replication
237
The recovery was done successfully based on the catalog entry, as shown in Figure 7-51.
Figure 7-51 Output of LISTC command for TESTFR.DCOLLECT.ESDS01
CLUSTER ------- TESTFR.DCOLLECT.ESDS01
IN-CAT --- UCAT.TESTFR
HISTORY
DATASET-OWNER-----(NULL) CREATION--------2007.058
RELEASE----------------2 EXPIRATION------0000.000
SMSDATA
STORAGECLASS ------HSMFR MANAGEMENTCLASS---MCDB22
DATACLASS --------(NULL) LBACKUP ---0000.000.0000
BWO STATUS------00000000 BWO TIMESTAMP---00000 00:00:00.0
BWO---------------(NULL)
RLSDATA
LOG ----------------(NULL) RECOVERY REQUIRED --(NO) FRLOG -------
----(NULL)
VSAM QUIESCED -------(NO) RLS IN USE ---------(NO)
LOGSTREAMID-----------------------------(NULL)
RECOVERY TIMESTAMP LOCAL-----X'0000000000000000'
RECOVERY TIMESTAMP GMT-------X'0000000000000000'
DATA ------- TESTFR.DCOLLECT.ESDS01.DATA
IN-CAT --- UCAT.TESTFR
HISTORY
DATASET-OWNER-----(NULL) CREATION--------2007.058
RELEASE----------------2 EXPIRATION------0000.000
ACCOUNT-INFO-----------------------------------(NULL)
VOLUMES
VOLSER------------MHL0A0 DEVTYPE------X'3010200F'
VOLSER------------MHL0A1 DEVTYPE------X'3010200F'

238
z/OS V1R8 DFSMS Technical Update
The fast replication dump that we requested is listed in Figure 7-52.
Figure 7-52 LIST CP of the fast replication backup version to be used for recovery
The excerpt from the DFSMShsm dump log (see Example 7-43) shows that now restore
commands instead of copy commands are being executed. The processing is at the physical
data set level as well as with copy. The entire process is driven by the catalog entry of the
data set being requested for recovery. As you can see in Example 7-43, there are two restore
processes being executed according to the volume pointers in the catalog entry (MHL0A0
and MHL0A1).
Example 7-43 DFSMShsm dump log while processing FRRECOV for TESTFR.DCOLLECT.ESDS01
ARC0640I GDSN01 - PAGE 0001 5695-DF175 DFSMSDSS V1R08.0 DATA SET SERVICES 2007.082
17:40
ARC0640I GDSN01 - ADR035I (SCH)-PRIME(06), INSTALLATION EXIT ALTERED BYPASS FAC CLASS CHK
DEFAULT TO YES
ARC0640I GDSN01 - ADR035I (SCH)-PRIME(03), INSTALLATION EXIT ALTERED WORKUNIT DEFAULT TO
ARC0640I GDSN01 - RESTORE DS(INCLUDE(TESTFR.DCOLLECT.ESDS01 )) -
ARC0640I GDSN01 - INDDNAME(SYS03708) OUTDDNAME(SYS03709) REPLACE CANCELERROR -
ARC0640I GDSN01 - BYPASSACS(TESTFR.DCOLLECT.ESDS01 ) -
ARC0640I GDSN01 - STORCLAS(STANDARD ) FORCECP(0) -
ARC0640I GDSN01 - MGMTCLAS(MCDB22 )
ARC0640I GDSN01 - ADR101I (R/I)-RI01 (01), TASKID 001 HAS BEEN ASSIGNED TO COMMAND 'RESTORE '
ARC0640I GDSN01 - ADR109I (R/I)-RI01 (01), 2007.082 17:40:34 INITIAL SCAN OF USER CONTROL
STATEMENTS COMPLETED.
ARC0640I GDSN01 - ADR050I (001)-PRIME(01), DFSMSDSS INVOKED VIA APPLICATION INTERFACE
ARC0640I GDSN01 - ADR016I (001)-PRIME(01), RACF LOGGING OPTION IN EFFECT FOR THIS TASK
ARC0640I GDSN01 - ADR006I (001)-STEND(01), 2007.082 17:40:34 EXECUTION BEGINS
ARC0640I GDSN01 - ADR780I (001)-TDDS (01),
-- DFSMShsm CONTROL DATASET --COPY POOL--LISTING --------- AT 18:02:44 ON 07/03

COPYPOOL=CP1
VERSION VTOCENQ DATE TIME FASTREPLICATIONSTATE DUMPSTATE
006 Y 2007/03/12 16:01:45 RECOVERABLE ALLCOMPLETE
TOKEN(C)=C'EX6'
TOKEN(H)=X'C5E7F6'
TOTAL NUM OF VOLUMES=00002,INCREMENTAL=N

SGNAME SOURCE - TARGET SOURCE - TARGET SOURCE - TARGET SOURCE - TARGET
SG1 MHL0A0 - MHL0AF MHL0A1 - MHL1AC

DUMPCLASS REQUIRED DUMPSTATE VOLSSUC EXPDATE AVAILABLE
DCREDB18 Y COMPLETE 00002 2008/03/02 Y

HWCOMP ENCRYPT ENCTYPE RSAKEY/KPWD
NO NONE ********* ***************************************

SOURCE DUMPVOLS DEVICE TYPE
MHL0A0 TST007 3590-1
FILE SEQ=01, DSNAME=HSM.DMP.DCREDB18.VMHL0A0.D07071.T450116
MHL0A1 TST007 3590-1
FILE SEQ=02, DSNAME=HSM.DMP.DCREDB18.VMHL0A1.D07071.T450116

Chapter 7. DFSMShsm fast replication
239
THE INPUT DUMP DATA SET BEING PROCESSED IS IN FULL VOLUME FORMAT AND WAS CREATED BY DFSMSDSS
VERSION 1
ARC0640I GDSN01 - RELEASE 8 MODIFICATION LEVEL 0
ARC0640I GDSN01 - ADR378I (001)-TDDS (02), THE FOLLOWING DATA SETS WERE SUCCESSFULLY PROCESSED
FROM VOLUME MHL0A0
ARC0640I GDSN01 - TESTFR.DCOLLECT.ESDS01
TESTFR.DCOLLECT.ESDS01.DATA
ARC0640I GDSN01 - ADR006I (001)-STEND(02), 2007.082 17:46:36 EXECUTION ENDS
ARC0640I GDSN01 - ADR013I (001)-CLTSK(01), 2007.082 17:46:36 TASK COMPLETED WITH RETURN CODE
0000
ARC0640I GDSN01 - ADR012I (SCH)-DSSU (01), 2007.082 17:46:36 DFSMSDSS PROCESSING COMPLETE.
HIGHEST RETURN CODE IS 0000
ARC0640I GDSN01 - PAGE 0001 5695-DF175 DFSMSDSS V1R08.0 DATA SET SERVICES 2007.082
17:46
ARC0640I GDSN01 - ADR035I (SCH)-PRIME(06), INSTALLATION EXIT ALTERED BYPASS FAC CLASS CHK
DEFAULT TO YES
ARC0640I GDSN01 - ADR035I (SCH)-PRIME(03), INSTALLATION EXIT ALTERED WORKUNIT DEFAULT TO
ARC0640I GDSN01 - RESTORE DS(INCLUDE(TESTFR.DCOLLECT.ESDS01 )) -
ARC0640I GDSN01 - INDDNAME(SYS03708) OUTDDNAME(SYS03712) REPLACE CANCELERROR -
ARC0640I GDSN01 - BYPASSACS(TESTFR.DCOLLECT.ESDS01 ) -
ARC0640I GDSN01 - STORCLAS(STANDARD ) FORCECP(0) -
ARC0640I GDSN01 - MGMTCLAS(MCDB22 )
ARC0640I GDSN01 - ADR101I (R/I)-RI01 (01), TASKID 001 HAS BEEN ASSIGNED TO COMMAND 'RESTORE '
ARC0640I GDSN01 - ADR109I (R/I)-RI01 (01), 2007.082 17:46:36 INITIAL SCAN OF USER CONTROL
STATEMENTS COMPLETED.
ARC0640I GDSN01 - ADR050I (001)-PRIME(01), DFSMSDSS INVOKED VIA APPLICATION INTERFACE
ARC0640I GDSN01 - ADR016I (001)-PRIME(01), RACF LOGGING OPTION IN EFFECT FOR THIS TASK
ARC0640I GDSN01 - ADR006I (001)-STEND(01), 2007.082 17:46:36 EXECUTION BEGINS
ARC0640I GDSN01 - ADR780I (001)-TDDS (01),
THE INPUT DUMP DATA SET BEING PROCESSED IS IN FULL VOLUME FORMAT AND WAS CREATED BY DFSMSDSS
VERSION 1
ARC0640I GDSN01 - RELEASE 8 MODIFICATION LEVEL 0
ARC0640I GDSN01 - ADR378I (001)-TDDS (02), THE FOLLOWING DATA SETS WERE SUCCESSFULLY PROCESSED
FROM VOLUME MHL0A1
ARC0640I GDSN01 - TESTFR.DCOLLECT.ESDS01
TESTFR.DCOLLECT.ESDS01.DATA
ARC0640I GDSN01 - ADR006I (001)-STEND(02), 2007.082 17:49:39 EXECUTION ENDS
ARC0640I GDSN01 - ADR013I (001)-CLTSK(01), 2007.082 17:49:39 TASK COMPLETED WITH RETURN CODE
0000
ARC0640I GDSN01 - ADR012I (SCH)-DSSU (01), 2007.082 17:49:39 DFSMSDSS PROCESSING COMPLETE.
HIGHEST RETURN CODE IS 0000
7.4.4 Hints and tips for the recovery of data sets that no longer exist
If the data set that you need to recover was deleted and you want to recover it from a fast
replication backup, you need at least a catalog entry for the data set. The most simple and
common way to meet this requirement is to allocate an empty dummy data set on the
volumes.
You do not need to remember the exact DCB or size because if it does not fit, physical copy
deletes and reallocates the data set according to the actual needs. There are two important
things that you must know:
VSAM (including type, like KSDS, ESDS, and so on) or non-VSAM
Volumes where the data set resided

240
z/OS V1R8 DFSMS Technical Update
If you recover a data set from a fast replication dump copy you could find the information in a
VTOCCOPY data set that was created under control of DFSMShsm (see Figure 7-53).
If you want to recover from a DASD fast replication copy a VTOCCOPY data set is not
available.
You will find all this information in DCOLLECT output (see Figure 7-55 on page 241) of the
volumes that are being backed up by using fast replication.
Figure 7-53 Dump VTOCCOPY data set created by DFSMShsm based on dump class setting
When you run a job (Figure 7-54) right before you start a fast replication for a copy pool, you
can preserve the volume pointers as well as the information of the type of VSAM for future
use.
Figure 7-54 Example JCL for execution of a DCOLLECT run
We recommend reusing a data set for output that is allocated on one of the copy pool
volumes. This makes sure that you never become confused about which data set provides
accurate information for a particular fast replication backup version that you intend to use for
recovery.
You must be prepared to not be able to open a copy of this data set from one of the target
volumes. So, if you need to access more than the current content of such a data set, start
BROWSE HSM.DUMPVTOC.T450116.VMHL0A0.D07071 Line 00000000 Col 001 080
Command ===> Scroll ===> CSR
********************************* Top of Data **********************************
SYS1.VTOCIX.MHL0A0 ..Å......... ...h....{..............
TESTFR.DCOLLECT.D997 ..,.....,... ..Ø{....Ø..............
SYS1.VVDS.VMHL0A0 ..Å...........Ø.Ø....Ø..............
TESTFR.DCOLLECT.D999 ..,.....,... ..Ø{....Ø..............
TESTFR.DCOLLECT.ESDS0.DATA ..,.....,.....Ø.{....Ø..............
TESTFR.CNTL.JCL ..,.....,......ØØ....μ..............
TESTFR.DCOLLECT.ESDS01.DATA ..,.....,.....Ø.{....Ø..............
TESTFR.SRCHFOR.LIST ..,.....,... ..Ø&..À.Ø..............
TESTFR.MHLRES2.MVOL.DATA ..,.....,... ..ØØ..À.Ø..............
UCAT.TESTFR ..,.....,.å...{.{....Ø..............
UCAT.TESTFR.CATINDEX ..,...........{.h....Ø..............
******************************** Bottom of Data ********************************
//STEP1 EXEC PGM=IDCAMS
//*
//*
//SYSPRINT DD SYSOUT=*
//DCOUT DD DSN=MHLRES2.DCOLLECT.SG1,DISP=(MOD,CATLG),
// SPACE=(CYL,(1,1),RLSE),RECFM=VB,LRECL=644,
// STORCLAS=HSMFR
//SYSIN DD *
DCOLLECT -
OUTFILE(DCOUT) -
STORAGEGROUP( -
SG1 -
) -

/* END OF DCOLLECT COMMAND

Chapter 7. DFSMShsm fast replication
241
FRRECOV for this particular data set from the desired fast replication backup or dump
generation. This should never be a problem if you really reuse (not reallocate) the data set
when you run the next DCOLLECT since the volume pointers in the catalog entry should not
change.
Figure 7-55 Output of DCOLLECT STORAGEGROUP(SG1)
In order to find information about the data set names that you are interested in it is easiest to
browse the DCOLLECT data set and to use the find command.
When all required information is available you can allocate a dummy data set. You should
consider using a storage class that was defined with the guaranteed space attribute in order
to make sure that the dummy data set is allocated on the same volumes as it was at time of
backup. Do not forget to alter the storage class after allocation of the data set. During
FRRECOV DSNAME processing there is no call for ACS routine services, so the storage
class is not redetermined automatically.
After you have finished the preparation activities you can try the FRRECOV DSNAME
command for the data set. While the physical data set copy is a physical process that works
at the tracks level to copy a data set, you never need to worry about the physical location of a
data set’s extents. DFSMSdss is able to use any free extents on a volume for output of the
data set copy process. You might get into trouble if the current free space on the volumes is
not sufficient to restore the data set. If this is the case, recovery fails, so you must retry the
FRRECOV DSNAME command after you have moved some data sets off the volumes.
7.5 Reporting on the DFSMShsm fast replication environment
In this section we discuss reporting in the fast replication environment.
BROWSE MHLRES2.DCOLLECT.SG1 Line 00000016 Col 005 084
Command ===> Scroll ===> CSR
D ..SC64.(..........TESTFR.DCOLLECT.ESDS01.DATA . ......{.MHL0A0
A ..SC64.(..........TESTFR.DCOLLECT.ESDS01.DATA TESTFR.DCOLLECT.
D ..SC64.(..........UCAT.TESTFR .â¦.....{.MHL0A0
A ..SC64.(..........UCAT.TESTFR UCAT.TESTFR
D ..SC64.(..........SYS1.VTOCIX.MHL0A0 . ±... .Ø.MHL0A0
D ..SC64.(..........TESTFR.MHLRES2.MVOL.DATA . ±... .&.MHL0A0
V ..SC64.(..........MHL0A1X...........¢...¢................h...?3390 Ø.....SG
D ..SC64.(..........SYS1.VVDS.VMHL0A1 . ........MHL0A1
A ..SC64.(..........SYS1.VVDS.VMHL0A1 SYS1.VVDS.VMHL0A
D ..SC64.(..........TESTFR.DCOLLECT.D987 .à.... .&.MHL0A1
D ..SC64.(..........TESTFR.DCOLLECT.D989 .à.... .&.MHL0A1
D ..SC64.(..........TESTFR.CMD.CLIST .è......°.MHL0A1
D ..SC64.(..........TESTFR.MHLRES2.DUMMY . .... .&.MHL0A1
D ..SC64.(..........TESTFR.MHLRES2.MVOL.DATA . .... .&.MHL0A1
D ..SC64.(..........TESTFR.DCOLLECT.ESDS1.DATA . ......{.MHL0A1
A ..SC64.(..........TESTFR.DCOLLECT.ESDS1.DATA TESTFR.DCOLLECT.
D ..SC64.(..........TESTFR.SUPERC.LIST . .... .m.MHL0A1
D ..SC64.(..........TESTFR.DCOLLECT.ESDS01.DATA . ......{.MHL0A1
A ..SC64.(..........TESTFR.DCOLLECT.ESDS01.DATA TESTFR.DCOLLECT.
D ..SC64.(..........SYS1.VTOCIX.MHL0A1 . ±... .Ø.MHL0A1

242
z/OS V1R8 DFSMS Technical Update
7.5.1 Statistic records
DFSMShsm writes its own statistic records. This is not a statistic record created by a z/OS
function such as dynamic allocation. The command SETSYS SMF or NOSMF tells
DFSMShsm to write or not write SMF records.
The optional parameters SMF(smfid) and NOSMF are mutually exclusive. They specify
whether DFSMShsm writes SMF records that contain DFSMShsm statistics.
SMF(smfid) specifies that DFSMShsm write SMF records in the SYS1.MANX or
SYS1.MANY system data sets.
For smfid, substitute a record identification. Use SMF user codes for the record
identification in the range of 128 to 255.
If you specify smfid, DFSMShsm writes records with SMF identifications of smfid and
smfid+1. Records with an identification of smfid contain daily statistics (DSR) and volume
statistics (VSR). Records with an identification of smfid+1 contain function statistics (FSR)
and ABARS function statistics (WWFSR).
NOSMF specifies that no SMF records are to be written.
7.5.2 REPORT command
The REPORT command has been updated with z/OS DFSMS V1.5 in support of DFSMShsm
fast replication.
REPORT DAILY command
The REPORT DAILY command has been updated with z/OS DFSMS 1.5 to report the total
number of volumes for which a fast replication backup or recovery has been requested. The
total number of failures is also reported. The two lines in Example 7-44 have been added to
the REPORT DAILY heading.
There were no updates with z/OS DFSMS 1.8. Processing of dumps of copy pool volumes is
counted in the existing fields:
FULL VOLUME DUMPS=
DUMP COPIES=
FULL VOLUME RESTORES=
Example 7-44 Updates to the REPORT DAILY heading with z/OS DFSMS V1.5
FAST REPLICATION VOLUME BACKUPS = 00000004 REQUESTED, 00000000 FAILED
FAST REPLICATION VOLUME RECOVERS = 00000001 REQUESTED, 00000000 FAILED
REPORT VOLUMES command
The REPORT VOLUMES command has been updated with z/OS DFSMS V1.5 to report the
total number times that a fast replication backup was successfully created for the volume and
the number of failed attempts for that volume. Recover statistics are also reported. The two
lines in Example 7-45 on page 243 have been added to the REPORT VOLUMES heading.
Note: When a DFSMShsm function executes, only selected fields within the FSR
record are set. The fields that are set depend on the function that is being performed
and the method that is used to request the function. The fields that are set are required
by the DFSMShsm REPORT command.

Chapter 7. DFSMShsm fast replication
243
There were no updates with z/OS DFSMS V1.8. Processing of dumps of copy pool volumes
is reported in the existing fields:
VOLUME DUMP=
DUMP COPIES=
FULL VOLUME RESTORES=
Note that dumps of copy pool volumes are created by processing volumes of copy pool
backup storage groups, but you will never find statistics about volumes of copy pool backup
storage groups in DFSMShsm.
Example 7-45 Updates to the REPORT VOLUMES heading with z/OS DFSMS 1.5
FAST REPLICATION BACKUP = 0000 REQUESTED, 0000 FAILED
FAST REPLICATION RECOVER = 0000 REQUESTED, 0000 FAILED
REPORT statistics
The two REPORT options show the current statistics (DSR and VSR) for the DFSMShsm
address space. The statistics are accumulated in the DFSMShsm work space and are written
to the MCDS and to SMF under the following conditions:
For DSR
– At the start of a new hour
– When a REPORT command is entered
– When a QUERY STATISTICS command is entered
– When DFSMShsm shuts down
– At the start of a new day (the previous day’s record is written)
For VSR
– Whenever a daily statistics record is written
– At the end of secondary space management
– After space management has been performed on a volume
For FSR
– The FSR is both one of the DFSMShsm statistic records and a control block that
contains statistics for a particular function that is performed. It is maintained in the
DFSMShsm workspace until a function such as FRBACKUP has completed.
– Upon completion of the function, the record is written to the DFSMShsm log and
accumulated by category into the daily and volume statistics records in the migration
control data set (MCDS).
– The FSR is also written to the SMF data sets (SYS1.MANx or SYS1.MANy) if SETSYS
SMF is specified.
Note: DFSMShsm activity (recalls, migrations, backups, recovers, dumps, recycles,
restores, and deletions of migrated data sets) must occur in order for DFSMShsm to
recognize the start of a new hour or day.
Tip: If statistics are needed for a week or several days, they may be extracted more
easily from the SMF data, not only out of the address space or MCDS. It is also easier
to use REXX™ or other programs to analyze them.

244
z/OS V1R8 DFSMS Technical Update
7.5.3 DSR records
There are no updates in the DSR records with z/OS DFSMS V1.8.
7.5.4 Updated VSR records
There are no updates in the VSR records with z/OS DFSMS V1.8.
7.5.5 New fields in FSR records
Fields related to fast replication are shown in bold in Table 7-2.
Table 7-2 Function statistics record extract of the new and updated fields
Offsets Type Length Name Description
42 (2A) FIXED 1 FSRTYPE DFSMShsm function type. The function types
are as follows:
1=Primary to level 1 migration
2=Level 1 to level 2 migration,
or level 1 to level 1 migration
or level 2 to level 2 migration
3=Primary to level 2 migration
4=Recall from level 1 to primary
5=Recall from level 2 to primary
6=Delete a migrated data set
7=Daily backup
8=Spill backup
9=Recovery
10=Recycle backup volume
11=Data set deletion by age
12=Recycle migration volume
13=Full volume dump
14=Volume or data set restore
15=ABACKUP function (see WWFSR control
block)
16=ARECOVER function (see WWFSR
control block)
17=Expire primary or migrated data sets
18=Partrel function
19=Expire or roll off incremental backup
version
20=(H)BDELETE an incremental backup
version
21=Fast replication backup function
22=Fast replication recover function
23=Fast replication delete function
131 (83) BITSTRING 1 FSRMFLGS Flags from the MWE.
1... ....FSRFRTRY When set to 1, the backup copy was made
during a retry, after the first try failed because
the data set was in use.
.1.. ....FSRF_REMO
TE
When set to 1, this request completed
successfully on a remote system.

Chapter 7. DFSMShsm fast replication
245
..1. ....FSRFPIGB When set to 1, the request was completed
using a tape already mounted.
...1 ....FSRF_REMO
TE_ HOST_
PROCESSED
When set to 1, MWE processed by remote
host.
.... 1...FSRF_DASD When set to 1, the DASD copy of the version
was deleted.
.... .1..FSRF_DUMP
CPY
When set to 1, the dump class of the copy pool
dump was deleted.
.... ..1.FSRF_DUMP
VER
When set to 1, the entire dump version of the
copy pool version was deleted.
.... ...1 FSRF_COPY
POOL _
FRDUMP
A value of 1 indicates a fast replication dump
or restore.
191 (BF) FIXED 1 FSR_COPYM
ETHOD
Requested method of fast replication. The
valid methods are as follows:
1=None
2=Preferred
3=Required
223 (DF) BITSTRING 1 FSRFLG4 Flags.
1... ....FSRF_FRRE
COV_
DSNAME
When set to 1, fast replication recovery was
requested for a data set through the
FRRECOV DSNAME command.
.1.. ....FSRF_FRRE
COV_
FROMDISK
When set to 1, fast replication recovery will be
performed from disk. This flag is set only when
FSRF_FRRECOV_DSNAME is set to 1.
..1. ....FSRF_MULT_
DSNAMES
When set to 1, the fast replication recovery
request specified more than one data set
name.
...1 ....FSRF_MULTI
VOLUME
When set to 1, the fast replication recovery
request was performed on part of a
multi-volume data set. This flag is set only
when FSRF_FRRECOV_DSNAME is set to 1.
.... 1...FSRF_ALTER
PRI
When set to 1, the priority of this request was
altered through the ALTERPRI command.
.... .1..FSRF_ALTER
PRI_HI
When set to 1, the HIGH keyword was
specified on the ALTERPRI command. When
set to 0, the LOW keyword was specified. This
flag applies only when FSRF_ALTERPRI is
set to 1.
.... ..xx * Reserved.
Offsets Type Length Name Description

246
z/OS V1R8 DFSMS Technical Update
The FSR SMF record number depends on the setting of SETSYS SMF(smfid). If smfid is
specified as 240, then the FSRs will be 241 SMF records. The FSR SMF record type is
always equal to smfid+1.
For more information about SMF records or DFSMShsm FSR records, refer to the following
publications:
z/OS MVS System Management Facilities (SMF), SA22-7630
DFSMShsm Diagnosis, GC52-1083
DFSMShsm Implementation and Customization Guide, SC35-0418
Gathering the data
Prior to producing reports for DFSMShsm fast replication, gather the appropriate data. The
first step is to determine the SMF ID that is related to DFSMShsm Functional Statistics
Records. You can do this several ways:
Method 1: Use the QUERY SETSYS command. The output in the ARC0150I message
contains the SMFID, as shown in Example 7-46.
Example 7-46 SMFID in QUERY SETSYS output
ARC0150I JOURNAL={NONE | SPEED | RECOVERY}, LOG={YES | NO | HELD}, TRACE={YES | NO},
SMFID={smfid | NONE}, DEBUG={YES | NO}, EMERG={YES | NO}, JES={2 | 3}, SYS1DUMP={YES | NO},
RACFIND={YES | NO}, ERASEONSCRATCH={YES | NO}
Method 2: Review the system ARCCMDxx member of SYS1.PARMLIB.
Look for SETSYS SMFID(smfid) and use it to determine whether SMF records are being
generated, and if so which one to use:
– DSR and VSR = smfid.
– FSR and WWFSR = smfid + 1.
(WWFSRs are the ABARSs function statistics)
For example, an SMFID=240 means that FSR records are type 241.
– SMFID=NONE means there are no records being collected.
248 (F8) CHARACTER 40 FSR_CPNAM
E
Copy pool name. This flag is set only when
FSRF_FRRECOV_DSNAME is set to 1.
248 (F8) CHARACTER 8(5) FSRDCLAS A 5-element array consisting of 8-byte fields
containing the names of dump classes.
292 (124) FIXED 4 FSR_FR_REA
S
Fast replication return code.
Offsets Type Length Name Description

Chapter 7. DFSMShsm fast replication
247
All FSR types are shown in Figure 7-56:
Function types 1 through 14 and 17 through 23 are FSR records.
Function types 15 and 16 are ABARS WWFSR records.
For details, see Section 1.2.1 in the manual DFSMShsm Implementation and Customization
Guide, SC35-0418.
Figure 7-56 FSR type records
Preparing for analysis
To begin analyzing the data, first combine SMF data for all of the systems that share the
same DFSMShsm Control Data Sets (MCDS, BCDS, OCDS) into a single file. We eventually
want to convert the SMF data to variable-blocked (VB) format to be processed by a REXX
exec.
The SMF dump program (IFASMFDP) is used to transfer the contents of the SMF data set to
another data set and to reset the status of the dumped data set to empty so that SMF can use
it again for recording data. The SMF dump program dumps the contents of multiple VSAM or
QSAM data sets to sequential data sets on either tape or direct-access devices. The SMF
dump program enables the installation to route different records to separate files and produce
a summary activity report.
The IFASMFDP program typically produces data as RECFM=VBS. In order for the REXX tool
FSRSTAT to process the IFASMFDP output, you must convert the format to RECFM=VB.
You can use IDCAMS REPRO to convert the DFSMShsm SMF data to RECFM=VB,
LRECL=4096. Do not be tempted to use the IFASMFDP program to create RECFM=VB,
LRECL=4096 data directly.
DFSMShsm function type (FSRTYPE). The function types are:
1=Primary to level 1 migration
2=Level 1 to level 2 migration,
or level 1 to level 1 migration
or level 2 to level 2 migration
3=Primary to level 2 migration
4=Recall from level 1 to primary
5=Recall from level 2 to primary
6=Delete a migrated data set
7=Daily backup
8=Spill backup
9=Recovery
10=Recycle backup volume
11=Data set deletion by age
12=Recycle migration volume
13=Full volume dump
14=Volume or data set restore
15=ABACKUP function
16=ARECOVER function
17=Expire primary or migrated data sets
18=Partrel function
19=Expire or roll off incremental backup version
20=(H)BDELETE an incremental backup version
21=Fast replication backup function
22=Fast replication recover function
23=Fast replication delete function

248
z/OS V1R8 DFSMS Technical Update
Sample JCL is provided in Example 7-47 and in Example 7-49.
Example 7-47 IFASMFDP JCL example
//STEP1 EXEC PGM=IFASMFDP
//INDD1 DD DSN=SYS1.MANX,DISP=SHR
//OUTDD1 DD DSN=MHLRES2.DATA,UNIT=3390,VOL=SER=WORK01,
// DISP=(NEW,CATLG,KEEP),SPACE=(CYL,(10,1))
//SYSPRINT DD SYSOUT=A
//SYSIN DD *
INDD(INDD1,OPTIONS(DUMP))
OUTDD(OUTDD1,TYPE(241))
DATE(2003120,2003126)
/*
use TYPE(241) - select this to match FSR records
or use TYPE(240:241) - select this to match DSR,VSR and all FSR records
DATE(2003120,2003126) - select a period of 7 or 14 days,
using YYYYddd Julian date format.
2003020 is Year 2003, 120th day
The output data set that is specified in the OUTDD1 DD statement in Example 7-47 contains
the summary activity report. Example 7-48 shows a sample report.
Example 7-48 SMF dump summary activity report
SUMMARY ACTIVITY REPORT
START DATE-TIME 10/22/2003-13:10:01 END DATE-TIME 10/22/2003-13:35:08
RECORD RECORDS PERCENT AVG. RECORD MIN. RECORD MAX. RECORD RECORDS
TYPE READ OF TOTAL LENGTH LENGTH LENGTH WRITTEN
240 7 .15 % 1,036.00 1,036 1,036 7
241 4 .09 % 300.00 300 300 4
TOTAL 4,591 100 % 888.24 18 32,720 13
NUMBER OF RECORDS IN ERROR 0
Example 7-49 IDCAMS REPRO example
//* INPUT FILE IS RECFM=VBS FROM IFASMFDP
//STEP1 EXEC PGM=IDCAMS
//SYSPRINT DD SYSOUT=A
//INDATA DD DISP=OLD,DSNAME=MHLRES2.DATA
//OUTDATA DD DSNAME=MHLRES2.FSR,UNIT=3390,
// DISP=(NEW,CATLG),VOL=SER=WORK01,
// SPACE=(CYL,(1,1)),RECFM=VB,LRECL=4096
//SYSIN DD *
REPRO -
INFILE(INDATA) -
OUTFILE(OUTDATA)
/*
Sample REXX execs
The FSRSTAT program is a REXX sample program that reads DFSMShsm FSR records and
generates a statistical summary report. The FRSTAT program is shipped with DFSMShsm in
SYS1.SAMPLIB member ARCTOOLS. It has to be modified so that it only selects fast
replication FSRTYPE records.

