IBM TotalStorage Migration Guide for the SAP User

An IBM Redbook Publication
IBM Redbook Form Number: SG24-6400-00
ISBN: 0738491659
ISBN: 9780738491653
Publication Date: 28-Feb-2005
Last Update Date: 03-Mar-2005
Find Similar Download

Related People

Dave Lovelace - Author [+3] [-3]
Sabine Kaschca - Author
Bernd Mueller - Author
Joe Meerscheidt - Author

Abstract

This IBM Redbooks publication demonstrates the relative ease with which SAP users can migrate to the IBM TotalStorage Enterprise Storage Server (ESS) 800. It describes how this can be accomplished in numerous ways, across a variety of hardware and software platforms.
The book offers migration scenarios for customers who want to migrate from OEM storage subsystems to the ESS, and it covers a variety of migration tools and methodologies you can use to perform these migrations.
In some cases, the use of software migration tools (such as FDRPAS) from third party vendors is described, while in other cases the use of the IBM SAN Volume Controller (SVC) to migrate users in an open systems environment is detailed, demonstrating the added value of IBM's preferred migration tools, including the SVC and ISV partner tools.

Language

English

Table of Content

Chapter 1. Overview and project summary
Chapter 2. Environment 1 SAP migration experiences using FDRPAS
Chapter 3. Environment 2 SAP migration experiences
Chapter 4. SAP operational activities after migration to ESS
Appendix A. INNOVATION Data Processing Corporation product information
ibm.com/redbooks
Front cover
IBM TotalStorage e
Migration Guide for the de for the
SAP UserUser
Dave Lovelace
Sabine Kaschta
Joe Meerscheidt
Bernd Mueller
Learn about SAP operational tasks
after migrating to IBM ESS
Become familiar with tools like
FDRPAS to migrate easily
Learn how to use the IBM SAN
Volume Controller to migrate


International Technical Support Organization
IBM TotalStorage Migration Guide for the SAP User
January 2005
SG24-6400-00

© Copyright International Business Machines Corporation 2005. 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 (January 2005)
Note: Before using this information and the product it supports, read the information in “Notices” on
page vii.

© Copyright IBM Corp. 2005. All rights reserved.
iii
Contents
Notices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .vii
Trademarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix
The team that wrote this Redbook. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix
Become a published author. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .x
Comments welcome. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .x
Chapter 1. Overview and project summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Objective of this book . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 SAP migration environment 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2.1 Software used for environment 1 SAP migration . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2.2 Hardware used in the environment 1 SAP migration . . . . . . . . . . . . . . . . . . . . . . . 2
1.3 SAP migration environment 2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.3.1 Software used in environment 2 SAP migration . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3.2 Hardware used in the environment 2 SAP migration . . . . . . . . . . . . . . . . . . . . . . . 3
1.4 IBM TotalStorage ESS 800 overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.4.1 IBM TotalStorage Resiliency Family. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.4.2 ESS 800 heterogeneous support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.4.3 ESS 800 highlights . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.4.4 Benefits of the ESS 800 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.4.5 ESS Copy Services. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.4.6 FlashCopy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.4.7 ESS Specialist . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.5 IBM TotalStorage SAN Volume Controller overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.5.1 Storage virtualization and the SVC. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.5.2 SVC physical components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.5.3 SVC compatibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.5.4 The SAN. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.5.5 SVC multi-pathing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.5.6 SVC logical components. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.5.7 SVC configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.5.8 Explanation of terms. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Chapter 2. Environment 1 SAP migration experiences using FDRPAS. . . . . . . . . . . . 15
2.1 FDRPAS overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.1.1 Supported hardware. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.1.2 Supported software. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.1.3 How does it work?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.1.4 Single system versus multiple system operation . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.1.5 FDRPAS Tasks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.2 Install FDRPAS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.2.1 Basic installation steps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.2.2 The FDRPAS INSTALL ISPF dialog. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.2.3 Additional installation tasks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
2.3 FDRPAS concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
2.4 FDRPAS SWAP via batch job. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
2.4.1 Job control requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
2.4.2 SWAP/SWAPDUMP/SIMSWAP statement. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

iv
IBM TotalStorage Migration Guide for the SAP User
2.4.3 EXCLUDE statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
2.4.4 HISTORY statement. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
2.4.5 LICENSE statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
2.4.6 MONITOR statements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
2.4.7 MOUNT statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
2.4.8 SWAPBUILDIX statement. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
2.4.9 SIMSWAP scenario . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
2.4.10 FDRPAS SWAP: one volume, single system . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
2.4.11 FDRPAS sequential swap of five volumes on four systems. . . . . . . . . . . . . . . . 52
2.4.12 FDRPAS concurrent swap of five volumes on four systems. . . . . . . . . . . . . . . . 54
2.4.13 FDRPAS concurrent swap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
2.5 FDRPAS through ISPF Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
2.6 FDRERASE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
2.6.1 Considerations for erasing data from modern disks . . . . . . . . . . . . . . . . . . . . . . . 65
2.6.2 Types of erase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
2.6.3 FDRERASE JCL requirements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
2.6.4 FDRERASE control statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
2.6.5 Usage of FDRERASE via ISPF panels. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
Chapter 3. Environment 2 SAP migration experiences. . . . . . . . . . . . . . . . . . . . . . . . . 79
3.1 Environment 2 SAP migration using SVC. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
3.1.1 ESS 800 Open Systems configuration for AIX . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
3.2 FlashCopy process. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
3.3 Using IBM TotalStorage SAN Volume Controller to migrate. . . . . . . . . . . . . . . . . . . . . 97
3.4 General SVC planning guidelines. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
3.4.1 Sizing the SVC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
3.4.2 Physical planning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
3.4.3 Network and IP addresses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
3.4.4 Naming convention. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
3.5 Client host preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
3.5.1 Device drivers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
3.5.2 Hardware support. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
3.5.3 SDD installation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
3.6 General SAN planning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
3.6.1 SAN zones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
3.6.2 SVC Master Console zone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
3.7 SVC planning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
3.8 Back-end disk support. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
3.8.1 ESS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
3.8.2 DS4000(FAStT) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
3.9 Sample configuration planning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
3.9.1 Sample vdisk mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
3.10 Configuring the ESS 800 and SVC for migration . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
3.10.1 Modify hosts systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
3.10.2 Modify Volume Assignments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
3.10.3 Configuring the SAN Volume Controller . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
3.10.4 Start the SVC application . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
3.10.5 Creating a host . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
3.10.6 Creating managed disks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
3.10.7 Creating managed disk groups. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
3.11 SVC migration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
3.11.1 Migrating a vDisk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
3.11.2 Migrating a vDisk between managed disk groups. . . . . . . . . . . . . . . . . . . . . . . 119

Contents
v
3.12 Install and configure the TSM Client. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
3.13 Installation of the TSM Server. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
3.13.1 Basic configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
3.13.2 Installation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
3.13.3 Configuration of the server . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
3.13.4 Starting and Stopping the Server . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
3.13.5 Starting and stopping an administrative client or the Web interface. . . . . . . . . 126
3.13.6 Increasing disk storage pool sizes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
3.13.7 Back up data for migration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
3.13.8 Restore to migrate data to the IBM ESS Storage Subsystem. . . . . . . . . . . . . . 130
3.14 Environment 2 migration experiences: FDR/UPSTREAM. . . . . . . . . . . . . . . . . . . . . 131
3.14.1 Supported client environments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
3.14.2 Supported databases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
3.14.3 Client-server application . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
3.14.4 FDR/UPSTREAM Reservoir server components . . . . . . . . . . . . . . . . . . . . . . . 133
3.14.5 FDR/UPSTREAM client-based components. . . . . . . . . . . . . . . . . . . . . . . . . . . 134
3.14.6 Installation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
3.15 Configuring FDR/UPSTREAM Reservoir . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
3.15.1 Removable Storage Manager. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
3.15.2 FDR/UPSTREAM Reservoir overall parameter configuration. . . . . . . . . . . . . . 145
3.15.3 Starting FDR/UPSTREAM Reservoir service . . . . . . . . . . . . . . . . . . . . . . . . . . 145
3.16 Start the client service. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
3.17 Backup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
3.18 Restore the data to IBM ESS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
3.18.1 Use FDR/UPSTREAM Reservoir for restoring . . . . . . . . . . . . . . . . . . . . . . . . . 152
3.18.2 Monitor your tasks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
3.18.3 Summary of the backup process. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
Chapter 4. SAP operational activities after migration to ESS. . . . . . . . . . . . . . . . . . . 157
4.1 Tools to manage and monitor your database tasks . . . . . . . . . . . . . . . . . . . . . . . . . . 158
4.1.1 SAP R/3 DBA Planning Calendar. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
4.1.2 IBM DB2 UDB Control Center. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
4.2 Online backup of the database. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
4.2.1 Use SAP R/3 DBA Planning Calendar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
4.2.2 Use the IBM DB2 UDB Control Center. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
4.2.3 Use the IBM DB2 command line. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166
4.3 Offline backup of the database. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168
4.3.1 Use SAP R/3 DBA Planning Calendar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168
4.3.2 Use IBM DB2 UDB Control Center. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168
4.3.3 Use the IBM DB2 UDB command line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
4.4 Archive inactive logs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
4.4.1 Use SAP R/3 DBA calendar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
4.5 Back up tablespaces. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
4.5.1 Use the IBM DB2 UDB Control Center. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
4.5.2 Use the IBM DB2 UDB Command Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
4.6 Restore databases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174
4.6.1 Use the IBM DB2 Control Center . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174
4.6.2 Use the IBM DB2 command line. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174
4.7 Restore tablespaces. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176
4.7.1 Use the IBM DB2 UDB Control Center. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176
4.7.2 Use the IBM DB2 UDB command line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177
4.8 Other daily activities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179

vi
IBM TotalStorage Migration Guide for the SAP User
4.8.1 Maintain statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
4.8.2 Update table statistics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
4.8.3 Reorganize tables. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
4.8.4 Status of activities and tasks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182
4.8.5 Database performance: Tables and indexes . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
4.8.6 Overview of backup status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
4.8.7 Database Alert Monitor. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
4.8.8 Database check . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
4.8.9 Database Alert Monitor/System Check. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
4.8.10 Database performance monitor. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194
4.8.11 Database snapshot analysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
4.8.12 Deadlocks analysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196
4.8.13 Exclusive Lock Waits analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198
4.8.14 Monitor scheduled jobs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198
4.8.15 DB2 UDB diag log. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200
4.8.16 Transactions in an SAP R/3 environment with IBM DB2 UDB on zOS. . . . . . . 200
4.8.17 Summary of regular database activities and tasks . . . . . . . . . . . . . . . . . . . . . . 204
Appendix A. INNOVATION Data Processing Corporation product information. . . . . 205
FDRPAS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206
FDR/UPSTREAM Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
FDRERASE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266
Syntax For FDR Products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281
How to read the syntax diagrams. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281
Related publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283
IBM Redbooks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283
Other publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283
Online resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283
How to get IBM Redbooks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284
Help from IBM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285

© Copyright IBM Corp. 2005. All rights reserved.
vii
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 on the products and services currently available in your area.
Any reference to an IBM product, program, or service is not intended to state or imply that only that IBM
product, program, or service may be used. Any functionally equivalent product, program, or service that does
not infringe any IBM intellectual property right may be used instead. However, it is the user's responsibility to
evaluate and verify the operation of any non-IBM product, program, or service.
IBM may have patents or pending patent applications covering subject matter described in this document. The
furnishing of this document does not give you any license to these patents. You can send license inquiries, in
writing, to:
IBM Director of Licensing, IBM Corporation, North Castle Drive Armonk, NY 10504-1785 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 illustrates 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. 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 IBM's application programming interfaces.

viii
IBM TotalStorage Migration Guide for the SAP User
Trademarks
The following terms are trademarks of the International Business Machines Corporation in the United States,
other countries, or both:
Eserver®
Eserver®
Redbooks (logo) ™
eServer™
ibm.com®
iSeries™
pSeries®
xSeries®
z/OS®
zSeries®
AIX®
AS/400®
Domino®
DB2 Universal Database™
DB2®
Enterprise Storage Server®
FlashCopy®
Informix®
IBM®
Lotus Notes®
Lotus®
MVS™
Notes®
OS/2®
OS/390®
Redbooks™
RACF®
RAMAC®
RS/6000®
S/390®
Seascape®
Tivoli®
TotalStorage®
Virtualization Engine™
The following terms are trademarks of other companies:
Java and all Java-based trademarks and logos are trademarks or registered trademarks of Sun
Microsystems, Inc. in the United States, other countries, or both.
Microsoft, Windows, and the Windows logo are trademarks of Microsoft Corporation in the United States,
other countries, or both.
FDRPAS, FDR Plug and Swap, FDRINSTANT are trademarks of INNOVATION Data Processing Corporation.
UPSTREAM Reservoir, UPSTREAM RESERVOIR, FDRERASE are copyrighted by INNOVATION Data
Processing Corporation.
Intel and Intel Inside (logos) are trademarks of Intel Corporation in the United States, other countries, or both.
UNIX is a registered trademark of The Open Group in the United States and other countries.
Linux is a trademark of Linus Torvalds in the United States, other countries, or both.
Other company, product, and service names may be trademarks or service marks of others.

© Copyright IBM Corp. 2005. All rights reserved.
ix
Preface
This IBM® Redbook demonstrates the relative ease with which SAP users can migrate to the
IBM TotalStorage® Enterprise Storage Server® (ESS). It describes how this can be
accomplished in numerous ways, across a variety of hardware and software platforms.
The book offers migration scenarios for customers who want to migrate from OEM storage
subsystems to the ESS, and it covers a variety of migration tools and methodologies you can
use to perform these migrations.
In some cases, the use of software migration tools (such as FDRPAS) from third party
vendors is described, while in other cases the use of the IBM SAN Volume Controller (SVC)
to migrate users in an open systems environment is detailed, demonstrating the added value
of IBM's preferred migration tools, including the SVC and ISV partner tools.
The team that wrote this Redbook
This Redbook was produced by a team of specialists from around the world.
Dave Lovelace is a Project Leader at the International Technical Support Organization, San
Jose Center. Before joining the ITSO, Dave worked in IBM’s Storage Systems Group as
Program Director of Business Development and Strategic Alliances. In that capacity Dave
worked closely with Independent Software vendors to create remarketing agreements and he
was also responsible for a number of Software OEM activities. With more than 30 years in the
IT industry, Dave has held positions as an IBM Systems Engineer, an MVS™ Systems
Programmer, as well as in OS/390® marketing, Software Packaging (CBIPO) and several
years working in client server environments with the US Military Academy at West Point.
Sabine Kaschta is a DB2® Specialist working for IBM IT Education Services in Germany as
an education consultant. She has 13 years of experience working with DB2. Before she
joined IBM in 1998, she worked for a third-party vendor providing second-level support for
DB2 utility products. She is also experienced in DB2 system programming and client/server
implementations within an insurance practice in Germany. She is a co-author of four other
IBM Redbooks™ about DB2 topics.
Joe Meerscheidt is a Storage I/T Specialist with Solutions-Advanced Technical Support
(ATS) Americas, supporting Storage Alliance ISVs. Before becoming a Storage I/T Specialist
three years ago, Joe worked as an xSeries® server I/T Specialist supporting the
eServer/Ariba ISV alliance. He was also a Project Lead/Team Lead for IGS supporting IGS
SAP outsourcing contracts. Joe has more than 20 years of experience in the IT industry, and
he has worked on many IBM hardware and software platforms including Intel/xSeries,
System38, AS/400® - iSeries™, RS/6000® - pSeries®, MVS, and Mapics.
Bernd Mueller is an IT specialist in presales e-business Technical Sales Support for IBM
eServer™ pSeries. He is an SAP Technical Consultant and provides support for customer
machines running in an SAP environment. He has over 8 years experience with AIX® and
pSeries in SAP landscapes. Bernd is a member of the EMEA industrial focus group team
supporting IBM sales teams. His areas of expertise include pSeries, ORACLE, DB2 UDB,
and SAP Basis.

x
IBM TotalStorage Migration Guide for the SAP User
Thanks to the following people for their contributions to this project:
The Dallas team was very responsive to our project needs and their help is greatly
appreciated.
George Clark, Dan Pack, Tim Raley
IBM Developer Technical Support Center, Dallas, Texas
Jon Tate, Sangam Rachlera, Mary Lovelace
International Technical Support Center
Tom Meehan, Lou D’Agnola, Fred Lama, Eva Minnocci, Bruce Black
INNOVATION Data Processing Corporation
Linda K Paschal
ISG Marketing, IBM Systems Group
Eva Lau
ISV Alliances & Business Development
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'll team with IBM technical professionals, Business Partners and/or
customers.
Your efforts will help increase product acceptance and customer satisfaction. As a bonus,
you'll 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
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 redbook form found at:
ibm.com/redbooks
Send your comments in an email to:
redbook@us.ibm.com
Mail your comments to:
IBM Corporation, International Technical Support Organization
Dept. QXXE Building 80-E2
650 Harry Road
San Jose, California 95120-6099

© Copyright IBM Corp. 2005. All rights reserved.
1
Chapter 1.
Overview and project summary
This chapter describes the objectives of the migration project and the various hardware and
software environments used to migrate SAP users from non-IBM storage subsystems to IBM
TotalStorage ESS 800.
1

2
IBM TotalStorage Migration Guide for the SAP User
1.1 Objective of this book
The objective of this book is to demonstrate the relative ease with which SAP users can
migrate to IBM TotalStorage ESS 800. We describe how this can be accomplished in
numerous ways across a variety of hardware and software platforms.
In this chapter we describe the test environments we set up in the lab to provide practical
migration scenarios.
1.2 SAP migration environment 1
This environment consisted of an SAP database server using IBM DB2 on z/OS® with the
SAP application server on an RS/6000 running AIX. In this migration environment the
database for SAP was residing on a non-IBM storage device and was then migrated to an
IBM TotalStorage ESS 800 non-disruptively using an ISV product from Innovation Data
Corporation called FDRPAS. For purposes of this discussion the term
non-disruptive
means
that the normal workload continues while the data is migrated to the IBM TotalStorage ESS
800. End users and running applications continue to operate as normal, there is no need to
re-IPL the z/OS system, and the SAP workload runs normally during the migration.
1.2.1 Software used for environment 1 SAP migration
The software used in the environment 1 migration is identified in this section.
Software used for the SAP database server
z/OS
IBM DB2 R7
FDRPAS
FDRERASE
Software used for the SAP application server
AIX 5.1
SAP
1.2.2 Hardware used in the environment 1 SAP migration
The following hardware was used in the environment 1 migration:
A non-IBM storage subsystem was used to contain the SAP database prior to migration
The SAP database was then migrated to an IBM TotalStorage ESS 800
The database host was an IBM Z800 processor
The SAP application server was hosted on an IBM RS/6000 processor
1.3 SAP migration environment 2
Environment 2 used an AIX operating system to control the SAP database server and the
SAP application server. In environment 2 we followed several migration scenarios using a
Note: We used FDRPAS, a non-disruptive migration software product from Innovation data
Processing, because IBM has this product under contract for is in certain migration
offerings for IBM customers.