Chapter 7. DFSMShsm fast replication
249
The installation of DFSMShsm places a member called ARCTOOLS in the SYS1.SAMPLIB
data set. When you supply a valid jobcard and run the ARCTOOLS program, it creates the
following partitioned data sets:
HSM.SAMPLE.TOOL Sample REXX execs and JCL to manage your DFSMShsm
data
HSM.ABARUTIL.JCL JCL used by ABARS utilities
HSM.ABARUTIL.PROCLIB JCL PROCs used by ABARS utilities
HSM.ABARUTIL.DOCS Documentation for ABARS utilities
The HSM.SAMPLE.TOOL data set contains the FSRSTAT PGM along with other tools.
FRSTAT is written in REXX, so the following considerations apply:
Does not require any special programs or languages (such as SAS or MICS)
Can be modified easily and customized to meet your needs
Can be slow, so consider running it in batch using PGM=IKJEFT01
Requires input data to be converted to RECFM=VB format
Creating reports
You can run the FSRSTAT exec in batch using PGM=IKJEFT01. A sample JCL is shown in
Example 7-50, where:
The SYSEXEC DD statement references the location of the FSRSTAT REXX program.
The output file name is the same as the input file name with the added low-level qualifier
of FSRSTAT. For example:
– Input file name: MHLRES2.FSR
– Output file name: MHLRES2.FSR.FSRSTAT
Example 7-50 IKJEFT01 FSRSTAT report JCLl
//STEP1 EXEC PGM=IKJEFT01
//SYSEXEC DD DSN=HSM.SAMPLE.TOOL,DISP=SHR
//SYSTSPRT DD SYSOUT=*
//SYSTSIN DD *
FSRSTAT 'MHLRES2.FSR' FSRID(241)
/*
FSRSTAT report output is shown in Example 7-51.
Example 7-51 FSRSTAT report output
Data Set Analyzed: 'MHLRES2.FSR'
Number of lines read: 13

FSR records by type
021 4

FSR records by host
SC65 4 100.00% 100.00%

FSR records by Return Code
0000 4 100.00% 100.00%

FSR records by Date
0103295F 4

250
z/OS V1R8 DFSMS Technical Update
Customizing reports
You can use DFSORT to produce customized reports. In our example, we first extract the
FSR type 21, 22, and 23 records in separate data sets. In the SYSIN statements:
X’15’ is the value for FSR record type 21.
X’16’ is the value for FSR record type 22.
X’17’ is the value for FSR record type 23.
Example 7-52 shows a DFSORT job.
Example 7-52 DFSORT example jcl
//STEPNN EXEC PGM=SORT
//SYSPRINT DD SYSOUT=*
//PLDTRACE DD DUMMY
//SYSOUT DD SYSOUT=*
//SORTIN DD DISP=SHR,DSN=MHLRES2.FSR
//SORTOF1 DD LIKE=MHLRES2.FSR,
// DISP=(NEW,CATLG),UNIT=SYSDA,
// DSN=MHLRES2.FSR.TYPE21
//SORTOF2 DD LIKE=MHLRES2.FSR,
// DISP=(NEW,CATLG),UNIT=SYSDA,
// DSN=MHLRES2.FSR.TYPE22
//SORTOF3 DD LIKE=MHLRES2.FSR,
// DISP=(NEW,CATLG),UNIT=SYSDA,
// DSN=MHLRES2.FSR.TYPE23
//SYSIN DD *
OPTION COPY,VLSHRT
OUTFIL FILES=1,
INCLUDE=(43,1,BI,EQ,X'15')
OUTFIL FILES=2,
INCLUDE=(43,1,BI,EQ,X'16')
OUTFIL FILES=3,
INCLUDE=(43,1,BI,EQ,X'17')
You can use an ICETOOL job, as shown in Example 7-53, to generate a report about specific
data that you want to analyze. A good field for performance reports might be the FSRCPU
value.
Example 7-53 ICETOOL sample to extract FSR data
//MHLRES2Y JOB (999,POK),'MHLRES2',CLASS=A,MSGCLASS=K,
// NOTIFY=&SYSUID,TIME=1440,MSGLEVEL=(1,1)
/*JOBPARMBL=999,SYSAFF=*
//JOBLIB DD DSN=SYS1.SICELINK,DISP=SHR
//ICETL2 EXEC PGM=ICETOOL,REGION=4096K
//TOOLMSG DD SYSOUT=*
//DFSMSG DD SYSOUT=*
//LIST1 DD SYSOUT=*
//DFSPARM DD *
//IN01 DD DSN=MHLRES2.FSR.TYPE21,DISP=SHR
//OUT01 DD DSN=MHLRES2.FSR.TYPE21.CREATE,DISP=SHR
//* DISP=(NEW,CATLG,DELETE),UNIT=SYSDA,
//* SPACE=(CYL,(5,1),RLSE),
//* DCB=(*.IN01)
//TOOLIN DD *
COPY FROM(IN01) USING(CPY1)
DISPLAY FROM(OUT01) LIST(LIST1) BLANK -
TITLE('FSRTYPES ') -
ON(1,4,CH)

Chapter 7. DFSMShsm fast replication
251
//CPY1CNTL DD *
INCLUDE COND=(43,1,BI,EQ,X'15')
OUTFIL FNAMES=OUT01,
OUTREC=(1,4,C'FSRTYPE',43,1,FI,44,1,HEX,
45,44,
89,6,
99,6,
105,4,HEX,
179,2,HEX,
181,4,HEX)
/*
Example 7-54

shows the output that is generated from the job in the previous example.
Example 7-54 Output of an ICETOOL job
BROWSE MHLRES2.FSR.TYPE21.CREATE
Command ===>
*********************************************************** Top of Data ******************
FSRTYPE 21 00DSN$DB8B$DB MHL213MHL12500000000300000000001
FSRTYPE 21 00DSN$DB8B$DB MHL214MHL12600000000300000000000
FSRTYPE 21 00DSN$DB8B$LG MHL225MHL03700000000300000000000
FSRTYPE 21 00DSN$DB8B$LG MHL226MHL14700000000300000000000
FSRTYPE 21 00CP1 MHL0ACMHL0A000000000300000000000
FSRTYPE 21 00CP1 MHL0ADMHL0A100000000300000000000
FSRTYPE 21 00CP2 MHL1ACMHL1A000000000300000000000
FSRTYPE 21 00CP2 MHL1ADMHL1A100000000300000000000
FSRTYPE 21 00CP3 MHL2AEMHL1A000000000300000000000
FSRTYPE 21 00CP3 MHL2AFMHL1A100000000300000000000
FSRTYPE 21 00DSN$DB8B$DB MHL013MHL12500000000300000000001
FSRTYPE 21 00DSN$DB8B$DB MHL014MHL12600000000300000000001
FSRTYPE 21 00DSN$DB8B$LG MHL025MHL03700000000300000000000
FSRTYPE 21 00DSN$DB8B$LG MHL026MHL14700000000300000000000
FSRTYPE 21 00CP1 MHL0AEMHL0A000000000300000000001
7.6 Security for DFSMShsm fast replication
This section describes the RACF support for DFSMShsm commands and the RACF Facility
Class profiles that are used to protect the DFSMShsm fast replication function.
This support is available since DFSMShsm V1.5. For systems prior to z/OS V1.5, the level of
authorization was defined using the AUTH command in the DFSMShsm PARMLIB member.
DFSMShsm provides two levels of commands to be protected by RACF Facility Class
profiles:
Storage administrator commands
End-user commands
User commands can be issued by any TSO user if that user has appropriate RACF authority
to the data sets to be processed.
Storage administrator commands can only be issued by a DFSMShsm authorized user. No
RACF checking is performed to confirm your authority to access the data sets to be
processed. In many installations, the storage administrator is different from the security
administrator. The security administrator would like to control and monitor the activities
against all sensitive data.

252
z/OS V1R8 DFSMS Technical Update
A storage administrator who only needs to issue DFSMShsm configuration type commands,
such as ADDVOL, can also issue other commands, such as RECOVER datasetname
NEWNAME. This user might not be allowed by RACF to access this data set, but he can copy
or rename it to a new data set name to which he has read or alter authority. He can use
DFSMShsm storage administrator commands to access or delete data that is restricted to
him.
As with storage administrator commands, users have access to all user commands and
parameters. There is no granularity at the command level.
The RACF FACILITY class
Table 7-3 shows the base set of FACILITY class profiles for DFSMShsm command protection.
Table 7-3 Base STGADMIN profiles
Note that the STGADMIN.* profile in this table is a general profile that can be used by other
applications, so deleting it may affect other applications.
Table 7-4 shows the RACF profiles that are required to protect the DFSMShsm fast replication
commands FRBACKUP, FRRECOV, and FRDELETE.
Table 7-4 STGADMIN profiles for fast replication
Note: The resource names in the RACF FACILITY class for protection of the DFSMShsm
fast replication commands has changed with DFSMShsm V1.8 (see Table 7-4 on page 252
for the new format).
Profile Protects
STGADMIN.* All DFSMS commands including all DFSMShsm
commands
STGADMIN.ARC.* All DFSMShsm commands
STGADMIN.ARC.cmd Specific storage administrator command
STGADMIN.ARC.cmd.prm Specific storage administrator command with
specific parameter
STGADMIN.ARC.ENDUSER.* All user commands
STGADMIN.ARC.ENDUSER.h_cmd Specific user command
STGADMIN.ARC.ENDUSER.h_cmd.prm Specific user command with specific parameter
Profile Protects
STGADMIN.ARC.FB.* Any FRBACKUP command
STGADMIN.ARC.FB.cpname FRBACKUP COPYPOOL only for cpname
STGADMIN.ARC.FR.* Any FRRECOV command
STGADMIN.ARC.FR.cpname FRRECOV COPYPOOL only for cpname
STGADMIN.ARC.FD.* Any FRDELETE command
STGADMIN.ARC.FD.cpname FRDELETE COPYPOOL only for cpname
STGADMIN.ARC.LC.* Any LIST COPYPOOL and
LIST DSNAME(dsname) COPYPOOL command

Chapter 7. DFSMShsm fast replication
253
For a complete listing of DFSMShsm commands that can be protected, refer to the manual
DFSMShsm Implementation and Customization Guide, SC35-0418.
7.7 Audit and error recovery
This section discusses auditing and basic error recovery in the fast replication environment.
7.7.1 FRDELETE considerations
This section contains considerations for using the DFSMShsm Fast Replication FRDELETE
command.
The FRDELETE command was enhanced by new parameters with DFSMShsm 1.8:
DASDONLY Delete only a DASD backup copy (works like
pre-DFSMShsm 1.8).
DUMPONLY(DCLASS(...) Delete only a dump copy (of a particular dump class).
BOTH Delete DASD backup copy and dump tape copy of that
particular version (this is the default).
Figure 7-57 FRDELETE command syntax
When you use an FRDELETE command, as shown in Example 7-55, you receive some
output in the backup log, as shown in Figure 7-58 on page 254.
Example 7-55 FRDELETE DUMPONLY command
HSEND FRDELETE COPYPOOL(CP1) VERSION(1) DUMPONLY
STGADMIN.ARC.LC.cpname LIST COPYPOOL(cpname) only for cpname
Profile Protects
--- |-BOTH----------------|
>>-FRDELETE-COPYPOOL(cpname)--|-VERSIONS(vernum,...)-|-------------------------|-->
|-TOKEN(token)------------||-DUMPONLY(-A-)-|
|-ALL---------------------------||-DASDONLY-------|

A>>-------------------------------------------------------------------->
|-DUMPCLASS(dclass1, ...,dclass5)-|
|-DCLASS------|

254
z/OS V1R8 DFSMS Technical Update
Figure 7-58 Output of FRDELETE COPYPOOL DUMPONLY command
FRDELETE is used to delete unneeded fast replication backup or dump versions. New DASD
backup copy versions replace older backup copy versions through normal processing. Dump
copy versions are deleted based on dump class settings during automatic dump.
There are some instances, other than normal roll off, that cause unneeded copy versions to
be deleted:
The number of fast replication backup versions (as being reported by the LIST CP(...)
ALLVOLS(ALLVERS) command) reaches the maximum of 85 versions. If this happens,
during creation of a new generation by using the FRBACKUP command, the gen(84)
dump copy is rolled off (see Figure 7-59).
Figure 7-59 Roll-out of dump copies because of reaching the architectural limit of 85 generations
The number of versions specified is decreased by modifying the SMS copy pool definition.
This causes subsequent FRBACKUP commands to remove unneeded DASD copy
versions.
DFSMSHSM DUMP LOG, TIME 15:38:38, DATE 07/03/12
ARC0639I DUMP VTOC COPY SCRATCHED, DATA SET NAME=HSM.DUMPVTOC.T204914.VMHL0A0.D07050
ARC0262I DUMP COPY INVALIDATED FOR VOLUME MHL0A0, CREATION DATE=07/02/19
ARC0262I DUMP GENERATION INVALIDATED FOR VOLUME MHL0A0, CREATION DATE=07/02/19
ARC0639I DUMP VTOC COPY SCRATCHED, DATA SET NAME=HSM.DUMPVTOC.T204914.VMHL0A1.D07050
ARC0261I TAPE VOLUME TST019 NEEDS TO BE REINITIALIZED
ARC0263I DUMP VOLUME TST019 DELETED, VOLUME STATUS=PURGED
ARC0262I DUMP COPY INVALIDATED FOR VOLUME MHL0A1, CREATION DATE=07/02/19
ARC0262I DUMP GENERATION INVALIDATED FOR VOLUME MHL0A1, CREATION DATE=07/02/19
Note: Depending on your settings you may end up in a different number of fast replication
DASD backup generations and fast replication tape dump generations. This is because
DASD backup generations are strictly rolled off by generation processing (based on the
number of replicate backup version setting in the copy pool construct), while tape dump
generations are primarily rolled off by retention period (based on dump class setting).
Generation
Version
DASD
Tape (Dump)
+1
586
yes
not yet
0
585
yes
yes
1
584
yes
yes
2
583
yes
yes
3
582
no
yes
...
83
502
no
yes
84
501
no
yes

Chapter 7. DFSMShsm fast replication
255
When a copy pool is renamed and is no longer needed, it must be deleted by the storage
administrator. Use the FRDELETE COPYPOOL command and:
– Specify the ALL keyword when all copies of the version are to be deleted.
– Specify the VERSIONS keyword when an individual or group of individual versions is
to be deleted.
– Specify the TOKEN keyword to delete the version that is associated with the token. If
the token is not unique to an individual version, the delete fails.
– If none of the above are specified, the delete fails.
Before a version is deleted, any outstanding FlashCopy relationships are withdrawn.
Never

withdraw the relationship outside of DFSMShsm. This causes the backup version to be
invalidated, but DFSMShsm thinks that it is valid. It also could result in a data integrity
exposure.
7.7.2 Decreasing copy pool backup volumes
When decreasing the number of volumes in a copy pool backup storage group, take the
following into consideration:
You must take special care when removing a volume from a copy pool backup storage
group. Before doing so, you must ensure that the volume is not the target of a valid
DFSMShsm backup version.
You can verify this by examining the output from the DFSMShsm command LIST
COPYPOOLBACKUPSTORAGEGROUP.
If a target volume must be removed from a valid source to target pairing, you should use
the FRDELETE command to delete the DFSMShsm backup version that the volume is a
part of before you remove the volume.
7.7.3 Reuse of invalid backup versions
New DASD copy versions replace invalid DASD copy versions through the next FRBACKUP
processing.
7.7.4 FIXCDS display and repair
Five new record types have been introduced by fast replication within the DFSMShsm BCDS:
F, H, I, J, and K.
New MCH keys
There are five control record types in use in order to describe fast replication backups:
X'23' = FRD Code K (new with DFSMShsm 1.8)
X'25' = FRSV Code J (new with DFSMShsm 1.5)
X'2D' = FRTV Code I (new with DFSMShsm 1.5)
X'2E' = FRB Code F (new with DFSMShsm 1.5)
X'2F' = FRVP Code H (new with DFSMShsm 1.5)

256
z/OS V1R8 DFSMS Technical Update
K is the fast replication dump record (FRD)
The key for a type K FRD record is the name of the copy pool and the version number the
dumps were made for.
Example 7-56 FIXCDS output of a K (FRD) record
HSEND FIXCDS K CP1...........................005
MCH= 01DC2300 C04E338A 3CC35B0E 00000000 00000000
* C$ *
+0000 00010000 00000000 C4C3D9C5 C4C2F1F8 00010000 0108049F 00028000 00000000
* DCREDB18 *
+0020 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
* *
+0040 00000000 D7D3E4E2 C3D6D7E8 00010000 0108049F 00020000 00000000 00000000
* PLUSCOPY *
+0060 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
* *
+0080 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
* *
+00A0 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
* *
+00C0 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
* *
+00E0 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
* *
+0100 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
* *
+0120 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
* *
+0140 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
* *
+0160 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
* *
+0180 00000000 00000000 00000000 00000000 00000000 00000000 00000000
* *
ARC0197I TYPE K, KEY CP1...........................005, FIXCDS DISPLAY
ARC0197I (CONT.) SUCCESSFUL
F is the fast replication backup record (FRB)
The key for a type F FRB record is the name of the copy pool that the record represents.
Example 7-57 shows an example of the key that is used with a type F FRB record.
Example 7-57 FIXCDS output of a F (FRB) record
FIXCDS F DSN$DB8B$DB ODS(MHLRES2.FIXCDS.CB)
MCH= 01782E00 BA2CE223 4A4F5E4C 00000000 00000000 * S
+0000 00000004 00000004 00000000 00000000 00000000 00000000 00000000 00000000 *
+0020 00000000 00000000 00000004 00000000 D4C8D3D9 C5E2F160 E3C5E2E3 00000000 * MHLRES1 TEST
+0040 00000000 00000000 00000000 00000000 00000000 00000000 14224253 0103288F *
+0060 00010100 00000000 00000000 00000003 00000000 D4C8D3D9 C5E2F160 E3C5E2E3 * MHLRES1 TEST
+0080 00000000 00000000 00000000 00000000 00000000 00000000 00000000 11564119 *
+00A0 0103288F 00010100 00000000 00000000 00000002 00000000 00000000 00000000 *
+00C0 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 *
+00E0 22101134 0103287F 00010100 00000000 00000000 00000005 00000000 00000000 *
+0100 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 *
+0120 00000000 00000000 00000000 00000400 00000000 00000000 *
ARC0197I TYPE F, KEY DSN$DB8B$DB, FIXCDS DISPLAY SUCCESSFUL

Chapter 7. DFSMShsm fast replication
257
H is the fast replication volume pairs record (FRVP)
Table 7-5 shows the fields that make up a key for a type H FRVP backup record.
Table 7-5 Type H record key fields
Example 7-58 shows the key (in hexadecimal) that is used for a type H FRVP backup record.
In this example, we use a key that consists of:
A copy pool name of COPYPL1 (followed by periods to fill the field)
A B for a backup record request
A version number of 1
A storage group name of SRCSG1 (followed by periods to fill the field)
An extent number of 0 (zero)
Example 7-58 FIXCDS output of a H (FRVP) record
FIXCDS H X'C3D7F14B4B4B4B4B4B4B4B4B4B4B4B4B4B4B4B4B4B4B4B4B4B4B4B4B4B4BC20008E2C7F14B4B4B4B4B0000'
ODS('MHLRES2.DISPLAY.H.OUT')
MCH= 00942F00 00000000 00000000 00000000 00000000 *
+0000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 *
+0020 D4C8D3F0 C1F0D4C8 D3F0C1C5 80000000 00000000 00000000 0000D4C8 D3F0C1F1 *MHL0A0MHL0AE MHL0A1
+0040 D4C8D3F0 C1C6A000 00000000 00000000 00000000 *MHL0AF
ARC0197I TYPE H, KEY CP1...........................B..SG1......., FIXCDS DISPLAY SUCCESSFUL H

Example 7-59 shows a sample IDCAMS PRINT job to print all H records. It can be used for
printing the other records as well.
Example 7-59 Sample IDCAMS PRINT job to print all H records
//PRINT EXEC PGM=IDCAMS
//********************************************************
//* *
//* PRINT new Fast replication records *
//* X'25' = FRSV (Code J) *
//* X'2D' = FRTV (Code I) *
//* X'2E' = FRB (Code F) *
//* X'2F' = FRVP (Code H) *
//********************************************************
//SYSPRINT DD SYSOUT=*
//DD1 DD DSN=HSM.BCDS,
// DISP=SHR
//DD2 DD DSN=MHLRES3.BCDS.PRINT3,DISP=(,CATLG,DELETE),
// SPACE=(TRK,5),DCB=(LRECL=250,RECFM=VB),UNIT=3390
//SYSIN DD *
PRINT -
INFILE(DD1) OUTFILE(DD2) FROMKEY(X'2F') TOKEY(X'2F')
/*
Characters
1 to 30 The copy pool name suffixed with periods if the copy pool name is less than 30
characters in length.
31 The type of record. P if the record is a prepare record. B if the record is a backup record.
32 and 33 The version number in hexadecimal format.
34 to 41 The storage group name suffixed with periods if the storage group name is less than
eight characters in length.
42 and 43 The extent number.

258
z/OS V1R8 DFSMS Technical Update
Output from the IDCAMS job is shown in Example 7-60.
Example 7-60 Output from IDCAMS PRINT job to print all H records
IDCAMS SYSTEM SERVICES TIME: 21:02:52 10/20/03 PAGE 1
LISTING OF DATA SET -HSM.BCDS
KEY OF RECORD - 2FC4E2D55BC4C2F8C25BC4C24B4B4B4B4B4B4B4B4B4B4B4B4B4B4B4B4B4B4BC20009C4C2F8C24B4B4B4B0001
000000 2FC4E2D5 5BC4C2F8 C25BC4C2 4B4B4B4B 4B4B4B4B 4B4B4B4B 4B4B4B4B 4B4B4BC2
*.DSN$DB8B$DB...................B
000020 0009C4C2 F8C24B4B 4B4B0001 00942F00 00000000 00000000 00000000 00000000
*..DB8B..........................
000040 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
*................................
000060 D4C8D3F1 F2F5D4C8 D3F0F1F3 80000000 00000000 00000000 0000D4C8 D3F1F2F6
*MHL125MHL013..............MHL126
000080 D4C8D3F0 F1F4A000 00000000 00000000 00000000
*MHL014..............
KEY OF RECORD - 2FC4E2D55BC4C2F8C25BC4C24B4B4B4B4B4B4B4B4B4B4B4B4B4B4B4B4B4B4BC2000AC4C2F8C24B4B4B4B0001
000000 2FC4E2D5 5BC4C2F8 C25BC4C2 4B4B4B4B 4B4B4B4B 4B4B4B4B 4B4B4B4B 4B4B4BC2
*.DSN$DB8B$DB...................B
000020 000AC4C2 F8C24B4B 4B4B0001 00942F00 00000000 00000000 00000000 00000000
*..DB8B..........................
000040 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 *................................
000060 D4C8D3F1 F2F5D4C8 D3F2F1F3 80000000 00000000 00000000 0000D4C8 D3F1F2F6
*MHL125MHL213..............MHL126
000080 D4C8D3F2 F1F4A000 00000000 00000000 00000000
*MHL214..............
KEY OF RECORD - 2FC4E2D55BC4C2F8C25BD3C74B4B4B4B4B4B4B4B4B4B4B4B4B4B4B4B4B4B4BC20009C4C2F8C2D3D6C7F10001
000000 2FC4E2D5 5BC4C2F8 C25BD3C7 4B4B4B4B 4B4B4B4B 4B4B4B4B 4B4B4B4B 4B4B4BC2
*.DSN$DB8B$LG...................B
000020 0009C4C2 F8C2D3D6 C7F10001 007A2F00 00000000 00000000 00000000 00000000
*..DB8BLOG1...:..................
000040 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 *................................
000060 D4C8D3F0 F3F7D4C8 D3F0F2F5 A0000000 00000000 00000000 0000
*MHL037MHL025..............
KEY OF RECORD - 2FC4E2D55BC4C2F8C25BD3C74B4B4B4B4B4B4B4B4B4B4B4B4B4B4B4B4B4B4BC20009C4C2F8C2D3D6C7F20001
000000 2FC4E2D5 5BC4C2F8 C25BD3C7 4B4B4B4B 4B4B4B4B 4B4B4B4B 4B4B4B4B 4B4B4BC2
*.DSN$DB8B$LG...................B
000020 0009C4C2 F8C2D3D6 C7F20001 007A2F00 00000000 00000000 00000000 00000000
*..DB8BLOG2...:..................
000040 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
*................................
000060 D4C8D3F1 F4F7D4C8 D3F0F2F6 A0000000 00000000 00000000 0000
*MHL147MHL026..............
I is the fast replication target volume record (FRTV)
The key for a type I FRTV record is the volume serial number. Example 7-61 shows an
example of the key that is used with a type I FRTV record.
Example 7-61 FIXCDS output of an I (FRTV) record
FIXCDS I MHL214 ODS(MHLRES2.FIXCDS.CB)
MCH= 00662D00 BA2C090A 71FB6688 BA2C090A 71FB6688 *
+0000 D4C8D3F1 F2F60000 00000000 00000000 00000000 00000000 00000000 00000000 *MHL126
+0020 00000000 00000000 *
ARC0197I TYPE I, KEY MHL214, FIXCDS DISPLAY SUCCESSFUL
J is the fast replication source volume record (FRSV)
The key for a type J FRSV record is the volume serial number. An example of the key that is
used with a type J FRSV record is shown in Example 7-62.
Example 7-62 FIXCDS output of a J (FRSV) record
FIXCDS J MHL126 ODS(MHLRES2.FIXCDS.CB)
MCH= 07AC2500 BA2CE221 E11F1B08 B9CC2CCD 05D66242 * S O
+0000 C4E2D55B C4C2F8C2 5BC4C240 40404040 40404040 40404040 40404040 40400B03 *DSN$DB8B$DB
+0020 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 *
+0040 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 *
+0060 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 *
+0080 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 *

Chapter 7. DFSMShsm fast replication
259
+00A0 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 *
+00C0 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 *
+00E0 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 *
+0100 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 *
...
+0720 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 *
+0740 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 *
+0760 00000000 00000000 00000000 *
ARC0197I TYPE J, KEY MHL126, FIXCDS DISPLAY SUCCESSFUL
7.7.5 AUDIT COPYPOOLCONTROLS
AUDIT was enhanced in order to check the fast replication CDS record relationships. You can
use a command, as shown in Example 7-63.
Example 7-63 AUDIT COPYPOOLCONTROLS command
HSEND AUDIT COPYPOOLCONTROLS(CP1) ODS(MHLRES2.AUDITCP.LIST)
If you do not specify the name of a copy pool, DFSMShsm checks the fast replication CDS
record relationships of all copypools.
Figure 7-60 shows an example for the output of the AUDIT COPYPOOLCONTROLS. We
deleted two fast replication CDS records in order to demonstrate a simple fix process by using
your own-built FIXCDS commands.
Figure 7-60 Output of AUDIT COPYPOOLCONTROLS command
You can find a complete description of all AUDIT error types in DFSMSdfp Storage
Administration Reference, SC26-7402. In Table 7-6 we listed those error types that are
related with fast replication CDS records.
Table 7-6 AUDIT COPYPOOLCONTROLS error types
Note: AUDIT COPYPOOLCONTROLS only provides detection support, so there is no
option available to specify the FIX parameter.
-DFSMSHSM AUDIT- ENHANCED AUDIT -- LISTING - AT 18:08:13 ON 07/03/16 FOR
COMMAND ENTERED:
AUDIT COPYPOOLCONTROLS(CP1)

/* ERR 177 I RECORD WAS EXPECTED FOR VOLUME MHL0AF AND WAS NOT FOUND
/* ERR 177 I RECORD WAS EXPECTED FOR VOLUME MHL1AC AND WAS NOT FOUND
- END OF - ENHANCED AUDIT - LISTING -
Description Troubleshooting Hints
*ERR 170 SPECIFIED COPY POOL cpname
DOES NOT EXIST
*ERR 171 F RECORD INDICATES x DUMP
VERSIONS, ONLY y VERSIONS WERE FOUND
*ERR 172 H RECORD INDICATES x DASD
VERSIONS, ONLY y VERSIONS WERE FOUND

260
z/OS V1R8 DFSMS Technical Update
*ERR 173 F (FRB) RECORD INDICATES x H
(FRVP) RECORDS FOR COPY POOL cpname,
ONLY y RECORDS WERE FOUND
*ERR 174 F (FRB) RECORD FOR COPY POOL
cpname WAS NOT FOUND, ORPHANED H
(FRVP) RECORDS EXIST
Use the FIXCDS DELETE command to remove
the orphaned entries from the BCDS.
*ERR 175 H (FRVP) RECORD INDICATES IT IS
A ’PREPARE’ RECORD, THE
CORRESPONDING F (FRB) RECORD DOES
NOT
You need to determine which indicator is
correct. Use the FIXCDS PATCH command
to correct the problem.
You can use the LIST COPYPOOL
command to help you determine which
indicator is correct.
Either the ’Prepare’ flag should be on in the F
(FRB) record or the FRVP records are
orphan entries and should be deleted using
the FIXCDS DELETE command.
*ERR 176 H (FRVP) RECORD FOR VERSION x
EXISTS, BUT VERSION CANNOT BE FOUND
IN CORRESPONDING F (FRB) RECORD FOR
COPY POOL cpname
An FRVP record may be an orphan and should
be deleted using the FIXCDS DELETE
command.
ERR 177 I (FRTV) RECORD WAS EXPECTED
FOR VOLUME volser AND WAS NOT FOUND
A F (FRB) record indicates this is a valid version.
The volume represented by this FRTV record
needs to be made available and an FRTV record
should be created to represent it or the fast
replication backup of this copy pool will fail
The FIXCDS CREATE command can be used to
recreate the FRTV record.
*ERR 178 I (FRTV) RECORD INDICATED BY H
(FRVP) RECORD CANNOT BE FOUND.
NEEDS TO EXIST FOR COPY POOL cpname
VERSION vsn
*ERR 179 H (FRVP) RECORD HAS WRONG
TARGET VOLUME tvolser, OR I (FRTV)
RECORD HAS WRONG SOURCE VOLUME
svolser
Determine which one is incorrect and correct the
record using the FIXCDS PATCH command.
*ERR 180 J (FRSV) RECORD NOT FOUND
FOR VOLUME volser, WHICH BELONGS TO
COPY POOL cpname, VERSION vsn
*ERR 181 F (FRB) RECORD FOR cpname NOT
FOUND, POSSIBLY ORPHANED J (FRSV)
RECORD FOR VOLUME volser
Remove cpname from the FRSV cpname
array.
If there are no cpnames left in the array,
delete the FRSV record for VOLUME volser.
*ERR 182 NO P (MCP) RECORD FOUND FOR
VOLUME volser, IS THIS AN ERROR?
*ERR 183 VOLUME volser FROM THE J (FRSV)
RECORD L0 FIELD COULD NOT BE FOUND IN
A CORRESPONDING H (FRVP) RECORD
H (FRVP) record was not found or volser was not
found in the corresponding H (FRVP) records.
*ERR 184 ORPHANED J (FRSV) RECORD
DETECTED
Delete the record using the FIXCDS DELETE
command.
Description Troubleshooting Hints

Chapter 7. DFSMShsm fast replication
261
In order to correct the error type that was reported in Figure 7-60 on page 259, we find the
following description in Table 7-6 on page 259:
“ERR 177 I (FRTV) RECORD WAS EXPECTED FOR VOLUME volser AND WAS NOT FOUND
*ERR 185 COPY POOL cpname COULD NOT
BE FOUND IN THE J (FRSV) RECORD COPY
POOL ARRAY FOR VOLUME volser
You need to use the FIXCDS PATCH command
to patch the FRSV record with the cpname, or if
the FRSV record contains no other copy pool
names, delete it with the FIXCDS DELETE
command.
*ERR 186 ORPHANED K (FRD) RECORD
FOUND, COPY POOL cpname NO LONGER
EXISTS
F (FRB) record could not be found for copy pool
cpname.
*ERR 187 K (FRD) RECORD EXISTS, F (FRB)
RECORD INDICATES NO DUMP COPY
The FRB record indicates that there is no dump
copy for this copy pool version but an FRD record
exists for it.
Either the FRD record should be deleted using
the FIXCDS DELETE command or the
FRD_EXISTS flag in the FRB_BVI should be
patched on.
*ERR 188 K (FRD) RECORD NOT FOUND, F
(FRB) RECORD FOR COPY POOL cpname
INDICATES THAT IT DOES EXIST
The FRB record indicates a dump for this copy
pool exists, but the record representing the dump
cannot be found.
*ERR 189 H (FRVP) RECORD NOT FOUND
COPY POOL cpname
During the AUDIT VOLUMECONTROLS
(RECOVERABLE), an error was detected.
You should consider running AUDIT
COPYPOOLCONTROLS (cpname) to learn
more regarding this discrepancy.
*ERR 190 VOLUME volser COULD NOT BE
FOUND IN ANY ASSOCIATED H (FRVP)
RECORD FOR COPY POOL cpname
You should consider running AUDIT
COPYPOOLCONTROLS (cpname) to learn
more regarding this discrepancy.
*ERR 191 P RECORD FOUND FOR VOLUME
volser, IS NOT INDICATED AS A COPY POOL
VOLUME
*ERR 192 F RECORD FOUND FOR
COPYPOOL cpname, WHICH VOLUME volser
BELONGS TO
Audit the specified copy pool to determine the
recoverability of this volume.
*ERR 193 J RECORD NOT FOUND FOR
VOLUME volser
The MCP or DGN record for volume volser
indicates that an FRSV record should exist
for this volume.
Expect this error to be issued for each
dumped version of this volume.
The MCPF_COPYPOOL bit should be
patched off if it is determined that this is no
longer a copypool volume, as is indicated by
the missing FRSV record.
FIXCDS P volser PATCH(X’9’ BITS(.......0)).
Description Troubleshooting Hints

262
z/OS V1R8 DFSMS Technical Update
A FRB record indicates that this is a valid version. The volume represented by this FRTV
record needs to be made available and an FRTV record should be created to represent it or
the fast replication backup of this copy pool fails.
The troubleshooting hint says:
The FIXCDS CREATE command can be used to recreate the FRTV record.
The structure of a FRB record (see Example 7-61 on page 258) is rather simple. We just need
to learn which are the source volumes of the volumes mentioned in the messages of error
type 177. This can be done by a command, as shown in Example 7-64.
Example 7-64 LIST COPYPOOL command
HSEND LI CP(CP1)
The output shown in Figure 7-61 gives a quick answer and allows to create the commands as
shown in Example 7-65.
Figure 7-61 Output of LIST CP(CP1)
Example 7-65 FIXCDS commands to create FRTV records
HSEND FIXCDS I MHL1AC CREATE(0 MHL0A1)
HSEND FIXCDS I MHL0AF CREATE(0 MHL0A0)
Figure 7-62 Messages in response to FIXCDS I MHL1AC CREATE command
-- DFSMShsm CONTROL DATASET --COPY POOL--LISTING --------- AT 15:32:31 ON 07/03

COPYPOOL=CP1

VERSION VTOCENQ DATE TIME FASTREPLICATIONSTATE DUMPSTATE
006 Y 2007/03/12 16:01:45 RECOVERABLE ALLCOMPLETE
TOKEN(C)=C'EX6'
TOKEN(H)=X'C5E7F6'
TOTAL NUM OF VOLUMES=00002,INCREMENTAL=N

SGNAME SOURCE - TARGET SOURCE - TARGET SOURCE - TARGET SOURCE - TARGET
SG1 MHL0A0 - MHL0AF MHL0A1 - MHL1AC
DATA BEFORE PATCH
+0000 00000000 00000000
* *
DATA AFTER PATCH
+0000 D4C8D3F0 C1F10000
*MHL0A1 *

© Copyright IBM Corp. 2008. All rights reserved.
263
Chapter 8.
DFSMSrmm enhancements
The DFSMSrmm enhancements in DFSMS V1.8 provide improvements in the areas of
enterprise level interface, UTC implementation, tape data set authorization, vital record
specification policy management simplification, and usability items.
The following topics are covered:
Support true e-mail address for the RMM NOTIFY function
Setting up DFSMSrmm common time support
DFSMSrmm VRS policy management simplification
– Separation of Data Set Name Mask from the Policy
– Release options applied if VRS matched
– Special ABEND and OPEN via DSNAME match
– Find unused VRSs
– Incomplete VRS chains - dummy VRS *broken*
– Toleration and removal of old functions
– Conversion to DFSMSrmm from other tape management systems
DFSMSrmm usability items
– Updates to RMM TSO SEARCHVOLUME subcommand
– ISPF lists show retention information
– SELECT primary command in RMM dialog search results
– ISPF lists show retention information
– Rexx variable constraint relief
Enabling ISPF Data Set List (DSLIST) support
Prepare for future releases
– Set a DFSMSrmm control data set ID
– Re-allocate your DFSMSrmm control data set
– Update LRECL for REPORT, BACKUP, and JRNLBKUP DD
– Migrate from VRSEL(OLD) to VRSEL(NEW)
8

264
z/OS V1R8 DFSMS Technical Update
8.1 Support true e-mail address for the RMM NOTIFY function
The established DFSMSrmm notification feature is extended to optionally send mails to
Internet addresses. E-mail notification is provoked in the same way as the known
notifications. There is no difference in the current processing, only the result (e-mail instead
of the established messages) is different.
Prerequisites for e-mail notification:
SMTP server available in the customer environment and known to DFSMSrmm
Owner attribute‚ e-mail address is set
If you specified an e-mail address, you must have an SMTP server configured and started.
Information about how to set up an SMTP server can be found in the manual z/OS
Communications Server: IP Configuration Guide, SC31-8775.
By default, DFSMSrmm tries to use an SMTP server called SMTP on the current JES node.
To tell DFSMSrmm to use any other SMTP server, use the restricted owner SMTP.
You use the NODE operand to identify the node that runs the SMTP server, and the USERID
operand to identify the SMTP server.
Figure 8-1 shows the difference between the use of a notification message and a true e-mail
address.
Figure 8-1 Difference between notification message and true e-mail address
The trigger for using e-mail notification is the owner‘s e-mail address. As long as there is no
e-mail address defined to DFSMSrmm, DFSMSrmm will send the notification to node/user ID.
If an e-mail address is defined, DFSMSrmm sends the notification to this address, whether a
node/user ID is defined or not.
Restriction: The owner SMTP is now a reserved owner name value that you can use to
configure the node name and SMTP server address space or machine name.




Electronic mail:
Email . :
Userid . : SCHLUM Node . . . . : IBMDE
notification
message
email




Electronic mail:
Email . : SCHLUM@DE.IBM.COM
Userid . : SCHLUM Node . . . . : IBMDE




Electronic mail:
Email . :
Userid . : SCHLUM Node . . . . : IBMDE
notification
message
email




Electronic mail:
Email . : SCHLUM@DE.IBM.COM
Userid . : SCHLUM Node . . . . : IBMDE

Chapter 8. DFSMSrmm enhancements
265
8.2 Support true e-mail address implementation
To set up e-mail notification, five topics need to be addressed.
8.2.1 Basic setup
Basic setup items that are common for normal notification and e-mail notification are:
PARMLIB option NOTIFY(YES)
Specifies whether DFSMSrmm should automatically notify volume owners when the
volumes they own become eligible for release or when software product volumes are
added.
Volume RELEASEACTION(NOTIFY)
Must be specified as part of the GETVOLUME, ADDVOLUME, or CHANGEVOLUME
subcommands to get notifications sent, when a volume is released.
DFSMSrmm under JES2 or JES3
You must be running DFSMSrmm under the JES2 or JES3 subsystem and not the master
subsystem.
8.2.2 Defining an owner’s e-mail address
The e-mail address of an owner can be defined using the DFSMSrmm dialog, as shown in
Figure 8-2, or using the appropriate TSO subcommands, like ADDOWNER or
CHANGEOWNER.
Figure 8-2 Specifying a true e-mail address using the add owner dialog panel
Panel Help
______________________________________________________________________________
DFSMSrmm Add Owner Details - SCHLUM
Command ===>


Owner id . . SCHLUM
Surname . . Schlumberger
Forenames Norbert
Department INTERNATIONALE PROJEKTE (6304)

Address:
Line 1 . . IBM DEUTSCHLAND INFORMATIONSSYSTEME GMBH
Line 2 . . AM KELTENWALD 1
Line 3 . . 71139 EHNINGEN (GERMANY)

Telephone:
Internal 919-3579 External . . . +49-(0)7056-964522

Electronic mail:
Email . . SCHLUM§DE.IBM.COM
Userid . . SCHLUM Node . . . . . IBMDE

Press ENTER to ADD owner details, or END command to CANCEL.

266
z/OS V1R8 DFSMS Technical Update
Use Example 8-1 to add a new owner using the RMM TSO ADDOWNER or CHANGE
OWNER subcommand.
Example 8-1 Specifying a true e-mail address using the RMM TSO subcommands
RMM ADDOWNER SCHLUM +
ADDR1('IBM DEUTSCHLAND INFORMATIONSSYSTEME GMBH') +
ADDR2('AM KELTENWALD 1') +
ADDR3('71139 EHNINGEN (GERMANY)') +
DEPARTMENT('INTERNATIONALE PROJEKTE (6304)') +
EMAIL(’SCHLUM§DE.IBM.COM’) +
EXTEL('07034-15-3579') +
FNAME('NORBERT') +
INTEL('919-3579') +
NODE(IBMDE) +
SNAME('SCHLUMBERGER') +
USER(SCHLUM)

or

RMM CHANGEOWNER SCHLUM +
EMAIL(’SCHLUM§DE.IBM.COM’)
If you specified an e-mail address, you must have an SMTP server configured and started.
Information about how to set up an SMTP server can be found in the manual z/OS
Communications Server: IP Configuration Guide.
By default, DFSMSrmm tries to use an SMTP server called SMTP on the current JES node.
8.2.3 SMTP server
To tell DFSMSrmm to use any other SMTP server, use the restricted owner named SMTP.
The Owner SMTP is now a reserved owner name value that you can use to configure the
node name and SMTP server address space or machine name.
You use the NODE operand to identify the node that runs the SMTP server, and the USERID
operand to identify the SMTP server. The default used by DFSMSrmm is:
node Current JES node
name SMTP
Use the restricted owner SMTP to change the default values:
node The node that runs the SMTP server
userid The SMTP server task name
Use the TSO NETSTAT command, as shown in Example 8-2, to get the server task name of
your SMTP.
Example 8-2 Sample NETSTAT command
TSO NETSTAT
Note: If you add the owner SMPT you must specify both values.

Chapter 8. DFSMSrmm enhancements
267
Figure 8-3 shows you the result of the NETSTAT command.
Figure 8-3 NETSTAT output
To get the node name use the display JES2 Network Environment command, as shown in
Example 8-3.
Example 8-3 Display the JES2 Network Environment
/$D NJEDEF
or
/$D NJEDEF
EZZ2350I MVS TCP/IP NETSTAT CS V1R7 TCPIP Name: TCPIP 13:32:28
EZZ2585I User Id Conn Local Socket Foreign Socket State
EZZ2586I ------- ---- ------------ -------------- -----
EZZ2587I BPXOINIT 00000044 0.0.0.0..10007 0.0.0.0..0 Listen
EZZ2587I CNMPEGZB 00013534 0.0.0.0..4080 0.0.0.0..0 Listen
EZZ2587I CNMPEGZB 00013536 0.0.0.0..4020 0.0.0.0..0 Listen
EZZ2587I CNMPEGZB 00013535 0.0.0.0..4092 0.0.0.0..0 Listen
EZZ2587I DB2BDIST 000048EF 0.0.0.0..5022 0.0.0.0..0 Listen
EZZ2587I DB2BDIST 000048F2 0.0.0.0..5021 0.0.0.0..0 Listen
EZZ2587I DFSKERN 00000031 0.0.0.0..139 0.0.0.0..0 Listen
EZZ2587I INETD1 00000057 0.0.0.0..923 0.0.0.0..0 Listen
EZZ2587I INETD1 00000054 0.0.0.0..512 0.0.0.0..0 Listen
EZZ2587I INETD1 00000056 0.0.0.0..514 0.0.0.0..0 Listen
EZZ2587I INETD1 00000055 0.0.0.0..513 0.0.0.0..0 Listen
EZZ2587I NPM 00000019 0.0.0.0..1965 0.0.0.0..0 Listen
EZZ2587I OMPROUTE 00000033 127.0.0.1..1026 127.0.0.1..1027 Establ
EZZ2587I TAPEMAN 00000046 9.149.157.65..35043 0.0.0.0..0 Listen
EZZ2587I TCPFTPH1 00000039 0.0.0.0..21 0.0.0.0..0 Listen
EZZ2587I TCPIP 00000016 127.0.0.1..1024 127.0.0.1..1025 Establ
EZZ2587I TCPIP 0000001F 0.0.0.0..423 0.0.0.0..0 Listen
EZZ2587I TCPIP 00000015 127.0.0.1..1025 127.0.0.1..1024 Establ
EZZ2587I TCPIP 000186C1 9.149.157.65..23 9.157.16.221..2005 Establ
EZZ2587I TCPIP 00000012 127.0.0.1..1024 0.0.0.0..0 Listen
EZZ2587I TCPIP 00018178 9.149.157.65..23 9.143.10.29..1695 Establ
EZZ2587I TCPIP 0000001E 0.0.0.0..623 0.0.0.0..0 Listen
EZZ2587I TCPIP 00000020 0.0.0.0..23 0.0.0.0..0 Listen
EZZ2587I TCPIP 0000002E 127.0.0.1..1027 127.0.0.1..1026 Establ
EZZ2587I TCPSMTP 0000003E 0.0.0.0..25 0.0.0.0..0 Listen
EZZ2587I WWW 000186F3 9.149.157.65..80 9.157.16.217..2796 FinWai
EZZ2587I WWW 00017ED9 0.0.0.0..80 0.0.0.0..0 Listen
EZZ2587I DFSKERN 00000030 0.0.0.0..137 *..* UDP
EZZ2587I DFSKERN 00000032 0.0.0.0..138 *..* UDP
EZZ2587I OMPROUTE 000186F5 0.0.0.0..60380 *..* UDP
EZZ2587I TCPSMTP 0000003F 0.0.0.0..1039 *..* UDP
EZZ2587I VTAM 0000000D 9.149.157.65..12003 *..* UDP
EZZ2587I VTAM 0000000B 9.149.157.65..12001 *..* UDP
EZZ2587I VTAM 0000000E 9.149.157.65..12004 *..* UDP
EZZ2587I VTAM 0000000C 9.149.157.65..12002 *..* UDP
Note: Normally, port TCP 25 is reserved for SMTP. Verify that the name of the member
containing the SMTP cataloged procedure has been added to the PORT statement in
hlq.PROFILE.TCPIP.

268
z/OS V1R8 DFSMS Technical Update
Figure 8-4 shows you the result of the JES2 Network Environment display command.
Figure 8-4 Display JES2 Network Environment result
8.2.4 E-mail message configuration
DFSMSrmm supports two notification tasks:
Volume release notification - Notify the owner when the owner’s volumes are eligible for
release.
Product notification - Notify the designated product owner when a program product
volume is added.
For each task EDGMTAB contains two different sets of messages for normal and e-mail
notification. This enables you to design two completely independent notification messages,
one for the known notification way and one for e-mails. Figure 8-5 shows you a sample e-mail
notification if a volume is changed to status pending release.
Figure 8-5 Sample volume release notification e-mail
The default messages (2450-2463) delivered in EDGMTAB create an e-mail as shown above
if a volume is pending release. The established messages (2405–2409) generate plain text,
as shown in Figure 8-6.
Figure 8-6 Sample volume release notification message
RESPONSE=SC70
$HASP831 NJEDEF
$HASP831 NJEDEF OWNNAME=WTSCPLX2,OWNNODE=1,DELAY=300,
$HASP831 HDRBUF=(LIMIT=100,WARN=80,FREE=100),
$HASP831 JRNUM=1,JTNUM=1,SRNUM=7,STNUM=7,
$HASP831 LINENUM=40,MAILMSG=YES,MAXHOP=0,
$HASP831 NODENUM=999,PATH=1,RESTMAX=0,
$HASP831 RESTNODE=100,RESTTOL=0,TIMETOL=0
Subject: DFSMSrmm volume expiration
Volume T10003 assigned to owner D008210 on 2005/355 at 02:50:04
is now pending release.
If you wish the volume to be retained, please take immediate action.
You can use the dialog functions or the RMM CHANGEVOLUME TSO command.

Chapter 8. DFSMSrmm enhancements
269
To prevent that, DFSMSrmm changes the volume status to the status SCRATCH and retains
the volume for an additional 90 days using the command shown in Example 8-4.
Example 8-4 Use the RMM TSO CHANGEVOLUME subcommand to change the expiration date
RMM CHANGEVOLUME T10003 RETPD(90)
or
RMM CV T10003 RETPD(90)
Figure 8-7 Messages 2456 - 2461 in EDGMTAB
These messages can contain HTML code to format the text, as shown in Example 8-5.
Example 8-5 Sample HTML code
('<font size="+2" color=blue><b>take immediate action '</font></b>.<br>')
....
EDGMSGB 2456,TYPE=I,MOD=YES,MSGID=NO,
('Volume <FONT COLOR=RED>¢¢¢¢¢¢</FONT> ',
'assigned to owner <b>¢¢¢¢¢¢¢¢</b> ')
SPACE 2
EDGMSGB 2457,TYPE=I,MOD=YES,MSGID=NO,
'on ¢¢¢¢¢¢¢¢¢¢ at ¢¢:¢¢:¢¢'
SPACE 2
EDGMSGB 2458,TYPE=I,MOD=NO,MSGID=NO,
('is now pending release.<p>',
'If you wish the volume to be retained, please ')
SPACE 2
EDGMSGB 2459,TYPE=I,MOD=NO,MSGID=NO,
('<font size="+2" color=blue><b>take immediate action ‘,
'</font></b>.<br>')
SPACE 2
EDGMSGB 2460,TYPE=I,MOD=NO,MSGID=NO,
('You can use the <b>dialog functions</b> ',
'or the <b>RMM CHANGEVOLUME</b>')
SPACE 2
EDGMSGB 2461,TYPE=I,MOD=NO,MSGID=NO,
' TSO command.'
....

270
z/OS V1R8 DFSMS Technical Update
Figure 8-8 shows you a sample e-mail notification if a new software product volume is added.
Figure 8-8 Sample add software product volume notification e-mail
The default messages (2720–2739) delivered in EDGMTAB create an e-mail, as shown
above, if a program product volume is added. The established messages (2700–2713)
generate plain text, as shown in Figure 8-9.
Figure 8-9 Sample add software product volume notification message
8.3 Setting up DFSMSrmm common time support
Before DFSMSrmm common time support (UTC), also known as GMT, is enabled, all dates
and times are stored in the DFSMSrmm control data set in local time. When the control data
set is shared, and the sharing systems are set to run in different time zones, the local dates
and times in the control data set may be from any of your systems. When you display
information or extract records, you need to be aware of how the records were created, on
which system, and where they may have been updated in order to interpret the dates and
times shown. The same consideration also applies for records created or updated prior to
enabling common time support because DFSMSrmm assumes that they are times local to the
system running the DFSMSrmm subsystem and converts the values based on that
assumption.
When you enable common time support, DFSMSrmm maintains the records in the control
data set in common time. Most date and time fields are paired together to enable an accurate
conversion to and from common time and between different time zones. In some cases,
DFSMSrmm has date fields in control data set records, and there is no associated time field.
Subject: Volume T10000 has been added for software product 4730
A volume has been added on 2005/355 at 02:51:06 to a DFSMSrmm
software product which you own:
Product Number = 4730 Level = V01R01M00
Name = PROD
Description = TEST

Volume Rack Feature Code Description
------ ------ ------------ ------------------------------
T10000 789 2323 TESTVOLUME

Chapter 8. DFSMSrmm enhancements
271
For these date fields, DFSMSrmm uses an internal algorithm that approximates conversion
between time zones based on the time zone offsets involved.
8.3.1 DFSMSrmm - using the date and time
Dates and times in DFSMSrmm are used mostly for reference, but also to calculate retention
periods. Examples are:
Journal record prefix
Contains date and time that the journal records were written, compared for forward
recovery.
Record creation date and time
In any record stored in the DFSMSrmm control data set.
Expiration date
In volume and data set records. Shows the date that the entity is expired, but not the time.
Tokens
In control data set records. Used for various purposes but usually as a unique value to
associate multi-volumes multi-files, or logical volumes and stacked volumes together.
Reference dates
In control data set records.
All current uses of date and time in DFSMSrmm use the TIME DEC option of the TIME
macro. This retrieves the local date and time. DFSMSrmm assumes that all systems that
share a control data set (CDS), and therefore the JOURNAL as well, run on a synchronized
time source.
Although in theory there is supposed to be a common date and time function available in
DFSMSrmm and used for all date conversions, in practice it is used only when a date
conversion is required or to get the current date in DFSMSrmm internal format. It is not used
where the time is wanted or when the date is wanted not in internal format.
DFSMSrmm uses an internal date format that is derived via TIME DEC from the TOD — a
4-byte, 7-character packed decimal number YYYYDDDs. Using this value DFSMSrmm has
no known limit to the dates that can be handled. The internal time is a seven-character
packed decimal number also, HHMMSSTs. DFSMSrmm does not need to upgrade to using a
128-bit timer value, nor have any problems with the TOD clock rolling over in 2043.
8.3.2 Date and time in a DFSMSrmm client server environment
All dates and times used on the client system are local time. All dates and times used on the
server are in local time. When data is retrieved from the DFSMSrmm control data set (CDS)
Warning: Using the SET system command with either the DATE or the TIME keyword, or
both, or replying to message IEA888A to run the system on future or past dates can affect
the way that DFSMSrmm calculates local times. In order to get the correct results from
DFSMSrmm processing when you need to test with future or past dates, you should alter
the TOD clock and keep the time zone offset as before.
Note: Dates and times are always displayed to the user in the same format that they were
recorded, in local time.

272
z/OS V1R8 DFSMS Technical Update
and displayed by subcommand processing, the dates and times used are exactly as stored in
the DFSMSrmm CDS. No conversion from server time zone to client time zone is performed.
Figure 8-10 shows an example of when there is a need to use the DFSMSrmm common time
support.
Figure 8-10 DFSMSrmm client server implementation
8.3.3 Enable common time support
Before you can enable the DFSMSrmm common time support you must check that the (time
of Day (TOD) clocks of all systems in the RMMPLEX are set to Universal Time, Coordinated
(UTC).
To enable common time support:
1.Ensure that all systems in the RMMplex have toleration maintenance installed or are at
z/OS V1R8 or later, and all applications dependent on the correct date and time
information from DFSMSrmm are updated to support the new time zone support if
required. DFSMSrmm subcommand output remains in local time, so most applications do
not need to change unless they are to exploit the availability of the time zone offset.
2.Ensure that the system time of day clock is set to GMT on all systems in the RMMplex. It
is common practice for the system to use local time based either on the TIMEZONE value
in the CLOCKxx member of PARMLIB or from an external time source.
3.Run the EDGUTIL utility with UPDATE with the UTC(YES) operand on the CONTROL
statement of the SYSIN file to enable common time support.
UTC(YES) enables DFSMSrmm common time support. Prior to enabling this support all
systems in the RMMplex should have toleration maintenance installed or be z/OS V1.8 or
later, and applications dependent on the correct date and time information from DFSMSrmm
should be updated to support the new time zone support if required.
Client
2
Client
1
IP-Net
Server
Client
2
Client
1
IP-Net
Server
Client
2
Client
1
IP-Net
Server
Important: You cannot disable common time support once it is enabled.

Chapter 8. DFSMSrmm enhancements
273
Figure 8-11 shows you the result of a RMM TSO LISTCONTOL if you do not have
DFSMSrmm common time support enabled.
Figure 8-11 Common time support disabled
Use Example 8-6 to enable DFSMSrmm common time support.
Example 8-6 Enable UTC support
//UTIL EXEC PGM=EDGUTIL,PARM=’UPDATE’
//SYSPRINT DD SYSOUT=*
//MASTER DD DISP=SHR,DSN=RMM.CONTROL.DSET
//SYSIN DD *
CONTROL UTC(YES) CDSID(SC70)
/*
Where:
CDSID Specifies one-to-eight alphanumeric characters that identify the
control data set by name. There is no default. A CDSID is required in
z/OS 1.9 or later.
UTC Enables DFSMSrmm common time support.
Control record:
Type = MASTER Create date = 2002/064 Create time = 15:46:09
Update date = 2007/082 Update time = 17:49:39
Journal: Utilization = 17% (75% threshold) STATUS: = ENABLED
CDS: Utilization = 81%
Exit status: Options:
EDGUX100 = ENABLED Stacked Volumes = NONE
EDGUX200 = NONE Extended Bin = DISABLED
Common Time = DISABLED
CDSID ENQ name = ENABLED
Last backup: Last expiration processing:
Date = 2007/079 Time = 13:25:42 Date = 2007/079 Time = 13:25:39
Last journal backup: Last store update:
Date = 2007/079 Time = 13:25:42 Date = 2007/072 Time = 18:30:39
Last report extract: Last VRS processing:
Date = 2007/072 Time = 18:30:39 Date = 2007/079 Time = 13:25:39
Last scratch procedure: Last Catalog synchronize:
Date = Time = Date = Time =
Rack numbers = 6258 Empty racks = 4212
LOCAL store bins = 0 Empty LOCAL bins = 0
DISTANT store bins = 0 Empty DISTANT bins = 0
REMOTE store bins = 0 Empty REMOTE bins = 0
Control functions in progress:
Backup = N Restore = N
Verify = N Expiration = N
Report Extract = N Disaster Store = N
VRS = N Synchronize = N
Client/Server:
host name =
IP address =

274
z/OS V1R8 DFSMS Technical Update
Once you enable common time support DFSMSrmm starts to record CDS record dates and
times in common time and converts existing values, as required, from local times to common
time. Figure 8-12 shows you an example of a display data set.
Each user can set his own common time zone values to display the date and time information
correctly. Example 8-8 on page 284 shows you the DFSMSrmm dialog user option selection
where the time zone can be specified.
Figure 8-12 Set your own common time zone value
All dates and times displayed or entered in the DFSMSrmm dialog are values local to this
time zone. To change the time zone you must specify an offset value and a text string to
identify that zone to you. The offset value is the time that your selected time zone is ahead of
or behind universal time (UTC/GMT). Changes you make only affect future dialog actions and
displays.
The report extract data set contains date and time values in the local time of the running
system. The extract header record includes a field that lists the time zone offset, as shown in
Figure 8-13.
Figure 8-13 Report extract data set time zone offset
Panel Help
______________________________________________________________________________
EDGP§OP1 DFSMSrmm Dialog User Options
Command ===>

Date format . . . . . JULIAN ( American, European, Iso or Julian )
Time zone . . . . . . MST -04:00:00 ( zone offsetHH:MM:SS )

Confirm deletes . . . YES ( Yes or No )

Processing option . . F F - Foreground, B - Background

Eject option . . . . C C - Convenience, B - Bulk

Variable reuse . . . Y Y - Yes, N - No

Job statement information:-

===> //RMMJOB JOB ,RMM,NOTIFY=&SYSUID,
===> // MSGCLASS=H,CLASS=A,MSGLEVEL=(1,1),REGION=6M
===> //*
===> //*

Enter END command to save changes, or CANCEL to end without saving.
H 2007/072 183039SC64 JN-04:00:00
Recommendation: We recommend that you have the system TOD clock set to GMT and
enable DFSMSrmm to use UTC. You do this by using the EDGUTIL utility with the
UPDATE parameter. Once you do this, any newly recorded dates and time will be stored in
common time and existing records are converted to be in all common time as they are
updated. You will continue to see dates and times presented in local time because
DFSMSrmm handles the conversion from common time to your local time.