Chapter 1. Overview and project summary
3
variety of software and hardware options. Two scenarios were performed with the use of
backup/restore software products. We also describe how we migrated SAP data from
non-IBM storage to the IBM TotalStorage ESS 800 using an IBM SAN Volume Controller.
1.3.1 Software used in environment 2 SAP migration
This section describes the software used for backup/restore scenarios.
IBM Tivoli® Storage Manager was used to back up the SAP database when it was hosted on
a non-IBM storage device and then restored to an IBM TotalStorage ESS 800.
In another migration we demonstrated how to use Innovation Data Corporation’s
FDR/Upstream to back up the SAP database from a non-IBM storage subsystem and restore
the database to the IBM TotalStorage ESS 800.
1.3.2 Hardware used in the environment 2 SAP migration
The following hardware was used in the environment 2 SAP migration:
RS/6000 processor
IBM San Volume Controller
Various tape devices
1.4 IBM TotalStorage ESS 800 overview
In this section we introduce the IBM TotalStorage Enterprise Storage Server Model 800
(ESS 800), the IBM Seascape® architecture, and the IBM TotalStorage Resiliency family. We
also discuss some of the benefits that can be achieved when using the IBM TotalStorage
Enterprise Storage Server Model 800. Finally, we define some basic terms used in this
redbook.
For detailed information on the IBM TotalStorage Enterprise Storage Server Model 800, refer
to IBM TotalStorage Enterprise Storage Server Model 800, SG24-6424.
E-business is driving a data explosion, generating exponential growth in the need for
immediately accessible, always available, and highly functional storage capacity. The IBM
TotalStorage Enterprise Storage Server Model 800 provides extensive heterogeneous server
connectivity and supports rapid universal access to vast quantities of data through many
advanced functions and features. Because of its enterprise-wide support and management
scope, the ESS is tailor made to help provide consistent, efficient, and effective enterprise
resource planning. The ESS 800 is a solution that has the outboard intelligence required by
Storage Area Network (SAN) solutions for offloading key functions for the host servers, which
frees up valuable CPU for the applications. As a comprehensive SAN-based storage solution,
the ESS provides considerable management flexibility to meet the fast-paced requirements of
today’s businesses.
The IBM TotalStorage Enterprise Storage Server Model 800 provides significantly improved
levels of performance, throughput, and scalability over its predecessor E and F models. The
ESS 800 provides enhancements that increase the overall value of the ESS in the
marketplace and improves on the strategic SAN initiatives. The move to e-business on
demand presents customers with both significant opportunities and critical challenges. To
address the unique requirements of the on demand world, companies must deploy high
performance, autonomic, intelligent storage technologies that can offer data protection and
functions to support near continuous availability. The IBM TotalStorage ESS 800 is designed
to help set new standards in performance, automation, and integration; as well as capabilities

4
IBM TotalStorage Migration Guide for the SAP User
to support near continuous availability to data for the on demand world. The ESS 800 also
supports many advanced copy functions to help increase data availability during planned
outages and protect data from planned and unplanned outages.
Some of the more significant enhancements are:
2Gb Fibre Channel/Ficon host adapter
Up to 64 GB of cache
2 GB NVS (non-volatile storage) with double the bandwidth
More powerful processors with turbo option
Improvements in interoperability
Improved and expanded advanced copy functions and features
Figure 1-1 Enterprise System Storage Server (ESS) Model 800
1.4.1 IBM TotalStorage Resiliency Family
The IBM TotalStorage Resiliency Family is a comprehensive set of integrated solutions that
are automated and optimized to address any customer’s business continuity needs. The
TotalStorage Resiliency Family is designed to provide server awareness and exploitation as
well as application and database awareness and exploitation of core technologies to help
improve business continuity. The IBM TotalStorage Resiliency Family is designed to provide
total cost of ownership savings through high application availability and improved productivity
from storage IT personnel. With information on demand, businesses can respond with
flexibility and speed to any customer requirement, market opportunity, or threat.
The IBM TotalStorage Resiliency Family is designed to:
Maintain access to data at all times
Protect critical business data
Match disaster recovery costs to the value of the application
Help reduce backup window times
Help improve storage IT personnel productivity
Refer to Figure 1-2 for an overview of the IBM TotalStorage Resiliency Family.

Chapter 1. Overview and project summary
5
Figure 1-2 IBM TotalStorage Resiliency Family
1.4.2 ESS 800 heterogeneous support
The ESS 800 supports storage sharing for a wide variety of heterogeneous operating
environments. Many types of server platforms can concurrently attach to the ESS 800,
including IBM eServer iSeries and AS/400; Linux®, Novell Netware, Windows® NT, Windows
2000, Microsoft® Windows Server 2003; IBM eServer zSeries® and S/390®, IBM eServer
pSeries and xSeries, and many other types of UNIX® Servers.
Refer to Figure 1-3 for a diagram of the ESS 800 heterogeneous support.
For a complete list of platforms that are supported on the ESS 800 refer to the ESS
interoperability matrix at:
http://www.storage.ibm.com/disk/ess/pdf/interop.pdf

6
IBM TotalStorage Migration Guide for the SAP User
Figure 1-3 ESS heterogeneous support
1.4.3 ESS 800 highlights
Among the capabilities of the ESS 800 are that it:
Supports storage for a wide variety of heterogeneous environments
Is designed to provide outstanding performance scaling up to 55.9TB physical capacity
Offers high availability and resiliency to enterprise mission-critical business applications
Supports fast data transfer rates for mixed environments
Is SAN ready
Supports RAID5 or RAID10
Helps increase administrative productivity through efficient, centralized management
Provides significant improvements in ESS Copy Services, including:
– Global Mirror (Asynchronous PPRC)
– Metro Mirror (Synchronous PPRC)
– FlashCopy® to a PPRC Primary volume
1.4.4 Benefits of the ESS 800
Benefits of using the ESS 800 include the following:
Extensive connectivity - The ESS provides simultaneous attachment to a wide variety of
host server with storage capacity partitioned among the attached servers.
Advanced Copy Services functions - The ESS supports many advanced copy functions for
the purposes of mirroring for disaster recovery and providing a near instantaneous copy of
data.
HP
Compaq/
DEC
Sun
Enterprise
Storage Server
Web/GUI
Storage
Management
Network
xSeries
zSeries
iSeries
AS400
pSeries
RS6000
RS/6000
SP
DG
PCs
Windows NT and
Windows 2000
Novell
Other platforms
to be announced
NUMA-Q
Netfinity
SGI
Compaq

Chapter 1. Overview and project summary
7
Availability for applications and e-business 24x7-365 - The ESS is designed with
redundant power, cooling, adapters, buses, and processors. Redundancy of all
components means the ESS has virtually no single point of failure in regard to hardware
components.
Performance - The ESS can be configured to provide the capacity and performance
required for applications to support over 75,000 I/Os per second. The new OS/390 options
allow multiple simultaneous I/O operations to the same logical volume, thereby
significantly reducing IOS queue time.
Low total cost of ownership - Key features such as advanced business continuance
functions, performance, scalability, the ability to mix and match disk drive capacities and
speeds, and heterogeneous connectivity and flexibility make the ESS an excellent choice
to help lower the cost of ownership.
Scalability - The ESS can adapt to growing storage needs created by e-business, with
scalability from .5GB up to 55.9TB.
Storage Area Network (SAN) - The ESS offers customers the leading storage subsystem,
supporting the key server platforms in the industry and enabling them to realize the
promise of non-proprietary SANs.
1.4.5 ESS Copy Services
The ESS 800 supports many advanced hardware copy functions for the purposes of mirroring
for disaster recovery and copy functions that provide an instate copy of the data. ESS Copy
Services is an optional feature of the IBM TotalStorage ESS 800.
The major copy functions provided by the ESS 800 are the following:
FlashCopy V1
FlashCopy V2
Peer-to-Peer Remote Copy (PPRC) V1
– PPRC Metro Mirror (Synchronous)
– PPRC Global Copy (Extended distance)
Peer-to-Peer Remote Copy (PPRC) V2
– PPRC Global Mirror (Asynchronous)
– PPRC Metro/global Mirror (Asynchronous cascading)
Extended Remote Copy (XRC)
– zSeries Global Mirror
– zSeries Metro/Global Mirror
1.4.6 FlashCopy
FlashCopy makes a single point-in-time or time-zero copy of a LUN. A point-in-time copy
(target) gives you an instantaneous copy of the original data (the source). When the
FlashCopy task is invoked, the process of establishing the FlashCopy Pair and the creation of
the control bitmaps takes only a few seconds to complete. Once complete, you have access
to a point-in-time (time-zero) copy of the (source) data or volume. As soon as the FlashCopy
pair has been established, you can read and write to both the source and target volumes.

8
IBM TotalStorage Migration Guide for the SAP User
1.4.7 ESS Specialist
The ESS Specialist is the Web user interface that is included with the ESS and is used for
viewing the current status of the ESS and modifying the logical configuration. The ESS
Specialist and ESS Copy Services interfaces are sets of Java™ applets that execute within a
Web browser. The Java applets communicate with the microcode running on the ESS
clusters to retrieve the current configuration data, submit the requested configuration change,
and display the outcome of a request. A maximum of 8 ESSs can be defined in the Specialist.
See ESS Specialist Example below Figure 1-4
Figure 1-4 ESS (Storwatch) Specialist
Use the following steps to access and log on to the ESS Specialist:
1.In the browser address field enter either of the URLs (IP addresses) that are assigned to
the two ESS clusters.
2.Accept the security certificates.
3.Click ESS Specialist in the navigation frame of the ESS Launch page. A login dialog box
will appear.
4.If accessing the ESS Specialist for the first time use the following the default userid and
password:
a.Enter storwatch in the User Name field.
b.Enter specialist in the Password field.
c.Click OK.
1.5 IBM TotalStorage SAN Volume Controller overview
This section describes the IBM TotalStorage SAN Volume Controller (SVC).
Note: You should change the User Name and Password to something else as soon as
possible.

Chapter 1. Overview and project summary
9
1.5.1 Storage virtualization and the SVC
The IBM TotalStorage SAN Volume Controller is a scalable hardware and software solution
that allows aggregation of storage from different disk subsystems. It provides storage
virtualization and thus a consistent view of storage across a Storage Area Network (SAN).
See Figure 1-5 for a visual representation of the IBM Virtualization Engine™.
Figure 1-5 IBM Virtualization Engine
The SAN Volume Controller provides in-band storage virtualization by creating a pool of
managed disks from attached back-end disk storage subsystems. These managed disks are
then mapped to a set of virtual disks for use by various host computer systems. The hosts
see the virtual disks as available storage volumes within which they will direct the placement
of a client’s data. The hosts access their data from these storage volumes, not knowing that
they are actually SVC-managed virtual disks. This is illustrated in Figure 1-7 on page 12.
1.5.2 SVC physical components
A
node
is a single storage engine; the IBM TotalStorage SAN Volume Controller Storage
nodes are main hardware elements of the SVC, and up to four nodes can be installed
in
pairs
. These four nodes are managed as a set (cluster) and present a single point of control
to the administrator for configuration and service activity. In future releases of the SVC, the
cluster size (or number of nodes) will be increased to permit greater scalability.
Within each cluster, one node will be defined as the
configuration node
. This node will be
assigned the cluster IP address and will be responsible for transitioning additional nodes into
the cluster.
For I/O purposes, SVC nodes within the cluster are grouped into pairs called
I/O Groups
, with
a single pair being responsible for serving I/O on a given virtual disk.
One node within the I/O Group will represent the preferred path for I/O to a given virtual disk.
The other node becomes the non-preferred or alternate path for the virtual disk. The selection
of the node that will become the preferred node alternates between each of the nodes within
an I/O Group as each virtual disk is created, thus balancing the workload evenly between the
two nodes.

10
IBM TotalStorage Migration Guide for the SAP User
The SVC is supplied with a pair of Uninterruptible Power Supplies (UPS), as well as voltage
regulation to protect valuable electronic components within the SVC configuration in the
event of a main power outage. The UPS provides enough power to de-stage data to the SVC
internal disk and shut down the nodes within the SVC cluster gracefully.
The SAN Volume Controller provides a Master Console that is used as a single platform to
configure, manage, and service the SAN Volume Controller. The console comes
pre-configured with a number of software products. The Master Console allows a system
administrator to monitor the configuration of the SVC system and all of the internal
components, for example, SNMP trap management, Call Home and Remote Service facilities,
diagnostic utilities, and logical configuration.
1.5.3 SVC compatibility
The SVC is capable of supporting many Host platforms, including Windows NT/2000, AIX,
Red Hat Linux, Sun Solaris, and HP-UX hosts, as well as others. The SAN switch support is
also very broad and includes the IBM TotalStorage SAN Switches and members of Cisco,
McDATA, and CNT families. Support of disk subsystems includes, but is not limited to, the
IBM TotalStorage DS4000(DS4000(FAStT)) Storage Server and the IBM TotalStorage,
Enterprise Storage Server (ESS), as well as HP MA8000 and HDS 9200. Future releases of
the SVC will include additional support for third party disk subsystems.
For the latest supported configurations refer to the following Web site:
http://www.ibm.com/storage/support/2145
1.5.4 The SAN
A Storage Area Network (SAN) is a high-speed dedicated network for sharing storage
resources. The SAN allows the establishment of direct connections between storage devices,
the SVC, and host servers.
A
redundant SAN
is a SAN configuration in which any one single component may fail, and
connectivity between the devices within the SAN is maintained, possibly with degraded
performance. This is normally achieved by splitting the SAN into two independent counterpart
SANs (two SAN fabrics). This provides multiple paths to the same storage device or LUN.
IBM highly recommends that a redundant SAN be used with the SAN Volume Controller (a
non-redundant SAN is supported) and the metadata servers (MDS).
The SVC, its back-end storage, and SAN File System Host Clients, are all connected to the
same redundant SAN. The different components’ view of the SAN is carefully controlled by
using various SAN zones.
To install a SAN Volume Controller into an existing SAN, you must first ensure that the switch
zoning is set to isolate the new SAN Volume Controller connections from the currently active
parts of the SAN.
1.5.5 SVC multi-pathing
Each SVC node presents a vdisk to the host bus adapters (HBAs) through multiple SAN
paths, usually four. Most operating systems are not able to resolve multiple images of the
same device through multiple paths, so IBM provides a multi-pathing device driver for the
ESS called the Subsystem Device Driver (SDD). For the DS4000(FAStT), IBM provides a
multi-pathing driver called Redundant Disk Array Controller (RDAC).

Chapter 1. Overview and project summary
11
Most operating systems present each path to a virtual disk as a separate storage device. The
SAN Volume Controller, therefore, needs the IBM Subsystem Device Driver software to be
running on the host. This software handles the many paths that are available to the virtual
disk, and presents a single storage device to the operating system. SDD will not coexist on a
host with other multi-pathing drivers such as the LSI RDAC driver for DS4000(FAStT).
SDD manages the multiple paths from the host to the SVC, making use of the preferred paths
in a round-robin manner before using any non-preferred path. SDD performs data path
failover in the event of a failure within the SVC, or the host path, while also masking out the
additional disks that would otherwise be seen by the hosts due to the redundant paths
through the SAN fabric.
1.5.6 SVC logical components
The SVC provides block aggregation and volume management for disk storage within the
SAN. In simpler terms, this means that the SVC manages a number of back-end storage
controllers and maps the physical storage within those controllers to logical disk images that
can be seen by application servers and workstations in the SAN.
The SAN is zoned in such a way that the application servers cannot see the back-end
storage, preventing any possible conflict between the SVC and the application servers both
trying to manage the back-end storage.
The SVC is connected to the SAN in such a way that all back-end storage is visible to all of
the SVC node Fibre Channel (FC) ports. All of the SVC FC ports are also visible to the
application servers.
The SVC will detect the storage presented to the SAN by the back-end controllers as a
number of disks, known as
managed disks
or
mdisks
. Because the SVC does not attempt to
provide recovery from physical disk failures within the back-end controllers, mdisks are
usually, but not necessarily, part of a RAID array. The application servers, on the other hand,
do not see the mdisks at all.
Instead, they see a number of logical disks, known as
virtual disks
or
vdisks
, which are
presented to the SAN by the SVC.
Mdisks are collected into groups, known as
managed disk groups
(MDG). The mdisks that are
used in the creation of a particular vdisk must all come from the same managed disk group.
Each mdisk is then divided up into a number of
extents
(each with a default minimum size of
16 MB, maximum size 512 MB), which are numbered sequentially from the start to the end of
the mdisk. It is these extents that are used to create a virtual disk. This is shown in Figure 1-6.
Figure 1-6 Mdisk extents within a group create striped vdisks

12
IBM TotalStorage Migration Guide for the SAP User
The virtualization function in the SVC maps the vdisks seen by the application servers onto
the mdisks provided by the back-end controllers. I/O traffic for a particular vdisk is, at any one
time, handled exclusively by the nodes in a single I/O Group. Thus, although a cluster can
have many nodes within it, the nodes handle I/O in independent pairs. This means that the
I/O capability of the SVC scales well (almost linearly), since additional throughput can be
obtained by simply adding additional I/O Groups.
1.5.7 SVC configuration
The mapping of LUNs to mdisks, mdisks to vdisks, and vdisks to the hosts is shown in
Figure 1-7. The diagram does not show the SAN fabric components.
Figure 1-7 SVC mapping configuration
1.5.8 Explanation of terms
This section defines some of the important terms used throughout this redbook.
Cluster
The nodes in an SVC configuration are clustered together for the purposes of failover, and to
facilitate administration of the SVC installation. Nodes are paired together in I/O Groups for
failover, so SVC cluster members are not entirely interchangeable.
Boss node
A single node acts as the boss node for overall management of the cluster. If the boss node
fails, another node in the cluster will take over the responsibilities.
Virtual Disks: Associated
with an I/O Group and
provisioned to hosts. They
are created from extents
in Managed Disk Groups.
Managed Disks Group:
Pool of extents created
from managed disks.
Managed Disks: Allocated
to SVC from RAID
controllers.
User creates Managed
Disk Groups (MDGs) and
assigns Managed Disks
based on performance
and reliability
characteristics desired
User creates virtual disks
(VDISKs), associated with
I/O Group, and assigns to
host
Automatic discovery of
LUNs allocated to SVC
from RAID controllers
Mapping from Virtual Disks Managed Disks:
1.Stripe extents across multiple Managed Disks
2.Sequentially group across one or more MDisks
3.Image mode: one-to-one mapping of MDisks to VDisks
RAID Controller(s)
SCSI LUNs
RAID Arrays
Managed

Disks
Managed Disk Groups
Virtual Disks
Virtual disks appear to the
hosts as SCSI targets
supporting the SCSI
command set
Storage Virtualization Engine
MDG 1
MDG 2
Hosts

Chapter 1. Overview and project summary
13
Configuration node
At any one time, a single node in the cluster is used to manage logical configuration activity.
This configuration node manages a cache of the configuration information that describes the
cluster configuration, and provides a focal point for configuration commands.
Extent
An extent is a fixed size unit of data that is used to manage the mapping of data between
mdisks and vdisks.
Back-end storage
The SVC mdisks are logically located on the “back-end” storage sub-system, for example, an
ESS or DS4000(FAStT).
I/O Group
A group of two SVC nodes defined by the configuration process. Each SVC node is
associated with exactly one I/O Group. The nodes in the I/O Group provide access to the
vdisks in the I/O Group.
Image mode vdisk
Image mode provides a direct block-for-block translation from the storage subsystem LUN
(mdisk) to the virtual disk. This mode is used to allow virtualization of mdisks that already
contain data and were written by a “pre-virtualized” subsystem,
not
through an SVC node.
When an image mode vdisk is created, it maps directly to the mdisk it is created from.
Image mode allows a customer to insert an SVC into the data path of an existing storage
configuration without moving any data, therefore, with minimal downtime.
LUN
In this redbook, when we refer to a logical unit number (LUN) we are referring to the unit of
storage that is presented by a storage subsystem.
This means it is used interchangeably for a unit defined in a back-end storage subsystem
such as an ESS or DS4000(FAStT), and is presented to the SVC or a vdisk defined on the
SVC and presented to a host.
Managed disk (mdisk)
This is a SCSI Disk (LUN) presented by a RAID storage controller and managed by the SVC.
The mdisk is not visible to host systems on the SAN.
Managed Disk Group
A collection of mdisks that jointly contain all the data for a specified set of vdisks. It can be
considered a pool of extents from which the vdisks are created.
Note: A logical unit number (LUN) is a unique identifier used on a SCSI bus that enables it
to differentiate between up to eight separate devices (each of which is a logical unit (LU)).
Each LUN is a unique number that identifies a specific logical unit, which may be an end
user, a file, or an application program. The LUN is a 3-bit identifier used on a SCSI bus to
distinguish between LUs with the same SCSI ID. The term “LUN” has been used
somewhat interchangeably and it is always worthwhile considering the use that is being
applied to it.

14
IBM TotalStorage Migration Guide for the SAP User
Master Console
The Master Console is the platform on which the software used to manage the SVC and
optionally MDS runs.
Preferred node
The node within an I/O Group (a pair of nodes) that has the preferred SDD path.
SVC node
The name given to the individual server engines in an SVC cluster on which the SVC software
runs.
SAN Volume Controller
The SVC is a SAN appliance, designed for attachment to a variety of host computer systems,
which carries out block-level virtualization of disk storage.
Virtual disk (vdisk)
An SVC device that appears to host systems attached to the SAN as a SCSI Disk or LUN.
Each vdisk is associated with exactly one I/O Group.