Chapter 8. DFSMSrmm enhancements
275
Set the TIMEZONE in the CLOCKxx PARMLIB member
TIMEZONE d.hh.mm.ss specifies the difference between the local time and the Coordinated
Universal Time (UTC). If ETRMODE YES and ETRZONE YES are specified (and an
operational Sysplex Timer® is available), the system ignores the TIMEZONE parameter.
d Specifies the direction from UTC.
Value Range: E for east of UTC or W for west of UTC.
Default: W.
hh.mm.ss Specifies the number of hours (hh), minutes (mm), and seconds (ss)
that the local time differs from the UTC.
Value Range: The value for hh must be between 00 and 15. The value
for mm and ss must be between 00 and 59. mm.ss values are
optional.
In addition, the combined hh.mm.ss value must be within the range
00:00:00–15:00:00. This means that a value like 15.59.59 is not valid
because it is outside the range, even though the hh portion is between
00 and 15 and the mm and ss portions are between 00 and 59. If the
mm portion or ss portion or both are omitted, a default value of 00 is
applied to the omitted portion, and appears in message IEA598I at IPL
time. For example, if CLOCKxx contains TIMEZONE W.15, then at IPL
time, message IEA598I indicates:
IEA598I TIME ZONE = W.15.00.00
Default: 00.00.00
See Example 8-7 for how you can use the DISPLAY T command to display the local time of
day and date and the UTC of day and date.
Example 8-7 Displaying the local and coordinated universal time and date
D T
The local time of day and date and the coordinated universal time of day and date are to be
displayed (message IEE136I), as shown in Figure 8-14.
Figure 8-14 Result of the displaying the local and coordinated universal time and date
8.3.4 Potential problems using local time
Some potential problems are:
Systems sharing the CDS but in different time zones create records using their own local
date and time values.
Journal records might appear to not be in sequence. However, the VSI count and record
numbers show the correct sequence. Tokens vary but will at least always increment.
Users see a date and time and believe it to be in their local time zone, but it could reflect
that used on another system.
Data sets or volumes could be released early.
RESPONSE=SC70 IEE136I LOCAL: TIME=12.56.19 DATE=2007.068 UTC:
RESPONSE=TIME=17.56.19 DATE=2007.068

276
z/OS V1R8 DFSMS Technical Update
8.4 DFSMSrmm VRS policy management simplification
In this new release the VRS processing changed and the RMM TSO subcommand and the
reporting are enhanced as follows:
1.Separation of Data Set Name Mask from the Policy.
2.Release options applied if VRS matched.
3.Special ABEND and OPEN via DSNAME match.
4.Find unused VRSs.
5.Incomplete VRS chains - dummy VRS *broken*.
6.Toleration and removal of old functions.
7.Conversion to DFSMSrmm from other tape management systems.
8.4.1 Separation of Data Set Name Mask from the Policy
Figure 8-15 shows the structure of a DSNAME VRS and how the concept of separating the
data set name mask from the policy itself is put into effect.
The data set name VRS now allows a COUNT of zero so that this first VRS in a chain has no
retention specification. The NEXTVRS in the chain and subsequent VRSs now specify the
entire policy.
Figure 8-15 DSNAME VRS concept
Setting up DFSMSrmm retention and movement policies can require large numbers of VRSs
to be created because each data set name mask VRS also contains the initial retention and
movement information.
Attention: The DFSMSrmm PARMLIB option VRSEL(OLD) will be removed in a future
release. We recommend migrating from VRSEL(OLD) to VRSEL(NEW) before moving to
z/OS V1.8. Each time you run VRSEL processing and VRSEL(OLD) is in use, the new
message EDG2317E and a minimum return code of 4 are issued.
ANDVRS \NEXTVRS
mask
DSNAME
JOBNAME
GDG\NOGDG
policy
COUNT
STORENUMBER
DAYS \CYCLES \...
DELAY
WHILECATALOG
UNTILEXPIRED
LOCATION
DELETEDATE
DESCRIPTION
OWNER
RELEASE
PRIORITY
DSNAME VRS
DSNAME (mask) VRS NAME (policy) VRS
DELETEDATE
DESCRIPTION
OWNER
RELEASE
PRIORITY
policy
COUNT(0)
mask
DSNAME
JOBNAME
GDG\NOGDG
DELETEDATE
DESCRIPTION
OWNER
policy
COUNT
STORENUMBER
DAYS \CYCLES \...
WHILECATALOG
UNTILEXPIRED
LOCATION
ANDVRS \NEXTVRS
mask
NAME
NEXTVRS

Chapter 8. DFSMSrmm enhancements
277
Separating the data set name mask from the policy itself, as shown in Figure 8-16, enables
clear and well-defined service levels to be set up for tape management.
These policy/service-level VRSs can then be easily modified as required without changing
the filters that select them.
Figure 8-16 Separation of DSNAME mask from the policy
If you run DFSMSrmm with VRSEL(OLD), for each VRS where COUNT is set to 0,
DFSMSrmm issues the message EDG2225I. The processing continues, and DFSMSrmm
sets a minimum return code of 4 and ignores the VRS. No data sets can match to or use the
VRS.
8.4.2 Release options applied if VRS matched
Release options for matching VRSs can be applied regardless of whether the data set
actually is ever VRS retained.
Note: VRSEL(NEW) is required to implement the
separating the data set name mask from
the policy
processing.
Mask
Policy
Mask
Policy
Name 1
Policy
Name 2
Policy
Name 1
Policy
Mask 1
Policy
Mask 2
Policy
Mask 3
Policy

278
z/OS V1R8 DFSMS Technical Update
The combination of COUNT(0) and release option changes ensures that a volume can be
scratched the same day that a data set on it is created. Figure 8-17 shows you how you can
specify COUNT(0) in a DSNAME VRS definition.
Figure 8-17 VRS count zero processing
When a data set is no longer retained by a vital record specification, DFSMSrmm releases
the volume on which the data set resides only if no data set and the volume is retained by a
vital record specification. If you use the DFSMSrmm EDGRMMxx PARMLIB OPTION
command VRSEL(NEW) option and the RMM ADDVRS RELEASE(EXPIRYDATEIGNORE)
operand, DFSMSrmm ignores the volume expiration date and uses information in a vital
record specification to control retention. There are two special RELEASE options available,
and you can select one or both of these options together:
EXPIRYDATEIGNORE
SCRATCHIMMEDIATE
DFSMSrmm does not immediately return a volume to scratch status or to its owner when a
volume reaches its expiration date and is not retained by a vital record specification. You
must run expiration processing two times to return a volume to scratch status or to its owner.
The first run of expiration processing sets the volume status to pending release. The second
run of expiration processing completes the return. Running expiration processing two times
gives you time to make changes to the volume status before the volume is released.
Note: Sometimes DFSMSrmm cannot make the return in a single run, for example, there
may be other release actions required.
COUNT(0)
– ADDVRS subcommand
• New range for COUNT and STORENUMBER is
0 to 99999.
– COUNT(0) can be specified in data set name and
retention name VRSs
– When you specify COUNT(0) it does not matter
what location or retention criteria you specify for
the VRS because these are never considered by
DFSMSrmm
– You can now specify that a VRS is NOT to retain
a data set by use of COUNT(0)
for VRSEL(NEW) only
for VRSEL(NEW) only
ANDVRS \NEXTVRS
mask
DSNAME
JOBNAME
GDG\NOGDG
policy
COUNT(0)
STORENUMBER
DAYS \CYCLES \...
DELAY
WHILECATALOG
UNTILEXPIRED
LOCATION
DELETEDATE
DESCRIPTION
OWNER
RELEASE
PRIORITY

Chapter 8. DFSMSrmm enhancements
279
If you do not need to run expiration processing in two runs, specify the DFSMSrmm
EDGRMMxx PARMLIB OPTION command VRSEL(NEW) option and the RMM ADDVRS
RELEASE(SCRATCHIMMEDIATE) operand. This enables you to return volumes to scratch
in a single run of expiration processing. For more information about the RELEASE option
refer to the DFSMSrmm Implementation and Customization Guide, SC26-7405.
Figure 8-18 VRS release option
8.4.3 Special ABEND and OPEN via DSNAME match
The support for ABEND and OPEN is extended to allow selection of the appropriate policy
using the data set name mask. You can use the reserved data set or job names ABEND and
OPEN to specify policies for:
Data sets closed as a result of an abnormal end (ABEND flag in the data set record ON) in
a task
Data sets that are left open (OPEN flag in the volume record ON) or are in use during
inventory management.
Figure 8-19 shows you how you can specify the new reserved ABEND and OPEN masks in
the ADD VRS subcommand.
Figure 8-19 Add new ABEND or OPEN VRS
 Release options can now be
applied to all data sets that
match to a VRS, not just to
those retained by a VRS
– With COUNT(0) drop data sets on
creation day (ABEND VRS for
example)
– Multi-File Volumes
• VRS retained data sets override
‘Matched’ data sets
for VRSEL(NEW) only
for VRSEL(NEW) only
ANDVRS \NEXTVRS
mask
DSNAME
JOBNAME
GDG\NOGDG
policy
COUNT
STORENUMBER
DAYS \CYCLES \...
DELAY
WHILECATALOG
UNTILEXPIRED
LOCATION
DELETEDATE
DESCRIPTION
OWNER
RELEASE
PRIORITY
DSNAME VRS
———ADDVRS——————DSNAME(data_set_name_mask)—————————JOBNAME(jobname_mask)———

| | | | | |
|——AS————| |—DSNAME(’ABEND’)————————————| |—JOBNAME(’ABEND’)——————|
| | | |
|—DSNAME(’OPEN’)—————————————| |—JOBNAME(’OPEN’)———————|

280
z/OS V1R8 DFSMS Technical Update
This allows you to use either data set name masks or job name masks to manage open or
abend data sets.
The data set name mask or job name can even be used to match to data sets via
management class or vital record specification management value as long as the data set
name mask specified is not more than a single qualifier.
8.4.4 Find unused VRSs
With this release DFSMSrmm helps you manage your VRSs by VRSEL maintaining the last
reference date (DLR) and last reference time (TLR) for each VRS. This new function also
includes the following new functions:
Counts the number of unused VRSs
Identifies which VRS policy chains are not being used
You can use this information to identify and delete VRSs that are no longer required.
The VRS last reference date and time are now externalized in the report extract file in the
output of the LISTVRS subcommand, and the last reference date is also shown in the
REPORT file, as shown in Figure 8-20. Also, the REPORT file contains a list of VRS chains
not used in this VRSEL run.
Figure 8-20 Vital records retention report
The MESSAGE file contains an additional message that shows the number of VRSs that are
not used by this VRSEL run.
The maintenance of the VRS last reference date and time is supported only if you have
specified VRSEL(NEW). In the REPORT file the LASTREF column is listed in any case:
VRSEL(NEW) It will be filled with dates.
VRSEL(OLD) It will be filled with blanks.
REMOVABLE MEDIA MANAGER VITAL RECORDS RETENTION REPORT PAGE 120
(C) IBM CORPORATION 1993,2006 ----- ------- --------- ------ TIME 11:21:02 DATE 03/22/2006
JOB MASK DATA SET OR VOLUME MASK OWNER TYPE RETN C X DELETE DLY COUNT STNUM LOCATION RLSE
LASTREF
____________________________________________________________________________________________________________________________________
*.GDG.*.ICRMT.** X002 DSN CYCLES Y N 12/31/1999 0 99999 99999 LOCAL
03/22/2001
JOB NAME DATA SET NAME 2ndVRS 2ndNAMED FSEQ DSEQ VOLUME VSEQ OWNER CURRENT REQUIRED PRTY RETDATE RETNAME
____________________________________________________________________________________________________________________________________
X015IJIC DT04.GDG.DSNDB06.ICRMT.SYSSTR.G0001V00 12 12 L01699 1 X015 LOCAL LOCAL 300 WHILECATLG *
NUMBER OF DATA SETS RETAINED (GROUP STORE) = 1 1
X015IJIC DT04.GDG.DSNDB06.ICRMT.SYSUSER.G0001V00 11 11 L01699 1 X015 LOCAL LOCAL 300 WHILECATLG *
NUMBER OF DATA SETS RETAINED (GROUP STORE) = 1 1
X015IJIC DT04.GDG.DSNDB06.ICRMT.SYSVIEWS.G0001V00 13 13 L01699 1 X015 LOCAL LOCAL 300 WHILECATLG *
NUMBER OF DATA SETS RETAINED (GROUP STORE) = 1 1
REMOVABLE MEDIA MANAGER
UNUSED VRS CHAINS REPORT
PAGE 121
(C) IBM CORPORATION 1993,2006
------ --- ------ ------
TIME 11:21:02 DATE 03/22/2006
JOB MASK DATA SET OR VOLUME MASK OWNER TYPE RETN C X DELETE DLY COUNT STNUM LOCATION RLSE
LASTREF
____________________________________________________________________________________________________________________________________
*.BACKUP.** LIB DSN CYCLES Y N 12/31/1999 0 99999 99999 LOCAL
01/23/1992
*.IMGCPY.IMSDB.** LIB NAME DAYS N Y 12/31/1999 0 99999 99999 HOME
12/12/2004
VITAL.** LIB DSN CYCLES Y N 12/31/1999 0 99999 99999 CURRENT IX
05/27/2003
VITAL RECORDS RETENTION REPORT
----- ------- --------- ------
VITAL RECORDS RETENTION REPORT
----- ------- --------- ------
LASTREF
----------
03/22/2001
LASTREF
----------
03/22/2001
UNUSED VRS CHAINS REPORT
------ --- ------ ------
UNUSED VRS CHAINS REPORT
------ --- ------ ------
LASTREF
----------
01/23/1992
12/12/2004
05/27/2003
LASTREF
----------
01/23/1992
12/12/2004
05/27/2003

Chapter 8. DFSMSrmm enhancements
281
8.4.5 Incomplete VRS chains - dummy VRS *broken*
If VRSEL processing finds an incomplete chain it is reported with EDG2230I, but the
processing continues. In this case DFSMSrmm sets a minimum return code of 4 and any data
set that matches this incomplete chain is retained by a special Name VRS *broken*.
Figure 8-21 shows you an example of an incomplete VRS chain.
Figure 8-21 Incomplete VRS chain
The two new messages that you can get if the VRSEL processing detects errors are:
EDG2230I NEXTVRS name_vrs DOES NOT EXIST. CHAINING vrs_type VRS IS vrs_mask.
Where:
vrs_name The NAME vital record specification defined by NEXTVRS.
vrs_type DSN - DSNAME vital record specification.
VOL - VOLUME vital record specification.
NAME - NAME vital record specification.
vrs_mask The mask that uniquely identifies the vital record
specification with the chaining error. For DSNAME vital
record specifications the mask includes the data set name
and optionally the job name.
Severity Information.
Explanation During inventory management vital record processing,
DFSMSrmm checks all vital record specification chains by
following the chain using the NEXTVRS values. The vital record
specification displayed in the message does not exist in the
DFSMSrmm control data set.
Source DFSMSrmm.
Detecting Module EDGVREC0.
System Action Processing continues and DFSMSrmm sets a minimum return
code of 4. For VRSEL(OLD) DFSMSrmm retains additional data
sets or volumes, in the home location, up to the COUNT value
specified in the initial VRS in the chain. For VRSEL(NEW)
additional data sets are retained in the current location,
permanently.
Operators Response None.
Sysprogr Response ´ Add the missing vital record specification or correct the
NEXTVRS value specified on the vital record specification
displayed in the message.
Routing Codes N/A.
Descriptor Codes None.
DSN1
Policy
NAME1
Policy
NAME2
Policy
*broken*
Policy
count (99999)
days
location(current)
ANDVRS(NAME1)
NEXTVRS(NAME2)
NEXTVRS(NAME3)

282
z/OS V1R8 DFSMS Technical Update
EDG2317E MIGRATION FROM VRSEL(OLD) TO VRSEL(NEW) IS RECOMMENDED.
Where:
Severity Warning.
Explanation During inventory management vital record processing,
DFSMSrmm checks the VRSEL option that you defined in the
EDGRMMxx PARMLIB. The VRSEL(OLD) option will be
removed in a future release of z/OS. Migrate using
VRSEL(NEW).
Source DFSMSrmm.
Detecting Module EDGMHKP.
System Action Processing continues. A minimum return code of 4 is set.
Operators Response None.
Sysprogr Response´ Plan a migration to VRSEL(NEW). Refer to the migration
planning steps documented in the DFSMSrmm Implementation
and Customization Guide, SC26-7405.
Routing Codes 11.
Descriptor Codes None.
8.4.6 Toleration and removal of old functions
IBM recommends that you perform a migration to VRSEL(NEW). Refer to the migration
planning steps documented in the DFSMSrmm Implementation and Customization Guide.
This prevents use of old VRS operands STARTNUMBER, LOCATION(BOTH), and
STORENUMBER(xx,yy).
The toleration APAR OA13355 includes a ++HOLD(ACTION) and requires that VRSs are
cleaned up before you are able to run EDGHSKP with VRSEL on a z/OS V1R8 system.
The cleanup actions are documented under message EDG2221E. EDG2221E sets return
code 12 instead of 4.
Error messages
A description of the new message you get if the VRSEL processing detects some errors is:
EDG2222E type VRS FOR mask SPECIFIES UNSUPPORTED OPTIONS - SOME RETENTION
OPTIONS IGNORED.
Where:
type type Is the type of vital record specification. It can be one of:
DSN - DSNAME type vital record specification
VOL - VOLUME type vital record specification
mask This is the vital record specification data set name or volume
serial number.
Severity Information.
Explanation During vital record processing, DFSMSrmm found a vital record
specification that contains unsupported options. The
unsupported options are STARTNUMBER and
LOCATION(BOTH).
Source DFSMSrmm.
Detecting Module EDGVREC0.
System Action Processing ends. A return code of 12 is set.
Important: As long as you do not implement COUNT(0) or JOBNAME(ABEND\OPEN)
you can run VRSEL processing on either a toleration system or z/OS V1R8.

Chapter 8. DFSMSrmm enhancements
283
Operators Response None.
Sysprogr Response You must replace the vital record specification with other vital
record specifications that provide the retention options that you
require. For example, if you use LOCATION(BOTH) you can
replace it with use of the NEXTVRS operand. This example
shows first an unsupported vital record specification and then the
equivalent supported vital record specifications that you might
use. Unsupported:
RMM ADDVRS DSNAME(data_set_name_mask) -
CYCLES COUNT(5) LOCATION(BOTH) -
STORENUMBER(2,1)
Supported:
RMM ADDVRS DSNAME(data_set_name_mask) -
CYCLES COUNT(5) LOCATION(LOCAL) -
STORENUMBER(2) NEXTVRS(DIST1C)
RMM ADDVRS NAME(DIST1C) LOCATION(DISTANT) -
STORENUMBER(1)
If you use STARTNUMBER you can replace it with the use of the
NEXTVRS operand. This example shows first an unsupported
vital record specification and then the equivalent supported vital
record specifications that you might use.
Unsupported:
RMM ADDVRS DSNAME(data_set_name_mask) -
CYCLES COUNT(3) LOCATION(VAULT1) -
STORENUMBER(2) STARTNUMBER(1)
Supported:
RMM ADDVRS DSNAME(data_set_name_mask) -
CYCLES COUNT(3) LOCATION(HOME) -
STORENUMBER(1) NEXTVRS(VLT12C)
RMM ADDVRS NAME(VLT12C) LOCATION(VAULT1) -
STORENUMBER(2)
In addition, IBM suggests that you perform a migration to
VRSEL(NEW). See the migration planning steps documented in
the DFSMSrmm Implementation and Customization Guide,
SC26-7405.
Routing Codes 11.
Descriptor Codes 7.

284
z/OS V1R8 DFSMS Technical Update
8.4.7 Conversion to DFSMSrmm from other tape management systems
Before release V1R8, DFSMSrmm converted only DSN, LABEL, and JOB information to
UXTABLE entries. Now it also supports MGMTCLAS and ABEND keywords, as shown in
Figure 8-22.
Figure 8-22 MGMCLAS and ABEND RDS keywords support
EDGCRFMT and EDGRSRDS conversion programs
If you have a CA-1 retention data set (RDS) entry, then this RDS entry is converted into two
steps to create a UXTABLE entry that can be used in the EDGUX100 user exit:
1.EDGCRFMT creates input for EDGCSRDS.
2.EDGCSRDS builds the UXTABLE entries.
Example 8-8 shows you some CA-1 retention data set (RDS) entries.
Example 8-8 CA-1 retention data set (RDS) entries
M=SMSCLAS1,LABEL=RETPD=2010/011,JOB=TEST
MGMTCLAS=SMSCLAS2,LABEL=RETPD=2020/011
DSN=RXXXX.XXXX,ABEND=RETPD=2010/001,JOB=TESTTEST
DSN=RXXXX.XXXX,LABEL=RETPD=2010/001,JOB=TESTTEST
DSN=R3650.*,LABEL=RETPD=3650,JOB=COMBIN*
{ }{ {keyword }}
{ }{ {preferred date}}[,J=create.jobname ]
{D=dsname[-] }{,LABEL=EXPDT={Julian date }}[,JOB=create.jobname]
{DSN=dsname[-] }{,LABEL=RETPD=nnnn }[,JOBNAME=create.jobname]
MGMTCLAS=smsclass {,LABEL=WRETPD=nnnn }[,SELECT=ALL ]
M=smsclass {,ABEND=RETPD=nnnn }
{,ABEND=WRETPD=nnnn }
{,ABEND=EXPDT={keyword }}
{preferred date}
{Julian date }

Chapter 8. DFSMSrmm enhancements
285
These CA-1 retention data set entries are converted to UXTABLE, as shown in Figure 8-23.
For more information about a CA-1 to DFSMSrmm conversion refer to the IBM Redbooks
publication Converting to DFSMSrmm from CA-1, SG24-6241. In this book the conversion is
described in detail.
Figure 8-23 Sample UXTABLE entries
EDGCSVDS conversion program
EDGCSVDS converts the vaulting policies from the VPDD into K-Records representing the
DFSMSrmm storage location movement policies.
Processing is changed to implement data set filter VRSs for all DSN entries found in the
VPDD. Instead of storing data set and job names in memory until the vault statements are
reached, and then looping through them to create K records, we can now create a K record
for the filter as we read the entry from the VPDD and no longer have memory limitations in
EDGCSVDS.
* start of RDS entries
EDGCVRSG DSN=RXXXX.XXXX.G%%%%V%%, X
JOB=ABEND, X
RO=001, X
RETPD=2010
EDGCVRSG DSN=RXXXX.XXXX, X
JOB=ABEND, X
RO=001, X
RETPD=2010
EDGCVRSG DSN=RXXXX.XXXX.G%%%%V%%, X
JOB=TESTTEST, X
RO=001, X
RETPD=2010
EDGCVRSG DSN=RXXXX.XXXX, X
JOB=TESTTEST, X
RO=001, X
RETPD=2010
EDGCVRSG DSN=R3650.*, X
JOB=COMBIN*, X
RO=NO, X
RETPD=3650
EDGCVRSG DSN=SMSCLAS1, X
JOB=TEST, X
RO=011, X
RETPD=2010
EDGCVRSG DSN=SMSCLAS2, X
RO=011, X
RETPD=2020
* default RP value
EDGCVRSG DSN=*, X
RO=NO, X
RETPD=3

286
z/OS V1R8 DFSMS Technical Update
EDGCNVT conversion program
EDGCNVT is updated to allow COUNT(0) in the count filed of the EDGCKREC record. This
program converts the output of the data extraction programs to DFSMSrmm format (CDS
records) or ADDVRS commands, as shown in Figure 8-24.
Figure 8-24 EDGCNVTRMM TSO ADDVRS commands
8.5 DFSMSrmm usability items
With this new release there are some new, easy-to-use ISPF dialog and RMM TSO
subcommand enhancements. These simplify the tasks performed by the storage
administrator and simplify the analysis of data set and volume retention.
8.5.1 Updates to RMM TSO SEARCHVOLUME subcommand
There are enhancements to the TSO RMM SEARCHVOLUME subcommand.
CLIST operand
When you specify the TYPE(LOGICAL) operand and CLIST, DFSMSrmm returns more
information in the output file if the obtained logical volume resides on a stacked volume. In
such a case, DFSMSrmm returns the first six characters of the container name, the logical
volume serial number, and the status value. The status value can be:
SCRATCH If the volume is in scratch status or ready to return to scratch with the
SCRATCHIMMEDIATE release option set
INITIALIZE If the volume is in scratch status and contains no valid data
Blank If status is not available
Operand JOBNAME or NOJOBNAME
Use this operand to search for volumes created by the specified job name. A job name is one
to eight alphanumeric characters or $, #, or @, and must start SEARCHVOLUME
Separating data set name mask from the policy itself
DSN=A1.TEST
DSN=A2.TEST
V=MINSK,C=1
RMM ADDVRS DSN('A1.TEST ') NOGDG
CYCLES
COUNT( 1)
DELAY( 0)
OWNER(K ) DELETE(1999/365)
LOCATION(MINSK )
STORENUM( 1) PRIORITY( 0)
RMM ADDVRS DSN('A2.TEST ') NOGDG
CYCLES
COUNT( 1)
DELAY( 0)
OWNER(K ) DELETE(1999/365)
LOCATION(MINSK )
STORENUM( 1) PRIORITY( 0)
RMM ADDVRS DSN('A1.TEST ') NOGDG
CYCLES
COUNT( 0)
DELAY( 0)
OWNER(K ) DELETE(1999/365)
LOCATION(CURRENT )
STORENUM( 0) PRIORITY( 0)
NEXTVRS(VRS00001)
RMM ADDVRS NAME(VRS00001) CYCLES
COUNT( 1)
OWNER(K ) DELETE(1999/365)
LOCATION(MINSK )
STORENUM( 1)
NEXTVRS(VRS00002)
RMM ADDVRS DSN('A2.TEST ') NOGDG
CYCLES
COUNT( 0)
DELAY( 0)
OWNER(K ) DELETE(1999/365)
LOCATION(CURRENT )
STORENUM( 0) PRIORITY( 0)
NEXTVRS(VRS00001)
now
before
V1R8
VPD

Chapter 8. DFSMSrmm enhancements
287
Subcommand with an alphabetic character, $, #, or @. You can also use a generic job name.
Use % in your generic job name mask to match any one character and * to match any
character string in the job name. If you do not specify JOBNAME, jobname is not used as a
selection. If you specify JOBNAME(*), DFSMSrmm returns all volumes that match the search
values specified and that have a job name. Volumes that do not have a job name are not
listed.
JOBNAME is mutually exclusive with NOJOBNAME.
Operand LOCATION
LOCATION(SHELF | LOCAL | DISTANT | REMOTE | library_name | LOCDEF_location_name |
generic_location_name)
Specify to list volumes residing in a specific location. Specify one of the following:
SHELF
Volumes stored in shelf locations in a non-system-managed library.
LOCAL, DISTANT, or REMOTE
DFSMSrmm built-in storage locations. Use the LOCATION operand together with
INTRANSIT to limit the list to only those volumes residing in or moving from the specific
location. Use the LOCATION operand together with HOME to limit the list to only those
volumes residing in a specific location that has the same home location.
library_name
Volumes stored in shelf locations in a specific system-managed library. A library name is
one-to-eight alphanumeric characters starting with a non-numeric character. You cannot
specify a distributed library name.
LOCDEF_location_name
Volumes moving to a storage location that was defined using the LOCDEF command. You
can enter any value, as no checking is done against the current list of locations defined to
DFSMSrmm.
generic_location_name
You can use a generic location name. You can use % in your generic mask to match any
one character and * to match any character string.
Operand RELEASEACTION
RELEASEACTION(ALL, ERASE, INIT, NOTIFY, REPLACE, RETURN, SCRATCH)
Release actions are those that will be set as pending actions when a volume is released. Use
this operand to search for volumes that have the specified release action set. Volumes are
returned if any of the values you specify are set in the volume. Also see the ACTION operand
for how to search for volumes with pending actions.
Specify a value of ALL, or one or more of the following, separated by commas:
ALL To list all volumes with any pending action
ERASE To list only volumes that require erasing
INIT To list only volumes that require initialization
RETURN To list only volumes that should be returned to their owner
REPLACE To list only volumes that must be replaced by new volumes and
returned to the scratch pool

288
z/OS V1R8 DFSMS Technical Update
NOTIFY To list only volumes for which owners must be notified
SCRATCH To list all volumes to be returned to scratch status
8.5.2 ISPF lists show retention information
The ISPF dialog has been updated to shown more information about the search data set
result list and search volume result list panels.
Volume search result list
In the volume search result list (Figure 8-25) there are now two useful enhancements
available:
The Set Retained column shows whether the volume is retained because the
RETAINBY(SET) option is in use and other volumes in the set are either VRS retained or
retained by volume expiration date.
The Status column has been corrected to display volume availability instead of volume
status if volume availability is not blank.
Figure 8-25 DFSMSrmm search volume result panel 1 of 2
Where:
The Set Retained (SR) column shows whether the volume is:
Y Yes for set retained
(blank) Not set retained
The Volume Status can be one of:
Availability
LOAN The volume is on loan.
RELEASE The volume is pending release.
OPEN A file on the volume is opened for output.
Panel Help Scroll
______________________________________________________________________________
DFSMSrmm Volumes (Page 1 of 2) Row 1 to 11 of 11
Command ===> Scroll ===> PAGE

Enter HELP or PF1 for the list of available line commands
Use the RIGHT command to view other data columns
Volume Assigned Expiration S Dest- Tr- Data
S serial Owner date date R Status Location ination ans sets
-- ------ -------- ---------- ---------- - ------- -------- -------- --- -----
TST008 STC 2007/058 2007/058 VRS LIB1 N 1
TST009 HAIMO 2004/296 1999/365 Y MASTER LIB1 N 1
TST010 STC 2007/060 1999/365 Y MASTER LIB1 N 1
TST011 STC 2007/059 2007/061 VRS LIB1 N 1
TST014 2007/060 ENTRY SHELF LIB1 Y 0
TST020 2007/072 SCRATCH LIB1 N 2
TST021 STC 2007/058 2007/060 VRS LIB1 N 1
TST023 STC 2007/059 1999/365 LOAN LIB1 N 1
TST026 PAOLOR3 2003/302 2003/365 RELEASE LIB1 N 1
TST029 MHLRES2 2007/059 2007/071 OPEN LIB1 N 2
TS4284 SIEGEL 2007/073 2007/073 MASTER SHELF N 99
******************************* Bottom of data ********************************

Chapter 8. DFSMSrmm enhancements
289
VRS The volume is being retained by a vital record specification.
Status MASTER, SCRATCH, USER, INIT, or ENTRY
MASTER The volume contains valid user data and cannot be overwritten
unless the data set names match.
USER The volume is assigned to a user and is in use. It contains any
type of data and can be overwritten, rewritten, and used
repeatedly until the volume's expiration date.
INIT The volume is awaiting initialization before becoming available
for use as a scratch volume.
ENTRY The volume has been predefined to DFSMSrmm prior to entry
into an ATLDS, for use as a scratch volume.
SCRATCH The volume is free and available for use. It contains no data or
the data on the volume has expired or is not valid.
In the right section of the volume search result panel there are no changes, as shown in
Figure 8-26
Figure 8-26 DFSMSrmm search volume result panel 2 of 2
Panel Help Scroll
______________________________________________________________________________
DFSMSrmm Volumes (Page 2 of 2) Row 1 to 11 of 11
Command ===> Scroll ===> PAGE