© Copyright IBM Corp. 2005. All rights reserved.
15
Chapter 2.
Environment 1 SAP migration
experiences using FDRPAS
This chapter describes how we non-disruptively migrated the databases for an SAP system
from one disk system to another. We performed this migration using the product FDRPAS.
After a successful migration from one disk system to another, you might want to return the old
disks to your vendor, or reuse them for another purpose within your company. In either case,
it is essential to completely erase the old data from your source disks. A good way to do this
is to use FDRERASE, which is also described in this chapter.
The following topics are covered in this chapter:
An overview of FDRPAS
Installing FDRPAS
Things to consider before you actually start using FDRPAS
Invoking an FDRPAS swap task in batch mode
Using the FDRPAS ISPF interface
How to quickly and safely erase disk volumes
2

16
IBM TotalStorage Migration Guide for the SAP User
2.1 FDRPAS overview
If you plan to move your data from one storage subsystem to another, independently of
whether both are from the same vendor of from different ones, FDRPAS might be a good
solution for you because it lets you move active volumes to new disk devices. Some of the
advantages that FDRPAS offers are the following:
Applications do not need to be closed.
Data sets do not need to be closed.
Jobs and users are not disturbed at all. They do not even realize that their data has been
moved.
You can move your data during daytime hours. This means less weekend and midnight
hours for your sysprogs and storage administrators.
You can move volumes from old disk hardware to new disk hardware.
You can move volumes for I/O load balancing.
You can create point-in-time backups of volumes.
You can rebuild indexed VTOCs without varying volumes offline.
2.1.1 Supported hardware
FDRPAS can be used on any disk hardware, such as:
IBM
– 3990
– RAMAC®, RVA, 2105 Shark
EMC
– any Symmetrix
StorageTek
– Iceberg, SVA, Eclipse, V2x
HDS
– 7700, 7700E, 9900
Any other appliances that provide 3990 or 2105 emulation
The non-disruptive solution is only possible for like-to-like volumes such as 3390 to 3390,
3990 to 3990, and so on. If, however you plan to migrate three 3390-3 volumes to one 3990
volumes, this cannot be done non-disruptively.
Large disks of up to 32760 cylinders are supported.
2.1.2 Supported software
In this section, we basically want to point you to the appropriate FDRPAS manuals from the
Innovation Data Group. Our reason for not being too detailed here is the rapid changes that
are taking place in the technology. There are a few things that we can mention here since
they are not likely to change too soon.
FDRPAS supports:
All current S/390 operating systems (OS/390 and z/OS).
Multiple systems and Sysplexes

Chapter 2. Environment 1 SAP migration experiences using FDRPAS
17
All volume and data set types, such as:
– System
– Application
– TSO
– Databases
– Catalog
2.1.3 How does it work?
In this section we give you an overview of the swap process and the single phases the
process goes through. To keep this discussion as simple as possible the first time through,
we assume that the volume to be swapped is only defined in one system’s I/O configuration.
The same is true for the target volume as shown in Figure 2-1. FDRPAS can also swap
volumes attached to multiple systems or LPARs. (We describe the difference from this simple
approach in Section 2.1.4, “Single system versus multiple system operation” on page 20.)
Figure 2-1 Single system swap starting point
The aim of FDRPAS is to copy all data from the Online source volume to the Offline target
volume, vary SAP001 offline, assign volume serial SAP001 to unit=9999, and vary unit=9999
online in a last step. The goal is to non-disruptively change the picture to what we show in
Figure 2-2.
Figure 2-2 Single system swap ending point
CPUE
Online
source
volume
Offline target
volume
Processes changing data
I/O
SAP001, Unit=8000
Unit=9999
CPUE
Offline
source
volume
Online target
volume
Processes changing data
I/O
Unit=8000
SAP001, Unit=9999

18
IBM TotalStorage Migration Guide for the SAP User
The operation of FDRPAS is divided into 5 phases:
Phase 1 - Initialization
This phase begins immediately after the FDRPAS batch job or ISPF dialog requests the
swap of one volume to another. In this phase, FDRPAS verifies that the specified source
and target devices are valid for a swap, meaning that it checks whether:
– Both volumes are the same disk device type.
– The target device is offline.
– The source is eligible to be swapped.
Phase 2 - Activation
FDRPAS temporarily suspends all application and system I/O to the source device and
installs an I/O intercept to monitor updates to the source volume as shown on Figure 2-3.
Figure 2-3 Swap Phase 2 - Activation
Phases 3 and 4 - Copy and Consolidation
In the first pass of this phase, FDRPAS copies all tracks on the source system. To save
time and resources, it only copies the tracks which:
– Are currently allocated to a data set. This means that FDRPAS does not touch empty
tracks.
Note: The phase where I/Os are suspended usually does not take any longer than one
to two seconds. So this should really not have any negative impact on your data
processing.
CPUE
Online
source
volume
Offline target
volume
SAP001, Unit=8000
Unit=9999
Processes changing data
I/O
End
Activation
Start
Activation
CPUE
Online
source
volume
Offline target
volume
SAP001, Unit=8000
Unit=9999
Processes changing data
I
/
O

i
n
t
e
r
c
e
p
t

Chapter 2. Environment 1 SAP migration experiences using FDRPAS
19
– Contain the VTOC, VTOC index, VVDS, and volume label.
– Are actually used within a physical sequential, partitioned, and VSAM data set, unless
those data sets are currently allocated to some job or task at the beginning of the
swap, in which case FDRPAS copies all allocated tracks.
While the first pass of this phase is copying the tracks just mentioned, the I/O intercept
shown on Figure 2-3 monitors all I/O activities on the source volume and keeps the
associated tracks in a consolidated list. FDRPAS uses this list for an additional pass, this
time copying all the tracks on that list. While FDRPAS copies these tracks, the I/O
intercept continues to collect updated tracks and so on.
These iterative steps within phase 3 continue until the number of tracks that must be
copied in a last step is small. The system considers a number of 150, that is, a remaining
copy time of one to two seconds, to be small enough to suspend any writes on the old
source files and to skip over to step 4. If FDRPAS realizes after the first iteration that the
number of tracks which still need to be copied exceed 150 tracks, this threshold is raised
by 10 tracks in every iteration. According to Innovation’s experience, even in systems with
heavy traffic, the number of iterations that FDRPAS must go trough is usually very small.
Copying 150 tracks only takes one to two seconds, so even if FDRPAS has to go through
15 iterations, the last copy phase would just take around four seconds. This should not
negatively impact your access to your DB2 subsystem. This iterative process is also
shown in Figure 2-4.
Figure 2-4 Phase 3 - Copy
Phase 5 - Swap completion
At the beginning of phase 5, the source and target devices are completely synchronized.
FDRPAS invokes operating system services to finally swap the devices. When this has
been performed successfully, all I/O intercepts are removed and the operative volume is
now re-enabled for I/O operation. As shown in Figure 2-2, the former target unit now has
the swapped volume serial associated to it and appears online. Starting from this point, all
upcoming I/O is directed to the former target device. The former source device is now
Start phase 3 - COPY
Copy eligible tracks
# updated
tracks too
high?
Suspend updates
Copy remaining
updated tracks
End phase 3 and 4
YES
NO
Phase 3 and 4 coalesced

20
IBM TotalStorage Migration Guide for the SAP User
offline and its volume label is modified so that it cannot be accidentally placed online
again.
2.1.4 Single system versus multiple system operation
In the description of the swap phases in the previous section, we assumed that the volume to
be swapped is only varied online on one operating system. This was a valid assumption to
explain the basics of how FDRPAS performs a swap, although we are well aware that your
real existing configurations probably have multiple systems attached to each volume. If this is
also the case in your specific environment, you need to be aware of the differences in
behavior we describe in this section.
Single system
FDRPAS operation is very simple when only a single z/OS image can access the DASD
volume to be moved, as we explained previously.
Multiple system
The complexity of an FDRPAS swap rises when multiple z/OS system images can access the
DASD volume to be moved because FDRPAS must coordinate the swap throughout the
attached system images. This means that all attached system images must be monitored for
updates during the swap and FDRPAS must conduct the final swap simultaneously on all
images. Refer to “FDRPAS Tasks” on page 21 for a discussion about monitor tasks.
FDRPAS supports up to 128 connected systems. When your volumes are on disk systems
such as the following ones, FDRPAS can determine the number of systems with access to
the source volume and the CPUID of each of them:
IBM 3990-6
2105 ESS SHARK
Other vendors disk systems that emulate the first two in this list
These disk subsystems can be considered the “good guys” as far as FDRPAS is concerned,
because FDRPAS can verify that a monitor task is running on each identified system. As
described earlier, it is essential for FDRPAS to know which systems are attached to your
volumes and on which it must have an eye in order not to miss any updates that are being
made during the swap.
This is
not
the case for the following disk systems:
IBM 3990-3/RVA
StorageTek SVA/Eclipse
Others that emulate the first two in this list
Important: If your data is stored on one of these disk systems, then FDRPAS completely
depends on you to accurately specify the number of systems with access to the source
volume. When you start a swap for one of these volumes, FDRPAS verifies that there is a
monitor task running on the number of systems that you have specified, but it cannot verify
the CPU IDs of the systems. This means that if you have specified #SYSTEMS=5 and you
also have monitor tasks running on 5 systems, but your disks are actually online to 10
systems, FDRPAS starts the swap and does not track any updates coming from the
remaining 5 systems, nor does it swap the volume on those systems.

Chapter 2. Environment 1 SAP migration experiences using FDRPAS
21
2.1.5 FDRPAS Tasks
There are two types of FDRPAS tasks: SWAP tasks and MONITOR tasks. You always have at
least one swap task active when you start the swap of one volume to another. This is
independent of the question about single or multiple systems operation. In contrast, monitor
tasks only come into play when you run swap tasks for shared DASD, that is, in a multiple
systems environment.
Swap tasks
A swap task always just runs on one system. One swap task can perform the swap of any
number of disk volumes. However, the number of volumes which one swap task can perform
in parallel is limited to 32. As described in “MAXTASKS=” on page 41, MAXTASKS is the
operand that controls the number of swaps which a single swap task can run in parallel.
A swap task can run as a batch job or as a started task. Refer to 2.2, “Install FDRPAS” on
page 24 for an example of how the swap task’s procedure JCL might look.
You can initiate swap tasks either as a batch job or using the FDRPAS ISPF interface. If you
initiate the swap through ISPF, the dialog invokes a started task that you must have created
on a valid procedure library. This is also explained in 2.2, “Install FDRPAS” on page 24.
Monitor tasks
There are different types of monitor statements and therefore monitor tasks available when
talking about FDRPAS. The monitor task that we refer to in this section is just one of them.
Refer to the appropriate Innovation documentation to learn about other available monitor
tasks and their syntax.
The update monitor task, which is the topic of this section, monitors one or more offline
potential target disk devices. Its main functions are:
Detect that a swap is starting for a given target device and start an internal subtask or
external started task to service the swap.
Verify that the indicated source device is online.
Register to the swap to indicate that it is ready to participate.
Install the I/O intercept on the source.
Transmit the list of updated tracks on the system it is running to the swap task.
Call the system’s swap service to complete the swap when asked for by the swap task.
Figure 2-5 illustrates how it could look if everything is defined, started, and works the way it is
supposed to.
Tip: To gain the best performance, it is recommended that you run the swap task on the
system which has the most update activity on the source volume.

22
IBM TotalStorage Migration Guide for the SAP User
Figure 2-5 Monitor tasks I
In contrast to the situation shown in Figure 2-5, Figure 2-6 shows a situation where volume
SAP001 is online on all three LPARs CPUA, CPUB, and CPUC. The target volume specified
in the swap is part of the I/O configuration on these three LPARs as well. As expected, the
target volume has been varied offline. So far everything is OK.
Once your start the SWAP task on CPUA, the initialization phase starts and indicates that a
swap waits for the monitor tasks on the other systems to acknowledge that they are ready to
participate in the swap.
On CPUB, a monitor task has been started for all offline volumes whose unit numbers start
with 99. The unit number of the target volume that is specified in the swap is 9999, that is, the
running monitor task is responsible for monitoring this volume as well. The monitor task
acknowledges its participation and everything is OK for CPUB.
If we now look at CPUC, we can see that there is also a monitor task sitting and waiting to be
called, but instead of looking at unit 9999, the monitor task’s mount statement points to unit
9990. The result is, that the monitor task does not feel responsible for unit 9999 and does not
acknowledge the swap task of its participation in the swap.
The swap task waits for a certain time for the acknowledgements of the other systems and
finally issues error message FDRW68 associated with an “open reply.” You must now decide
how you want the swap task to proceed. You can answer the open reply with one of the
following options:
RETRY
NO
YES
Online
source
volume
SAP001,Unit=8000
Offline target
volume
Unit=9999
CPUA
SWAP TYPE=FULL
MOUNT VOL=SAP001,
SWAPUNIT=9999
CPUB CPUC
MONITOR TYPE=SWAP
MOUNT SWAPUNIT=99*
MONITOR TYPE=SWAP
MOUNT SWAPUNIT=9999
SWAP has started for unit 9999
SAP001 online at CPUB
install I/O intercept
Transmit list of updates to SWAP
Call SWAP service when done
SAP001online at:
CPUA
CPUB
CPUC
Unit 9999 offline at:
CPUA
CPUB
CPUC
monitor !!
I/O intercept
Updates tracks
SWAP has started for unit 9999
SAP001 online at CPUC
install I/O intercept
Transmit list of updates to SWAP
Call SWAP service when done

Chapter 2. Environment 1 SAP migration experiences using FDRPAS
23
If you specify RETRY, the swap waits again for a certain time and runs into this error again if
nothing changes on CPUC. NO means that you want to stop the swap and YES means that
you ask FDRPAS to continue regardless of the known problems.
Figure 2-6 Monitor tasks II
Monitor tasks can handle 64 concurrent swaps with internal subtasks. If more than 64 are
requested, it starts an external started task, that is PASPROC that you have created during
the installation of FDRPAS. Refer to Section 2.2, “Install FDRPAS” on page 24 for more
information about the started task.
If you do not stop a running monitor task on any of the CPUs, it continues to run forever. It
does, however, terminate automatically when:
All the target devices it is monitoring are online
It has been idle for a specific duration, which is the number of minutes that you specified
for parameter DURATION on the MONITOR statement.
Important: Never reply YES to FDRW68! Instead you should reply with NO, take all
necessary steps and resubmit the swap JCL. If you answer FDRW68 with YES, you tell
FDRPAS explicitly that you do not care about any updates which might occur on SAP001
coming from CPUC. Therefore, the updated records are not tracked and are not copied to
the target in phase 3 and 4 of the swap. In addition to that, the swap does not take place on
CPUC. That means that this system continues to store data on unit 8000 instead of unit
9999, which all other systems continue to work on. Data inconsistency and a loss of data is
most likely to occur and is your responsibility in this case.
Online
source
volume
SAP001,Unit=8000
Offline target
volume
Unit=9999
CPUA
SWAP TYPE=FULL
MOUNT VOL=SAP001,
SWAPUNIT=9999
CPUB CPUC
MONITOR TYPE=SWAP
MOUNT SWAPUNIT=99*
MONITOR TYPE=SWAP
MOUNT SWAPUNIT=9990
SWAP has started for unit 9999
SAP001 online at CPUB
install I/O intercept
Transmit list of updates to SWAP
Call SWAP service when done
SAP001online at:
CPUA
CPUB
CPUC
Unit 9999 offline at:
CPUA
CPUB
CPUC
monitor !!
I/O intercept
FDRW68!!

24
IBM TotalStorage Migration Guide for the SAP User
2.2 Install FDRPAS
Before we show you how easy it is to migrate from one DASD system to another without any
impact to your daily operations, we take you through the steps we used to install the FDRPAS
software. This information is presented to give you just a brief overview of the process; when
you actually plan to install FDRPAS on your system, refer to the installation instructions from
Innovation Data to ensure that you have the most up-to-date information.
2.2.1 Basic installation steps
In order to keep the installation process as easy as possible for you, FDRPAS does not use
SMP/E to provide the necessary data sets to actually start the installation of the product.
There are two possible ways to get the product files:
1.Order installation tapes
2.Receive them via the network
We decided to use option two, so we requested an e-mail that contained all necessary
information to perform the installation. In fact you receive two e-mails if you decide to use this
method. One piece of mail contains the order acknowledgement, that is your order number,
installation password, and some other information. The second e-mail contains the actual
instructions for you to get to a binary file from which the whole installation process is started.
Download FDRPAS.BIN from the network
One way to get FDRPAS.BIN, the initial file to start the installation, is to click a link provided in
the mail describing how to start the installation. This link opens a File Download window that
allows you to save the downloaded file on your computer. Specify an appropriate folder in
which to store the binary file.
Upload FDRPAS.BIN to z/OS to start the installation
There are several ways to upload a file from a Windows subsystem to z/OS. The method that
we used was the file transfer feature of our 3270 emulation. You could also use FTP to do
this. The only things that you must take care of are:
1.The target file on z/OS must have the following allocation attributes:
Primary space 5000 blocks
Secondary space 1000 blocks
DSORG PS
RECFM FB
LRECL 80
BLKSIZE 3120
2.The transfer must be in BINARY mode.
Expanding the distribution file on z/OS
Once you have uploaded the binary file from your workstation, you must now expand the
distribution file on the z/OS host. To do this, use either of the two following methods:
1.Go to ISPF option 6 and enter the following command:
RECEIVE INDATASET(xmit.data)
Be sure to replace xmit.data with the name that you chose for the upload of the binary file
FDRPAS.BIN.