Enter HELP or PF1 for the list of available line commands
Use the LEFT command to view other data columns
Volume Rack Media Home Media Recorded Compac- Attri-
S serial number name location type Label format tion butes
-- ------ ------ -------- -------- -------- ----- -------- -------- --------
TST008 TST008 MEDIA3 LIB1 HPCT SL 128TRACK YES NONE
TST009 TST009 MEDIA3 LIB1 HPCT SL 128TRACK YES NONE
TST010 TST010 MEDIA3 LIB1 HPCT SL 128TRACK YES NONE
TST011 TST011 MEDIA3 LIB1 HPCT SL 128TRACK YES NONE
TST014 TST014 MEDIA3 SHELF HPCT SL 128TRACK * NONE
TST020 TST021 MEDIA3 LIB1 HPCT SL 128TRACK YES NONE
TST021 TST021 MEDIA3 LIB1 HPCT SL 128TRACK YES NONE
TST023 TST023 MEDIA3 LIB1 HPCT SL 128TRACK YES NONE
TST026 TST026 MEDIA3 LIB1 HPCT SL 128TRACK YES NONE
TST029 TST028 MEDIA3 LIB1 HPCT SL 128TRACK YES NONE
TST284 TST029 MEDIA3 LIB1 HPCT SL 128TRACK YES NONE
******************************* Bottom of data ********************************

290
z/OS V1R8 DFSMS Technical Update
Data set search result list
In the right section of the data set search result list shown in Figure 8-27 you now can directly
see whether a data set is vital record selected. You must press the right button to show the
right panel information of the of the data set search result. The VRS Retained column
displays whether data sets are retained by DFSMSrmm VRS retention. It shows whether the
data set is:
Y Yes if data set VRS retained
(blank) If data set is not VRS retained
Figure 8-27 DFSMSrmm data set display 1 of 2
Panel Help Scroll
______________________________________________________________________________
EDGPD020 DFSMSrmm Data Sets (Page 1 of 2) Row 1 to 16 of 16
Command ===> Scroll ===> PAGE


Enter HELP or PF1 for the list of available line commands
Use the RIGHT command to view other data columns
Volume File
S Data set name serial Owner seq
-- -------------------------------------------- ------ -------- -----
HSM.BACKTAPE.DATASET TST010 STC 1
HSM.BACKTAPE.DATASET TST027 STC 1
HSM.COPY.HMIGTAPE.DATASET TST002 STC 1
HSM.COPY.HMIGTAPE.DATASET TST004 STC 1
HSM.COPY.HMIGTAPE.DATASET TST016 STC 1
HSM.DMP.CRYPCOPY.VMHL0A0.D07072.T141318 TST019 STC 1
HSM.DMP.CRYPCOPY.VMHL0A1.D07072.T141318 TST019 STC 2
HSM.DMP.MHLRES.VMHL0A0.D07059.T425215 TST015 STC 1
HSM.DMP.MHLRES.VMHL0A1.D07059.T425215 TST023 STC 1
HSM.DMP.MHLRES.VMLD00B.D06110.T385618 TST013 STC 1
HSM.DMP.PLUSCOPY.VMHL0A0.D07058.T060317 TST024 STC 1
HSM.DMP.PLUSCOPY.VMHL0A1.D07058.T060317 TST024 STC 2
HSM.HMIGTAPE.DATASET TST001 STC 1
HSM.HMIGTAPE.DATASET TST012 STC 1

Chapter 8. DFSMSrmm enhancements
291
Press the right button to show the right panel of the search data set result, as shown in
Figure 8-28, and to see whether a data set is VRS retained.
Figure 8-28 DFSMSrmm data set display 2 of 2
8.5.3 SELECT primary command in RMM dialog search results
Within the DFSMSrmm ISPF dialog, a new SELECT primary command, as well as changes
to the search results lists for data sets, volumes, and vital record specifications, were added.
The SELECT command is supported on all results lists. You can use this primary command
to apply the same line command to all selected entries in a search results list table. For
example, all volumes in the volume result list are ejected when you issue the command
shown in Example 8-9 in a volume search results list.
Example 8-9 Using an asterisk to select all
Command ===> SELECT * E
You can also specify SEL or S, instead of SELECT. When you do not specify a line
command, S is the default.
You can specify a generic value, as shown in Example 8-10, for the first parameter, or a
specific value. For example, issue the command shown in Example 8-10 to view volume data
for all volumes starting with AB0, or issue a specific volser as shown in Example 8-11 on
page 292to release the volume MW0001, if found in the list.
Example 8-10 Using a generic volume mask
Command ===> SELECT AB0* V
Panel Help Scroll
ssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssss
EDGPD030 DFSMSrmm Data Sets (Page 2 of 2) Row 1 to 24 of 24
Command ===> Scroll ===> PAGE


Enter HELP or PF1 for the list of available line commands
Use the LEFT command to view other data columns
Create Expiration V
S Data set name date date R
-- -------------------------------------------- ---------- ---------- -
HSM.BACKTAPE.DATASET 2007/060 1999/365 Y
HSM.BACKTAPE.DATASET 2007/053 1999/365 Y
HSM.COPY.HMIGTAPE.DATASET 2007/054 1999/365
HSM.COPY.HMIGTAPE.DATASET 2007/060 1999/365
HSM.COPY.HMIGTAPE.DATASET 2007/060 1999/365
HSM.DMP.CRYPCOPY.VMHL0A0.D07072.T141318 2007/072 1999/365
HSM.DMP.CRYPCOPY.VMHL0A1.D07072.T141318 2007/072 1999/365
HSM.DMP.MHLRES.VMHL0A0.D07059.T425215 2007/059 1999/365
HSM.DMP.MHLRES.VMHL0A1.D07059.T425215 2007/059 1999/365
HSM.DMP.MHLRES.VMLD00B.D06110.T385618 2006/110 1999/365
HSM.DMP.PLUSCOPY.VMHL0A0.D07058.T060317 2007/059 1999/365
HSM.DMP.PLUSCOPY.VMHL0A1.D07058.T060317 2007/059 1999/365
HSM.HMIGTAPE.DATASET 2007/054 1999/365
HSM.HMIGTAPE.DATASET 2006/087 1999/365

292
z/OS V1R8 DFSMS Technical Update
Example 8-11 Specifying a full volume serial number
Command ===> SELECT MW0001 R
The first parameter is applied on the first key of the table. If you want another column of the
table to be taken as the select criteria, issue a SORT command before the select command.
For example, first SORT by owner, then issue the command shown in Figure 8-12 on
page 274 to eject all volumes of owner SCHLUM.
Example 8-12 Select all information assigned to owner SCHLUM
Command ===> SELECT SCHLUM E
For example, we use the
Search for data sets
dialog to search for data sets starting with
high-level qualifier MHLRES, as shown in Figure 8-29.
Figure 8-29 DFSMSrmm search data set dialog
We get a lot of data set information and use the SORT command to sort the data set list by
OWNER in descending order to show all data sets have an owner first. Figure 8-30 on
page 293 shows you the result and the command that we use, where:
SORT This is the command.
D This is the direction in which it can be A for ascending or D for
descending:
ASCENDING ( A )
DESCENDING ( D )
O This is the identifier in a shortened form of one of the data column
names in the data sets list. It can be one of the following:
CREATE ( C ) - Create date
OWNER ( O ) - Owner
Panel Help
DFSMSrmm Data Set Search
Command ===>

Enter fully qualified or partial data set name and job name:

Data set name . . . . 'mhlres*.**'
Job name . . . . . . . Specific or generic name

Enter optional parameters to qualify search

Owner . . . . . . . . * Owner of volumes (Default is your userid)
Volume serial . . . . OR Volume serial
List entire set . . . YES, For all data sets in the
multi-volume set, otherwise NO
Status . . . . . . . . PRIVATE, SCRATCH, or blank for all
Retained by VRS . . . YES, NO, or blank for all
Create date . . . . . Created since YYYY/DDD
Physical file seq . . Relative position on the volume
Limit . . . . . . . . * Limit search to first nnnn data sets
Clist . . . . . . . . YES to create a data set, or NO, or blank
Program name . . . . . Specific or generic name

Enter HELP or PF1 for the list of available line commands

Chapter 8. DFSMSrmm enhancements
293
DATASET ( D ) - Data set name
VOLUME ( VO ) - Volume serial
FILESEQ ( F ) - Physical file sequence number
EXPIRY ( E ) - Expiration date
VR ( VR ) - VRS retained
Figure 8-30 DFSMSrmm search data set result
Panel Help Scroll
______________________________________________________________________________
DFSMSrmm Data Sets (Page 1 of 2) Row 2,079 to 2,110 of 2,110
Command ===> SORT D O Scroll ===> CSR

Enter HELP or PF1 for the list of available line commands
Use the RIGHT command to view other data columns
Volume File
S Data set name serial Owner seq
-- -------------------------------------------- ------ -------- -----
MHLRES1.TEST.DATA DV2083 1
MHLRES1.TEST.DATA DV2084 1
MHLRES1.TEST.DATA DV2085 1
MHLRES1.TEST.DATA DV2086 1
MHLRES1.TEST.DATA DV2087 1
MHLRES1.TEST.DATA DV2088 1
MHLRES1.TEST.DATA DV2089 1
MHLRES1.TEST.DATA DV2090 1
MHLRES1.TEST.DATA DV2091 1
MHLRES1.TEST.DATA DV2092 1
MHLRES1.TEST.DATA DV2093 1
MHLRES1.TEST.DATA DV2094 1
MHLRES1.TEST.DATA DV2095 1
MHLRES1.TEST.DATA DV2096 1
MHLRES1.TEST.DATA DV2097 1
MHLRES1.TEST.DATA DV2098 1
MHLRES1.TEST.DATA DV2099 1
MHLRES4.ABARS.OUTPUT.C.C01V0001 TST022 STC 4
MHLRES4.ABARS.OUTPUT.D.C01V0001 TST022 STC 1
MHLRES4.ABARS.OUTPUT.I.C01V0001 TST022 STC 3
MHLRES4.ABARS.OUTPUT.O.C01V0001 TST022 STC 2
MHLRES5.TEST.FILE1.D130 MARY01 1
MHLRES5.TEST.FILE1.D130 MARY02 1
MHLRES5.TEST.FILE1.D130 MARY03 1
MHLRES5.TEST.FILE1.D130 MARY04 1
MHLRES5.TEST.FILE2.D130 MARY01 2
MHLRES5.TEST.FILE2.D130 MARY02 2
MHLRES5.TEST.FILE2.D130 MARY03 2
MHLRES5.TEST.FILE2.D130 MARY04 2
MHLRES5.TEST.TAPE TST007 1
MHLRES5.TEST.TAPE TST029 MHLRES2 2
MHLRES6.TEST.TAPE TST029 MHLRES2 1

294
z/OS V1R8 DFSMS Technical Update
Figure 8-31 shows you the results after the panel displays the information in descending
order sorted by OWNER and the use of the SELECT command to select all data sets owned
by user STC to changing anything.
Figure 8-31 Result after using the SELECT command
Where:
SELECT DFSMSrmm-supported primary command
STC Specific owner should be used to set the line command
C Line command to change the data set information
Panel Help Scroll
DFSMSrmm Data Sets (Page 1 of 2) 4 rows updated
Command ===> SELECT STC C Scroll ===> PAGE

Enter HELP or PF1 for the list of available line commands
Use the RIGHT command to view other data columns
Volume File
S Data set name serial Owner seq
-- -------------------------------------------- ------ -------- -----
C MHLRES4.ABARS.OUTPUT.C.C01V0001 TST022 STC 4
C MHLRES4.ABARS.OUTPUT.D.C01V0001 TST022 STC 1
C MHLRES4.ABARS.OUTPUT.I.C01V0001 TST022 STC 3
C MHLRES4.ABARS.OUTPUT.O.C01V0001 TST022 STC 2
MHLRES6.TEST.TAPE TST029 MHLRES2 1
MHLRES5.TEST.TAPE TST029 MHLRES2 2
MHLRES1.TEST.DATA DV1897 1
MHLRES1.TEST.DATA DV1896 1
MHLRES1.TEST.DATA DV1901 1
MHLRES1.TEST.DATA DV1900 1
MHLRES1.TEST.DATA DV1899 1
MHLRES1.TEST.DATA DV1898 1
MHLRES1.TEST.DATA DV1906 1
MHLRES1.TEST.DATA DV1905 1
MHLRES1.TEST.DATA DV1904 1
MHLRES1.TEST.DATA DV1903 1
MHLRES1.TEST.DATA DV1902 1
MHLRES1.TEST.DATA DV1910 1
MHLRES1.TEST.DATA DV1909 1
MHLRES1.TEST.DATA DV1908 1
MHLRES1.TEST.DATA DV1907 1
MHLRES1.TEST.DATA DV2002 1
MHLRES1.TEST.DATA DV2001 1
MHLRES1.TEST.DATA DV1999 1
MHLRES1.TEST.DATA DV1998 1
MHLRES1.TEST.DATA DV1997 1
MHLRES1.TEST.DATA DV1996 1
MHLRES1.TEST.DATA DV1995 1
MHLRES1.TEST.DATA DV1994 1
MHLRES1.TEST.DATA DV1993 1
MHLRES1.TEST.DATA DV1992 1
MHLRES1.TEST.DATA DV1991 1

Chapter 8. DFSMSrmm enhancements
295
8.5.4 RMM TSO CHANGEVRS subcommand
There is an RMM TSO CHANGEVRS subcommand available, like all other RMM CHANGE
commands, to update previously created VRS definitions defined to DFSMSrmm. The syntax
is similar to other change RMM TSO subcommands.
Syntax
Figure 8-32 shows you the syntax of the RMM TSO CHANGEVRS subcommand.
Figure 8-32 CHANGEVRS syntax diagram
Operands
The operands are:
DSNAME(data_set_name_mask)
Identifies the type of vital record specification and specifies the mask of a data set name,
management class, or management value of an existing vital record specification. The
mask can have a fully qualified or a generic name. It can also be one of the reserved
words ABEND or OPEN. The data set name mask is 1 to 44 characters, enclosed in
quotation marks if any special characters are included. If the data set name mask is not
enclosed in quotation marks, PROFILE PREFIX is applied. This operand is required and
must immediately follow the CHANGEVRS subcommand. DSNAME is mutually exclusive
with the NAME and VOLUME operands.
Note: To use the RMM TSO CHANGEVRS subcommand, you need CONTROL access to
the STGADMIN.EDG.VRS resource profile to change vital record specifications.

———CHANGEVRS——————————DSNAME(—data_set_name_mask—)————————————————————————
| | | |
|—————CS——————| |——NAME(————search_limit—)——————|



————————————————————————————————————————————————————————————————————————————
| |
|———JOBNAME(—jobname_mask—)——COUNT(—days/cycles—)——|



——————————————————————————————————————————————————————————————————————————
| | | |
|———NEXTVRS(—next_VRS_name—)——| |——TZ(—{+|-}HH[:MM[:SS]]—)——|
Note: DFSMSrmm no longer folds data set names to uppercase letters when you
specify quoted data set names. When you specify data set names or data set name
masks, be sure to specify the correct case for each character. If you create VRS data
set name masks with lowercase or mixed case letters, these will not match to data sets
with all uppercase characters.

296
z/OS V1R8 DFSMS Technical Update
NAME(VRS_name)
Identifies the vital record specification type and specifies a name for the vital record
specification. A vital record specification name is eight alphanumeric characters chosen
by your installation.
JOBNAME(jobname_mask)
Identifies the job name for the vital record specification. A job name is one-to-eight
alphanumeric characters, $, #, or @. You can specify a specific job name or a job name
mask. It can also be one of the reserved words ABEND or OPEN.
This operand is optional. You must specify the operand, though, if the vital record
specification that you want to change has the JOBNAME operand specified.
If you have data sets with job names that include symbols other than alphanumeric
characters, $, #, or @, use a job name mask to cover them.
COUNT(days/cycles)
Specifies a retention amount, based on the retention type of the existing vital record
specification. Specify COUNT(number_of_days) to request that DFSMSrmm retains all
cycles or copies of a data set. Specify COUNT(number_of_cycles) to request that
DFSMSrmm retains the number of data set cycles that you specify.
The value range for data set name vital record specification and retention name vital
record specification is 0 to 99999. A value of 99999 indicates that DFSMSrmm retains all
cycles of a data set.
If count() is not specified, the count value in the existing vital record specification record is
not changed.
DFSMSrmm validates the COUNT value as follows:
– If EXTRADAYS is specified, COUNT must equal STORENUMBER:
(STORENUMBER) = (COUNT).
– Regardless of whether NEXTVRS and ANDVRS are used, COUNT can be
(STORENUMBER) <= (COUNT).
– If DAYS or LASTREFERENCEDAYS retention is used, (STORENUMBER) + (DELAY)
<= (COUNT).
NEXTVRS(next_VRS_name)
Specifies the name of the next vital record specification in a chain of vital record
specifications. If you specify the name of a vital record specification that does not exist,
DFSMSrmm uses a dummy vital record specification with the name *broken* instead. This
keeps the volume or data set in its current location.
TZ({+|-}HH[:MM[:SS]])
Specifies the time zone offset when date and time values are specified. The format is
{+|-}HH[:MM[:SS]], where:
– +|- is the offset direction. Specify + to indicate that the offset is east of the zero median
(UT). Specify - to indicate that the offset is west of the zero median (UT). The offset
direction is required.
– HH is hours.
– MM is minutes.
– SS is seconds.
An optional colon (:) separates hours from optional minutes and optional seconds. You
can specify a time in the range of 00:00:00–15:00:00 for HH:MM:SS. The MM and SS
value range is 00–59.

Chapter 8. DFSMSrmm enhancements
297
8.5.5 RMM TSO SEARCHOWNER subcommand
Like other RMM TSO SEARCH subcommands, the SEARCHOWNER command is used to
create a list of resources defined to DFSMSrmm.
Syntax is similar to other RMM TSO subcommands, but no dialog support is added yet.
Use the SEARCHOWNER subcommand to create a list of owners defined to DFSMSrmm.
You can restrict how many owners DFSMSrmm displays by specifying the LIMIT or END
operand. DFSMSrmm searches until it reaches your limit or endpoint, or until it lists all owners
that match your search criteria. If you do not specify a search limit, DFSMSrmm lists a
maximum of ten.
The information DFSMSrmm returns for each owner in the list is:
Owner’s ID
Owner’s last name
Owner’s first names
Owner’s internal telephone number
Number of volumes owned
Note: To use the RMM TSO SEARCHOWNER subcommand, you need READ access to
the STGADMIN.EDG.MASTER resource profile.

298
z/OS V1R8 DFSMS Technical Update
Syntax
Figure 8-33 shows you the syntax of the RMM TSO SEARCHOWNER subcommand.
Figure 8-33 SEARCHOWNER syntax diagram
Operands
The operands are:
ADD
Specify this operand to request that new records written to the CLIST data set are added
after any existing records in the data set. When the CLIST data set is empty or
DFSMSrmm creates the CLIST data set during command execution, specifying ADD is
the same as specifying START.
ADD is mutually exclusive with START.
You can easily build a set of commands from CLIST processing using multiple SEARCH
subcommands of the same or different resource types. For variable length records, the
minimum record length can cause the LRECL to be increased. For fixed length records, if
the minimum length cannot be accommodated, the subcommand fails.
———SEARCHOWNER————————————————————————————————————————————————————————————
| | | |
|—————SO——————| | |——————10——————| |
|——LIMIT(————search_limit———)—|
| |——————*———————| |
| |
|——END(end_owner)—————————————|


————————————————————————————————————————————————————————————————————————————
| |
| |———LIST———| |
|———CLIST(—————————————————————,—————————————————————)————|——————————|——|
|——prefix_string——| |——sufffix_string——| |——NOLIST——|



————————————————————————————————————————————————————————————————————————————
| |
| |——START———| |
|———CONTINUE(————————————————————————)————————————|——————————|—————|
|——OWNER(ownerid)——| |———ADD————|


|——*———————————————————————|
——————OWNER(——————————————————————————————)———————————————————————————————
| |
| |——START———| |
|————|——————————|—————|
|———ADD————|


Chapter 8. DFSMSrmm enhancements
299
CLIST(prefix_string,suffix_string)
Specifies a CLIST to create a data set of executable commands. You can edit the data set
to remove any owners that you do not want in the list. Then you can run the CLIST at your
convenience.
DFSMSrmm returns the owner serial number for each record if you do not specify
(prefix_string and suffix_string). When the owner serial number contains special
characters the value is returned within quotation marks.
You can add RMM TSO subcommands and operands to the records in the CLIST data set
by specifying (prefix_string and suffix_string). These text strings cannot exceed 255
characters. Separate the prefix_string and suffix_string using a blank or a comma
between the text strings. Insert blanks in the prefix and suffix values to prevent
DFSMSrmm from concatenating the strings with the data that DFSMSrmm returns. To
enter a null prefix_string, add a pair of separator characters such as ’’ to the text string (for
example, CLIST(’’,’ suffix_string’)).
CONTINUE(OWNER(ownerid))
Specify the CONTINUE operand without any value to notify DFSMSrmm SEARCH
subcommand processing that you want to break down the search results based on the
LIMIT value and request that DFSMSrmm return the search continue information for use
with the next command. For TSO, the continue information is returned either as a REXX
variable or as a line mode message. When the subcommand is issued from the
DFSMSrmm API, the continuation information may be either a line mode message or an
SFI or XML attribute.
CONTINUE is an optional operand. Use the LIMIT operand to control the maximum
number of entries to be returned each time that you start or continue the search.
To continue a previous search subcommand, the CONTINUE operand value includes the
following value to identify the current search position: OWNER(ownerid). ownerid is one to
eight characters enclosed in single quotation marks if it contains any special characters, or
blank.
The information required to continue a search subcommand is returned by each search
subcommand that specifies the CONTINUE operand and must be passed back to
DFSMSrmm unchanged in order to continue the previous search. You should specify the
exact same subcommand unchanged. To do this, just change the CONTINUE operand
value on each additional command required.
END(end_owner)
Specify END as an alternative to the LIMIT operand to enable you to specify both the
starting and ending point of the owner search. END is mutually exclusive with LIMIT.
LIMIT(search_limit | *)
Specifies the number of entries that DFSMSrmm lists. The maximum allowable decimal
value is 9999. Specify an asterisk to request a list of all entries matching your search
criteria. LIMIT is mutually exclusive with END. The default value is 10.
LIST
Specifies that DFSMSrmm produce a list when the CLIST operand is used. LIST is
mutually exclusive with the NOLIST operand. LIST is the default.
NOLIST
Specifies that DFSMSrmm not produce a list when the CLIST operand is used.
DFSMSrmm produces only the CLIST output file. NOLIST is mutually exclusive with the
LIST operand. LIST is the default.

300
z/OS V1R8 DFSMS Technical Update
OWNER(full_or_generic_owner | *)
Specifies an owner ID. DFSMSrmm only lists volumes belonging to the owner ID that you
specify. Specify a specific owner ID to list volumes belonging to that owner. Specify an
asterisk to list all volumes that match the other search criteria regardless of their owner.
An owner ID is one-to-eight alphanumeric characters or to six alphanumeric characters, $,
#, or @. The first character must not be a number. The default is your TSO user ID.
START
Specify this operand to request that records written to the CLIST data set start from the
beginning of the data set. START is mutually exclusive with ADD. START is the default
value.
Examples
Use the RMM TSO SEARCHOWNER subcommand, as shown in Example 8-13, to create a
list of all owners defined to DFSMSrmm.
Example 8-13 List all owners defined to DFSMSrmm
TSO RMM SO OWNER(*) LIMIT(*)
or
TSO RMM SEARCHOWNER OWNER(*) LIMIT(*)
DFSMSrmm displays a list, such as the one shown in Figure 8-34.
Figure 8-34 DFSMSrmm search owner result
8.5.6 Rexx variable constraint relief
A new operand, VARSTORAGE, was added to the PROFILE command to specify whether
variables in the CLIST or authorized REXX variable pools can use storage above the 16 MB
line.
There are no changes to RMM for this support.
If you have a large environment and you have a need to list a big number of volumes,
normally you get a return code 4 reason code 10 Insufficient storage for search
processing. More records might exist is avoided.
Owner Last Name First Names Internal Volumes
-------- -------------------- -------------------- -------- ----------
HAIMO Haimowitz Bob 349-5456 1
HSM STC DFSMShsm 0
MARY Lovlace Mary 475-3231 0
MHLRES1 Fletcher Anthony 0
MHLRES2 Weisshaar Gerhard 1
MHLRES2D 0
MHLRES2E 0
MHLRES3 Perkin Daniel 0
MHLRES4 Coelho Andre 1
MHLRES5 Schlumberger Norbert 119-3579 1286
PAOLOR2 0
PAOLOR3 1
STC 24
VAINI 0
14 ENTRIES LISTED

Chapter 8. DFSMSrmm enhancements
301
If your TSO LOGON session parameter size is 30,000, then approximately 2,400 volumes are
retrieved with the setting VARSTORAGE(LOW). If you change your settings to
VARSTORAGE(HIGH) you can retrieve approximately 20,000 volumes.
VARSTORAGE specifies the storage location to be used for CLIST variables or REXX
OUTTRAP variables containing output from authorized commands. A CLIST or REXX exec
uses the VARSTORAGE setting of the PROFILE command when the it starts. This setting
then remains unchanged for the life of the CLIST or REXX exec, even if the CLIST or REXX
exec issues a new PROFILE command with a different VARSTORAGE setting. The new
setting only applies when a new CLIST or REXX exec begins.
VARSTORAGE (HIGH) Indicates that CLIST variables and REXX OUTTRAP variables
containing output from authorized commands invoked by REXX
can be kept in storage above the 16 M line.
VARSTORAGE (LOW) Indicates that CLIST variables and REXX OUTTRAP variables
containing output from authorized commands invoked by REXX
can only be kept in storage below the 16 M line. If you specify
VARSTORAGE with no operands, VARSTORAGE(LOW) is the
default. This is the default value when your user profile is created.
To show your current setting use the TSO PROFILE command, as shown in Figure 8-35,
without any additional operands.
Figure 8-35 Display your current profile settings
Figure 8-36 shows that your current profile settings include the new VARSTORAGE
information. If you have not set the VARSTORAGE, you can see that the default setting LOW
is set.
Figure 8-36 Your current profile settings
TSO PROF
CHAR(0) LINE(0) PROMPT INTERCOM NOPAUSE NOMSGID MODE WTPMSG
NORECOER PREFIX(MHLRES5) PLANGUAGE(ENU) SLANGUAGE(ENU) VARSTORAGE(LOW)
DEFAULT LINE/CHARACTER DELETE CHARACTERS IN EFFECT FOR THIS TERMINAL

302
z/OS V1R8 DFSMS Technical Update
If you have requested to search for all volumes in your DFSMSrmm database, normally you
see the message more volumes exists on the panel, as shown in Figure 8-37, and the
search ended before all volumes were listed.
Figure 8-37 Display your incomplete search result
To update this value use the TSO PROFILE command, as shown in Figure 8-38.
Figure 8-38 Update your TSO profile settings
Figure 8-39 shows you your new profile settings include the new VARSTORAGE information.
Figure 8-39 Your new profile settings
Panel Help Scroll
______________________________________________________________________________
DFSMSrmm Volumes (Page 1 of 2) More volumes may exist
Command ===> Scroll ===> CSR
There is not enough storage available to list all the volumes
Enter HELP or PF1 for the list of available line commands
Use the RIGHT command to view other data columns
Volume Assigned Expiration Dest- Tra- Data
S serial Owner date date Status Location ination nsit sets
-- ------ -------- ---------- ---------- ------- -------- -------- ---- -----
AFS201 VGRMMLIB 2004/303 1999/365 USER SHELF N 0
AFS202 VGRMMLIB 2004/303 1999/365 USER SHELF N 0
BTS001 VGRMMLIB 2005/005 1999/365 USER SHELF N 0
BTS005 VGRMMLIB 2005/005 1999/365 USER SHELF N 0
BTS006 VGRMMLIB 2005/005 1999/365 USER SHELF N 0
BTS007 VGRMMLIB 2005/005 1999/365 USER SHELF N 0
BTS008 VGRMMLIB 2005/005 1999/365 USER SHELF N 0
BTS010 VGRMMLIB 2005/005 1999/365 USER SHELF N 0
BTS011 VGRMMLIB 2005/005 1999/365 USER SHELF N 0
BTS012 VGRMMLIB 2005/005 1999/365 USER SHELF N 0
TSO PROFILE VARSTORAGE(HIGH)
CHAR(0) LINE(0) PROMPT INTERCOM NOPAUSE NOMSGID MODE WTPMSG
NORECOER PREFIX(MHLRES5) PLANGUAGE(ENU) SLANGUAGE(ENU) VARSTORAGE(HIGH)
DEFAULT LINE/CHARACTER DELETE CHARACTERS IN EFFECT FOR THIS TERMINAL

Chapter 8. DFSMSrmm enhancements
303
After the update of your profile settings you should no longer get the more volumes exists
message, except you have a really large environment. Figure 8-40 shows you the search
results with 19.713 entries in the list.
Figure 8-40 Display your search result with up to 20.000 entries
8.6 Enabling ISPF Data Set List (DSLIST) support
To enable direct entry into the DFSMSrmm ISPF dialog from the ISPF Data Set List Utility,
use the ISPF Configuration Utility to update the ISPF Configuration Table. To enable this
function, select the Enable RM/Tape Commands option. For details of how to use the ISPF
Configuration Utility, refer to z/OS ISPF Planning and Customizing, GC34-4814. Figure 8-41
on page 304 shows that the data set list support is enabled, and also shows the default
values for the RM/Tape Command EDGRPD34 and Command APPLID EDG. You do not
need to change these values.
8.6.1 Implementation steps
The ISPF Data Set List support can be implemented for a single user or it can be a
installation-wide setting. To implement ISPF Data Set List support for a single user:
1.Create a new PDS with VB 255 to store the ISPF default settings.
2.Under ISPF, call 'TSO ISPCCONF' in any command line.
3.Select option 1, Create/Modify Settings and Regenerate Keyword File, in the ISPF
Configuration Utility primary panel, and added the name to the keyword file dsname field
that you have allocated before, and added a member name.
You should get the message Keyword file loaded.
4.Select option 3, PDF Exits and Other PDF Settings” in the ISPF Configuration Utility
primary panel.
5.Scroll to the correct place and add a slash (/) to Enable RM/Tape Commands.
Panel Help Scroll
______________________________________________________________________________
DFSMSrmm Volumes (Page 1 of 2) Row 1 to 33 of 19,713
Command ===> Scroll ===> CSR

Enter HELP or PF1 for the list of available line commands
Use the RIGHT command to view other data columns
Volume Assigned Expiration Dest- Tra- Data
S serial Owner date date Status Location ination nsit sets
-- ------ -------- ---------- ---------- ------- -------- -------- ---- -----
AFS201 VGRMMLIB 2004/303 1999/365 USER SHELF N 0
AFS202 VGRMMLIB 2004/303 1999/365 USER SHELF N 0
BTS001 VGRMMLIB 2005/005 1999/365 USER SHELF N 0
BTS005 VGRMMLIB 2005/005 1999/365 USER SHELF N 0
BTS006 VGRMMLIB 2005/005 1999/365 USER SHELF N 0
BTS007 VGRMMLIB 2005/005 1999/365 USER SHELF N 0
BTS008 VGRMMLIB 2005/005 1999/365 USER SHELF N 0
BTS010 VGRMMLIB 2005/005 1999/365 USER SHELF N 0
BTS011 VGRMMLIB 2005/005 1999/365 USER SHELF N 0
BTS012 VGRMMLIB 2005/005 1999/365 USER SHELF N 0

304
z/OS V1R8 DFSMS Technical Update
6.Exit the dialog by pressing the PF3 key tow times. You should get the message Keyword
file saved.
7.Select option 4, Build Configuration Table Load Module.
8.Specify any library in your ISPLLIB concationation as the
Output Configuration Table
Load Module Data Set
.
9.Enter to load the module built.
10. Exit ISPF. Ensure that the specified load library is ISPLLIB concatenated.
11. Start ISPF and it works (just for a single user, not for all users).
To implementing the ISPF Data Set List support for all users, follow the steps as described
before, but the configuration table load module should be stored in the SISPLPA library. To
load the module to the LPA you must IPL your system. For detailed information refer to 8.6.4,
“Move the ISPCFIGU module to the SISPLPA library (optional)” on page 313.
8.6.2 Use the ISPF Configuration Utility
Use the ISPF Configuration table to change site-wide defaults and to indicate that installation
exit routines are provided for some of the ISPF functions. The ISPF functions that allow
installation-written exit routines are data set allocation, print utility, data set compression, data
set list utility, member list filter, and data set name change. ISPF checks the configuration
table to determine, first, whether exit routines are provided, and second, whether those
routines are programs or CLISTs. If you specify both a CLIST and a program, ISPF uses the
program.
Enter the command TSO ISPCCONF to start the ISPF Configuration Utility, as shown in
Figure 8-41. You can chose the command in the ISPF command shell or in any available
command line.
Figure 8-41 Start the ISPF conversion utility
Menu Utilities Compilers Options Status Help
______________________________________________________________________________
z/OS Primary Option Menu
Option ===> TSO ISPCCONF

0 Settings Terminal and user parameters User ID . : SCHLUM
1 View Display source data or listings Time. . . : 18:40
2 Edit Create or change source data Terminal. : 3278
3 Utilities Perform utility functions Screen. . : 1
4 Foreground Interactive language processing Language. : ENGLISH
5 Batch Submit job for language processing Appl ID . : PDF
6 Command Enter TSO or Workstation commands TSO logon : IKJACCT
7 Dialog Test Perform dialog testing TSO prefix: MHLRES5
9 IBM Products IBM program development products System ID : SC70
10 SCLM SW Configuration Library Manager MVS acct. : ACCNT#
11 Workplace ISPF Object/Action Workplace Release . : ISPF 5.9
12 z/OS System z/OS system programmer applications
13 z/OS User z/OS user applications

Enter X to Terminate using log/list defaults

Chapter 8. DFSMSrmm enhancements
305
Refer to Figure 8-42 to see how you can select the Create/Modify Settings and Regenerate
Keyword File after the ISPF conversion utility is up and running.
Figure 8-42 Create/Modify Settings and Regenerate Keyword File selection
ISPF Configuration Utility Enter option
______________________________________________________________________________
Option ===> 1

1 Create/Modify Settings and Regenerate Keyword File
2 Edit Keyword File Configuration Table
3 Verify Keyword Table Contents
4 Build Configuration Table Load Module
5 Convert Assembler Configuration Table to Keyword File
6 Build SMP/E USERMOD


Keyword File Data Set
Data Set . . . 'RMM.ADDONS.CEXECVB'
Member . . . . SETTING

Configuration Table Assembler Source Data Set
Data Set . . .
Member . . . .

Output File Content for Keyword File
2 1. Include only non-default values
2. Include defaults as comments
3. Include all values


Current Configuration Table
Keyword File : not available
Identifier . : ISPCFIGU Level . . . : 480R8001
Compile Date : 2003/04/04 Compile Time :

306
z/OS V1R8 DFSMS Technical Update
After the keyword file is loaded select the PDF Exits and Other PDF Settings menu, as
shown in Figure 8-43.

Figure 8-43 PDF Exits and Other PDF Settings selection
Create/Modify ISPF Configuration Keyword file loaded
______________________________________________________________________________
Option ===> 3

General ISPF Settings System Profile (ISPSPROF) Settings
1 Editor Settings 6 Log and List Defaults
2 Edit/View/Browse VSAM Settings 7 Terminal and User Defaults
3 PDF Exits and Other PDF Settings 8 Workstation Defaults
4 ISPF Site-wide Defaults 9 Workstation Download Defaults
5 ISPDFLTS, CUA Colors, and Other
DM Settings


Output Keyword File
Data Set . . . 'RMM.ADDONS.CEXECVB'
Member . . . . SETTING


Instructions:
Enter option to change configuration settings,
END or EXIT command to generate keyword file, or
CANCEL command to exit without keyword file generation

Chapter 8. DFSMSrmm enhancements
307
The Modify PDF Configuration Settings panel is displayed in a srollable panel and you have
scroll down until you can see DSLIST Removable Media Settings, as shown in Figure 8-44.
Figure 8-44 Modify PDF Configuration Settings
Modify PDF Configuration Settings
______________________________________________________________________________
Command ===>
More: -
2. Use IEBCOPY for PDSEs only

When to use COPY or COPYMOD
2 1. Use COPY if the target block size is equal to or greater than the
source block size, COPYMOD otherwise
2. Use COPY if the target block size is equal to the source block size,
COPYMOD otherwise
3. Always use COPYMOD

DSLIST Removable Media Settings
Enter "/" to select option
/ Enable RM/Tape Commands

RM/Tape Command . . %EDGRPD34
Command APPLID . . . EDG

Other PDF Settings
Default PDF Unit . . . . . . . . . . . . SYSALLDA
Volume for Migrated Data Sets . . . . . MIGRAT
Delete Command for Migrated Data Sets HDELETE
Allowed Allocation Units . . . . . . . . ANY
Maximum IEBCOPY Return Code . . . . . . 0
Pathname Substitution Character . . . . Ü

Enter "/" to select option
Allocate Before Uncatalog
/ Verify Expiration Dates
/ Use SuperC Program Interface
Monitor Edit Macro Commands via the Activity Monitoring Exit
/ Allow SUBMIT from Browse
/ Allow SUBMIT from View
/ Warn when rename target could be a GDG
/ Default Edit/Browse/View member list from Option 3.4
/ Enable View
Use Panel ISRTSOA in Option 6
Print using ICF
Disallow wildcards in the high level qualifier for Data Set List
Disable all ENQ displays
/ Fail on LMF lock requests

308
z/OS V1R8 DFSMS Technical Update
After you have finished your updates you have to verify your new keyword table using
selection 3, in then you have to build a new configuration table using selection 4, as you can
see in Figure 8-45.
Figure 8-45 Build Configuration Table Load Module
ISPF Configuration Utility
________________________________________________________________________
Option ===> 4

1 Create/Modify Settings and Regenerate Keyword File
2 Edit Keyword File Configuration Table
3 Verify Keyword Table Contents
4 Build Configuration Table Load Module
5 Convert Assembler Configuration Table to Keyword File
6 Build SMP/E USERMOD


Keyword File Data Set
Data Set . . . 'RMM.ADDONS.CEXECVB'
Member . . . . SETTING

Configuration Table Assembler Source Data Set
Data Set . . .
Member . . . .

Output File Content for Keyword File
2 1. Include only non-default values
2. Include defaults as comments
3. Include all values


Current Configuration Table
Keyword File : not available
Identifier . : ISPCFIGU Level . . . : 480R8001
Compile Date : 2003/04/04 Compile Time :


Chapter 8. DFSMSrmm enhancements
309
A small pop-up window is dispayed to specifiy the input keyword file data set and member
and the output configuration table load module data set.
Figure 8-46 Build Configuration Table Load Module pop-up window
Is your target load library is a concatenated library of your LINKLIST you have to refresh the
the LINKLIST, as shown in Example 8-14.
Example 8-14 Using the LLA refresh command
F LLA,REFRESH
Figure 8-47 shows you the successful refresh of your linklist.
Figure 8-47 Successful refresh of linklist
After you have created the new configuration table load module ensure that the specified load
library is concatenated to your ISPLLIB, exit ISPF, and start ISPF again.
EsssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssN
e Build Configuration Table Load Module e
e Command ===> e
e e
e Input Keyword File Data set e
e Data Set . . . 'RMM.ADDONS.CEXECVB' e
e Member . . . . SETTING e
e e
e Output Configuration Table Load Module Data Set e
e Data Set . . . 'SCHLUM.ISPF.LLIB' e
e e
e Optional fields (leave blank for ISPF to use defaults) e
e Object data set . . . e
e Configuration member (Defaults to ISPCFIGU) e
e VSAM member . . . . . (Defaults to ISPCFIGV) e
DsssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssM
CSV210I LIBRARY LOOKASIDE REFRESHED
Important: If the DFSMSrmm REXX exec library, normally called SYS1.SEDGEXE1, is
not concatenated in your logon procedure, you must move the two members EDGRMLIB
and EDGRPD34 to a concatenated REXX Exec library and update member EDGRPD34 to
specify that the RMM ISPF environment is allocated:
UseIspfLibdef = true /* << true or false */

310
z/OS V1R8 DFSMS Technical Update
8.6.3 Using the ISPF Data Set List Utility support
After you have enabled the ISPF Data Set List (DSLIST) support and restarted your ISPF
session, you can use this support. The line commands supported by DFSMSrmm are I, S, M,
and D:
I Displays a search results list showing all data sets in the multivolume set for the
selected data set.
S Displays the individual data set details. DFSMSrmm determines the first file on the
selected volume that matches the selected data set. If other data sets of the same
name exist on the volume, the wrong details may be displayed. In that case, use the M
line command and then the DFSMSrmm I line command from that results list.
M Displays a search results list showing all data sets defined to DFSMSrmm that match
the selected data set name.
D Releases the volume. If the volume is part of a multivolume set, there is the option to
release all volumes in the set.
Figure 8-48 shows you the use of the display of individual data set details (S).
Figure 8-48 Use of the display of individual data set details
Menu Options View Utilities Compilers Help
______________________________________________________________________________
DSLIST - Data Sets Matching SCHLUM.LI* Row 1 of 5
Command ===> Scroll ===> PAGE

Command - Enter "/" to select action Message Volume
-------------------------------------------------------------------------------
SCHLUM.LIBDUMP SBOX03
SCHLUM.LIBDUMP.BIN.XMIT SBOXFG
SCHLUM.LINEITEM.PUNCH SBOX81
I SCHLUM.LINEITEM.UNLOAD Info - I TST026
SCHLUM.LISTDEF.INPUT SBOX09
***************************** End of Data Set list ****************************

Chapter 8. DFSMSrmm enhancements
311
The result is shown in Figure 8-49.
Figure 8-49 Result of display of individual data set details
Panel Help
____________________________________________________________________________
DFSMSrmm Data Set Details
Command ===>

Data set name . . : 'SCHLUM.LINEITEM.UNLOAD'
Volume serial . . : TST026 Physical file sequence number . . : 1
Owner . . . . . . : SCHLUM Data set sequence number . . . . : 1
More:
Job name . . . . : SCHLUMR3U
Step name . . . . : STEP010 Record format . . . . : VB
Program name . . : DSNUTILB Block size . . . . . : 32760
DD name . . . . . : SYSREC Logical record length : 151
Create date . . . : 2003/300 YYYY/DDD Block count . . . . . : 36140
Create time . . . : 14:51:16 Total block count . . : 36140
Expiration date . : 2003/300 YYYY/DDD Percent of volume . . : 2
Original . . . . : YYYY/DDD Device number . . . . : 0B92
System id . . . . : SC63

Last job name . . : PAOLOR3T Last DD name . . . . : TAPEVOL
Last step name . : STEP010 Last device number . : 0B92
Last program name : IDCAMS
Date last read . : 2003/302 VRS management value :
Date last written : 2003/300 Management class . . : MCDB22
Data class . . . . . :
Retention date . : Storage class . . . . : SCLIB1
VRS retained . . : NO Storage group . . . . : SGLIB1

Security name . . :
Classification . :

Primary VRS details:
VRS name . . . :
Job name . . . : VRS type . . . . . :
Subchain name : Subchain start date :

Secondary VRS details:
Value or class :
Job name . . . :
Subchain name : Subchain start date :

Catalog status . : YES

312
z/OS V1R8 DFSMS Technical Update
Figure 8-50 shows all data sets defined to DFSMSrmm that match the selected data set
name using function M.
Figure 8-50 Result if you are selecting function M
Use selection D to releases the volume, as shown in the data set. If you select this function
DFSMSrmm shows the confirm volume release panel, as shown in Figure 8-51.
Figure 8-51 Result if you are selecting function release the volume
Panel Help Scroll
______________________________________________________________________________
DFSMSrmm Data Sets (Page 1 of 2) Row 1 to 1 of 1
Command ===> Scroll ===> PAGE


Enter HELP or PF1 for the list of available line commands
Use the RIGHT command to view other data columns
Volume File
S Data set name serial Owner seq
-- -------------------------------------------- ------ -------- -----
SCHLUM.LINEITEM.UNLOAD TST026 SCHLUM PAOLOR3
******************************* Bottom of data ********************************
Panel Help
______________________________________________________________________________
DFSMSrmm Confirm Volume Release
Command ===>

Volume . . . . . . : TST026 Location . . . . . . . . : LIB1
VOL1 volser . . : In container . . . . . . :
Volume type . . . : PHYSICAL Rack number . . . . . . : TST026
Media name . . . . : MEDIA3 Expiration date . . . . : 2003/365

Original expiration date :
Status . . . . . . : USER
Description . . . :
Data set name . . : 'SCHLUM.LINEITEM.UNLOAD'
Release actions:
Media type . . . . : HPCT Return to SCRATCH pool : NO
Label . . . . . . : SL Replace volume . . . . : NO
Density . . . . . : IDRC Return to owner . . . : YES
Recording format . : 128TRACK Initialize volume . . : NO
Compaction . . . . : YES Erase volume . . . . . : NO
Attributes . . . . : NONE Notify owner . . . . . : NO
Availability . : PENDING RELEASE Expiry date ignore . . : NO
Scratch immediate . . : NO
Press ENTER to NOFORCE volume, or END command to CANCEL.


Chapter 8. DFSMSrmm enhancements
313
Figure 8-52 displays a search result if you have selected function I that includes all volumes
residing on the same tape or all data sets residing on the multi-volume set.
Figure 8-52 SEARCH volume result
8.6.4 Move the ISPCFIGU module to the SISPLPA library (optional)
After you have installed and verified ISPF, you can enhance its performance by adding the
LPA-eligible load modules (in the SISPLPA library) to the LPA list in an LPALSTxx member of
PARMLIB. Add those load modules not eligible for LPA (in the SISPLOAD library) to the link
list in an LNKLSTxx member of PARMLIB. You can then remove these data sets from the
STEPLIB in your TSO LOGON procedure. After adding SISPLPA to LPALST and SISPLOAD
to LNKLST, specify CLPA as an initial program load (IPL) parameter to force the SISPLPA
modules into the link pack area and to have SISPLOAD added as a system link library.
Finally, after you tested all, you can move the ISPCFIGU member to your ISP. Re-IPL your
system to refresh your linklist, as shown in Example 8-15, and then you must re-log in on to
your system before you can use this new service.
Example 8-15 Using the LLA refresh command
F LLA,REFRESH
Figure 8-53 shows you the successful refresh of your linklist.
Figure 8-53 Result of the LLA REFRESH
8.7 Prepare for future releases
In the following sections are suggestions for preparing for future DFSMSrmm releases.
Panel Help Scroll
______________________________________________________________________________
DFSMSrmm Data Sets (Page 1 of 2) Row 1 to 10 of 10
Command ===> Scroll ===> PAGE


Enter HELP or PF1 for the list of available line commands
Use the RIGHT command to view other data columns
Volume File
S Data set name serial Owner seq
-- -------------------------------------------- ------ -------- -----
SCHLUM.LINEITEM.UNLOAD TST026 MHLRES5 1
SCHLUM.TESTSTAC.TESTMF02 TST026 MHLRES5 2
SCHLUM.TESTSTAC.TESTMF03 TST026 MHLRES5 3
SCHLUM.TESTSTAC.TESTMF04 TST026 MHLRES5 4
SCHLUM.TESTSTAC.TESTMF05 TST026 MHLRES5 5
SCHLUM.TESTSTAC.TESTMF06 TST026 MHLRES5 6
SCHLUM.TESTSTAC.TESTMF07 TST026 MHLRES5 7
SCHLUM.TESTSTAC.TESTMF08 TST026 MHLRES5 8
SCHLUM.TESTSTAC.TESTMF09 TST026 MHLRES5 9
SCHLUM.TESTSTAC.TESTMF10 TST026 MHLRES5 10
******************************* Bottom of data ********************************
CSV210I LIBRARY LOOKASIDE REFRESHED

314
z/OS V1R8 DFSMS Technical Update
8.7.1 Set a DFSMSrmm control data set ID
Specify one to eight alphanumeric characters that identify the control data set by name,
because a CDSID is required in future DFSMSrmm releases. To update your DFSMSrmm
control data set control record use the job as shown in Example 8-16.
Example 8-16 Specifying a CDSID
//UTIL EXEC PGM=EDGUTIL,PARM=’UPDATE’
//SYSPRINT DD SYSOUT=*
//MASTER DD DISP=SHR,DSN=RMM.CONTROL.DSET
//SYSIN DD * D DDDDD
CONTROL CDSID(PROD)
/*
After you have specified a CDSID in the DFSMSrmm control data set control record you must
update your EDGRMMnn member in your corresponding PARMLIB data set, as shown in
Example 8-17.
Example 8-17 Sample EDGRMMnn PARMLIB member
OPTION OPMODE(R) /* Record-Only Mode */ -
ACCOUNTING(J) /* Accounting from JOB */ -
BACKUPPROC(EDGBKUP) /* Name of BACKUP-proc */ -
BLP(RMM) /* DFSMSrmm controls BLP */ -
CATRETPD(0012) /* catalog retention */ -
CATSYSID(*) /* all catalogs shared */ -
CDSID(PROD) /* control data set id */ -
COMMANDAUTH(OWNER) /* type of authorization */ -
DATEFORM(J) /* Date format */ -
DISPDDNAME(LOANDD) /* DISP ctrl DD card */ -
DISPMSGID(EDG4054I) /* DISP message number */ -
DSNAME(RMM.PROD.CDS) /* CDS data set name */ -
IPLDATE(N) /* IPL date checking */ -
JRNLNAME(RMM.PROD.JRNL) /* JRNL data set name */ -
JOURNALFULL(75) /* Percentage JRNL full */ -
LINECOUNT(054) /* Lines per page */ -
MASTEROVERWRITE(LAST) /* Overwriting of a vol */ -
MAXHOLD(100) /* Number of I/O oper. */ -
MAXRETPD(NOLIMIT) /* Maximum retention */ -
MEDIANAME(3480) /* spec. how to move vols */ -
MOVEBY(VOLUME) /* spec. how to move vols */ -
MSG(M) /* case for message txt */ -
NOTIFY(Y) /* Notify volume owners */ -
PDA(OFF) /* PDA is disabled */ -
PDABLKCT(255) /* number of blocks */ -
PDABLKSZ(31) /* blocksize in K */ -
PDALOG(OFF) /* PDA output disabled */ -
PREACS(NO) /* Disable EDGUX100 ACS pr. */ -
RETAINBY(VOLUME) /* spec. how to retain vols */ -
RETPD(0005) /* Default retention */ -
REUSEBIN(STARTMOVE) /* reuse BIN as soon as pos.*/ -
SCRATCHPROC(EDGXPROC) /* ATL/MTL procedure */ -
SMFAUD(248) /* SMF audit records */ -
SMFSEC(249) /* SMF security records */ -
SMSACS(YES) /* enable MV ACS processing */ -

Chapter 8. DFSMSrmm enhancements
315
SMSTAPE(UPDATE(EXITS,SCRATCH,COMMAND),PURGE(YES)) ATL*/ -
SYSID(EGZB) /* Name of the system */ -
TPRACF(N) /* RACF tape support */ -
TVEXTPURGE(EXPIRE) /* set an expiration date */ -
UNCATALOG(N) /* Catalog support */ -
VRSCHANGE(INFO) /* No additional action */ -
VRSEL(NEW) /* New VRS processing */ -
VRSJOBNAME(2) /* DATASETNAME/JOBNAME */ -
VRSMIN(0000000100,WARN) /* Warn if < 100 VRSs */

If the CDSID was not previously set you get message EDG0237E MISSING IDENTIFIER
FOR THE CONTROL DATA SET. See the description below.
EDG0237E MISSING IDENTIFIER FOR THE CONTROL DATA SET.
Where:
Severity
Information.
Explanation
During initialization DFSMSrmm checks to ensure that you have a CDSID specified in the
EDGRMMxx PARMLIB member. A CDSID is mandatory.
Source
DFSMSrmm.
Detecting Module
EDGPARM.
System Action
DFSMSrmm initialization stops. This message is followed by message EDG0107A.
Operators Response
Notify the system programmer. Reply to message EDG0107A as directed when the
system programmer has corrected the error.
Application Programmer Response
You must add a CDSID to the OPTION statement in EDGRMMxx before attempting to
start DFSMSrmm on a release at z/OS R9 or later. Correct the error in the startup
parameters.
Routing Codes
3.
Descriptor Codes
3.
8.7.2 Re-allocate your DFSMSrmm control data set
Re-allocate your existing DFSMSrmm control data set and use a control interval size (CISZ)
of 26624 bytes and BUFFERSPACE of 829440 bytes for the data component of the control
data set to help improve inventory management run times through reduced I/O to the control
data set. Any suitable control interval size (CISZ) between 10240 and 26624 that meets your
needs can be used.

316
z/OS V1R8 DFSMS Technical Update
The JCL to allocate the DFSMSrmm control data set in Example 8-18 has been modified to
use the recommended values to allocate the VSAM KSDS correctly.
Example 8-18 Sample JCL to allocate the DFSMSrmm CDS
//LCLCDSAL JOB ,140.SCHLUMBERGER,NOTIFY=SCHLUM,
// MSGCLASS=H,CLASS=A,MSGLEVEL=(1,1)
//* ****************************************************************** *
//* STEP 1 Delete old DFSMSrmm control data set *
//* 2 Allocate a new DFSMSrmm control data set *
//* ****************************************************************** *
//STEP01 EXEC PGM=IDCAMS
//SYSPRINT DD SYSOUT=*
//SYSIN DD *
DELETE RMM.PROD.CDS CL
SET MAXCC = 0
/*
//STEP02 EXEC PGM=IDCAMS ALLOCATION MASTER FILE
//SYSPRINT DD SYSOUT=*
//MASTER DD DISP=SHR,UNIT=3390,VOL=SER=DFRMM4
//SYSIN DD *
DEFINE CLUSTER(NAME(RMM.CONTROL.DSET) -
FILE(MASTER) -
FREESPACE(15 0) -
KEYS(56 0) -
REUSE -
RECSZ(512 9216) -
SHR(3 3) -
KILOBYTES(4500 1500) -
STORAGECLASS(gspace) -
VOLUMES(DFRMM4)) -
DATA(NAME(RMM.CONTROL.DSET.DATA) -
BUFFERSPACE(829440) -
CISZ(26624)) -
INDEX(NAME(RMM.CONTROL.DSET.INDEX) -
CISZ(2048))
/*
After you have changed tour VSAM KSDS attributes for the DFSMSrmm control data set you
should change the region size parameters in your housekeeping JCL and in the DFRMM
procedure as well.
REGION
As you determine the REGION size for the DFSMSrmm started procedure, the amount of
virtual storage that DFSMSrmm uses depends on the resources that you have defined.
DFSMSrmm virtual storage usage can be affected by any REGION size controls or
restrictions that your systems might have in place such as in IEFUSI. The sample DFRMM
procedure specifies REGION=40M, which normally provides all the private region below 16
MB and 40 MB above 16 MB. To enable DFSMSrmm to use all available virtual storage,
specify REGION=0M. If you want to set a specific region size, consider the following tips
along with the current region size of your DFRMM started procedure, to determine whether
you need to make any changes to the REGION size.
Important: Usa a REGION=0M to use the maximum available region size on your system
or specify a minimum of REGION=40MB to speed up your daily housekeeping.

Chapter 8. DFSMSrmm enhancements
317
The VSAM local shared resources (LSR) buffer pool that is built by the DFSMSrmm
subsystem for the control data set is obtained above 16 MB. DFSMSrmm builds an LSR
buffer pool for the DFRMM started procedure, and also for the EDGUTIL utility batch address
space, which has a predetermined size. The LSR buffer pool is 800*data CISZ + 200*index
CISZ. Assuming 10 240 for the control data set data CISZ and 2 048 for the control data set
index CISZ, the value is 800*10 240+200*2 048=8.4 MB. If you use larger CI sizes, more
buffer space is required. For example, if you use a 26K data CISZ, a 21.2 MB buffer size is
required.
Here is an example of how you can resolve a S878 abend. As the buffer space increased, an
increase of the REGION size resolved the S878 abend. Since the DATA CISZ was increased
from 10240 to 26624, the buffer space increased from 8.6 MB to 21.2 MB. That is why
HSKPing worked with a REGION size of 40 MB before the CISZ change. The difference of
21.2 and 8.6 requires a REGION size of 52.6 MB, and you have to specify a REGION size of
60 MB.
8.7.3 Update LRECL for REPORT, BACKUP, and JRNLBKUP DD
Update all your JCL to use the new recommended LRECL for the Report, BACKUP, and
JRNLBKUP DD statements.
Backing up the control data set
Use the sample JCL in Example 8-19 to get no error until you are backing up the DFSMSrmm
control data set. The LRECL of the BACKUP file has be expanded to be increased to 9216
and 9248 bytes.
Example 8-19 Sample JCL to allocate the backup files
//EDGHSKP EXEC PGM=EDGHSKP,PARM=’BACKUP(DSS)’
//MESSAGE DD DISP=SHR,DSN=RMM.MESSAGE
//SYSPRINT DD SYSOUT=*
//BACKUP DD DISP=(,CATLG),UNIT=TAPE,DSN=BACKUP.CDS(+1),
// LABEL=(,SL)
// AVGREC=U,LRECL=9216,BLKSIZE=0,RECFM=U
//JRNLBKUP DD DISP=(,CATLG),UNIT=TAPE,DSN=BACKUP.JRNL(+1),
// DCB=(RECFM=VB,BLKSIZE=0,LRECL=9248),
// LABEL=(2,SL),
// VOL=REF=*.BACKUP
//DSSOPT DD *
CONCURRENT OPTIMIZE(1) VALIDATE
/*
Recommendation: If the buffer space is larger than 8.6 MB, add the difference to the
40 MB region size that is used by DFSMSrmm and use this value as the REGION size for
the DFRMM started procedure.
Note: Use this blocksize if you are using DSS or AMS to back up the DFSMSrmm control
data set and journal.