Chapter 2. Environment 1 SAP migration experiences using FDRPAS
25
2.On the ISPF command line, enter:
TSO RECEIVE INDATASET(xmit.data)
Be sure to replace xmit.data with the name that you chose for the upload of the binary file
FDRPAS.BIN.
If the receive command completes successfully, it creates one data set named
'youruserid.IDPREC.LOAD', which contains one load module.
If you prefer to have this data set stored using a different high-level qualifier, you can force
this by entering the fully qualified data set name, for example DSN(your.preferred.name)
when you are prompted for it as shown in Figure 2-7.
Figure 2-7 RECEIVE INDATASET command prompt
2.2.2 The FDRPAS INSTALL ISPF dialog
Once you have successfully received member IDPREC on the data set mentioned, you can
now proceed with the installation process. Go to the ready prompt of your ISPF session and
enter the following command:
CALL 'youruserid.IDPREC.LOAD(IDPREC)'
The information shown in Figure 2-8 is displayed on your 3270 emulation screen. Enter the
passcode, which has been provided to you just below the text. In our case, the password was
a 16 digit number provided in the Acknowledgement notice.
Figure 2-8 FDR product initial installation screen
If you entered the password correctly, Product Install Screen1 pops up as shown in
Figure 2-9.
When the data set list is displayed, enter either:
INMR901I Dataset FDRSYS.IDPREC.LOAD from FDRCFG on JESCPUA
INMR906A Enter restore parameters or 'DELETE' or 'END' +
*-------------------------------------------------------------------*
* Welcome to Innovation's - TRIAL install of FDR Product
* 'XTRTST2.FDR.SHIPPDS' will be allocated and loaded.
* When the load phase is complete a ISPF Menu will be displayed
* with an ISPF Menu titled 'IDP PRODUCT INSTALL (SCREEN1)'
* This Dialog has ISPF help with a further description of this
* install process
*-------------------------------------------------------------------*
Please enter PASSWORD to load install file or QUIT to end
1234567891234567

26
IBM TotalStorage Migration Guide for the SAP User
Figure 2-9 FDR product install screen1
You must enter the volume and unit information on this screen. You can change the variables
which show up in yellow on the screen (shown as bold in Figure 2-9). Those variables control
the names of the data sets which will be created by the installation procedure. If you decide
not to change anything, the resulting data set names are as shown in Figure 2-10. Press
Enter to accept the changes and F3 to continue with the installation.
Figure 2-10 FDRPAS distribution libraries
Once you press F3, Install screen two is displayed as shown in Figure 2-11. This screen
warns you not to continue with the installation process if the load of any of the files shown in
Figure 2-10 fails. If the process really fails, we recommend that you to start over with the
installation, because otherwise ABRALLOC clist is not set up properly. If all files could be
loaded successfully, additional information is added to the screen to inform you about the
success. Press Enter to get to the next screen.
-------------------------- IDP Product Install (Screen1) ----------------------
Command ===>
More: +
Press Enter to accept changes
Press End to continue install
Enter Can to quit

Product ==> FDR Version ==> 5.4/41
Input DSN (from FTP site) ==> XTRTST2.FDR.SHIPPDS
Output DSN:
High Level Qualifier ==> XTRTST2
2nd Level Qualifier ==> FDR5441
Low Level Qualifiers: ('X ' Will Bypass File Processing)
CLIST ==> CLIST SKELETON ==> SKELETON
LOADLIB ==> LOAD TABLES ==> NOT.SHIPPED
MESSAGES ==> MESSAGES ICL ==> ICL
PANELS ==> PANELS JCL ==> JCL


VOLUME ==> UNIT ==>

Resulting Dataset Names
CLIST ==> XTRTST2.FDR5441.CLIST
LOADLIB ==> XTRTST2.FDR5441.LOAD
Attention: Do not press any key such as F1 for help here. The dialog abends if you do so
and you must start from the beginning again.
XTRTST2.FDR.SHIPPDS
XTRTST2.FDR5441.CLIST
XTRTST2.FDR5441.ICL
XTRTST2.FDR5441.JCL
XTRTST2.FDR5441.LOAD
XTRTST2.FDR5441.MESSAGES
XTRTST2.FDR5441.PANELS
XTRTST2.FDR5441.SKELETON

Chapter 2. Environment 1 SAP migration experiences using FDRPAS
27
Figure 2-11 FDR product install screen 2
Now you have a chance to decide whether you want to continue with the installation dialog or
if you prefer to stop your activity for now and resume at this point again at a later time. Let’s
assume that you want to suspend the installation process here so that you can see how to
invoke the installation dialog in a different way at any time in case you want to, for example,
change any settings for your products.
As shown in Figure 2-12, you can do this by simply typing Can or Cancel on the command line.
Figure 2-12 FDR product install screen3
T
Invoking the ISPF installation dialog
To continue with the installation of FDRPAS, you can now launch the installation panels using
command TSO EXEC 'XTRTST2.FDR5441.CLIST(ABRALLOC)' from the command line. This
CLIST allocates all necessary libraries, which means that there is no need for you to change
LOGON procedures in order to be able to use the FDR ISPF platform. As a result, the FDR
primary options menu appears as shown in Figure 2-13.
-------------------------- IDP Product Install (Screen2) ----------------------
Command ===>
Press Enter to continue
Enter Can/Cancel to quit


The TSO RECEIVE command will be invoked to create each file
selected from the previous menu. Before each file is created
by using TSO RECEIVE the target dataset will be deleted.
The install will display messages below indicating install
progress and results. If a load fails the user is given
the option to continue or quit. If any dialog file fails to
load it is suggested the user NOT continue to next screen.
The next step will change member ABRALLOC in the List library
that is use to allocate and invoke product dialog.
IDPREC ** load beginning for ----> XTRTST2.FDR5441.MESSAG
-------------------------- IDP Product Install (Screen3) ----------------------
Command ===>
Press Enter to continue
Enter Can/Cancel to quit


The FDR installation detests have been received.

If you would like to display the FDR PRIMARY OPTIONS MENU
and continue with the Installation process, press ENTER;
otherwise press CANCEL.

(For more details go to Section 90.10 in manual)
Tip: The reference to Section 90.10 in the manual is not correct; refer to manual section
380 instead.

28
IBM TotalStorage Migration Guide for the SAP User
Figure 2-13 FDR primary options menu
This primary options menu is not only used for FDRPAS; it is also used during the installation
and customization of other FDR programs. Because of that, not all options that you see on
the panel apply to FDRPAS. For your installation, we recommend that you just use the
options described here.
Setting global options for FDRPAS
Select I - Install and then on the next screen select 4 - SETOPTS to set the installation
options in the FDR global options table. As shown in Figure 2-14, there are a lot of options
that can be set here, but right now only the following options apply to FDRPAS:
1 - Security Options
2 - General Options
14 - FDRPAS Options
Refer to the installation section of manual FDR Plug and Swap for a more detailed description
of the options.
. . . . . . . . . . . . . . . . . . . . . . . . . . .
---------- FDR TOTAL DASD MANAGEMENT SYSTEM -- FDR PRIMARY OPTIONS MENU -------
OPTION ===>
V 5.4/41
1 REPORTS - ABR REPORTING FUNCTIONS
2 RESTORE - ABR DATA SET RESTORE
3 ARCHIVE - ABR DATA SET ARCHIVE OR SUPERSCRATCH
4 BACKUP - ABR DATA SET BACKUP
5 REMOTE Q - ABR REMOTE QUEUE UTILITY FUNCTIONS

C COMPAKTOR - COMPAKTOR MAP AND SIMULATION REPORTS
R RELEASE - COMPAKTOR RELEASE

I INSTALL - INSTALLATION AND MAINTENANCE OF FDR AND OPTIONAL PRODUCTS
J JCL PARMS - SPECIFY FDR JCL AND SYSOUT DEFAULTS FOR SUBMITTED JOBS
K FORMAT - MODIFY FORMAT OF GENERATED REPORTS
M MESSAGES - FDR MESSAGES AND CODES QUERY FACILITY

P PLUG & SWAP - FDRPAS PLUG & SWAP
E FDRERASE - FDR DISK ERASE
Q QUERY - FDR/ABR STATISTICS QUERY
S SRS - SEARCH, REPORT, SERVICES DIALOG
T FDRTSEL - BACKUP FILE MANAGEMENT UTILITY

Chapter 2. Environment 1 SAP migration experiences using FDRPAS
29
Figure 2-14 FDR global options primary menu
On the FDRPAS Options panel, you must decide what name you want to associate to your
FDRPAS started task, and the HLQ that FDRPAS uses for the storage of history catalog
entries. Make sure that, whatever HLQ you use, the user associated to the monitor and swap
tasks has the authority to catalog data sets beginning with the PASINDEX. If this is not the
case, this does not negatively influence the actual swaps, but you do not get any history
records. If you do not specify a value for PASINDEX, FDRPAS writes a diagnostic message
to the syslog. If you do not want history catalog entries and if you want to avoid these
diagnostic messages, you must change PASINDEX to #BYPASS.
The history records are just catalog entries. If FDRPAS does not have access to this catalog
and you have not used the described bypass option, a volume swap does not cause history
records to be written.
As shown in Figure 2-15, we decided to accept the default name PASPROC for the started
task which is used when a FDRPAS monitor task has to dynamically invoke another monitor
task to handle the swap of a specific volume.
. . . . . . . . . . . . . . . . . . . . . . . . . . .
---------- FDR INSTALLATION -- SET FDR GLOBAL OPTIONS PRIMARY MENU ------------
OPTION ===>

1 - SECURITY OPTIONS 8 - MORE ABR GENERAL OPTIONS
2 - GENERAL OPTIONS 9 - ABR REPORT DEFAULTS
3 - COMPAKTOR OPTIONS 10 - MORE ABR REPORT DEFAULTS
4 - ABR GENERAL OPTIONS 11 - OPERATING SYSTEM EXITS
5 - ABR DATA SET NAMES 12 - FDRREORG OPTIONS
6 - ABR ARCHIVE UTILITY DEFAULTS 13 - RESERVED
7 - ABR DISK PROCESSING OPTIONS 14 - FDRPAS OPTIONS

SAVE - SAVE OPTION CHANGES COPY - COPY OPTIONS FROM A PRIOR LEVEL
CANCEL - EXIT WITHOUT SAVING CHANGES AUDIT - DISPLAY USER CHANGED OPTIONS
REFRESH - REFRESH OPTIONS TABLE IN LPA RESET - RE-INITIALIZE ALL OPTIONS

FDR PROGRAM LIBRARY DATA SET:
DATA SET NAME ===> 'XTRTST2.FDR5441.LOAD'
VOLUME SERIAL ===>

NOTE: TO REFRESH THE OPTIONS THAT ARE DYNAMICALLY INSTALLED IN THE ACTIVE LPA,
IT IS NECESSARY TO RUN FDRSTART - USE THE REFRESH CMD TO GENERATE FDRSTART JCL.

Attention: Customers outside the US must make sure that '#BYPASS' equals to
X'7B'BYPASS. If, for example, your 3270 emulation is using EBCDIC code page 297 and
you type a '#', then the hexadecimal representation of '#' is x'B1' instead of x'7B'. In this
case, in order to really bypass the history information, you have to type £BYPASS,
because the '£' (British pound sign) is represented by x'7B'. You can check the following
Web page to find out which paragraph you must type in order to generate x'7B':
http://www-1.ibm.com/servers/eserver/iseries/software/globalization/codepages.html

30
IBM TotalStorage Migration Guide for the SAP User
Figure 2-15 FDRPAS global options
This was the last step which we had to perform through the installation clist.
2.2.3 Additional installation tasks
There are some additional tasks you must go though before you can actually start using
FDRPAS. A detailed description of these tasks follows.
Authorizing FDRPAS programs
FDRPAS must execute as an APF-authorized program on all systems involved. This means
that you must APF-authorize the FDRPAS load library.
Authorizing the FDRPAS ISPF PROGRAM
Before you can use the FDRPAS ISPF dialogs, you must add program FDRPASA to the list
of TSO-authorized programs. These programs are listed in member IKJTSOxx in
SYS1.PARMLIB.
ACF2 Command limiting facility
If you are using the ACF2 Command limiting facility to limit the use of TSO command
processors, you must add FDRPASA and FDRPASIS to the list of authorized commands for
any user who will use the FDRPAS ISPF dialogs.
Create cataloged procedure PASPROC
You must now create a cataloged procedure on your system procedure library. Make sure that
the name of the member is the same as the one that you specified earlier, as shown in
Figure 2-15. You can use member PASPROC on your hlq.icl library as sample jcl for this
member. Figure 2-16 shows the contents of this member.
---------- FDR INSTALLATION -- SET FDR GLOBAL PLUG & SWAP OPTIONS ------------
COMMAND ===>


PASPROC NAME OF CATALOGED PROCEDURE USED FOR PLUG & SWAP........... PASPROC

PASINDEX HIGH LEVEL INDEX FOR FDRPAS CATALOG ENTRIES................ XTRTST1

Chapter 2. Environment 1 SAP migration experiences using FDRPAS
31
Figure 2-16 Sample PASPROC jcl
You can adjust the sample to your special needs:
Add the name of your FDR load library where it says FDRPAS PROGRAM LIBRARY.
Create a new data set for FDRPAS input statements, which can be either PDS or PS.
Alternatively you can stick with NULLFILE.
2.3 FDRPAS concepts
Be sure you understand the concepts explained in this section before you run FDRPAS.
Console messages
Make sure that your system console operators to not reply to any FDRPAS console
messages. Under no circumstances should you automate replies to any FDRPAS console
messages.
To avoid having any console message answered by any person who is not aware about the
implications of a wrong reply, you can use operand WTOR=NO of the SWAP statement.
Refer to “WTOR=” on page 42 for a more detailed discussion on this.
Target devices
You must define the target devices which are to be used for a swap in every subsystem that
uses the source volumes which are to be swapped.
//PASPROC PROC PROG=FDRPAS,
// LIB=library, <= FDRPAS PROGRAM LIBRARY
// EMAIL=NULLFILE, EMAIL STATEMENT INPUT DATA SET
// IN=NULLFILE, FDRPAS STATEMENT INPUT DATA SET
// OUT='SYSOUT=X', FDRPAS SYSOUT
// D=SHR INPUT DATASET DISPOSITION
//PAS EXEC PGM=&PROG,REGION=0M
//*********************************************************************
//* FDR PLUG AND SWAP *
//*********************************************************************
//STEPLIB DD DISP=SHR,DSN=&LIB
//SYSPRINT DD &OUT
//FDRSUMM DD &OUT
//SYSUDUMP DD &OUT
//FDREMAIL DD DSN=&EMAIL,DISP=SHR
//SYSIN DD DSN=&IN,DISP=&D

32
IBM TotalStorage Migration Guide for the SAP User
Figure 2-17 Definition of target DASDs
The target devices must be varied offline to all system images. This does not mean that they
can be offline in your I/O configuration.
Monitor listings
It is strongly recommended, that you plan to keep the FDRPAS swap and monitor listings for
at least one week after a swap. Doing this provides you with documentation of the swap, as
well as enabling Innovation to investigate in case you encounter any problems after the swap.
Verify volume integrity before a swap
It is recommended, that you check your VTOC, VTOCIX, VVDS, or data sets on a volume
prior to running a swap on these volumes. There are several tools that you can use for this
task, including some from Innovation Data. Contact Innovation Data for more information
about these products. For the purpose of this redbook, we describe two IDCAMS options that
might be used since these are readily available to everyone.
You can use IDCAMS to diagnose VVDS and catalog errors. Refer to Figure 2-18 for a
sample JCL.
Source1
Source2
LPAR1
LPAR2
LPAR3
Target1
Target2
Must be ONLINE
Must be varied OFFLINE


Chapter 2. Environment 1 SAP migration experiences using FDRPAS
33
Figure 2-18 IDCAMS JCL for diagnose of VVDS and catalog errors
When we used this job, we actually found an erroneous catalog entry. The output looked like
that shown in Figure 2-19.
Figure 2-19 DIAGNOSE ICFCATALOG output
System names
Many FDRPAS messages refer to “systems” or “system names.” To find out what the system
name of the LPAR is that you are currently working on, you can use command /D GRS. The
output looks like that shown in Figure 2-20. As you can see, CPUE is the system we are
currently working on.
//****************************************************************
//* CHECK VVDS AND CATALOG ERRORS
//****************************************************************
//DIAGVVDS EXEC PGM=IDCAMS,REGION=0M
//SYSPRINT DD SYSOUT=*
//VVDS DD UNIT=3390,VOL=SER=SAP001,DISP=SHR,
// DSN=SYS1.VVDS.VSAP001,AMP='AMORG'
//SYSIN DD *
DIAGNOSE VVDS INFILE(VVDS)
//DIAGCAT EXEC PGM=IDCAMS,REGION=0M
//CAT DD DSN=CATALOG.PMISC,DISP=SHR
//SYSPRINT DD SYSOUT=*
//SYSIN DD *
DIAGNOSE ICFCATALOG INFILE(CAT)
IDCAMS SYSTEM SERVICES TIME: 14:13:59

DIAGNOSE ICFCATALOG INFILE(CAT) 00150001
IDC21364I ERROR DETECTED BY DIAGNOSE:
ICFCAT ENTRY: NATIVE.VSAM.KSDS3.AIX1.DATA (T)
RECORD: NATIVE.VSAM.KSDS3.AIX1.DATA /00
OFFSET: X'0036'
REASON: 20 - ASSOCIATION NOT FOUND
IDC21365I ICFCAT RECORD DISPLAY:
RECORD: NATIVE.VSAM.KSDS3.AIX1.DATA /00
000000 00680034 E30000C4 2DD5C1E3 C9E5C54B E5E2C1D4 4BD2E2C4 E2F34BC1 C9E7F14
000020 C4C1E3C1 40404040 40404040 40404040 40404040 40000032 03000200 12D5C1E
000040 C9E5C54B E5E2C1D4 4BD2E2C4 E2F30000 17D5C1E3 C9E5C54B E5E2C1D4 4BD2E2C
000060 E2F34BC1 C9E7F100

IDC21363I THE FOLLOWING ENTRIES HAD ERRORS:
NATIVE.VSAM.KSDS3.AIX1.DATA (T) - REASON CODE: 20
IDC0001I FUNCTION COMPLETED, HIGHEST CONDITION CODE WAS 8

34
IBM TotalStorage Migration Guide for the SAP User
Figure 2-20 D GRS command output
CPU serial numbers
Some FDRPAS messages include CPU serial numbers so it is good to know which CPU is
associated to which serial number. You might also need to know these numbers if you want
to EXCLUDE some CPUs from your swap. Refer to 2.4.3, “EXCLUDE statement” on page 42
for a description of the EXCLUDE command.
You can use command D M=CPU to generate an output similar to that shown in Figure 2-21.
Figure 2-21 D M=CPU command output
d grs
ISG343I 14.35.06 GRS STATUS 943
SYSTEM STATE SYSTEM STATE
CPUB CONNECTED CPUA CONNECTED
CPUC CONNECTED CPUE CONNECTED
GRS STAR MODE INFORMATION
LOCK STRUCTURE (ISGLOCK) CONTAINS 1048576 LOCKS.
THE CONTENTION NOTIFYING SYSTEM IS CPUE
SYNCHRES: YES
Note: CPU and Controller serial numbers have been changed to fictitious numbers.
d m=cpu
IEE174I 14.45.17 DISPLAY M 962
PROCESSOR STATUS
ID CPU SERIAL
0 + xx11112066

CPC ND = 002066.0B1.IBM.02.000000013F1A
CPC SI = 2066.0B1.IBM.02.0000000000013F1A
CPC ID = 00

+ ONLINE - OFFLINE . DOES NOT EXIST W WLM-MANAGED
N NOT AVAILABLE

CPC ND CENTRAL PROCESSING COMPLEX NODE DESCRIPTOR
CPC SI SYSTEM INFORMATION FROM STSI INSTRUCTION
CPC ID CENTRAL PROCESSING COMPLEX IDENTIFIER

Chapter 2. Environment 1 SAP migration experiences using FDRPAS
35
2.4 FDRPAS SWAP via batch job
As explained in Section 2.1.5, “FDRPAS Tasks” on page 21, one way to initiate a swap from
one DASD volume to another is to use a batch job. Before we get into the initiation of a swap,
we first describe the operands you can use to influence the behavior of your swap and
monitor tasks, as well as other background information that is useful. Then we begin the
description of some swap scenarios in 2.4.10, “FDRPAS SWAP: one volume, single system”
on page 49.
2.4.1 Job control requirements
Our intent here is not to describe every single bit of the FDRPAS product. Instead we
describe just the main components that affect the way the product works and which you need
to understand to get started with your own swaps.
You can use the JCL shown in Figure 2-22 to run a simple swap on one system for one DASD
volume to another.
Figure 2-22 Simple FDRPAS batch job
EXEC statement
This statement specifies the program name FDRPAS and a region size. It is recommended
that you use 0M.
FDRSUMM
This is an optional output data set for an FDRERASE summary report. If you do not specify
this, the messages are written to SYSOUT. If you plan to specify it, the appropriate DCB
characteristics are:
RECFM=FBA
LRECL=121
STEPLIB
You can either work with STEPLIB of JOBLIB to specify the FDRPAS load library. Make sure
that the library specified here is APF-authorized.
//SWAP EXEC PGM=FDRPAS,REGION=0M
//STEPLIB DD DISP=SHR,DSN=XTRTST2.FDR5441.LOAD
//SYSPRINT DD SYSOUT=*
//SYSUDUMP DD SYSOUT=*
//SYSIN DD *
SWAP TYPE=FULL
MOUNT VOL=SAP001,SWAPUNIT=9905
Recommendation: If you have other FDR products installed on your systems, you should
not add the FDRPAS load library to your linklist. FDRPAS shares some of the load
modules, which are used by other products as well. If you do not follow this
recommendation and you are, for example, using different versions for other products,
FDRPAS might run into problems due to release incompatibilities.

36
IBM TotalStorage Migration Guide for the SAP User
SYSPRINT
Required DD statement SYSPRINT specifies the output message data set. If you prefer to
write the messages to a data set, its appropriate DCB characteristics are:
RECFM=FBA
LRECL=121
SYSUDUMP
Always specify a SYSUDUMP DD statement because it is very helpful in error situations. If
you do not specify a separate data set name, dump messages are written to SYSOUT. This
approach is reasonable in most situations.
SYSIN
Specifies the control statement data set. We used DD * along with an input stream for our
tests. You can, however, also specify an external data set and point to it if you prefer.
2.4.2 SWAP/SWAPDUMP/SIMSWAP statement
Apart from SWAP, which initiates an FDRPAS swap task to move data from one disk to
another, you can also use SWAPDUMP, which is similar to SWAP except that the volume will
not be swapped to the target device; and SIMSWAP, which performs a simulation of a SWAP
operation.
All of these statements must at least be followed by a MOUNT statement. We discuss the
usage of the MOUNT statement in Section 2.4.7, “MOUNT statements” on page 46.
All operands which are discussed in this section can be omitted. The only mandatory operand
is TYPE=FULL. If you specify any other operands, you must separate them with commas.
You can also refer to the syntax diagram shown in Figure 2-23 and continued in Figure 2-24.

Chapter 2. Environment 1 SAP migration experiences using FDRPAS
37
Figure 2-23 SWAP/SWAPDUMP/SIMSWAP syntax diagram part I

,
SYSTEMS

SIMSWAP

SWAP

TYPE

=

SWAPDUMP

nnn

,
BUFNO

nnn

=

,

CANCELPROT

=

CHECKTARGET

,

=

Yes

No

Yes

No

CONFIRMSPLIT

=

,

No

Yes

CONFIRMSWAP

,

=

No

Yes

CONFMESS

=

,

No

Yes

EMSG

,

=
(
OK

,
nn
)

LARGERSIZE

=

,

No

Yes

LOGMESS

=

,

No

Yes

FULL

38
IBM TotalStorage Migration Guide for the SAP User
Figure 2-24 SWAP/SWAPDUMP/SIMSWAP syntax diagram part II
TYPE=FULL
You always must specify this statement on the SWAP, SWAPDUMP or SIMSWAP statement.
#SYSTEMS=
Use this statement to specify the number of system images, meaning CPUs or LPARs, up to
a maximum of 128, which are involved in the swap of a disk.
If your source DASD is one of the following, you
must
specify #SYSTEMS:
IBM3990-3
IBM 2105
IBM RVA
StorageTek SVA
EMC Symmetrix 4xxx
Any other subsystem that emulates a 3990-3
If your DASD system is not one of those listed here, parameter #SYSTEMS is ignored by
FDRPAS.
Refer to the FDRPAS manual for a more detailed discussion on this parameter. It is extremely
important that you specify #SYSTEMS accurately. Following are four examples of what you
must specify in different configurations, assuming that you have five systems:
1.If all five systems can access both the source and target devices, and all five are started,
specify #SYSTEMS=5 and run monitors on all systems. Refer to “MOUNT in SWAP
TYPE=FULL” on page 46 for a more detailed description of the MONITOR parameter.

PACING

=

,

STATIC

DYNAMIC

ALL

=
PRINT
)

,

MAXTASKS

=

MIN#SYSTEMS

,

=

nn

nnn

NOTIFYERR

=

,

PACEDELAY

,

=

USERID

nnn

No

Yes

MAXCARDS

,

nnn

=

MAXPCTINVESWAPS

,

=

nnn

SWAPDELAY

=

,

SWAPIDERROR

=

,

NORETRY

RETRY

nnn

WTOR

=

,

YES

NO
MOUNT block

EXCLUDE block

HISTORY block

Chapter 2. Environment 1 SAP migration experiences using FDRPAS
39
2.If two systems are not started, run monitors on the remaining three and specify
#SYSTEMS=3.
3.If two of the systems do not have the source device in their configuration, but all five have
access to the target device, run monitors on all 5 and specify #SYSTEMS=5.
The swap would also work if you have specified #SYSTEMS=3 in this case, so logically it
is not necessary to set #SYSTEMS to 5.
4.If three of the systems have neither the source nor the target device in their configuration,
run monitors on the remaining two and specify #SYSTEMS=2.
BUFNO=
Specifies the number of I/O buffers that FDRPAS uses while copying data from the source to
the target device. Acceptable values are within a range of 2 to 32. The default value is 32. As
a rule of thumb, the lower the value for BUFNO, the longer the SWAP takes to complete and
less is the impact to your I/O system. If you do not experience problems due to the length of
the FDRPAS I/Os, we recommend not changing this value.
CANCELPROT=
Possible values are YES and NO. The default is NO.
YES protects the swap task from being cancelled accidently. In case somebody cancels the
swap task, Message FDR267 indicates that the active swaps can continue and that the tasks
terminate upon completion. If you want to immediately force the termination of active tasks,
you can enter a second CANCEL command.
A value of NO allows you to immediately terminate the swap task with a CANCEL command.
CHECKTARGET=
Possible values are YES and NO. The default is NO.
If you specify YES, the target device is checked prior to beginning a SWAP or SWAPDUMP
operation to ensure that the target is empty. If it is not empty, the Swap does not start.
If you specify NO, FDRPAS overlays the target volume regardless of its current contents.
CONFIRMSPLIT=
Possible values are YES and NO. The default is NO.
You can only use this parameter with the SWAPDUMP statement.
If you specify YES, FDRPAS waits for your GO to complete the swap or dump operation. In
contrast to that, with a value of NO being set, FDRPAS completes the operation as soon as
Important: For safety reasons it is highly recommended to specify #SYSTEMS=5 in
this scenario and let FDRPAS detect that there are two systems for which it does not
have to take care. If you follow the recommendation, you can see messages in the
SYSPRINT which states exactly this detection and you can use this as a kind of
documentation for your swap.
Attention: Customers outside the US must make sure that '#SYSTEMS' is equal to
X'7B'SYSTEMS. If it is not, you can use NSYSTEMS instead.

40
IBM TotalStorage Migration Guide for the SAP User
possible. As long as you do not have a special need to swap a set of volumes at the same
time, we recommend that you use NO.
CONFIRMSWAP=
This parameter has the same functionality as CONFIRMSPLIT. The only difference is that
CONFIRMSWAP is only used with the SWAP statement.
CONFMESS=
Possible values are YES and NO. The default is NO.
If CONFMESS is set to YES, you must reply a message as confirmation before a swap can
start. With NO specified, the swap begins immediately after the job is submitted.
ESMG=
This parameter belongs to the e-mail notification facility.
LARGERSIZE=
Possible values are YES and NO. The default is NO.
If you try to swap from one disk model to another one which has a larger number of cylinders,
you run into an error unless you have changed the default of this parameter from NO to YES.
If you specify YES, you can, for example, swap from a 3390-3 which has 3339 cylinders to a
3390-9 that consists of 10017 cylinders.
LOGMESS=
Possible values are YES and NO. The default is YES for the SWAP and NO for the
SWAPDUMP statement.
Specify YES if you want FDRPAS to write messages to SYSLOG.
MAXCARDS=
Possible values are 1 to 9999. The default is 250.
Using this statement, you can specify the maximum number of MOUNT statements that can
be present in this FDRPAS step.
MAXACTIVESWAPS=
Possible values are YES and NO. The default is NO.
Important: YES does not result in any console message or WTOR when the volumes are
ready to swap or split. You must use the ISPF panels to find out when the volumes are
ready for a confirmation. Also, you can only confirm the operation by using the ISPF panels
or the MONITOR TYPE=CONFIRMSWAP or CONFIRMSPLIT statement.
Attention: You may run into unpredictable results if you try to swap to a “large disk,” that is
between 10018 and 65520 cylinders in size, if you do not have proper IBM support for
large disks installed on all systems participating in the swap.
Note: FDRPAS acquires a table with a size of 160*MAXCARDS bytes below the line. Take
this into account if you intend to use very large values for this parameter.

Chapter 2. Environment 1 SAP migration experiences using FDRPAS
41
This parameter must be seen in conjunction with the MAXTASKS parameter described in the
following section. If MAXACTIVESWAPS is set to YES, it limits the overall number of swaps
which are concurrently running through the pass 1 copy phase to the number specified in
MAXTASKS. Overall this means that it all active FDRPAS jobs are limited by this parameter.
Once a swap has finished pass 1 and continues with its subsequent iterative passes to copy
all updated tracks, MAXACTIVESWAPS does not limit the number of volumes on which
FDRPAS concurrently works any more.
MAXTASKS=
Possible values are 1 to 32. The default is 1.
You can use this parameter to specify how many volume swaps can be processed in parallel.
If the number of MOUNT statements exceeds MAXTASKS, FDRPAS starts with the indicated
number of swaps. When the swap of one volume finishes, a new swap is initiated
immediately.
MIN#SYSTEMS=
Possible values are 1 to 128.
Using this parameter, you can specify the minimum number of system images which must
participate in the swap of the volumes in this step.
Let’s assume that the hardware reports that ten systems can access the volumes, but you are
certain that three of those systems have the source volume offline, so you can specify
MIN#SYSTEMS=7. You must then an FDRPAS monitor task on each of the systems that
participate in the swap. This allows the swap to proceed with operator intervention.
The purpose of this parameter is identical to that of the EXCLUDE parameter.
NOTIFYERR=
The default is NO TSO notification.
You can use this parameter to specify a TSO userid to which a notification about FDRPAS
problems is send immediately.
PACEDELAY=
Possible values are 1 to 32767 seconds. The default is 0, meaning no pacing delay.
You can use this value to specify the number of hundredths of a second that FDRPAS waits
between WRITE I/Os on the target device, which might be a good option to minimize the
impact of the background copy operation on other applications.
Recommendation: Although this parameter exists, its use is
not
recommended. Use
EXCLUDE, described in 2.4.3, “EXCLUDE statement” on page 42 instead!
Restriction: This option can only be used when the source volume is in a disk subsystem
where FDRPAS can determine which systems have access to the volumes. Refer to
“#SYSTEMS=” on page 38 for a list of those systems.
Attention: Customers outside the US must make sure that 'MIN#SYSTEMS' equals to
MINX'7B'SYSTEMS. If this is not the case, use MINNSYSTEMS instead.

42
IBM TotalStorage Migration Guide for the SAP User
PRINT=ALL
If you specify this option, you ask FDRPAS for additional printout from the swap task. This
additional printout includes a list of all data sets on the source volume, and a detailed list of
the tracks copied in each phase. You can find the generated output in SYSPRINT.
SWAPDELAY=
Possible values are 1 to 255 seconds. The default is 15 seconds.
Use this parameter to tell FDRPAS how long it is allowed to wait between checks to see if all
the system images have indicated that they are ready to swap the volume. As a maximum,
FDRPAS checks 30 times if all expected systems have joined the swap before it terminates
the swap.
This option is rarely used.
SWAPIOERR=
This is primarily intended for use with channel extenders. Since this occurs very seldom, we
do not describe this option here. Refer to the FDRPAS PLUG and SWAP manual for a
detailed description.
WTOR=
Possible values are YES and NO. YES is the default.
FDRPAS issues some messages which require user intervention, that is, they must be replied
to by an operator. There are some situations where a wrong reply can have massive impact
on your data integrity. To avoid your operators accidently entering wrong replies, you can
specify WTOR=NO. In this case, critical messages such as FDRW01 and FDRW68 are not
issued as WTOR messages any longer. Instead these messages appear as WTOs, meaning
operators cannot respond to them. To respond to these messages, you must use the
FDRPAS ISPF panels described in 2.5, “FDRPAS through ISPF Interface” on page 58.
2.4.3 EXCLUDE statement
As you can see from Figure 2-23 on page 37, the EXCLUDE statement optionally follows the
SWAP TYPE=FULL statement. You can use this statement to tell FDRPAS that the system
specified in this statement does not have to participate in a swap. The use of this statement is
only possible and necessary in environments where your disk subsystem is one of those that
was specified in “#SYSTEMS=” on page 38. The simple EXCLUDE syntax diagram is shown
in Figure 2-25.
Figure 2-25 EXCLUDE statement syntax diagram
You can repeat the EXCLUDE statement up to 127 times to exclude numerous systems from
a swap. However, it cannot exceed the number that you have specified for MAXCARDS.
Tip: Specify a value at least 3 times the SWAPDELAY value on the MONITOR statement.

EXCLUDE

CPUID

=

CPUID

Chapter 2. Environment 1 SAP migration experiences using FDRPAS
43
2.4.4 HISTORY statement
Use the HISTORY statement in a batch job to display the history records for one or more
volumes. For the specification of the volume names, you can use a trailing asterisk as a
wildcard.
The two additional options MAXSWAPHISTORY= and MAXSWAPHISTORY-DAYS= can be
used to purge some of the history catalog entries. If you do not specify a value here and you
also do not purge those entries manually, the existing history records are never purged and
stay there forever.
The use of this stand-alone statement in a batch job produces the same result as the ISPF
command HI on the FDRPAS ISPF panels. Refer to Figure 2-58 on page 61 for a sample
output.
Figure 2-26 shows the syntax diagram for the HISTORY statement.
Figure 2-26 HISTORY statement syntax
2.4.5 LICENSE statement
This standalone statement causes FDRPAS to quickly scan all of the disk storage online on
the system where you execute it. Refer to Figure 2-27 for the available syntax.
Figure 2-27 LICENSE syntax diagram
2.4.6 MONITOR statements
MONITOR TYPE =SWAP statement
This standalone statement initiates an FDRPAS monitor task. Refer to 2.1.5, “FDRPAS Tasks”
on page 21 for a more detailed description of the monitor tasks. The syntax diagram shown in
Figure 2-28 gives you an overview about the possible additional parameters that you can use
in conjunction with the MONITOR statement.

HISTORY

TYPE

SWAP

,

MAXSWAPHISTORYDAYS

,

=
MAXSWAPHISTORY

=

nnn

=

nnnnn

history MOUNT block

LICENSE

PRINT

TYPE

=

SWAP

=

ALL

44
IBM TotalStorage Migration Guide for the SAP User
Figure 2-28 MONITOR TYPE=SWAP statement syntax diagram
You can influence the time that a monitor stays active using the DURATION parameter,
which gives you the chance to specify a certain number of idle minutes. If you use this option,
the monitor task terminates after it has been idle for the specified duration. If you omit this
parameter, the task executes until it is terminated by a console STOP commend, cancelled,
or until all devices that it is monitoring have been swapped or are online.
Refer to the FDRPAS PLUG and SWAP manual for a description of the other parameters.
MONITOR TYPE=CONFIRMSWAP/CONFIRMSPLIT
Use this statement to monitor the swap of one or more volumes which have been started with
either CONFIRMSPLIT or CONFIRMSWAP, or both options.
As soon as all the volumes specified are synchronized, the monitor task signals the SWAP or
SWAPDUMP that it can complete its operation.
This monitor task may be started on any system that is involved in the swap of all of the
volumes specified.
The syntax diagram is shown in Figure 2-29 on page 45.
Attention: If you start a MONITOR task on a system where the target volume which you
intend to monitor is currently online, the task does not monitor the activities for this volume
even if you vary the volume offline before the actual swap task starts. If you want to make
sure that the MONITOR task monitors the changes, you must cancel and resubmit the
monitor task.

MONITOR

TYPE
SWAP

,

=
CANCELPRROT

Yes

No

,

DURATION

=

nnnn

LOGMESS

,

=

Yes

No

MAXTASKS

=

,

64

nn

PRINT

,

=

ALL
SWAPDELAY

,

=

nnn
MONITOR SWAPMOUNT BLOCK

Chapter 2. Environment 1 SAP migration experiences using FDRPAS
45
Figure 2-29 Syntax diagram of MONITOR TYPE=CONFIRMSWAP/CONFIRMSPLIT
MONITOR TYPE=VARYONLINE/RESETVOL
Although the name of this statement is MONITOR, this is not a monitor task, which takes part
in a swap.
Some offline disks (such as the original source device for a volume that was successfully
swapped to a new device, the target device of a successfully completed SWAPDUMP
operation, or the target device of an unsuccessful SWAP or SWAPDUMP) cannot be varied
online because these offline disk devices have a volume label which has been modified to
prevent this. If you plan to vary them online anyway, you can use the MONITOR
TYPE=VARONLINE/RESETVOL task to do so. Refer to Figure 2-30 for a syntax diagram for
this task.
VARYONLINE This option resets the volume label, changes the volume serial, and varies
the disks which are specified by the mandatory following MOUNT
statements. The syntax of the MOUNT block is shown in Figure 2-35 on
page 47.
RESETVOL This option does the same thing as VARYONLINE apart from the fact that it
does not vary the specified volume online.
Important: Make sure that you start this monitor task after you have started the associated
SWAP or SWAPDUMP task. If you try to start the monitor task prior to the SWAP, it fails.
Although mentioned earlier in “CONFIRMSWAP=” on page 40, we want to remind you that
it is highly recommended to keep the default CONFIRMSWAP=NO unless you have a
specific reason for wanting a number of disk volumes to complete their swap at the same
time.
Tip: The specified volume is only varied online on those systems on which the job is
issued. If you want to have the volume online on other systems as well, you must issue a
console VARY command.

CONFMESS

=

,

No

Yes
MONITOR


TYPE

=

CONFIRMSPLIT

CONFIRMSWAP

MAXCARDS

=

nnnn

,
monitor confirmsplit/confirmswap block

46
IBM TotalStorage Migration Guide for the SAP User
Figure 2-30 Syntax diagram of MONITOR TYPE=VARYONLINE/RESETVOL
2.4.7 MOUNT statements
There are several types and usages of the MOUNT statement. A MOUNT statement can
never be used alone. It can be part of:
SWAP TYPE=FULL
MONITOR TYPE=SWAP
MONITOR TYPE=CONFIRMSWAP/CONFIRMSPLIT
HISTORY TYPE=SWAP
MONITOR TYPE=VARYONLINE/RESETVOL
MOUNT in SWAP TYPE=FULL
In the SWAP TYPE=FULL statement, at least one MOUNT statement is mandatory. It is used
to specify an online volume that you want to swap to another target volume.
The syntax diagram for this mount statement in shown in Figure 2-31.
Figure 2-31 Syntax diagram of MOUNT in SWAP TYPE=FULL
MOUNT in MONITOR TYPE=SWAP
You can use this MOUNT statement to specify an offline target device or a set of devices to be
monitored to see if one or more of them is selected as the target of a swap by an FDRPAS
swap task on another system image. As you can see from the syntax diagram in Figure 2-32,
you can specify one or a list of volume units. You also have the option of using a trailing
asterisk as a wildcard.
Figure 2-32 Syntax diagram of MOUNT in MONITOR TYPE=SWAP
MOUNT in MONITOR TYPE=CONFIRMSWAP/CONFIRMSPLIT
Use this mount statement in conjunction with MONITOR TYPE CONFIRMSWAP or
CONFIRMSPLIT. With this statement, you specify a volume to be monitored until it is
synchronized with its offline target device. As you can see from the syntax diagram shown in
Figure 2-28 on page 44, you can specify multiple MOUNT statements using the syntax shown
in Figure 2-33.

MONITOR

RESETVOL

TYPE

=

SWAP

VARYONLINE

MOUNT in monitor varyonline/resetvol block

MOUNT

VOL

=

ser

,

SWAPUNIT

=

uuuu

MOUNT

SWAPUNIT

=

(

)

uuuu

,

Chapter 2. Environment 1 SAP migration experiences using FDRPAS
47
Figure 2-33 Syntax diagram of MOUNT in MONITOR TYPE=CONFIRMSWAP/CONFIRMSPLIT
In VOL=volser, you must specify the complete volume serial of the volume that you want to be
monitored. The use of wildcards is not supported here.
MOUNT in HISTORY TYPE=SWAP
You can specify the MOUNT syntax shown in Figure 2-34 if you want to specify a volume
serial for which history records are to be displayed and optionally purged. In contrast to the
previous MOUNT statement, the use of wildcards as suffixes is supported in this statement,
meaning that you can, for example, specify VOL=vol*. As shown in Figure 2-26 on page 43,
you can specify multiple MOUNT statements in this case.
Figure 2-34 Syntax diagram of MOUNT in HISTORY TYPE=SWAP
MOUNT in MONITOR TYPE=VARYONLINE/RESETVOL
This MOUNT statement must follow the MONITOR TYPE VARYONLINE or RESETVOL
statement described in “MONITOR TYPE=VARYONLINE/RESETVOL” on page 45. The
syntax shown in Figure 2-35 shows two operands:
VARYUNIT This operand specifies the 4-digit device address of an offline disk device to
be processed. You can use trailing asterisks, such as AB1* or 2C** as
wildcards to indicate a range of devices. In a range of devices, addresses
that do not correspond to a device are simply ignored by FDRPAS.
NVOL This operand specifies a new volume serial that is to be assigned to the
disk.
Figure 2-35 Syntax diagram of MOUNT in MONITOR TYPE=VARYONLINE/RESETVOL
2.4.8 SWAPBUILDIX statement
You can use this stand-alone statement to initiate an FDRPAS utility task to build or rebuild
the indexed VTOC, that is the VTOCIX on an online volume. Refer to the syntax diagram
shown in Figure 2-36 on page 48 for an idea of which parameters you can use for this
statement.
Attention: As a result of this operation, neither the VTOCIX nor the VVDS are renamed.
The VTOCIX still works correctly, but the VVDS is not accessible. Due to this, VSAM
clusters are not accessible and therefore not usable.