318
z/OS V1R8 DFSMS Technical Update
Running inventory management
Example 8-20 shows you the expanded logical record length to 255 bytes for the inventory
management data set REPORT DD to minimize the risk of having some of the information in
a next line.
Example 8-20 Sample JCL to allocate the REPORT file
//* *************************************************************
//STEP04 EXEC PGM=IEFBR14
//DD1 DD DSN=RMM.HSKP.VRSEL(+1),
// SPACE=(TRK,(300,30)),
// DISP=(,CATLG,DELETE),LRECL=255
Figure 8-54 shows you an example of how the inventory management report file looks if you
have specified a data set with a LRECL with 137 or fewer bytes.
Figure 8-54 REPORT file with a small LRECL
REMOVABLE MEDIA MANAGER VITAL RECORDS RETENTION REPOR
(C) IBM CORPORATION 1993,1998 ----- ------- --------- -----
JOB MASK DATA SET OR VOLUME MASK OWNER TYPE RETN C X
________________________________________________________________________________
BSYSMF.WEEK.** STCOPC DSN CYCLES Y N
JOB NAME DATA SET NAME 2ndVRS 2ndNAME FSEQ DSEQ VOLUME VSE
________________________________________________________________________________
SMFWEEK2 BSYSMF.WEEK.G0552V00 1 1 Q17058
CONT:-
SMFWEEK2 BSYSMF.WEEK.G0551V00 1 1 Q17191
CONT:-
SMFWEEK2 BSYSMF.WEEK.G0550V00 1 1 Q17084
CONT:-
SMFWEEK2 BSYSMF.WEEK.G0549V00 1 1 Q17357
CONT:-
SMFWEEK2 BSYSMF.WEEK.G0548V00 1 1 Q17287
CONT:-
SMFWEEK2 BSYSMF.WEEK.G0547V00 1 1 Q17356
CONT:-
SMFWEEK2 BSYSMF.WEEK.G0546V00 1 1 Q17321
CONT:-
SMFWEEK2 BSYSMF.WEEK.G0545V00 1 1 Q17068
CONT:-

Chapter 8. DFSMSrmm enhancements
319
In Figure 8-55 you can see the differences if you are using a larger logical record length to
allocate the REPORT file. In this case there are no lines starting with CONT:- included in the
report.
Figure 8-55 REPORT file with a enlarged LRECL
8.7.4 Migrate from VRSEL(OLD) to VRSEL(NEW)
Because the VRSEL(OLD) option is being removed in a future release, and there is a new
warning message when you run VRSEL processing on z/OS V1R8, we recommend that you
migrate from VRSEL(OLD) to VRSEL(NEW) before moving to z/OS V1R8. If you do not
perform this migration you get a warning message (EDG2221E) each time VRSEL is run and
EDGHSKP processing ends with job step return code 4.
Any data sets that match to an incomplete chain are retained by a special broken vital record
specification. The special broken vital record specification uses the name *broken* and is
listed in the REPORT and ACTIVITY files and in the data set matching vital record
specification information. With VRSEL(NEW), the broken vital record specification uses a
permanent retention date. With VRSEL(OLD), the broken vital record specification ensures
that the data set is retained to the maximum of the COUNT value in the first vital record
specification.
We recommend using VRSEL(NEW) to:
Have more flexibility in defining retention and movement policies.
Override a volume expiration date when a volume is dropped from vital record
specification retention and the data set retained on it has the release option
EXPIRYDATEIGNORE.
Return a volume to scratch status in a single inventory management run.
Before starting to use VRSEL(NEW), understand that your existing policies might be applied
differently under VRSEL(NEW). You can perform these steps to avoid problems that might
occur when the DFSMSrmm performs VRSEL(NEW) processing.
1.Before using VRSEL(NEW), back up your DFSMSrmm control data set using EDGBKUP
or EDGHSKP.
2.Perform cleanup on the name vital record specifications by making sure that any retention
information in them is correct. DFSMSrmm provides the EDGRVCLN REXX exec
described in “EDGRVCLN REXX Procedure to Clean Up Name Vital Record
REMOVABLE MEDIA MANAGER VITAL RECORDS RETENTION REPOR
(C) IBM CORPORATION 1993,1998 ----- ------- --------- -----
JOB MASK DATA SET OR VOLUME MASK OWNER TYPE RETN C X
________________________________________________________________________________
BSYSMF.WEEK.** STCOPC DSN CYCLES Y N
JOB NAME DATA SET NAME FSEQ DSEQ VOLUME VSEQ
________________________________________________________________________________
SMFWEEK2 BSYSMF.WEEK.G0552V00 1 1 Q17058 1
SMFWEEK2 BSYSMF.WEEK.G0551V00 1 1 Q17191 1
SMFWEEK2 BSYSMF.WEEK.G0550V00 1 1 Q17084 1
SMFWEEK2 BSYSMF.WEEK.G0549V00 1 1 Q17357 1
SMFWEEK2 BSYSMF.WEEK.G0548V00 1 1 Q17287 1
SMFWEEK2 BSYSMF.WEEK.G0547V00 1 1 Q17356 1
SMFWEEK2 BSYSMF.WEEK.G0546V00 1 1 Q17321 1
SMFWEEK2 BSYSMF.WEEK.G0545V00 1 1 Q17068 1

320
z/OS V1R8 DFSMS Technical Update
Specifications” on page 92 to report and clean up problems with name vital record
specifications.
3.Run DFSMSrmm inventory management vital record processing so that the DFSMSrmm
control data set reflects the cleanup that you have done.
4.Update the DFSMSrmm PARMLIB OPTION VRSEL(NEW) operand.
5.Make sure that all systems sharing DFSMSrmm control data sets have the same
PARMLIB options.
6.Run the inventory management VERIFY function against the control data set without
introducing any of the new vital record specification functions. When you run VERIFY,
changes are not actually made to the DFSMSrmm control data set so that you can look at
the results before any changes are made.
7.Inspect the inventory management VERIFY ACTIVITY file by looking at changes in
matching vital record specification information, vital record status, and retention date.
DFSMSrmm provides a sample job EDGJACTP that you can use with DFSORT to format
and print fields in the ACTIVITY file. If you cannot clean up to your satisfaction, you can
revert to VRSEL(OLD) at this time.
8.Correct vital record specifications as needed to make sure that the policies that you want
are in place.
9.Continue running the VERIFY function and inspect the results until you get the results that
you expect using the new functions
10.Begin defining vital record specifications that include the new release options or the use of
ANDVRS. We suggest starting slowly until you gain more experience with using the new
vital record specification functions. We recommend using the RMM ISPF dialog to add the
new vital record specifications or to make changes to existing vital record specifications.
11.Repeat the process from step 7, step 9, and step 10 until you are satisfied that the results
are what you expect.
12. Run inventory management production run processing.
13. Check the vital records retention report to make sure that data sets and volumes are
retained as you intended.
EDGRVCLN REXX exec
Before you can use VRSEL(NEW), you need to clean up the retention information in your
existing vital record specifications. Use the EDGRVCLN REXX procedure to report and
update existing vital record specifications that were created before VRSEL(NEW) was
introduced. The procedure is intended for use only during the implementation of name vital
record specification retention information and for cleanup of vital record specifications even if
name vital record specification retention information is not being implemented. EDGRVCLN
provides options that you can use to identify retention information that needs to be corrected
and options that you can use to correct the information.
The REXX exec has following functions:
LIST(DSNCHAIN) List all vital record specification chains.
LIST(CURRENT) List all vital record specifications that specify LOCATION(CURRENT).
LIST(CYCLES) List all vital record specifications that specify a retention type of
CYCLES.
LIST(ERROR) List all name vital record specifications that contain incorrect or
incomplete retention information.

Chapter 8. DFSMSrmm enhancements
321
LIST(FILTER) List all data set VRSs that are candidates for exploitation of the use of
COUNT(0).
FIX(CYCLEBYDAYS) Change all CYCLES vital record specifications to use the
CYCLEBYDAYS retention type.
FIX(ERROR) Correct all name vital record specifications that contain incorrect or
incomplete retention information as found by LIST(ERROR).
FIX(FILTER) Implement the changes recommend by LIST(FILTER).
EDGRVCLN LIST file
When you use one of the LIST parameters, the procedure produces the DFSMSrmm
ADDVRS subcommands that can be issued to define corrected vital record specifications.
When you use the LIST(CURRENT), LIST(ERROR), or LIST(CYCLES) parameters, the LIST
file includes an RMM DELETEVRS subcommand for each vital record specification in
addition to the ADDVRS subcommand. When you use the LIST(FILTER) parameter, the LIST
file includes an RMMCHANGEVRS and one or more ADDVRS subcommands for each vital
record specification that is a candidate for exploitation of the use of COUNT(0). If you do not
want to use the FIX parameters to correct all the identified vital record specifications, you can
use the LIST file as input to your own processing. You can edit the file to remove or modify
the commands to meet your specific requirements.
EDGRVCLN parameter LIST(DSNCHAIN)
Use this parameter to list all vital record specification chains.
The procedure searches for all data set vital record specifications. The procedure then
searches for all name vital record specifications in the chain.
Use JCL as shown in Example 8-21 to list all vital record specification chains.
Example 8-21 Sample JCL to use EDGRVCLN with parameter LIST(DSNCHAIN)
//EDGRVCLN JOB ,140.SCHLUMBERGER,MSGCLASS=H,REGION=6M,
// MSGLEVEL=(1,1),NOTIFY=SCHLUM
//CLEANUP EXEC PGM=IDCAMS
//SYSPRINT DD SYSOUT=*
//SYSIN DD *
DELETE RMM.EDGRVCLN.LIST.CHAINP NONVSAM PURGE
DELETE RMM.EDGRVCLN.LIST.CHAIN NONVSAM PURGE
SET MAXCC=0
/*
//TMPCHAIN EXEC PGM=IKJEFT01,DYNAMNBR=30
//SYSPROC DD DISP=SHR,DSN=RMM.ADDONS.CEXEC
//SYSTSPRT DD DISP=(,CATLG),DSN=RMM.EDGRVCLN.LIST.CHAINP,
// SPACE=(TRK,(15,5),RLSE),UNIT=SYSDA,
// DCB=(LRECL=255,RECFM=VB,BLKSIZE=0)
//LIST DD DISP=(,CATLG),DSN=RMM.EDGRVCLN.LIST.CHAIN,
// SPACE=(TRK,(15,5),RLSE),UNIT=SYSDA,
// DCB=(LRECL=255,RECFM=VB,BLKSIZE=0)
//SYSTSIN DD *
%EDGRVCLN LIST(DSNCHAIN)
/*

322
z/OS V1R8 DFSMS Technical Update
Figure 8-56 shows you the messages EDGRVCLN writes to the SYSTSPRT DD.
Figure 8-56 Sample SYSTSPRINT messages using EDGRVCLN with parameter LIST(DSNCHAIN)
The commands generated using parameter LIST(DSNCHAIN) are shown in Figure 8-57.
Figure 8-57 Commands created using EDGRVCLN with parameter LIST(DSNCHAIN)
EDGRVCLN Parameter LIST(CURRENT)
Use this parameter to list all vital record specifications that specify LOCATION(CURRENT).
The procedure searches for all vital record specifications and lists each one that includes the
LOCATION(CURRENT).
Use JCL as shown in Example 8-22 to list all vital record specifications that specify
LOCATION(CURRENT).
Example 8-22 Sample JCL to use EDGRVCLN with parameter LIST(CURRENT)
//EDGRVCLN JOB ,140.SCHLUMBERGER,MSGCLASS=H,REGION=6M,
// MSGLEVEL=(1,1),NOTIFY=SCHLUM
//CLEANUP EXEC PGM=IDCAMS
//SYSPRINT DD SYSOUT=*
//SYSIN DD *
DELETE RMM.EDGRVCLN.LIST.CURRENTP NONVSAM PURGE
DELETE RMM.EDGRVCLN.LIST.CURRENT NONVSAM PURGE
SET MAXCC=0
/*
//TMPCHAIN EXEC PGM=IKJEFT01,DYNAMNBR=30
//SYSPROC DD DISP=SHR,DSN=RMM.ADDONS.CEXEC
//SYSTSPRT DD DISP=(,CATLG),DSN=RMM.EDGRVCLN.LIST.CURRENTP,
// SPACE=(TRK,(15,5),RLSE),UNIT=SYSDA,
// DCB=(LRECL=255,RECFM=VB,BLKSIZE=0)
READY
%EDGRVCLN LIST(DSNCHAIN)
SEARCHVRS DSNAME 'SCHLUM.TAPE.**' CHAIN failed with Return Code 4, Reason Code
Error in the CDS
READY
END

RMM ADDVRS DSNAME('A.B.C') COUNT(99999) DELETEDATE(1999/365)+
PRIORITY(0) OWNER(SCHLUM) STORENUMBER(99999) +
LOCATION(HOME) CYCLES
...
RMM ADDVRS DSNAME('SCHLUM.RMMDEMO.**') COUNT(1) +
DELETEDATE(1999/365) +
DESCRIPTION('MVSSSC SYSTEM BACKUP (PROD)') +
PRIORITY(0) NEXTVRS(EXTRACAT) OWNER(SCHLUM) STORENUMBER(1) +
LOCATION(HOME) DAYS +
RMM ADDVRS NAME(EXTRACAT) DELETEDATE(1999/365) +
DESCRIPTION('TEST') OWNER(SCHLUM) STORENUMBER(99999) +
LOCATION(CURRENT)
RMM ADDVRS DSNAME('SCHLUM.RMMTEST.MOVE.**') COUNT(99999) +
DELETEDATE(1999/365) +
DESCRIPTION('RETAIN AND MOVE DATA SETS') +
PRIORITY(0) OWNER(SCHLUM) STORENUMBER(1) WHILECATALOG +
LOCATION(REMOTE) CYCLES

Chapter 8. DFSMSrmm enhancements
323
//LIST DD DISP=(,CATLG),DSN=RMM.EDGRVCLN.LIST.CURRENT,
// SPACE=(TRK,(15,5),RLSE),UNIT=SYSDA,
// DCB=(LRECL=255,RECFM=VB,BLKSIZE=0)
//SYSTSIN DD *
%EDGRVCLN LIST(CURRENT)
/*
Figure 8-58 shows you the messages EDGRVCLN writes to the SYSTSPRT DD.
Figure 8-58 Sample SYSTSPRINT messages using EDGRVCLN with parameter LIST(CURRENT)
The commands generated using parameter LIST(CURRENT) are shown in Figure 8-59.
Figure 8-59 Commands created using EDGRVCLN with parameter LIST(CURRENT)
EDGRVCLN parameter LIST(CYCLES)
Use this parameter to list all vital record specifications that specify a retention type of
CYCLES. This includes those vital record specifications that include WHILECATALOG,
where CYCLES is used as the default retention type.
The procedure searches for all data set vital record specifications and lists each one that
includes the CYCLE retention type.
Use JCL as shown in Example 8-23 to list all vital record specification that specify a retention
type of CYCLES.
Example 8-23 Sample JCL to use EDGRVCLN with parameter LIST(CYCLES)
//EDGRVCLN JOB ,140.SCHLUMBERGER,MSGCLASS=H,REGION=6M,
// MSGLEVEL=(1,1),NOTIFY=SCHLUM
//CLEANUP EXEC PGM=IDCAMS
//SYSPRINT DD SYSOUT=*
READY
%EDGRVCLN LIST(CURRENT)
4 VRSs with the LOCATION(CURRENT) LISTed.
READY
END
RMM DELETEVRS DSNAME('DATA.SET.BACKUP')
RMM ADDVRS DSNAME('DATA.SET.BACKUP') +
COUNT(99999) DELETEDATE(1999/365) DESCRIPTION('TEST') +
PRIORITY(0) OWNER(SCHLUM) STORENUMBER(99999) WHILECATALOG +
LOCATION(CURRENT) CYCLES
RMM DELETEVRS DSNAME('D99003')
RMM ADDVRS DSNAME('D99003') COUNT(3) +
DELETEDATE(1999/365) DESCRIPTION('TEST') +
PRIORITY(0) OWNER(SCHLUM) STORENUMBER(3) +
LOCATION(CURRENT) LASTREFERENCEDAYS
RMM DELETEVRS NAME(CATALOG)
RMM ADDVRS NAME(CATALOG) DELETEDATE(1999/365) +
DESCRIPTION('TEST') OWNER(SCHLUM) STORENUMBER(99999) +
LOCATION(CURRENT)
RMM DELETEVRS NAME(EXTRACAT)
RMM ADDVRS NAME(EXTRACAT) DELETEDATE(1999/365) +
DESCRIPTION('TEST') OWNER(SCHLUM) STORENUMBER(99999) +
LOCATION(CURRENT)

324
z/OS V1R8 DFSMS Technical Update
//SYSIN DD *
DELETE RMM.EDGRVCLN.LIST.CYCLESP NONVSAM PURGE
DELETE RMM.EDGRVCLN.LIST.CYCLES NONVSAM PURGE
SET MAXCC=0
/*
//TMPCHAIN EXEC PGM=IKJEFT01,DYNAMNBR=30
//SYSPROC DD DISP=SHR,DSN=RMM.ADDONS.CEXEC
//SYSTSPRT DD DISP=(,CATLG),DSN=RMM.EDGRVCLN.LIST.CYCLESP,
// SPACE=(TRK,(15,5),RLSE),UNIT=SYSDA,
// DCB=(LRECL=255,RECFM=VB,BLKSIZE=0)
//LIST DD DISP=(,CATLG),DSN=RMM.EDGRVCLN.LIST.CYCLES,
// SPACE=(TRK,(15,5),RLSE),UNIT=SYSDA,
// DCB=(LRECL=255,RECFM=VB,BLKSIZE=0)
//SYSTSIN DD *
%EDGRVCLN LIST(CYCLES)
/*
Figure 8-60 shows you the messages EDGRVCLN writes to the SYSTSPRT DD.
Figure 8-60 Sample SYSTSPRINT messages using EDGRVCLN with parameter LIST(CYCLES)
READY
%EDGRVCLN LIST(CYCLES)
24 DSNAME VRSs with CYCLES retention type LISTed.
READY
END

Chapter 8. DFSMSrmm enhancements
325
The commands generated using parameter LIST(CYCLES) are shown in Figure 8-61.
Figure 8-61 Commands created using EDGRVCLN with parameter LIST(CYCLES)
EDGRVCLN parameter LIST(ERROR)
Use this parameter to list all name vital record specifications that contain incorrect or
incomplete retention information. The procedure assumes that any name vital record
specification containing a retention type or count value is in error. This is the default value.
The procedure searches for all name vital record specifications and lists each one that
includes any retention type or count value.
Use JCL as shown in Example 8-24 to list all vital record specifications that contain incorrect
or incomplete retention information.
Example 8-24 Sample JCL to use EDGRVCLN with parameter LIST(ERROR)
//EDGRVCLN JOB ,140.SCHLUMBERGER,MSGCLASS=H,REGION=6M,
// MSGLEVEL=(1,1),NOTIFY=SCHLUM
//CLEANUP EXEC PGM=IDCAMS
//SYSPRINT DD SYSOUT=*
//SYSIN DD *
DELETE RMM.EDGRVCLN.LIST.ERRORP NONVSAM PURGE
DELETE RMM.EDGRVCLN.LIST.ERROR NONVSAM PURGE
RMM DELETEVRS DSNAME('A.B.C')
RMM ADDVRS DSNAME('A.B.C') COUNT(99999) +
DELETEDATE(1999/365) PRIORITY(0) OWNER(SCHLUM) +
STORENUMBER(99999) +
LOCATION(HOME) BYDAYSCYCLE
....
RMM DELETEVRS DSNAME('SCHLUM.EDG*.**')
RMM ADDVRS DSNAME('SCHLUM.EDG*.**') COUNT(6) +
DELETEDATE(1999/365) DESCRIPTION('TEST') +
PRIORITY(0) OWNER(SCHLUM) STORENUMBER(6) +
WHILECATALOG RELEASE(EXPIRYDATEIGNORE SCRATCHIMMEDIATE) +
LOCATION(HOME) BYDAYSCYCLE
RMM DELETEVRS DSNAME('SCHLUM.MULTIPLE.DATASET.**')
RMM ADDVRS DSNAME('SCHLUM.MULTIPLE.DATASET.**') +
COUNT(1) DELETEDATE(1999/365) DESCRIPTION('TEST')+
PRIORITY(0) OWNER(SCHLUM) STORENUMBER(1) +
LOCATION(HOME) BYDAYSCYCLE
RMM DELETEVRS DSNAME('SCHLUM.RMMTEST.MOVE.**')
RMM ADDVRS DSNAME('SCHLUM.RMMTEST.MOVE.**') COUNT(99999) +
DELETEDATE(1999/365) +
DESCRIPTION('RETAIN AND MOVE DATA SETS') +
PRIORITY(0) OWNER(SCHLUM) STORENUMBER(1) WHILECATALOG +
LOCATION(REMOTE) BYDAYSCYCLE
RMM DELETEVRS DSNAME('SCHLUM.TAPE.**')
RMM ADDVRS DSNAME('SCHLUM.TAPE.**') COUNT(1) +
DELETEDATE(1999/365) DESCRIPTION('TEST NEXT VRS') +
PRIORITY(0) NEXTVRS(EXTRA) OWNER(SCHLUM) STORENUMBER(1) +
LOCATION(HOME) BYDAYSCYCLE
RMM DELETEVRS DSNAME('SCHLUM.TEST1')
RMM ADDVRS DSNAME('SCHLUM.TEST1') COUNT(1) +
DELETEDATE(1999/365) PRIORITY(0) OWNER(SCHLUM) +
STORENUMBER(1) +
LOCATION(HOME) BYDAYSCYCLE
....

326
z/OS V1R8 DFSMS Technical Update
SET MAXCC=0
/*
//TMPCHAIN EXEC PGM=IKJEFT01,DYNAMNBR=30
//SYSPROC DD DISP=SHR,DSN=RMM.ADDONS.CEXEC
//SYSTSPRT DD DISP=(,CATLG),DSN=RMM.EDGRVCLN.LIST.ERRORP,
// SPACE=(TRK,(15,5),RLSE),UNIT=SYSDA,
// DCB=(LRECL=255,RECFM=VB,BLKSIZE=0)
//LIST DD DISP=(,CATLG),DSN=RMM.EDGRVCLN.LIST.ERROR,
// SPACE=(TRK,(15,5),RLSE),UNIT=SYSDA,
// DCB=(LRECL=255,RECFM=VB,BLKSIZE=0)
//SYSTSIN DD *
%EDGRVCLN LIST(ERROR)
/*
Figure 8-62 shows you the messages that EDGRVCLN writes to the SYSTSPRT DD.
Figure 8-62 Sample SYSTSPRINT messages using EDGRVCLN with parameter LIST(ERROR)
The commands that are generated using parameter LIST(ERROR) are shown in Figure 8-63.
Figure 8-63 Commands created using EDGRVCLN with parameter LIST(ERROR)
EDGRVCLN parameter LIST(FILTER)
Use this parameter to list all data set VRSs that are candidates for exploitation of the use of
COUNT(0).
The procedure searches for all data set name VRSs that do not specify COUNT(0) and
accumulates those that have common delay, retention, and movement requirements, and
NEXT/ANDVRS specified in the first VRS in the chain.
READY
%EDGRVCLN LIST(ERROR)

VRS NAME(CATALOG) is a valid location-only VRS

VRS NAME(D99000) is a valid location-only VRS

VRS NAME(EXTRACAT) is a valid location-only VRS
VRS NAME(TESTAND) had retention values and has been recreated as a
location-only VRS

VRS NAME(TESTNEXT) is a valid location-only VRS

VRS NAME(TESTX) is a valid location-only VRS
1 NAME VRSs in error are LISTed.
5 NAME VRSs were valid location-only VRSs.
READY
END
RMM DELETEVRS NAME(TESTAND)
RMM ADDVRS NAME(TESTAND) DELETEDATE(1999/365) +
DESCRIPTION('TEST LCLVRS1 AND LCLVRS2 FUNCT') +
ANDVRS(TESTX) OWNER(SCHLUM) STORENUMBER(99999) +
LOCATION(SHELF)

Chapter 8. DFSMSrmm enhancements
327
Use JCL as shown in Example 8-25 to list all data set VRSs that are candidates for
exploitation of the use of COUNT(0).
Example 8-25 Sample JCL to use EDGRVCLN with parameter LIST(DSNCHAIN)
//EDGRVCLN JOB ,140.SCHLUMBERGER,MSGCLASS=H,REGION=6M,
// MSGLEVEL=(1,1),NOTIFY=SCHLUM
//CLEANUP EXEC PGM=IDCAMS
//SYSPRINT DD SYSOUT=*
//SYSIN DD *
DELETE RMM.EDGRVCLN.LIST.FILTERP NONVSAM PURGE
DELETE RMM.EDGRVCLN.LIST.FILTER NONVSAM PURGE
SET MAXCC=0
/*
//TMPCHAIN EXEC PGM=IKJEFT01,DYNAMNBR=30
//SYSPROC DD DISP=SHR,DSN=RMM.ADDONS.CEXEC
//SYSTSPRT DD DISP=(,CATLG),DSN=RMM.EDGRVCLN.LIST.FILTERP,
// SPACE=(TRK,(15,5),RLSE),UNIT=SYSDA,
// DCB=(LRECL=255,RECFM=VB,BLKSIZE=0)
//LIST DD DISP=(,CATLG),DSN=RMM.EDGRVCLN.LIST.FILTER,
// SPACE=(TRK,(15,5),RLSE),UNIT=SYSDA,
// DCB=(LRECL=255,RECFM=VB,BLKSIZE=0)
//SYSTSIN DD *
%EDGRVCLN LIST(FILTER)
/*
Figure 8-64 shows you the messages that EDGRVCLN writes to the SYSTSPRT DD.
Figure 8-64 Sample SYSTSPRINT messages using EDGRVCLN with parameter LIST(FILTER)
READY
%EDGRVCLN LIST(FILTER)

Dsname VRSs with retention type CYCLE/BYDAYSCYCLE with the combination of
DELAY() and STORENUMBER() can not be converted.
The following have the same retention criteria and could be converted if
you manually change the VRSs not to have DELAY and STORENUMBER in one VRS:
JOBNAME DSNAME
TESTYY TESTYY
TEST* TESTZZ

RMM commands prepared for the exploitation of the use of COUNT(0) are LISTed.
READY
END

328
z/OS V1R8 DFSMS Technical Update
The commands that are generated using parameter LIST(FILTER) are shown in Figure 8-65.
Figure 8-65 Commands created using EDGRVCLN with parameter LIST(FILTER)
FIX(CYCLEBYDAYS)
Use this parameter to change all CYCLES vital record specifications to use the
CYCLEBYDAYS retention type.
FIX(ERROR)
Use this parameter to correct all name vital record specifications that contain incorrect or
incomplete retention information as found by LIST(ERROR). The corrections are made by
RMM CHANGEVRS DSNAME('TEST.LCLVRSC.FUNCTION') +
COUNT(0) +
NEXTVRS(A0000000)

RMM CHANGEVRS DSNAME('TEST.LCLVRSC.FUNCTION.**') +
COUNT(0) +
NEXTVRS(A0000000)

RMM ADDVRS NAME(A0000000) +
COUNT(99999) LASTREFERENCEDAYS +
LOCATION(DISTANT) STORENUMBER(12345) +
WHILECATALOG UNTILEXPIRED


RMM CHANGEVRS DSNAME('ABEND') +
JOBNAME(SCHLUM) +
COUNT(0) +
NEXTVRS(A0000004)

RMM CHANGEVRS DSNAME('ABEND') +
JOBNAME(SCHLUX) +
COUNT(0) +
NEXTVRS(A0000004)

RMM CHANGEVRS DSNAME('OPEN') +
JOBNAME(SCHLUM) +
COUNT(0) +
NEXTVRS(A0000004)

RMM ADDVRS NAME(A0000004) +
COUNT(1) DAYS +
LOCATION(HOME)


RMM CHANGEVRS DSNAME('SIEGEL.TEST.VRSEL.CATYN') +
COUNT(0) +
NEXTVRS(A0000005)

RMM CHANGEVRS DSNAME('SIEGEL.TEST1') +
COUNT(0) +
NEXTVRS(A0000005)

RMM ADDVRS NAME(A0000005) +
COUNT(3) DAYS +
LOCATION(HOME)

Chapter 8. DFSMSrmm enhancements
329
deleting and re-adding the vital record specifications that contain errors. Only use this option
when you are ready to correct the errors listed by the LIST(ERROR) option.
FIX(FILTER)
Use this parameter to implement the changes recommend by LIST(FILTER).
The procedure creates one or more retention NAME VRSs for each of the common groups of
retentions found, and then uses the CHANGEVRS subcommand to set COUNT(0) and the
NEXTVRS to chain to the new NAME VRS(es) created. If DELAY was in use, a retention
name VRS is created for the DELAY using DAYS since creation and a NEXTVRS to the
retention name VRS(es) for the retention and location pulled from the DSNAME VRS.

330
z/OS V1R8 DFSMS Technical Update

© Copyright IBM Corp. 2008. All rights reserved.
331
Chapter 9.
Tape security
DFSMS V1.8 provides new options for securing tape data sets using System Authorization
Facility (SAF). These are designed to allow you to define profiles to protect data sets on tape
using the DATASET class without the need to activate the TAPEDSN option or the TAPEVOL
class. DFSMS also provides options that you can use to specify that all data sets on a tape
volume should have common authorization and that users are authorized to overwrite
existing files on a tape volume.
In this chapter we describe the options available in DFSMS V1.8 for tape data set security
and how to implement it.
This chapter contains the following sections:
Tape data set authorization
Implementation
Removing TAPEVOL and TAPEDSN processing
Error messages
Testing various security settings
9

332
z/OS V1R8 DFSMS Technical Update
9.1 Tape data set authorization
DFSMS V1.8 provides new options for securing tape data sets using the system authorization
facility (SAF) to allow you to protect data sets on tape using the RACF DATASET class
without the need to activate the TAPEDSN option or the TAPEVOL class. This new tape data
set authorization checking allows you to specify that all data sets on a tape volume should
have common authorization. You can specify whether users are authorized to overwrite
existing files on a tape volume. In addition, you can specify that a user must have access to
the first file on a tape volume to add additional files on that tape volume.
9.1.1 Recommendations for tape security
For optimum tape security use the combined capabilities of DFSMSrmm, DFSMSdfp, and
RACF. We recommend that you specify the following parameters:
In DEVSUPxx PARMLIB member:
– TAPEAUTHDSN=YES
– TAPEAUTHF1=YES
– TAPEAUTHRC4=FAIL
– TAPEAUTHRC8=FAIL
In EDGRMMxx PARMLIB member:
OPTION TPRACF(N)
In RACF:
SETROPTS NOTAPEDSN NOCLASSACT(TAPEVOL)
The combination of DFSMSrmm, DFSMSdfp, and RACF ensures:
Full 44 character data set name validation.
Validation that the correct volume is mounted.
Control the overwriting of existing tape data sets.
Management of tape data set retention.
Control over the creation and destruction of tape volume labels.
No limitations caused by RACF TAPEVOL profile sizes and TVTOC limitations.
All tape data sets on a volume have a common authorization.
Use of generic DATASET profiles, enabling common authorization with DASD data sets.
Authorization for all tape data sets regardless of the tape label type.
Authorization for the use of bypass label processing (BLP).
Exploitation of RACF
erase on scratch
support.
Use of DFSMSrmm FACILITY class profiles for data sets unprotected by RACF
Your authorization to use a volume outside of DFSMSrmm control through
ignore

processing also enables authorization to the data sets on that volume.
To aid in the migration to this environment, DFSMSrmm provides the TPRACF(CLEANUP)
option, and DEVSUPxx provides TAPEAUTHRC8(WARN) and TAPEAUTHRC4(ALLOW).
The function in DFSMSdfp does not replace all the functional capabilities that the RACF
TAPEDSN option, TAPEVOL class, and TVTOC provide. However, together with the
functions that DFSMSrmm provides, you do have equivalent capability. The enhanced

Chapter 9. Tape security
333
DFSMSdfp function addresses the authorization requirements for tape data sets and relies on
your use of a tape management system such as DFSMSrmm to perform the following
operations:
Verify full 44 character data set names.
Control the overwriting of existing tape files.
Handle tape data set retention.
Control the creation and destruction of tape labels.
9.1.2 Overview of the TAPEVOL and TAPEDSN processing
The enhancement in authority checking that is available with z/OS DFSMS V1.8 takes into
account that today most customers are using a tape management system. A tape
management system like DFSMSrmm verifies during reuse of data sets and tape volumes the
data set names that were put in place at creation time. This ensures that the same data set
resources are always being checked in RACF whenever a particular data set is opened.
How protection of tape data sets works pre-DFSMS V1.8r
Processing in detail differs depending on:
The setting of SETROPTS TAPEDSN/NOTAPEDSN
The RACF TAPEVOL class being active or not
The label type of the tape volume
The requirement of bypass label processing (BLP)
SETROPTS NOTAPEDSN NOCLASSACT(TAPEVOL)
Tape data sets are not protected if neither RACF option TAPEDSN is being set nor the RACF
TAPEVOL class is active. This is true regardless of label type or BLP requirement.
SETROPTS TAPEDSN and SETROPTS CLASSACT(TAPEVOL)
If the RACF TAPEVOL class is active as well as the RACF option TAPEDSN set, tape data
are protected at the data set level.
In case of label type SL or AL, during the open of an existing data set a REQUEST=AUTH is
performed in the DATASET class with DSNTYPE=T:
1.If there is a matching TAPEVOL class profile, the caller’s authority according to the type of
open is being checked (READ for input and UPDATE for output).
– If the caller’s authority at the volume level is sufficient, no further checks are
performed.
– If the TAPEVOL profile contains a TVTOC, the data set name is verified at the full
length of 44 bytes.
• If the verification was successful, RACF looks for a matching data set profile and
determines whether the caller’s authority is sufficient.
• If verification was not successful the request ends with return code 8 (denied).
– If the TAPEVOL profile contains no TVTOC see step 2.
Important: With the DFSMS V1.8 tape security implementation, as opposed to a
TAPEVOL and TAPEDSN implementation, you can do the following:
Write more than 500 data sets to one tape or one tape volume set.
A tape volume set can expand more than 42 tape volumes.
Write duplicate data set names to one tape or one tape volume set.