MOUNT

VOL

=

ser

MOUNT

VOL

=

ser

MOUNT

,

NVOL

=

vvvvvv

VARYUNIT

=

uuuu

48
IBM TotalStorage Migration Guide for the SAP User
Figure 2-36 Syntax diagram of SWAPBUILDIX statement
2.4.9 SIMSWAP scenario
Before you run any swap task, it is highly recommended that you use the SIMSWAP
statement to check your SYSIN syntax and to see which systems are supposed to take part in
the swap. Let’s assume that you are operating in a environment as shown in Figure 2-37.
Figure 2-37 SIMSWAP scenario
Figure 2-38 shows a sample output of a successfully completed SIMSWAP job. As you can
see, the output clearly provides the information that your source volume SAP002 is attached
to four systems. As a result, you must start monitor tasks on all of them, but the one where

SWAPBUILDIX

TYPE

=

FULL

#SYSTEMS

=

nnn

,

LOGMESS

=

,

NO

YES

MIN#SYSTEMS

=

nnn

,
CPU SERIAL# xxyyzzzzzz
Source
Vol=SAP001
Unit=990A
Target
Unit=9902
CPU SERIAL# xx44442066
CPU SERIAL# xx22222066
CPU SERIAL# xx33332066
//XTRTST2S JOB ,XTRTST2,NOTIVY=XTRTST2,MSGCLASS=X,
// MSGLEVEL=(1,1)
//SWAP1 EXEC PGM=FDRPAS,REGION=0M
//STEPLIB DD DISP=SHR,DSN=XTRTST2.FDR5441.LOAD
//SYSPRINT DD SYSOUT=*
//SYSUDUMP DD SYSOUT=*
//SYSIN DD *
SIMSWAP TYPE=FULL
MOUNT VOL=SAP001,SWAPUNIT=9902

SWAP task as batch job

Chapter 2. Environment 1 SAP migration experiences using FDRPAS
49
you actually run the swap task. We think that it is a really good idea to run SIMSWAP prior to
each swap task to avoid open replies due to missing monitor tasks.
Figure 2-38 SIMSWAP output
2.4.10 FDRPAS SWAP: one volume, single system
This is the first scenario we used to test the FDRPAS swap mechanism. Refer to Figure 2-39
to get a better understanding about the configuration that was used in this scenario.
Figure 2-39 Swap scenario I
BROWSE - XTRTST2Y J0000310 SYSPRINT PARLSWAP - ROW 1 Col 1 80
COMMAND ===> SCROLL ===> CURSOR
******************************** Top of Data ***********************************
FDR001 FDR PLUG AND SWAP - FDRPAS VER. 5.4/41T - INNOVATION DATA PR
FDR303 CARD IMAGE -- SIMSWAP TYPE=FULL
FDR303 CARD IMAGE -- MOUNT VOL=SAP002,SWAPUNIT=9902
FDR233 CPU WITH (SERIAL# xx11112066) IS ATTACHED TO VOL=SAP002
FDR233 CPU WITH (SERIAL# xx22222066) IS ATTACHED TO VOL=SAP002
FDR233 CPU WITH (SERIAL# xx44442066) IS ATTACHED TO VOL=SAP002
FDR233 CPU WITH (SERIAL# xx33332066) IS ATTACHED TO VOL=SAP002
FDRW66 SWAP OF VOL=SAP002 TO UNIT=9902 NEEDS TO BE STARTED ON 4 SYSTEMS
FDR999 FDR SUCCESSFULLY COMPLETED
******************************* Bottom of Data ********************************
Note: Message FDRW66 might seem a little bit misleading. You could easily interpret that
the actual swap must be started on the 4 systems listed in messages FDR233. What
actually is meant here, is that these four systems are supposed to participate in the swap,
so you must start the swap on one of the above systems and monitor tasks on the
remaining three.
Unit=9900
Vol=SAP001
Unit=806D
SAP001
is online
Unit 9900
is offline,
in Config
No monitor task needed, because
just 1 system knows online volume
CPUE
//XTRTST2S JOB ,XTRTST2,NOTIVY=XTRTST2,MSGCLASS=X,
// MSGLEVEL=(1,1)
//SWAP1 EXEC PGM=FDRPAS,REGION=0M
//STEPLIB DD DISP=SHR,DSN=XTRTST2.FDR5441.LOAD
//SYSPRINT DD SYSOUT=*
//SYSUDUMP DD SYSOUT=*
//SYSIN DD *
SWAP TYPE=FULL
MOUNT VOL=SAP001,SWAPUNIT=9900
SWAP task as batch job
DB2 up and running on volume
SAP001 !

50
IBM TotalStorage Migration Guide for the SAP User
To start with a very simple scenario, our approach was to swap one volume, which is just
defined on one system, to another volume, which we refer to as the target volume. The target
volume was defined in the I/O configuration, but not varied online. The target volume must be
offline in order to be a valid candidate for a swap.
The source volume contains both DB2 for z/OS catalog and directory data, and SAP tables
and index spaces. Both DB2 and SAP were up and running while we started the swap task on
CPUE. As described in 2.1, “FDRPAS overview” on page 16, there was no need to start a
monitor task, because the target volume was only defined on CPUE.
As you can see from Figure 2-39, the batch job that we used to perform the swap is very short
and easy to read. The only things that we had to specify were the volume serial number of the
volume which we want to be swapped and the unit address of the offline target volume.
When the swap is complete, you can see a job output that looks similar to the one shown in
Figure 2-40. There are a few things in the job output which are worth mentioning so that you
understand the provided messages.
Message FDR158
The meaning of this message is:
Reason: FDRPAS has found that the data set named is active (ENQed to
another job or task on this system or another system).
Action: FDRPAS will copy all tracks allocated to the data set, and it will copy
this and all other active data sets on the volume last, in order to avoid
possibly re-copying updated tracks many times.
Updated tracks recopied
Refer to the lower third of Figure 2-40. As you can see there, FDRPAS had to start a second
pass in order to be able to copy all tracks of volume SAP001. 112 tracks had to be recopied,
because they were updated during the first pass.
CPU hardware serial number
Line four of the output shown in Figure 2-40 indicates which CPU takes part in the swap. In
our case, it is CPUE, which is associated to serial number xx11112066.

Chapter 2. Environment 1 SAP migration experiences using FDRPAS
51
Figure 2-40 Swap Scenario I job output
After successful completion of the swap, volume serial number SAP001 is associated to unit
number 9900. In addition, FDRPAS has varied unit 9900 online and unit 860B offline.
You can check this with the following commands:
Display the status of unit 806B.
As you can see from Figure 2-41, the status of the specified unit is not offline.
Figure 2-41 Display Unit output
Display information about volser SAP001.
As you can see from Figure 2-42, the unit of SAP001 is now 9900.
FDR001 FDR PLUG AND SWAP - FDRPAS VER. 5.4/41T - INNOVATION DATA
FDR303 CARD IMAGE -- SWAP TYPE=FULL
FDR303 CARD IMAGE -- MOUNT VOL=SAP001,SWAPUNIT=9900
FDR233 CPUE (SERIAL# xx11112066) ACKNOWLEDGES THE SWAP OF VOL=SAP001
FDRW66 SWAP OF VOL=SAP001 TO UNIT=9900 STARTED ON 1 SYSTEMS (CPUE)
FDR236 CPUE ACTIVATED I/O INTERCEPTS ON UNIT=806B
FDR007 STARTING TIME OF FULL VOL SWAP -- 10.46.16 -- UNIT=3390 ,IN=D#SAP001,O...
FDR158 DATA SET IS ACTIVE DSN=DB2SSYS.BSDS01
FDR158 DATA SET IS ACTIVE DSN=DB2SSYS.LOGCOPY2.DS03
FDR158 DATA SET IS ACTIVE DSN=DB2SSYS.DSNDBC.DSNDB01.DSNSCT02.I0001.A001
FDR158 DATA SET IS ACTIVE DSN=DB2SSYS.DSNDBC.DSNDB06.DSNDXX01.I0001.A001
FDR158 DATA SET IS ACTIVE DSN=DB2SSYS.DSNDBC.DSNDB06.DSNDYX01.I0001.A001
...
...
DATA SET IS ACTIVE DSN=DB2SSYS.DSNDBC.SX12X000.DD32SL0.I0001.A001
112 TRACKS UPDATED BY CPUE
112 TOTAL UNIQUE TRACKS UPDATED IN PASS 1 - RE-COPYING UPDATED TRACKS
CPUE DE-ACTIVATED I/O INTERCEPTS ON UNIT=806B 112 TRACKS UPDATED
FDRPAS SUCCESSFULLY COMPLETED SWAP OF VOL=SAP001 TO UNIT=9900 ON CPUE
ENDING TIME OF FULL VOL SWAP -- 11.37.21 -- UNIT=3390 ,IN=D#SAP001,O...
OPERATION STATISTICS FOR 3390 VOLUME..................SAP001
CYLINDERS ON VOLUME.............10,017
DATASETS PROCESSED...............1,010
BYTES READ FROM DASD.....6,068,234,058
DASD TRACKS SWAPPED............128,250
UPDATED TRACKS RECOPIED............112
DASD EXCPS.......................8,586
TARGET DASD EXCPS................8,709
CPU TIME (SECONDS)...............6.558
ELAPSED TIME (MINUTES).............9.0
SWAP TIME..........................8.9
FDR SUCCESSFULLY COMPLETED
d u,,,806b
IEE457I 15.24.58 UNIT STATUS 230
UNIT TYPE STATUS VOLSER VOLSTATE
806B 3390 OFFLINE /RSDNT

52
IBM TotalStorage Migration Guide for the SAP User
Figure 2-42 Display volser output
2.4.11 FDRPAS sequential swap of five volumes on four systems
After we successfully swapped one volume in the very simple environment in scenario I, we
increased the complexity of our scenario to make it more like one that you might encounter.
Our SAP and DB2 subsystem resides on five 3390-9 DASD devices. These devices are all
known in the I/O configuration of four systems, but are online to serial# xx11112066. The five
target volumes are all offline and also known in the I/O configuration of these four systems.
Refer to Figure 2-43, which shows the configuration graphically.
Figure 2-43 Swap scenario II
When you try to use the JCL shown in Figure 2-43, you encounter at least the error message
shown in Figure 2-44.
d u,vol=sap001
IEE457I 15.29.08 UNIT STATUS 234
UNIT TYPE STATUS VOLSER VOLSTATE
9900 3390 O SAP001 PRIV/RSDNT
Vol=SAP003
Unit=9907
All volumes are online
on SERIAL# xxyyzzzzzz
All volumes offline
CPU SERIAL# xxyyzzzzzz
Vol=SAP004
Unit=9908
Vol=SAP005
Unit=9904
Vol=SAP002
Unit=9901
Vol=SAP001
Unit=990A
Unit=990D
Unit=990E
Unit=990F
Unit=990C
Unit=990B
CPU SERIAL# xx44442066
CPU SERIAL#xx22222066
CPU SERIAL# xx33332066
//XTRTST2S JOB ,XTRTST2,NOTIFY=XTRTST2,MSGCLASS=X,
// MSGLEVEL=(1,1)
//SWAP1 EXEC PGM=FDRPAS,REGION=0M
//STEPLIB DD DISP=SHR,DSN=XTRTST2.FDR5441.LOAD
//SYSPRINT DD SYSOUT=*
//SYSUDUMP DD SYSOUT=*
//SYSIN DD *
SWAP TYPE=FULL
MOUNT VOL=SAP001,SWAPUNIT=990B
MOUNT VOL=SAP001,SWAPUNIT=990C
MOUNT VOL=SAP001,SWAPUNIT=990D
MOUNT VOL=SAP001,SWAPUNIT=990E
MOUNT VOL=SAP001,SWAPUNIT=990F
SWAP task as batch job

Chapter 2. Environment 1 SAP migration experiences using FDRPAS
53
Figure 2-44 Swap scenario II - FDR234 error message
When you check error message FDR234 with Reason M, you see the following explanation:
The system with the indicated CPU serial number did not respond during the indicated
phase of FDRPAS operation This will usually result in message FDRW68.
Message FDRW68 also came up in the sysmessages. The content of this message is shown
in Figure 2-45.
Figure 2-45 Swap scenario II - FDRW68 error message
The reason for that is, that no monitor task has been started for this swap on the CPUs with
serial numbers xx44442066, xx22222066, and xx33332066.
To solve this problem, you must start monitor tasks on all systems which are aware of the
offline target devices.You can, for example, use the JCL shown in Figure 2-46 to do this.
Figure 2-46 Sample MONITOR task JCL
After you have started the monitor tasks on CPUs with serial numbers xx44442066,
xx22222066, and xx33332066, you must answer an open reply which has been generated by
FDRPAS.
List the open replies using command /D R,ALL. There should be one open reply looking like
shown in Figure 2-47.
FDR001 FDR PLUG AND SWAP - FDRPAS VER. 5.4/41T - INNOVATION DATA PR
FDR303 CARD IMAGE -- SWAP TYPE=FULL
FDR303 CARD IMAGE -- MOUNT VOL=SAP001,SWAPUNIT=990B
FDR303 CARD IMAGE -- MOUNT VOL=SAP002,SWAPUNIT=990C
FDR303 CARD IMAGE -- MOUNT VOL=SAP003,SWAPUNIT=990D
FDR303 CARD IMAGE -- MOUNT VOL=SAP004,SWAPUNIT=990E
FDR303 CARD IMAGE -- MOUNT VOL=SAP005,SWAPUNIT=990F
FDR234** SWAP ERROR ON VOL=SAP005 - UNIT=990F REASON=M - CPU (SERIAL#=xx22222066) FAILED
TO RESPOND IN PHASE 1
FDR234** SWAP ERROR ON VOL=SAP005 - UNIT=990F REASON=M - CPU (SERIAL#=xx33332066) FAILED
TO RESPOND IN PHASE 1
FDR234** SWAP ERROR ON VOL=SAP005 - UNIT=990F REASON=M - CPU (SERIAL#=xx44442066) FAILED
TO RESPOND IN PHASE 1
FDRW68 CAUTION REQUEST TO SWAP VOL=SAP005 TO UNIT=990F ON 1 SYSTEMS- IGNORING
NON-RESPONDING CPUS REPLY YES, NO OR RETRY
/*ROUTE XEQ JESCPUC
//* FDRPAS WILL DISPLAY THE AMOUNT OF ONLINE STORAGE IN EVERY DISK
//* SUBSYSTEM ATTACHED TO THIS SYSTEM.
//*
//SWAP EXEC PGM=FDRPAS,REGION=0M
//STEPLIB DD DISP=SHR,DSN=FDRSYS.T5441.LOAD
//SYSPRINT DD SYSOUT=*
//SYSUDUMP DD SYSOUT=*
//SYSIN DD *
MONITOR TYPE=SWAP
MOUNT SWAPUNIT=(990*)

54
IBM TotalStorage Migration Guide for the SAP User
Figure 2-47 FDRW68 open reply
Answer the reply with /R 20,RETRY. RETRY gives FRDPAS additional wait time to see if the
additional systems indicate their participation.
The RETRY should be successful now, because the monitor tasks are now up and running.
Check the job messages again and you can see the new messages shown in Figure 2-48.
Message 3 tells you that the synchronization was successful, FDRPAS has got answers from
all participating systems, but systems other than serial# xx11112066 do not have to take part
in the swap activities, because SAP002 is not varied online on all the other systems.
Figure 2-48 Continuation of FDRPAS after starting monitor tasks
2.4.12 FDRPAS concurrent swap of five volumes on four systems
In the third scenario, which is basically the same as the second one, we show how you can
use the EXCLUDE statement in your swaps. In addition to that, we demonstrate the use of
operand MAXTASKS.
T TIME JOB ID MESSAGE TEXT
20 R 16.52.21 JOB09755 *20 FDRW68 CAUTION REQUEST TO SWAP
VOL=SAP002 TO UNIT=990C ON 1 SYSTEMS-
IGNORING NON-RESPONDING CPUS REPLY
YES, NO OR RETRY
FDR233 CPUE (SERIAL# xx11112066) ACKNOWLEDGES THE SWAP OF VOL=SAP002
FDR233 CPUB (SERIAL# xx22222066) ACKNOWLEDGES THE SWAP OF VOL=SAP002 AND WILL NOT
JOIN BECAUSE SOURCE IS NOT ONLINE
FDR233 CPUA (SERIAL# xx33332066) ACKNOWLEDGES THE SWAP OF VOL=SAP002 AND WILL NOT
JOIN BECAUSE SOURCE IS NOT ONLINE
FDR233 CPUC (SERIAL# xx44442066) ACKNOWLEDGES THE SWAP OF VOL=SAP002 AND WILL NOT
JOIN BECAUSE SOURCE IS NOT ONLINE
SWAP OF VOL=SAP002 TO UNIT=990C STARTED ON 1 SYSTEMS (CPUE)
FDR DISABLED FLASHCOPY UNIT=9906
CPUE ACTIVATED I/O INTERCEPTS ON UNIT=9906
STARTING TIME OF FULL VOL SWAP -- 17.01.54 -- UNIT=3390 ,IN=D#SAP002,O
DATA SET IS ACTIVE DSN=DB2SSYS.LOGCOPY1.DS02
DATA SET IS ACTIVE DSN=DB2SSYS.DSNDBC.DSNDB01.DSNLLX01.I0001.A001
DATA SET IS ACTIVE DSN=DB2SSYS.DSNDBC.DSNDB06.SYSSTR.I0001.A001
DATA SET IS ACTIVE DSN=DB2SSYS.DSNDBC.DSNDB07.DSN4K04.I0001.A001
CPUE DE-ACTIVATED I/O INTERCEPTS ON UNIT=9906 0 TRACKS UPDAT
FDRPAS SUCCESSFULLY COMPLETED SWAP OF VOL=SAP002 TO UNIT=990C ON CPUE
FDR RE-ENABLED FLASHCOPY UNIT=9906
ENDING TIME OF FULL VOL SWAP -- 17.05.22 -- UNIT=3390 ,IN=D#SAP002,O
OPERATION STATISTICS FOR 3390 VOLUME..................SAP002
CYLINDERS ON VOLUME.............10,017
DATASETS PROCESSED...............3,879
BYTES READ FROM DASD.....3,592,479,122
DASD TRACKS SWAPPED.............86,958
UPDATED TRACKS RECOPIED..............0
DASD EXCPS.......................5,936
TARGET DASD EXCPS................6,015
CPU TIME (SECONDS)...............8.713
ELAPSED TIME (MINUTES)............15.1
SWAP TIME..........................3.9
...
... ... continues for all other volumes ...

Chapter 2. Environment 1 SAP migration experiences using FDRPAS
55
The environment in which we are operating is exactly the same as that shown in Figure 2-43
on page 52. The assumption is, that you know for sure that volumes SAP001, SAP002,
SAP003, SAP004, and SAP005 are in the I/O configuration of systems with serial numbers
xx44442066, xx22222066, and xx33332066, but varied offline on all three systems. If this is
the case, you can exclude these three systems from the swap.
In scenario 2, the swap of the five volumes SAP001 - SAP005 was done in sequence, that is,
one after the other. In order to speed up the whole swap, assuming your overall system can
back it, you can tell FDRPAS to do the swaps in parallel. Use operand MAXTASKS as
described in “MAXTASKS=” on page 41 to do this. In our JCL shown in Figure 2-49, we
specify MAXTASKS 3. FDRPAS starts with 3 parallel swaps instantly. After the first swap is
finished, it continues with swap 4 and so on.
Figure 2-49 FDRPAS with EXCLUDE and MAXTASKS set
You can verify that the swaps have really been done in parallel when you look at the
produced messages as shown in Figure 2-50. The messages related to swap of volumes
Attention: If you use the EXCLUDE statement instead of starting monitor tasks on the
other systems, data integrity is completely your responsibility. You tell FDRPAS to ignore
whatever happens to these volumes while you perform the swap.
Tip: In the beginning, it is a little bit tricky to distinguish between a new statement and
operands to a statement. Our SWAP control statement can nicely be used to explain the
difference. MAXTASKS= is an operand which belongs to the SWAP statement and has to
be separated by a comma. In contrast to that, the EXCLUDE and the MOUNT statements
are considered to be separate, but not stand-alone statements. They can only be used in
conjunction with other statements such as SWAP in our case, but they must not be
separated by a comma. This should also be clear from the syntax diagrams provided in the
previous sections.
File Edit Edit_Settings Menu Utilities Compilers Test Help
sssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssss
EDIT------ XTRTST2.CNTL(PARLSWAP) - 01.00----------------- Columns 00001 00072
Command ===> Scroll ===> CSR
****** ***************************** Top of Data ******************************
000001 //XTRTST2Y JOB ,XTRTST2,NOTIFY=XTRTST2,MSGCLASS=X,MSGLEVEL=(1,1)
000002 //PARLSWAP EXEC PGM=FDRPAS,REGION=0M
000003 //STEPLIB DD DISP=SHR,DSN=XTRTST2.FDR5441.LOAD
000004 //SYSPRINT DD SYSOUT=*
000005 //SYSUDUMP DD SYSOUT=*
000006 //SYSIN DD *
000007 SWAP TYPE=FULL,
000008 MAXTASKS=3
000009 EXCLUDE CPUID=xx44442066
000010 EXCLUDE CPUID=xx22222066
000011 EXCLUDE CPUID=xx33332066
000012 MOUNT VOL=SAP001,SWAPUNIT=9901
000013 MOUNT VOL=SAP002,SWAPUNIT=9902
000014 MOUNT VOL=SAP003,SWAPUNIT=9903
000015 MOUNT VOL=SAP004,SWAPUNIT=9904
000016 MOUNT VOL=SAP005,SWAPUNIT=9905
000017 /*
****** **************************** Bottom of Data ****************************
. . . . . . . . . . . . . . . . . . . . . . . . . . .

56
IBM TotalStorage Migration Guide for the SAP User
SAP001 and SAP004 are in SYSPRIN1, those from SAP003 and SAP005 are in SYSPRIN2,
and the remaining one, for SAP002, is in SYSPRIN3.
Figure 2-50 FDRPAS - # of SYSPRINx equals to MAXTASKS
When you check the generated messages in MESSAGES, and SYSPRINx, you can find the
information shown in Figure 2-51. As you can see, the excluded system serial numbers are
explicitly mentioned in the output and the swap starts although no monitor tasks are running
on these systems.
Figure 2-51 FDRPAS EXCLUDED systems
2.4.13 FDRPAS concurrent swap
In our fourth scenario, illustrated in Figure 2-52, we used a configuration which is most likely
one that comes closest to what you encounter in real-world environments. All source volumes
are defined and varied online on all attached systems. All new target volumes are defined and
varied offline on all systems.
LOG * HELD X 1 H 30 L JESCPUE
JCL * HELD X 1 H 6 L JESCPUE
MESSAGES * HELD X 1 H 49 L JESCPUE
SYSPRINT PARLSWAP HELD SEL X 1 H 206 L JESCPUE
SYSPRIN1 PARLSWAP HELD SEL X 1 H 80 L JESCPUE
SYSPRIN2 PARLSWAP HELD X 1 H 76 L JESCPUE
SYSPRIN3 PARLSWAP HELD X 1 H 35 L JESCPUE
FDR001 FDR PLUG AND SWAP - FDRPAS VER. 5.4/41T - INNOVATION DATA PR
FDR233 CPU WITH (SERIAL# xx22222066) WAS EXCLUDED
FDR233 CPU WITH (SERIAL# 0144442066) WAS EXCLUDED
FDR233 CPU WITH (SERIAL# xx33332066) WAS EXCLUDED
FDR233 CPUE (SERIAL# xx11112066) ACKNOWLEDGES THE SWAP OF VOL=SAP005
FDRW66 SWAP OF VOL=SAP005 TO UNIT=9908 STARTED ON 1 SYSTEMS (CPUE)
FDR264 FDR DISABLED FLASHCOPY UNIT=990A
FDR236 CPUE ACTIVATED I/O INTERCEPTS ON UNIT=990A
FDR007 STARTING TIME OF FULL VOL SWAP -- 17.56.21 -- UNIT=3390 ,IN=D#SAP005,O
FDR158 DATA SET IS ACTIVE DSN=DB2SSYS.LOGCOPY2.DS01
FDR158 DATA SET IS ACTIVE DSN=DB2SSYS.LOGCOPY2.DS02
FDR158 DATA SET IS ACTIVE DSN=DB2SSYS.LOGCOPY1.DS03
FDR158 DATA SET IS ACTIVE DSN=DB2SSYS.DSNDBC.DSNDB01.DBD01.I0001.A001
FDR158 DATA SET IS ACTIVE DSN=DB2SSYS.DSNDBC.DSNDB01.SPT01.I0001.A001
...
...

Chapter 2. Environment 1 SAP migration experiences using FDRPAS
57
Figure 2-52 Swap scenario IV
In this example, we cannot safely say that any of the systems does not make changes to the
data on the source volumes. Therefore, it is not possible to exclude any of the systems from
the swap. If you start the swap task as shown in Figure 2-52 without having monitor tasks
started on the other systems, you run into the same error that we discussed in 2.4.11,
“FDRPAS sequential swap of five volumes on four systems” on page 52 and illustrated in
Figure 2-44.
Refer to the generated messages shown in Figure 2-53. Messages FDR233 now indicate that
all four systems take part in the swap.
Vol=SAP003
Unit=9907
All volumes are online
on all systems
All volumes offline
CPU SERIAL# xxyyzzzzzz
Vol=SAP004
Unit=9908
Vol=SAP005
Unit=9904
Vol=SAP002
Unit=9901
Vol=SAP001
Unit=990A
Unit=990D
Unit=990E
Unit=990F
Unit=990C
Unit=990B
CPU SERIAL# xx44442066
CPU SERIAL# xxzzzz2066
CPU SERIAL# xx33332066
Monitor task as batch job with SYSIN:
MONITOR TYPE=SWAP
MOUNT SWAPUNIT=(990*)

Monitor task as batch job with SYSIN:
MONITOR TYPE=SWAP
MOUNT SWAPUNIT=(990*)

Monitor task as batch job with SYSIN:
MONITOR TYPE=SWAP
MOUNT SWAPUNIT=(990*)

//XTRTST2S JOB ,XTRTST2,NOTIVY=XTRTST2,MSGCLASS=X
//SWAP1 EXEC PGM=FDRPAS,REGION=0M
//STEPLIB DD DISP=SHR,DSN=XTRTST2.FDR5441.LOAD
//SYSPRINT DD SYSOUT=*
//SYSUDUMP DD SYSOUT=*
//SYSIN DD *
SWAP TYPE=FULL,MAXTASKS=3
MOUNT VOL=SAP001,SWAPUNIT=990B
MOUNT VOL=SAP001,SWAPUNIT=990C
MOUNT VOL=SAP001,SWAPUNIT=990D
MOUNT VOL=SAP001,SWAPUNIT=990E
MOUNT VOL=SAP001,SWAPUNIT=990F
SWAP task as batch job

58
IBM TotalStorage Migration Guide for the SAP User
Figure 2-53 Swap scenario IV - job output
2.5 FDRPAS through ISPF Interface
The FDRPAS ISPF interface allows you to initiate, monitor, and control FDRPAS operations
on the system to which you are currently logged on. The interface allows you to:
Initiate SWAP and SWAPDUMP tasks
Monitor active swaps
Confirm swaps that specified CONFIRMSWAP=YES or CONFIRMSPLIT=YES
Suspend and resume active swaps
Reply to certain FDRPAS messages
Display FDRPAS history records
Display basic information about any disk devices in your installation
Issue command TSO EXEC 'fdrpas.clist.library(ABRALLOC)' from your ISPF command
line to get to the FDR ISPF primary options menu. The appearance of this menu looks like
that shown previously in Figure 2-13 on page 28. Enter P in the command line to open the
FDRPAS Plug and Swap ISPF dialog.
FDR001 FDR PLUG AND SWAP - FDRPAS VER. 5.4/41T - INNOVATION DATA PR
FDR233 CPUE (SERIAL# xx11112066) ACKNOWLEDGES THE SWAP OF VOL=SAP005
FDR233 CPUB (SERIAL# xx22222066) ACKNOWLEDGES THE SWAP OF VOL=SAP005 AND HAS JOINED
IN SWAP OF UNIT=9905 TO 990A
FDR233 CPUC (SERIAL# xx44442066) ACKNOWLEDGES THE SWAP OF VOL=SAP005 AND HAS JOINED
IN SWAP OF UNIT=9905 TO 990A
FDR233 CPUA (SERIAL# xx33332066) ACKNOWLEDGES THE SWAP OF VOL=SAP005 AND HAS JOINED
IN SWAP OF UNIT=9905 TO 990A
FDRW66 SWAP OF VOL=SAP005 TO UNIT=990A STARTED ON 4 SYSTEMS (CPUE CPUB CPUC C
FDR236 CPUB ACTIVATED I/O INTERCEPTS ON UNIT=9905
FDR236 CPUC ACTIVATED I/O INTERCEPTS ON UNIT=9905
FDR236 CPUA ACTIVATED I/O INTERCEPTS ON UNIT=9905
FDR264 FDR DISABLED FLASHCOPY UNIT=9905
FDR236 CPUE ACTIVATED I/O INTERCEPTS ON UNIT=9905
FDR007 STARTING TIME OF FULL VOL SWAP -- 13.55.35 -- UNIT=3390 ,IN=D#SAP005,O
FDR158 DATA SET IS ACTIVE DSN=DB2SSYS.LOGCOPY2.DS01
FDR158 DATA SET IS ACTIVE DSN=DB2SSYS.LOGCOPY2.DS02
FDR158 DATA SET IS ACTIVE DSN=DB2SSYS.LOGCOPY1.DS03
FDR158 DATA SET IS ACTIVE DSN=DB2SSYS.DSNDBC.DSNDB01.DBD01.I0001.A001
FDR158 DATA SET IS ACTIVE DSN=DB2SSYS.DSNDBC.DSNDB01.SPT01.I0001.A001
FDR158 DATA SET IS ACTIVE DSN=DB2SSYS.DSNDBC.DSNDB01.SYSLGRNX.I0001.A001
FDR158 DATA SET IS ACTIVE DSN=DB2SSYS.DSNDBC.DSNDB06.DSNDCX01.I0001.A001
...
... (cont.)

Chapter 2. Environment 1 SAP migration experiences using FDRPAS
59
To actually initiate a swap, you must enter, for example, SAP* as shown in Figure 2-54.
Figure 2-54 FDRPAS ISPF panel FDRPAS1
The result is a list of volumes that qualify for SAP*: SAP001, SAP002, SAP003, SAP004, and
SAP005 in our case. Figure 2-55 shows sample results. The options that are available to you
are listed at the top of the screen; one of them is SW or swap.
Figure 2-55 FDRPAS ISPF screen FDRPAS1 with selected volumes
Before you can really initiate the swap, you must tell FDRPAS to which units you want your
volumes to be swapped. Enter the unit addresses in the column “Swap to Offline Unit” and
specify swap in the command column. Press Enter.
Pressing Enter initiates the swap. If nothing seems to happen, check the messages in your
PASPROC procedure. If you have not yet created a procedure on an active PROCLIB, the
swap cannot start. This is also true if there is a JCL error in this procedure so it does not start
FDRPAS1----------------------- FDRPAS Plug & Swap ------------------ Row 1 of 1
COMMAND ===> SCROLL ===> PAGE

Active COnfirm SWap DUmp Msg SUspend REsume ABort OPtions HIstory SOrt
Panel: 1 of 4
Command Volume Unit Swap to Refresh 0
Serial Addr Offline
Mask Mask Unit Status
------- ------ ---- ------- ----------------------------------------------
''' sap*
Pass: % Tracks to copy: Copied: Updated:
Source - Reserve: Level: Pace: Type: Cyls:
Target - Reserve: Level: Pace: Type: Cyls:
Storgrp: SSID: CU Serial#:

******************************* Bottom of data ********************************
FDRPAS1------------------------ FDRPAS Plug & Swap ------------------ Row 1 of 5
COMMAND ===> SCROLL ===> PAGE

Active COnfirm SWap DUmp Msg SUspend REsume ABort OPtions HIstory SOrt
Panel: 1 of 4
Command Volume Unit Swap to Refresh 0
Serial Addr Offline
Mask Mask Unit Status
------- ------ ---- ------- ----------------------------------------------
SAP005 806B
Pass: % Tracks to copy: Copied: Updated:
Source - Reserve: 0 Level: 1 Pace: Type: 3390-9 Cyls: 10017
Target - Reserve: 0 Level: 1 Pace: Type: 3390-9 Cyls: 10017
Storgrp: DB2SGRP SSID: 0016 CU Serial#: 00111
SAP004 806A
Pass: % Tracks to copy: Copied: Updated:
Source - Reserve: 0 Level: 1 Pace: Type: 3390-9 Cyls: 10017
Target - Reserve: 0 Level: 1 Pace: Type: 3390-9 Cyls: 10017
Storgrp: DB2SGRP SSID: 0016 CU Serial#: 00111


60
IBM TotalStorage Migration Guide for the SAP User
properly. If you encountering such problems, refer to “Create cataloged procedure
PASPROC” on page 30 for some additional information about this procedure.
The first information that is shown in the Status column is SYNCHRONIZING, as shown in
Figure 2-56.
Figure 2-56 FDRPAS ISPF screen FDRPAS1 - SYNCHRONIZING
Let’s assume that the environment in which you are operating is as described in 2.4.13,
“FDRPAS concurrent swap” on page 56, but you have not yet started monitor tasks for your
volumes on all connected systems. In this case, FDRPAS does not get acknowledgements
from these systems, and message FDRW68 shown in Figure 2-57 pops up on your ISPF
screen. As discussed previously, FDRPAS is now awaiting your decision and reply in order to
either continue, RETRY, or terminate the swap of the first volume in the sequence.
FDRPAS1---------------------- FDRPAS Plug & Swap ------------------ Row 1 of 5
COMMAND ===> SCROLL ===> PAGE

Active COnfirm SWap DUmp Msg SUspend REsume ABort OPtions HIstory SOrt
Panel: 1 of 4
Command Volume Unit Swap to Refresh 0
Serial Addr Offline
Mask Mask Unit Status
------- ------ ---- ------- ----------------------------------------------
SAP005 806B 9905 SYNCHRONIZING
Pass: % Tracks to copy: Copied: Updated:
Source - Reserve: 0 Level: 1 Pace: Type: 3390-9 Cyls: 10017
Target - Reserve: 0 Level: 1 Pace: Type: 3390-9 Cyls: 10017
Storgrp: DB2SGRP SSID: 0016 CU Serial#: 00111

SAP004 806A 9904 SYNCHRONIZING
Pass: % Tracks to copy: Copied: Updated:
Source - Reserve: 0 Level: 1 Pace: Type: 3390-9 Cyls: 10017
Target - Reserve: 0 Level: 1 Pace: Type: 3390-9 Cyls: 10017
Storgrp: DB2SGRP SSID: 0016 CU Serial#: 00111

Recommendation/Important: Do
not
reply YES to message FDRW68.

Chapter 2. Environment 1 SAP migration experiences using FDRPAS
61
Figure 2-57 FDRPAS ISPF screen FDRPAS1 - FDRW68 message
You can now either reply to this message via console or you can enter MSG on either the
command line or in the command column. The screen shown in Figure 2-58 appears. Since
nothing has happened to your data so far, you can easily enter NO to terminate the swap.
Figure 2-58 FDRPAS ISPF screen FDRPAS5 - Reply
The status for the volume which was previously in synchronization changes to INACTIVE,
which indicates that you can start over with your activity.
You cannot start monitor tasks from the ISPF panels. Refer to 2.4.11, “FDRPAS sequential
swap of five volumes on four systems” on page 52 for a description of how to start monitor
tasks. In our case, we used the same job we had used previously to start monitor tasks on all
CPUs. Enter swap again in the command column when the monitor tasks are running and you
are ready to start the swap of your volume.
After a short phase of synchronization, the status should change to ACTIVE SWAP as shown
in Figure 2-59.
FDRPAS1---------------------- FDRPAS Plug & Swap ------------------ Row 1 of 5
COMMAND ===> SCROLL ===> PAGE

Active COnfirm SWap DUmp Msg SUspend REsume ABort OPtions HIstory SOrt
Panel: 2 of 4
Command Volume Unit Swap to Refresh 0
Serial Addr Offline
Mask Mask Unit Status
------- ------ ---- ------- ----------------------------------------------
SAP005 806B 9904 WAIT FOR CONSOLE REPLY FDRW68 CAUTION REQUEST
Pass: % Tracks to copy: Copied: Updated:
SAP004 806A INACTIVE
Pass: % Tracks to copy: Copied: Updated:
SAP003 8069 INACTIVE
Pass: % Tracks to copy: Copied: Updated:
SAP002 8068 INACTIVE
Pass: % Tracks to copy: Copied: Updated:
SAP001 8066 INACTIVE
Pass: % Tracks to copy: Copied: Updated:
******************************* Bottom of data ********************************
. . . . . . . . . . . . . . . . . . . . . . . . . . .
FDRPAS5 -------------------- FDRPAS Plug & Swap ------------------ Row 1 of 1
COMMAND ===> SCROLL ===> PAGE

The following operator console messages are waiting for reply, You may enter a
reply in this panel, or PF3 to return to the prior panel.
-------------------------------------------------------------------------------
Message: FDRW68 CAUTION REQUEST TO SWAP VOL=SAP001 TO UNIT=9901 ON 1 SYSTEMS
- IGNORING NON-RESPONDING CPUS REPLY YES, NO OR RETRY
Reply:
-------------------------------------------------------------------------------
******************************* Bottom of data ********************************

62
IBM TotalStorage Migration Guide for the SAP User
Figure 2-59 FDRPAS ISPF screen FDRPAS1 - ACTIVE SWAP
After the swap has finished, the status changes again from ACTIVE SWAP to SWAPPED.
To check which swaps have already been performed on your system, you can either enter
command HI on the command line, which results in a complete list of all former swaps, or
type HI in the command column for just a specific volume. We used the second option to just
verify the latest swap. The provided information is shown in Figure 2-60.
Figure 2-60 FDRPAS ISPF scree FDRPAS4 - History for latest swap
Use of options
The description of initiating a swap presented on the last few pages assumed that you do not
have any need to specify certain options described in previous sections. If you want to use
the ISPF screens and you
do
have a need for any specific parameter settings, you can make
these specifications by entering OP(tions) on the command line. The ISPF dialog comes up
with a screen as shown in Figure 2-61.
. . . . . . . . . . . . . . . . . . . . . . . . . . .
FDRPAS1 -------------------- FDRPAS Plug & Swap ------------------ Row 1 of 1
COMMAND ===> SCROLL ===> PAGE

Active COnfirm SWap DUmp Msg SUspend REsume ABort OPtions HIstory SOrt
Panel: 1 of 4
Command Volume Unit Swap to Refresh 0
Serial Addr Offline
Mask Mask Unit Status
------- ------ ---- ------- ----------------------------------------------
SAP001 9906 9901 ACTIVE SWAP (MAIN)
Pass: 1 26 % Tracks to copy: 74465 Copied: 19620 Updated: 0
Source - Reserve: 0 Level: 1 Pace: 0 Type: 3390-9 Cyls: 10017
Target - Reserve: 1 Level: 1 Pace: 0 Type: 3390-9 Cyls: 10017
Storgrp: DB2SGRP SSID: 0805 CU Serial#: 22222

******************************* Bottom of data ********************************
FDRPAS4 -------------------- FDRPAS Plug & Swap History ---------- Row 1 of 1
COMMAND ===> SCROLL ===> PAGE

Command Volume Unit Swapped System Date Time
Serial Addr to Unit
------- ------ ---- ------- -------- ---------- --------
SAP001 9906 9901 CPUE 08/26/2004 16:10:03
******************************* Bottom of data ********************************