334
z/OS V1R8 DFSMS Technical Update
2.If there is no matching TAPEVOL profile, RACF looks for a matching DATASET class
profile and determines whether the caller is authorized sufficiently (READ for input and
UPDATE for output).
With a label type of SL or AL, during open of a new data set a REQUEST=AUTH as well as a
REQUEST=DEFINE is performed in the DATASET class with DSNTYPE=T:
If the caller wants to append another data set.
– If there is a matching TAPEVOL class profile the caller must have at least UPDATE
authority, otherwise the request ends with return code 8 (denied).
– RACF is looking for a matching data set profile and determines that the caller is at least
authorized for UPDATE access.
If the caller wants to overwrite an existing data set.
– If the new data set name is different from the existing data set, then one of the
following is true:
• The security retention period of that particular existing data set (stored in the
protecting DATASET profile) and of all of the following existing data sets must have
expired.
• The caller must have at least UPDATE authority to the volume.
• RACF is looking for the matching profile of the new data set name and determines
that the caller is at least authorized for UPDATE access.
– If the new data set name is
not
different, then the caller must have UPDATE authority
either for the matching DATASET profile or for the matching TAPEVOL profile.
If the TAPEVOL profile contains a TVTOC, then for the new data set an entry in the
TVTOC is maintained.
In case of nonlabeled tapes (NL):
During open for input the caller must have READ authority to the volume or, if there is a
TVTOC for the volume, to the matching data set profile.
During open for output the caller must have at least UPDATE authority to both the volume
and the data set.
In case of nonstandard label tapes (NSL) data management does not do authorization
checking.
In case of a request for BLP during open of a data set the caller must be authorized for the
ICHBLP resource in the FACILITY class.
Note: Without a TVTOC there is no verification of the full 44 byte length of the data set
name.
Note: When a TAPEVOL class profile contains a TVTOC some restrictions apply:
The maximum number of entries for data sets that a TVTOC can contain is 500.
The maximum number of volumes that any data set on the tape with an entry in the
TVTOC can span is 42.
From a tape volume set you cannot delete a tape volume if a TVTOC entry indicates
that there is a protected data set on the volume.

Chapter 9. Tape security
335
SETROPTS TAPEDSN and SETROPTS NOCLASSACT(TAPEVOL)
If only the RACF option TAPEDSN is being set while the TAPEVOL class is not active, tape
data is protected at the data set level.
In case of label type SL, AL, or NL during open of a data set for reading, the caller needs at
least READ authority for the matching DATASET class profile.
In case of label type SL, AL, or NL during open of a data set for writing, the caller needs at
least the following for the matching DATASET class profile:
UPDATE authority to open an existing data set
ALTER authority to create a new data set
In the case of nonstandard label tapes (NSL) data management does not do authorization
checking.
In case of a request for BLP during open of a data set the caller must be authorized for the
ICHBLP resource in the FACILITY class.
SETROPTS NOTAPEDSN and SETROPTS CLASSACT(TAPEVOL)
If the RACF option TAPEDSN is not being set while the TAPEVOL class is active, tape data
are protected at the tape volume level.
In the case of label type SL, AL, or NL during open of a data set for reading, the caller needs
at least READ authority for the matching TAPEVOL class profile.
In the case of label type SL, AL, or NL during open of a data set for writing, the caller needs at
least UPDATE authority for the matching TAPEVOL class profile.
In the case of nonstandard label tapes (NSL) data management does not do authorization
checking.
In the case of a request for bypass-label processing during open of a data set the caller must
be authorized for the ICHBLP resource in the FACILITY class.
Summary
In order to achieve the highest level of security you should choose an environment that
provides discrete TAPEVOL class profiles with TVTOCs as well as the SETROPTS
TAPEDSN setting. This environment provides the following advantages:
Full 44 character data set name validation
Use of generic DATASET class profiles, enabling common authorization with DASD data
sets
Volume level control in order to guarantee exclusive volume use for applications
9.1.3 How the DFSMS V1.8 tape data set authority checking works
The alternate authority checking is enabled by modifying the contents of member DEVSUPxx
in the PARMLIB concatination. Four new keywords are available:
TAPEAUTHDSN
YES Enables tape authorization checks in the DATASET class but
without DSTYPE=T.
The system uses the data set name specified in the allocation or
JCL to check your authorization to read or write the specified file.

336
z/OS V1R8 DFSMS Technical Update
In addition, the system determines the RACF erase-on-scratch
setting from the RACF profile and passes it to your tape
management system.
When you request bypass label processing (BLP) and the mounted
volume uses standard labels, OPEN issues the authorization check
that the user is authorized to use BLP. This processing uses the
existing ICHBLP resource in the RACF FACILITY class. When you
specify TAPEAUTHDSN = YES only, it replaces the check that
RACF makes as part of tape volume authorization checking.
NO Indicates OPEN processing to issue RACROUTEs as it did before
based on the options set in RACF such as SETROPTS TAPEDSN
and SETROPTS CLASSACT(TAPEVOL). This is the default
setting.
TAPEAUTHF1
YES Enables additional tape authorization checks in the DATASET
class for existing files on the same tape volume when any other file
on the tape volume is opened.
This function depends on the tape management system returning
the 44-character data set name and data set sequence number to
OPEN/EOV through the IFGTEP during the volume mount exit
volume security function. If no data set name is returned by the
tape management system, processing is as though this keyword
had not been specified.
Although intended to enable an additional authorization check for
the first data set when any other data set on the tape volume is
opened, the implementation allows your tape management system
to request one or more additional authorization checks when any
data set on a tape volume is opened. Each additional data set
name and data set sequence number returned results in an
additional RACROUTE. Do not use this function unless you have a
tape management system and it can return a data set name and
data set sequence number. A data set sequence number is the
label number normally specified in the JCL LABEL keyword and
stored in the catalog.
When TAPEAUTHDSN=YES is in use, any additional RACROUTE
matches that issued for TAPEAUTHDSN except for the data set
name and data set sequence number. Otherwise, TAPEAUTHF1
uses a RACROUTE that matches that used for SETROPTS
TAPEDSN. When neither TAPEAUTHDSN nor SETROPTS
TAPEDSN is in use, TAPEAUTHF1 support is not provided.
NO Disables additional tape authorization checks in the DATASET
class for existing files on the same tape volume when any other file
on the tape volume is opened. This is the default setting.
TAPEAUTHRC4
ALLOW This applies to authorization checks in the DATASET class, and
applies only to the results of TAPEAUTHDSN = YES and
TAPEAUTHF1 = YES processing.
This allows accessing of data sets that are not protected by a
security profile. RC4 refers to the return code value of 4 returned
from SAF as a result of the RACROUTE issued by

Chapter 9. Tape security
337
OPEN/CLOSE/EOV. A return code of 4 in general means that the
resource is not protected.
FAIL Denies accessing of data sets that are not protected by a security
profile. Use this setting in a PROTECTALL(FAIL) environment.
This is the default setting.
TAPEAUTHRC8
WARN Enables warning mode for all tape authorization checks in the
DATASET class as a result of TAPEAUTHDSN = YES and
TAPEAUTHF1 = YES processing. Allows accessing of data sets
that typically cannot be accessed. RACF issues an ICH408I
message to indicate why access is not allowed. However,
OPEN/EOV allows access.
FAIL Denies accessing of data sets according to the result of the check
in DATASET class. This is the default setting.
In Table 9-1 you can see the different PROTECTALL settings in RACF and the equivalent
settings of the new TAPEAUTHRC4 and TAPEAUTHRC8 options for securing tape data sets
using the System Authorization Facility (SAF).
Table 9-1 Compare RACF and DEFSUP protection
Figure 9-1shows the RACF checking sequence when you have set the new tape security
options TAPEAUTHDSN and TAPEAUTHF1 and activated it. If you have not specified to
bypass DFSMSrmm processing by using the JCL parameter EXPDT=98000 or
ACCODE=XCANORES, then RACF checking is made. The data sets must be RACF
protected and the user must have access to the data set profile protecting the data sets.
Figure 9-1 New tape security flowchart
RACF settings DEVSUP settings Action
PROTECTALL(NONE) N/A None
PROTECTALL(WARN) TAPEAUTHRC4(ALLOW)
TAPEAUTHRC8(WARN)
Enables warning mode
PROTECTALL(FAIL) TAPEAUTHRC4(FAIL)
TAPEAUTHRC8(FAIL)
Denies accessing of data sets:
That are not protected
According to the result of the
check
bypass
DFSMSrmm?
(EXPDT=98000)
correct RACF
access to DSN
profile?
RACF protected?
DSN profile
available
OK
to process
requested
data set
yes
yes
ICH408I
Resource not
protected
bypass
DFSMSrmm/RACF
processing
yes
no
no
ABEND 913
user not
authorized
no

338
z/OS V1R8 DFSMS Technical Update
If you have specified to bypass DFSMSrmm processing by using the JCL parameter,
EXPDT=98000 or ACCODE=XCANORES DFSMSrmm makes additional security checks, as
shown in Figure 9-2. Instead of having different RACF resources to check the bypass
processing for volumes defined or not defined in the DFSMSrmm control data set, you can
specify the resource STGADMIN.EDG.IGNORE.TAPE.* in class FACILITY.
Figure 9-2 Bypass DFSMSrmm processing flowchart
9.2 Implementation
In this section we describe how you can implement the new tape data set authorization. This
section describes the new implementation of tape data set autorization using the new
DEVSUP settings to protect data sets on tape using the DATASET class.
Note: You can specify a profile in RACF class FACILITY for each single volume by using
the full volume serial number at the end of the resource name like:
STGADMIN.EDG.IGNORE.TAPE.RMM.V12345
or for a range of volumes using the asterisk as part of the volume serial number like the
following for all three different kinds of resources:
STGADMIN.EDG.IGNORE.TAPE.NORMM.ABC*
OK
to process
requested
data set
ICH408I Resource not protected
you cannot process data set
ICH408I Resource not protected
you cannot process data set
bypass
DFSMSrmm/RACF
processing
No
Yes
No
Yes
No
Yes
DFSMSrmm?
managed
volume?
FACILITY-Profile
STGADMIN.EDG.IGNORE.TAPE.NORMM.*
UACC(NONE)
Correct
access to FACILITY
resource
Correct
access to FACILITY
resource
FACILITY-Profile
STGADMIN.EDG.IGNORE.TAPE.RMM.*
UACC(NONE)

Chapter 9. Tape security
339
You should protect the use of the set MVS DEVSUP command so that only a small number of
people have access to this resource and are able to modify your tape data set authorization
settings. Example 9-1 shows you how you can protect the use of the MVS DEVSUP
command.
Example 9-1 Protect MVS SET DEVSUP command
//MHLRES5C JOB (999,POK),MSGLEVEL=1,NOTIFY=&SYSUID
//*
//* *******************************************************************
//* * TESTING Tape Data Set Authorization *
//* *******************************************************************
//DELETE EXEC PGM=IKJEFT01
//SYSPRINT DD SYSOUT=*
//SYSTSPRT DD SYSOUT=*
//SYSTSIN DD *
RDEF OPERCMDS MVS.SET.DEVSUP UACC(NONE)
PE MVS.SET.DEVSUP CL(OPERCMDS) ID(MHLRES5 MHLRES2) ACC(NONE)
SETR REFRESH RACLIST(OPERCMDS)
/*
Use the MVS SET DEVSUP command in Example 9-2 to test your RACF protection of the
MVS SET DEVSUP command.
Example 9-2 Unauthorized use of set DEVSUP
/T DEVSUP=18
This MVS SET COMMAND is not working because the user MHLRES5 is not permitted to
use this command. Figure 9-3 shows you the RACF error messages you receive.
Figure 9-3 RACF error messages
9.2.1 Check all high-level qualifiers on tape
You should check that each high-level qualifier on volumes in DFSMSrmm status MASTER or
USER are defined to RACF as a user or group. You can create different reports to check the
high-level qualifiers currently used:
Only on volumes in status SCRATCH
Only on volumes that have a status other than SCRATCH
Using all volumes in any volume status
Note: Before you start this new tape data set authorization implementation you should
search your DFSMSrmm control data set and check that for all data sets residing on tape,
at a minimum, the high-level qualifier is defined to RACF.
Note: Refresh the RACF OPERCMDS because the OPERCMDS are in the RACLIST
class.
RESPONSE=SC64
ICH408I USER(MHLRES5 ) GROUP(SYS1 ) NAME(MARY LOVELACE - RESI)
MVS.SET.DEVSUP CL(OPERCMDS)
INSUFFICIENT ACCESS AUTHORITY
ACCESS INTENT(UPDATE ) ACCESS ALLOWED(NONE )

340
z/OS V1R8 DFSMS Technical Update
9.2.2 Update DEVSUPnn PARMLIB member
Update the DEVSUPxx member in your PARMLIB to specify your installation default for
device support options. DEVSUPxx is processed during the NIP phase of IPL. After IPL, you
can use system command SET DEVSUP=XX to activate the DEVSUP changes.
To enable the new tape data set protection add the following parameters to the DEVSUPxx
PARMLIB member:
TAPEAUTHDSN To enable tape authorization checks in the DATASET class
TAPEAUTHF1 Enables additional tape authorization checks in the DATASET class
for existing files on the same tape volume when any other file on the
tape volume is opened.
TAPEAUTHRC4 Use this keyword to control PROTECTALL processing for tape data
sets. This applies to the results of RACROUTE processing when both
TAPEAUTHDSN=YES and TAPEAUTHF1=YES are specified.
TAPEAUTHRC8 Use this keyword as an aid to the implementation of TAPEAUTHDSN
and TAPEAUTHF1.
Provides a managed and controlled implementation of tape
authorization checks in the DATASET class, and applies only to the
results of TAPEAUTHDSN=YES and TAPEAUTHF1=YES processing.
Figure 9-4 shows a sample DEVSUPxx PARMLIB member including the default for
compaction for new data sets written to 3480, 3490, or 3590 tape subsystems. The category
used for this system for MEDIA1, MEDIA2, MEDIA3, and MEDIA5 tapes is the status scratch.
The PRIVATE category for tapes is the private status, and the ERROR category is for all
volumes in which OAM detects any error. This member also includes the final settings for the
new tape data set authorization implementation after you have successfully tested this new
function.
Figure 9-4 Sample DEVSUPxx PARMLIB member
Note: While TAPEAUTHRC4=FAIL and TAPEAUTHRC8=FAIL are specified to implement
the DFSMS V1.8 tape data set security enhancement, the TAPEAUTHRC4=ALLOW and
TAPEAUTHRC8=WARN options are provided to ease the migration to the new tape
security implementation. We recommend that you initially specify APEAUTHRC4=ALLOW
and TAPEAUTHRC8=WARN.
COMPACT = YES, /* INSTALLATION DEFAULT FOR IDRC */
MEDIA1 = 0021,
MEDIA2 = 0022,
MEDIA3 = 0023,
MEDIA5 = 0025,
ERROR = 002E,
PRIVATE = 002F,
TAPEAUTHDSN = YES,
TAPEAUTHF1 = YES,
TAPEAUTHRC4 = FAIL,
TAPEAUTHRC8 = FAIL

Chapter 9. Tape security
341
Where:
COMPACT
Specifies the default for compaction for new data sets written to 3480, 3490, or 3590 tape
subsystems.
MEDIAx
Used to specify category codes for library partitioning.
ERROR
Used to specify category codes for library partitioning.
PRIVATE
Used to specify category codes for library partitioning.
TAPEAUTHDSN=YES
Enables tape authorization checks in the DATASET class but without DSTYPE=T.
DSTYPE=T indicates to RACF that the check is for the data set on a tape volume and that
special RACF tape data set and a tape volume processing is to be performed. Without
DSTYPE=T RACF authorization checking considers only profiles in the DATASET class.
The system uses the data set name specified in the allocation or JCL to check your
authorization to read or write the specified file.
In addition, the system determines the RACF erase-on-scratch setting from the RACF
profile and passes it to your tape management system.
Use this option only when you have a tape management system, such as DFSMSrmm,
installed and actively checking that the 44-character data set name specified by the user
matches the data set name on tape. Without a tape management system, tape data set
open processing can only validate the last 17 characters of the data set name against the
tape volume labels.
When you request bypass label processing (BLP) and the mounted volume uses standard
labels, OPEN issues the authorization check that the user is authorized to use BLP. This
processing uses the existing ICHBLP resource in the RACF FACILITY class. When you
specify TAPEAUTHDSN=YES only, it replaces the check that RACF makes as part of
tape volume authorization checking.
TAPEAUTHF1=YES
Enables additional tape authorization checks in the DATASET class for existing files on
the same tape volume when any other file on the tape volume is opened.
This function depends on the tape management system returning the 44-character data
set name and data set sequence number to OPEN/EOV through the IFGTEP during the
volume mount exit volume security function. If no data set name is returned by the tape
management system, processing is as though this keyword had not been specified.
Although intended to enable an additional authorization check for the first data set when
any other data set on the tape volume is opened, the implementation allows your tape
management system to request one or more additional authorization checks when any
data set on a tape volume is opened. Each additional data set name and data set
sequence number returned results in an additional RACROUTE. Do not use this function
unless you have a tape management system and it can return a data set name and data
set sequence number. A data set sequence number is the label number normally specified
in the JCL LABEL keyword and stored in the catalog.
When TAPEAUTHDSN=YES is in use, any additional RACROUTE matches that issued
for TAPEAUTHDSN except for the data set name and data set sequence number.
Otherwise, TAPEAUTHF1 uses a RACROUTE that matches that used for SETROPTS

342
z/OS V1R8 DFSMS Technical Update
TAPEDSN. When neither TAPEAUTHDSN nor SETROPTS TAPEDSN is in use,
TAPEAUTHF1 support is not provided.
TAPEAUTHRC4
Denies accessing of data sets that are not protected by a security profile.
TAPEAUTHRC8
Denies accessing of data sets that typically cannot be accessed.
Use the MVS SET DEVSUP command to implement the new tape data set security settings.
Figure 9-5 shows the successful result of the command.
Figure 9-5 Result of the set DEVSUP command
9.3 Removing TAPEVOL and TAPEDSN processing
In this section we provide instructions for changing from using RACF TAPEVOL and
TAPEDSN to the new tape data set authorization checking.
9.3.1 Check and modify your RACF settings
First check your RACF settings using the RACF SETR LIST command in a TSO session, as
shown in Example 9-3.
Example 9-3 RACF SETR LIST command used in TSO

Command ===> SETR LIST

Example 9-4 shows you the same command used in a batch job.
Example 9-4 RACF SETR LIST command used in a batch job
//SCHLUM JOB (999,POK),MSGLEVEL=1,NOTIFY=&SYSUID
//*
//* *******************************************************************
//* * TESTING Tape Data Set Authorization *
//* *******************************************************************
//SETRLIST EXEC PGM=IKJEFT01
//SYSPRINT DD SYSOUT=*
T DEVSUP=18
IEE252I MEMBER DEVSUP18 FOUND IN SYS1.PARMLIB
IEE536I DEVSUP VALUE 18 NOW IN EFFECT
IEA253I DEVSUP 3480X RECORDING MODE DEFAULT IS COMPACTION.
IEA253I DEVSUP ISO/ANSI TAPE LABEL VERSION DEFAULT IS V3
IEA253I DEVSUP TAPE OUTPUT DEFAULT BLOCK SIZE LIMIT IS 32760
IEA253I DEVSUP COPYSDB DEFAULT IS INPUT
IEA253I DEVSUP TAPEAUTHDSN: YES
IEA253I DEVSUP TAPEAUTHF1: YES
IEA253I DEVSUP TAPEAUTHRC4: FAIL
IEA253I DEVSUP TAPEAUTHRC8: FAIL
Restriction: Only the new tape security parameters are updated using the SET DEVSUP
command. All existing parameters are not updated and cannot be changed without an IPL.

Chapter 9. Tape security
343
//SYSTSPRT DD SYSOUT=*
//SYSTSIN DD *
SETR LIST
/*
The following figures show the output with the current RACF settings. Figure 9-6 shows you
the RACF ATTRIBUTES, STATISTICS, and ACTIVE CLASSES.
Figure 9-6 RACF ATTRIBUTES, STATISTICS, and ACTIVE CLASSES
In Figure 9-7 you can see the active RACF GENERIC PROFILE CLASSES.
Figure 9-7 RACF GENERIC PROFILE CLASSES
ATTRIBUTES = INITSTATS WHEN(PROGRAM -- BASIC)
STATISTICS = DATASET DASDVOL GDASDVOL GTERMINL O2OMPE TAPEVOL TERMINAL
ACTIVE CLASSES = DATASET USER GROUP ACCTNUM ACICSPCT APPCLU APPCPORT APPCSERV
APPCTP APPL BCICSPCT CBIND CCICSCMD CDT CONSOLE CPSMOBJ
CPSMXMP CSFKEYS CSFSERV DASDVOL DCICSDCT DIGTCERT DIGTRING
DSNR ECICSDCT EJBROLE FACILITY FCICSFCT FIELD FSSEC GCICSTRN
GCPSMOBJ GCSFKEYS GDASDVOL GEJBROLE GMQADMIN GSDSF GXFACILI
HCICSFCT IBMOPC ILMADMIN JCICSJCT JESJOBS JESSPOOL KCICSJCT
KEYSMSTR LOGSTRM MCICSPPT MQADMIN NCICSPPT NETCMDS NETSPAN
NODES NODMBR OPERCMDS O2OMPE PCICSPSB PMBR PRINTSRV PROGRAM
PTKTDATA PTKTVAL QCICSPSB RACFVARS RACGLIST RRSFDATA
RVARSMBR SCICSTST SDSF SERVAUTH SERVER STARTED STORCLAS
SURROGAT SYSMVIEW TAPEVOL TCICSTRN TMEADMIN TSOAUTH TSOPROC
UCICSTST UNIXPRIV VCICSCMD VTAMAPPL WRITER XFACILIT
GENERIC PROFILE CLASSES = DATASET ACCTNUM ACICSPCT AIMS ALCSAUTH APPCLU
APPCPORT APPCSERV APPCSI APPCTP APPL CACHECLS
CBIND CCICSCMD CIMS CONSOLE CPSMOBJ CPSMXMP
CSFKEYS CSFSERV DASDVOL DBNFORM DCEUUIDS DCICSDCT
DEVICES DIGTCERT DIGTCRIT DIGTNMAP DIGTRING
DIRACC DIRAUTH DIRECTRY DIRSRCH DLFCLASS DSNADM
DSNR EJBROLE FACILITY FCICSFCT FIELD FILE FIMS
FSOBJ FSSEC GMBR IBMOPC ILMADMIN INFOMAN IPCOBJ
JAVA JCICSJCT JESINPUT JESJOBS JESSPOOL KEYSMSTR
LDAPBIND LFSCLASS LOGSTRM MCICSPPT MDSNBP MDSNCL
MDSNDB MDSNJR MDSNPK MDSNPN MDSNSC MDSNSG MDSNSM
MDSNSP MDSNTB MDSNTS MDSNUF MDSNUT MGMTCLAS
MQADMIN MQCHAN MQCMDS MQCONN MQNLIST MQPROC
MQQUEUE NDSLINK NETCMDS NETSPAN NODES NODMBR
NOTELINK NVASAPDT OIMS OPERCMDS O2OMPE PCICSPSB
PERFGRP PIMS PMBR PRINTSRV PROCACT PROCESS
PROPCNTL PSFMPL PTKTDATA PTKTVAL RACFVARS RACGLIST
RMTOPS RODMMGR ROLE RRSFDATA RVARSMBR SCDMBR
SCICSTST SDSF SECLMBR SERVAUTH SERVER SFSCMD
SIMS SMESSAGE SOMDOBJS STARTED STORCLAS SUBSYSNM
SURROGAT SYSMVIEW TAPEVOL TCICSTRN TEMPDSN
TERMINAL TIMS TMEADMIN TSOAUTH TSOPROC UNIXMAP
UNIXPRIV VMBATCH VMBR VMCMD VMMAC VMMDISK VMNODE
VMPOSIX VMRDR VMSEGMT VTAMAPPL VXMBR WRITER
XFACILIT

344
z/OS V1R8 DFSMS Technical Update
Figure 9-8 shows you the current active RACF GENERIC COMMAND CLASSES.
Figure 9-8 RACF GENERIC COMMAND CLASSES
Figure 9-8 shows you the RACF GENLIST, GLOBAL CHECKING and the settings of the
RACLIST classes.
Figure 9-9 RACF RACLIST CLASSES
GENERIC COMMAND CLASSES = DATASET ACCTNUM ACICSPCT AIMS ALCSAUTH APPCLU
APPCPORT APPCSERV APPCSI APPCTP APPL CACHECLS
CBIND CCICSCMD CIMS CONSOLE CPSMOBJ CPSMXMP
CSFKEYS CSFSERV DASDVOL DBNFORM DCEUUIDS DCICSDCT
DEVICES DIGTCERT DIGTCRIT DIGTNMAP DIGTRING
DIRACC DIRAUTH DIRECTRY DIRSRCH DLFCLASS DSNADM
DSNR EJBROLE FACILITY FCICSFCT FIELD FILE FIMS
FSOBJ FSSEC GMBR IBMOPC ILMADMIN INFOMAN IPCOBJ
JAVA JCICSJCT JESINPUT JESJOBS JESSPOOL KEYSMSTR
LDAPBIND LFSCLASS LOGSTRM MCICSPPT MDSNBP MDSNCL
MDSNDB MDSNJR MDSNPK MDSNPN MDSNSC MDSNSG MDSNSM
MDSNSP MDSNTB MDSNTS MDSNUF MDSNUT MGMTCLAS
MQADMIN MQCHAN MQCMDS MQCONN MQNLIST MQPROC
MQQUEUE NDSLINK NETCMDS NETSPAN NODES NODMBR
NOTELINK NVASAPDT OIMS OPERCMDS O2OMPE PCICSPSB
PERFGRP PIMS PMBR PRINTSRV PROCACT PROCESS
PROPCNTL PSFMPL PTKTDATA PTKTVAL RACFVARS RACGLIST
RMTOPS RODMMGR ROLE RRSFDATA RVARSMBR SCDMBR
SCICSTST SDSF SECLMBR SERVAUTH SERVER SFSCMD
SIMS SMESSAGE SOMDOBJS STARTED STORCLAS SUBSYSNM
SURROGAT SYSMVIEW TAPEVOL TCICSTRN TEMPDSN
TERMINAL TIMS TMEADMIN TSOAUTH TSOPROC UNIXMAP
UNIXPRIV VMBATCH VMBR VMCMD VMMAC VMMDISK VMNODE
VMPOSIX VMRDR VMSEGMT VTAMAPPL VXMBR WRITER
XFACILIT
GENLIST CLASSES = NONE
GLOBAL CHECKING CLASSES = O2OMPE
SETR RACLIST CLASSES = ACCTNUM APPCPORT APPCSERV APPCTP APPL CBIND CDT
CSFKEYS CSFSERV DIGTCERT DIGTRING FACILITY FIELD
ILMADMIN JESSPOOL NETCMDS NODES OPERCMDS O2OMPE
PRINTSRV PTKTDATA PTKTVAL RACFVARS RRSFDATA SDSF
SERVAUTH SERVER STARTED SURROGAT SYSMVIEW TSOAUTH
TSOPROC UNIXPRIV VTAMAPPL WRITER XFACILIT

Chapter 9. Tape security
345
In Figure 9-10 you can see that the tape data set protection is active (TAPEDSN) and that the
high-level qualifier PASSWORD will be added to each single-level qualifier data set that is
used on this system.
Figure 9-10 Other RACF-related information
GLOBAL=YES RACLIST ONLY = NONE
AUTOMATIC DATASET PROTECTION IS NOT IN EFFECT
ENHANCED GENERIC NAMING IS IN EFFECT
REAL DATA SET NAMES OPTION IS INACTIVE
JES-BATCHALLRACF OPTION IS INACTIVE
JES-XBMALLRACF OPTION IS INACTIVE
JES-EARLYVERIFY OPTION IS ACTIVE
PROTECT-ALL OPTION IS NOT IN EFFECT
TAPE DATA SET PROTECTION IS ACTIVE
SECURITY RETENTION PERIOD IN EFFECT IS 9999 DAYS.
ERASE-ON-SCRATCH IS INACTIVE
SINGLE LEVEL NAME PREFIX IS PASSWORD
LIST OF GROUPS ACCESS CHECKING IS ACTIVE.
INACTIVE USERIDS ARE NOT BEING AUTOMATICALLY REVOKED.
NO DATA SET MODELLING BEING DONE.
PASSWORD PROCESSING OPTIONS:
PASSWORD CHANGE INTERVAL IS 180 DAYS.
PASSWORD MINIMUM CHANGE INTERVAL IS 0 DAYS.
MIXED CASE PASSWORD SUPPORT IS NOT IN EFFECT
NO PASSWORD HISTORY BEING MAINTAINED.
USERIDS NOT BEING AUTOMATICALLY REVOKED.
NO PASSWORD EXPIRATION WARNING MESSAGES WILL BE ISSUED.