Chapter 2. Environment 1 SAP migration experiences using FDRPAS
63
Figure 2-61 FDRPAS ISPF screen FDRPASO - Options panel
You can now change any values that are needed for your next swap. For example, we
changed CONFIRMSWAP from NO to YES, then left the options panel by pressing the PF3
key.
We started the next swap on our machine after we applied this change. As before, the swap
first enters status SYNCHRONIZING. When all systems are ready to start the swap, the
status changes to ACTIVE SWAP. Instead of just changing to SWAPPED when the swap is
finished now, the process is supposed to wait for our input before if finally swaps volume
SAP002 from unit 9907 to unit 9902. As you can see from Figure 2-62, the fact that you must
confirm the final swap is already added to the ACTIVE SWAP information.
Figure 2-62 FDRPAS ISPF screen FDRPAS1 - ACTIVE SWAP CONFIRMSWAP
FDRPASO -------------------- FDRPAS Plug & Swap - Options --------------------
COMMAND ===> SCROLL ===> PAGE

Options for SWAP command: Options for SWAPDUMP command:
#SYSTEMS ===> 1 #SYSTEMS ===> 1
MIN#SYSTEMS ===> MIN#SYSTEMS ===>
CONFIRMSWAP ===> NO (yes no) CONFIRMSPLIT ===> NO (yes no)
CONFMESS ===> NO (yes no) CONFMESS ===> NO (yes no)
LOGMESS ===> YES (yes no) LOGMESS ===> NO (yes no)
PACEDELAY ===> 0 1/100 seconds PACEDELAY ===> 0 1/100 seconds
SWAPDELAY ===> 15 seconds SWAPDELAY ===> 15 seconds
CHECKTARGET ===> YES CHECKTARGET ===> NO
LARGERSIZE ===> NO (ok no) LARGERSIZE ===> NO (ok no)
PACING ===> STATIC (dynam static) PACING ===> STATIC (dynam static)

Other options:
Max Start ===> 10 maximum number of simultaneous Swap/Swapdump to start
Maxvols ===> 1 maximum number of volumes to generate per process
Maxtasks ===> 1 maximum number of concurrent volumes to process
Maxactive ===> NO limit the number of concurrent tasks in copy phase
Interval ===> 2 refresh interval in seconds
WTOR ===> YES prompt console operator for message reply

. . . . . . . . . . . . . . . . . . . . . . . . . . .
FDRPAS1 -------------------- FDRPAS Plug & Swap ------------------ Row 1 of 1
COMMAND ===> SCROLL ===> PAGE

Active COnfirm SWap DUmp Msg SUspend REsume ABort OPtions HIstory SOrt
Panel: 1 of 4
Command Volume Unit Swap to Refresh 0
Serial Addr Offline
Mask Mask Unit Status
------- ------ ---- ------- ----------------------------------------------
SAP002 9907 9902 ACTIVE SWAP CONFIRMSWAP (MAIN)
Pass: 1 88 % Tracks to copy: 101085 Copied: 89781 Updated: 11
Source - Reserve: 0 Level: 1 Pace: 0 Type: 3390-9 Cyls: 10017
Target - Reserve: 1 Level: 1 Pace: 0 Type: 3390-9 Cyls: 10017
Storgrp: DB2SGRP SSID: 0805 CU Serial#: 22222

******************************* Bottom of data ********************************

64
IBM TotalStorage Migration Guide for the SAP User
Once the swap reaches the point where it will actually swap from one unit to another, it places
the swap in status WAIT FOR CONFIRMSWAP as shown in Figure 2-63.
Figure 2-63 FDRPAS ISPF screen FDRPAS1 - WAIT FOR CONFIRMSWAP
We now have two choices to handle this waiting swap: confirm it or abort it. To confirm the
swap now, enter CO(nfirm) on either the command line of in the command column. The swap
enters the last few phases and finally ends up in status SWAPPED.
If you take the other option, AB(ort), the swap changes to a phase called CANCEL and turns
to INACTIVE afterwards.
2.6 FDRERASE
FDRERASE is a product you can use to erase all data from disk volumes before they are
reused or removed from a data center. It is particularly useful, after a successful swap from
one disk system to another, when you want to make sure that your data is definitely not
readable any more by an eventual new owner of the subsystem. Another typical situation
when a product such as FDRERASE is extremely useful is, for example, when you are
leaving a disaster recovery site after a test or real disaster recovery and you want to make
sure that no residual data remains on the disks.
Rules exist, at least for the US and some European countries, which require that various
corporate data remains secure even when it is not needed anymore. This applies, for
example, to health insurance records.
Due to time constraints, a significant number of data processing sites simply relabel old disk
volumes when they want to make the data unreadable, but relabelling does not really destroy
all data on the disk. Instead, it just wipes out the VTOC but leaves the data records on the
volume where they can be accessed by unauthorized users.
The same is true for data sets which you delete from a DASD. Deleting a data set does not
really erase the data from disk, but just deletes the VTOC pointers to the tracks containing the
data.
FDRPAS1 -------------------- FDRPAS Plug & Swap ------------------ Row 1 of 1
COMMAND ===> SCROLL ===> PAGE

Active COnfirm SWap DUmp Msg SUspend REsume ABort OPtions HIstory SOrt
Panel: 1 of 4
Command Volume Unit Swap to Refresh 0
Serial Addr Offline
Mask Mask Unit Status
------- ------ ---- ------- ----------------------------------------------
SAP002 9907 9902 WAIT FOR CONFIRMSWAP (MAIN)
Pass: 6 % Tracks to copy: 0 Copied: 0 Updated: 0
Source - Reserve: 0 Level: 1 Pace: 0 Type: 3390-9 Cyls: 10017
Target - Reserve: 1 Level: 1 Pace: 0 Type: 3390-9 Cyls: 10017
Storgrp: DB2SGRP SSID: 0805 CU Serial#: 22222

******************************* Bottom of data ********************************

Chapter 2. Environment 1 SAP migration experiences using FDRPAS
65
A solution for this is FDRERASE, because it is capable of securely erasing disk volumes
quickly and easily. FDRERASE offers the following three levels of data erasure, which we
describe in more detail on the next few pages:
QUICKERASE
ERASE
SECUREERASE
2.6.1 Considerations for erasing data from modern disks
Modern disk systems use internal Fixed Block Architecture (FBA) to emulate the
count-key-data (CKD) disks which are used by z/OS. FDRERASE is able to overwrite the
emulated CKD tracks and makes the original data unavailable.
There are a few things to consider when you plan to erase disks:
Usually, FBA disks can be removed from a CKD disk subsystem and attached to another
system as FBA disks. If you do not use ERASE or SECUREERASE to erase those disks,
it might be possible to recover your data directly from the FBA disks.
The mapping of the emulated CKD disk volumes onto the FBA disks can be changed. If
this mapping is changed, some of your old data may reside in areas of the FBA disks
which are no longer in use. If these FBA disks are removed from the subsystem, it may be
possible to recover this data. To make sure that this does not happen to your data, you
can use FDRERASE prior to remapping your disks.
There is one situation in which FDRERASE cannot erase data from disks. This may
happen if an FBA disk fails. In this case, the subsystem is able to assign an unused hot
spare FBA disk to replace the failed disk, and recreate the data that was on the failed disk.
When this failed disk is now returned to the vendor for diagnosis and repair, your data may
still be on this disk. In this case, you must contact your vendor to insure that your data is
secure.
There are some special considerations for disks from different vendors. Refer to the original
manuals for the most up-to-date information on this.
2.6.2 Types of erase
This section describes the three levels of data erasure that FDRERASE can perform.
QUICKERASE
This option of FDRERASE uses one of several hardware erase functions to erase data from
CKD disk volumes. QUICKERASE does not really overwrite existing data, but the tracks on
the disk are modified in such a way that no records are left on each track. The mechanism to
erase the data varies a little bit depending on the disk vendor, but in the end, data on the FBA
disks is overwritten and it is no longer possible for an MVS subsystem to recover any data
from the emulated CKD disk tracks.
As you can easily guess from the method’s name, this option is usually the fastest erase
option. Although it does not necessarily follow any official security standards, it can be safely
used when the disks are reused as CKD disks, such as at a disaster/recovery site or when
you reuse the disks within your own data center.
ERASE
This option of FDRERASE is not very much slower than QUICKERASE, but it meets the
guidelines for “clearing” a disk.

66
IBM TotalStorage Migration Guide for the SAP User
The default behavior of ERASE is to write a track-length record of binary zeros on every
track. This ensures that the data on the FBA disks is completely overwritten. Doing so,
FDRERASE actually only sends a few bytes per track down the disk channel, because the
control unit pads the record with additional zeros.
We used QUICKERASE and ERASE, both with the default options, for a first easy test. When
you compare the elapsed times of both executions in Figure 2-64 and Figure 2-65, you can
see that ERASE takes approximately 40% longer than QUICKERASE. Both results seem to
be decent times for the erase of a disk volume of 10 GB. FDRERASE always erases
complete volumes. You cannot use this product to just erase parts of it, such as single data
sets.
Figure 2-64 QUICK ERASE with default options
Figure 2-65 ERASE with default options
SECUREERASE
This option of FDRERASE meets the NCSC requirements for “purging” a disk.
Because these requirements are very stringent, SECUREERASE writes a minimum of three
passes of data on each track, with a varying pattern of data in each pass. After your disk has
gone through the process of SECUREERASE, even sophisticated recovery techniques for
recovering data from FBA disks can not bring back the original contents.
As you can imagine, using this approach consumes more time to finish. The results of a
SECUREERASE with default values are shown in Figure 2-66. The elapsed time was
approximately five times that of QUICKERASE and two and a half times the ERASE elapsed
time.
FDRERA1 ------------------------- FDRERASE ------------------------ Row 1 of 1
COMMAND ===> SCROLL ===> PAGE

Active QUick ERase SEcure SIm EMpty ABort SUspend REsume OPtions HIstory SOrt
Refresh 0
Command Unit Volume Elapsed Control SMS
Addr Serial Cmp Pass Time Unit Storage
Mask Mask % # Status MM:SS SSID Serial# Group Cyls
------- ---- ------ --- --- -------------- ------- ---- ------- -------- -----
9901 1/1 ERASED QUICK 2:22 0805 22222
******************************* Bottom of data ********************************
FDRERA1 ------------------------- FDRERASE ------------------------ Row 1 of 1
COMMAND ===> SCROLL ===> PAGE

Active QUick ERase SEcure SIm EMpty ABort SUspend REsume OPtions HIstory SOrt
Refresh 0
Command Unit Volume Elapsed Control SMS
Addr Serial Cmp Pass Time Unit Storage
Mask Mask % # Status MM:SS SSID Serial# Group Cyls
------- ---- ------ --- --- -------------- ------- ---- ------- -------- -----
9902 1/1 ERASED ERASE 3:56 0805 22222 10017
******************************* Bottom of data ********************************

Chapter 2. Environment 1 SAP migration experiences using FDRPAS
67
Figure 2-66 SECURE ERASE with default options
2.6.3 FDRERASE JCL requirements
Our intent is not to describe every detail of the FDRERASE product here. Instead we describe
the main components of the product and the basic information you need to get started on
your own erasure activities.
EXEC statement
This statement specifies the program name FDRERASE and a region size. It is
recommended that you use 0M.
STEPLIB
You can either work with STEPLIB or JOBLIB to specify the FDRPAS load library. Make sure
that the library specified here is APF-authorized.
FDRSUMM
This is an optional output data set for an FDRERASE summary report. If you do not specify
this, the messages are written to SYSOUT. If you do plan to specify it, the appropriate DCB
characteristics are:
RECFM=FBA
LRECL=121
SYSPRINT
The required DD statement SYSPRINT specifies the output message data set.
SYSUDUMP
Always specify a SYSUDUMP DD statement, because it is very helpful in error situations. If
you do not specify a separate data set name, dump messages are written to SYSOUT. This
approach is reasonable in most situations.
FDRERA1 ------------------------- FDRERASE ------------------------ Row 1 of 1
COMMAND ===> SCROLL ===> PAGE

Active QUick ERase SEcure SIm EMpty ABort SUspend REsume OPtions HIstory SOrt
Refresh 0
Command Unit Volume Elapsed Control SMS
Addr Serial Cmp Pass Time Unit Storage
Mask Mask % # Status MM:SS SSID Serial# Group Cyls
------- ---- ------ --- --- -------------- ------- ---- ------- -------- -----
9905 3/3 ERASED SECURE 11:02 0805 22222 10017
******************************* Bottom of data ********************************
Attention: If not otherwise specified, FDRERASE only operates on offline disks.
Nevertheless, FDRERASE can destroy data if misused. Therefore, you should limit access
to its functions. To do so, you can use your security system to limit access to the program
library that contains FDRERASE or the program name FDRERASE.

68
IBM TotalStorage Migration Guide for the SAP User
SYSIN
This specifies the control statement data set. We used DD * along with an input stream for our
tests. You can, however, also specify an external data set and point to it if you prefer.
2.6.4 FDRERASE control statements
In contrast to FDRPAS where you have a large number of control statements FDRERASE
only consists of a main statement, which is
RASE/QUICKERASE/SECUREERASE/SIMERASE/EMPTYVTOC and one or more MOUNT
statements, which we describe hereafter.
FDRERASE main statement
Refer to Figure 2-67 for an overview of the operands for the FDRERASE main statement.
Figure 2-67 FDRERASE main statement
SIMERASE
You can use this statement to validate your FDRERASE control statement and to check
which volumes qualify the specified syntax because it lists all disks that meet the selection
criteria.
QUICKERASE statement
Refer to “QUICKERASE” on page 65 for details regarding this statement.
ERASE statement
Refer to “ERASE” on page 65 for details regarding this statement.

ERASE

SIMERASE

TYPE

=

QUICKERASE

SECUREERASE

EMPTYVTOC

FULL

,
ACTIVETARGET

=

BYPSS

PROCESS

,
CONFERASE

=

NO

YES

,
CPYVOLID

=

NO

YES

,
ERASEPATTERN

=

,
ERASEPASS

n

=

nnn

,
MAXTASKS

=

,
LOGMESS

=

YES

NO

64

nn

,
ONLINE

=

BYPASS

VARYOFF

,
SELTERR

=

64

nn

YES

NO

,
VARYON

=

NOAFTER

AFTER

=

,
ALREADYERASED

=

BYPASS

PROCESS

,
CHECKTARGET

=

YES

NO

Chapter 2. Environment 1 SAP migration experiences using FDRPAS
69
SECUREERASE statement
“SECUREERASE” on page 66 discusses this option in detail.
EMPTYVTOC statement
When you use this statement, no data is really erased from your disks. This is a very quick
way of initializing an empty volume, because it just rewrites the VTOC as empty and rebuilds
the VTOCIX.
TYPE=FULL
This operand is mandatory on this statement.
ACTIVETARGET=
Valid options are PROCESS and BYPASS. The default is BYPASS.
This option influences FDRERASE’s behavior for disks which are active in PPRC. PROCESS
erases those active disks. BYPASS does not touch such volumes.
ALREADYERASED=
Valid options are PROCESS and BYPASS. The default is BYPASS.
To save time and resources, disks which have previously been erased by FDRERASE will be
bypassed if you stick with the default. This does not apply if the previous erase had
CPYVOLID set to YES, because in this case, FDRERASE can’t tell that the volume was
previously erased.
CHECKTARGET=
Valid options are YES and NO. The default is YES.
If you change the default, any offline disk volume can be erased by FDRERASE unless you
have specified other RACF® security to prevent anybody from doing this. If you use
CHECKTARGET=NO, FDRERASE erases the data on the disk, but does not uncatalog the
data sets.
The default option CHECKTARGET YES is preferred in most cases because FDRERASE
checks disk devices before beginning an erase operation to ensure that the disk is empty or
was created by FDRPAS and is the old source volume of a successful SWAP.
CONFERASE=
Valid options are YES and NO. The default is NO.
Change the setting to YES if you want FDRERASE to generate an open reply via console
message FDRW01, which you must answer before the erase can start.
CPYVOLID=
Valid options are YES and NO. The default is NO.
Specify YES if you plan to use the volume after the erase. In this case, FDRERASE invokes
ICKDSF to rebuild an empty VTOC and VTOCIX after the erase.
Attention: If you use CHECKTARGET=NO, FDRERASE cannot insure that the selected
devices are offline to all other sharing systems. This means that it is possible to erase a
disk which is in use on another system!

70
IBM TotalStorage Migration Guide for the SAP User
If you just accept the default, there will be no VTOC on the volume after the erase, meaning
that if you later decide to use this empty volume, you must initialize the volume using ICKDSF
manually at that point.
ERASEPASS=n
Valid values for n are between:
3 and 8 for SECUREERASE
1 and 8 for ERASE
The only acceptable value for QUICKERASE is one.
This option does not apply to SIMERASE, and EMPTYVTOC.
Use this parameter if you want to change the number of times FDRERASE overwrites each
track.
ERASEPATTERN=
Only applies for ERASE.
You can specify up to 8 bytes, that is, two hexadecimal digits per byte, which FDRERASE
then uses to override the currently existing data. The first byte is then used for the first pass,
the second byte for the second pass, and so on. If you specify fewer bytes than the value of
ERASEPASS=n, FDRERASE uses binary zeros for the extra passes.
There are two byte pattern with a special meaning:
01 During this pass, FDRERASE erases the track, which is similar to QUICKERASE.
This could be the last pattern to cause the track to be erased, leaving no records on
the track.
FE If you specify this pattern, it causes FDRERASE to use a random value.
LOGMESS=
Valid values are YES and NO. The default is YES.
Specify NO if you do not want any SYSLOG and console messages to be written.
MAXTASKS=
Valid values are between 1 and 64. The default is 64.
Analogous to FDRPAS, you can use this operand to specify the maximum number of volumes
which can be erased concurrently.
The value that you specify for MAXTASKS only applies to the disks on a single MOUNT
statement. This means that if you specify just 4 disks in a specific MOUNT statement, only 4
disks will be erased concurrently, even if MAXTASKS is set to 15 and there are additional
MOUNT statements following. Refer to “MOUNT statement” on page 71 for a more detailed
description of the MOUNT statement.
ONLINE=
Valid values are VARYOFF and BYPASS. The default is BYPASS.
If you decide to override the default value for this operand, disk devices which are online to
the system on which you run the FDRERASE program are varied offline and erased if they
are not currently allocated to any task on this system. To prevent you from accidently erasing

Chapter 2. Environment 1 SAP migration experiences using FDRPAS
71
online volumes, FDRERASE issues an additional WTOR message FDRW01 so that you
have to confirm the erasure of every single volume.
SELTERR=
Valid values are YES and NO. The default is YES unless overridden in the FDR Global
Option table. Refer to 2.2, “Install FDRPAS” on page 24 for a discussion of this table.
If you specify NO, you ask FDRERASE not to issue a condition code or ABEND when a step
terminates.
VARYON=
Valid values are AFTER and NOAFTER. The default is NOAFTER.
Use this operand with the non-default value AFTER if you want FDRERASE to vary the
erased volume online and remount it when done with the erase.
MOUNT statement
The specification of at least one MOUNT statement is mandatory after the main statement.
Use this statement to specify a disk device or a set of devices for FRDERASE. The syntax
looks as shown in Figure 2-68.
Figure 2-68 FDRERASE Mount statement syntax
ERASEUNIT=
Use this mandatory operand to specify one or more MVS device addresses of the disks that
you want to erase. A trailing asterisk is allowed as replacement for up to three digits.
Attention: ONLINE=VARYOFF is preliminary intended for use when erasing disks as you
are leaving a disaster recovery site. Be aware that unintended use of this operand can
cause the loss of data.

MOUNT

ERASEUNIT

=

,
STORAGEGROUP
VTOCTRK

=

,
CHANGEVOL

=

n

,
uuuu
)
(
VOLSER

,
CHANGEVOLNUM

=

,
VTOCCYL

=
cccc

,

O

tt

,
VTOCSIZE

=
nnnnn

,
VTOCLOC

=
nnnnn

72
IBM TotalStorage Migration Guide for the SAP User
The maximum number of devices or ranges of devices cannot exceed 255 subparameters
and the total number of devices that you want to erase with one MOUNT statement can be up
to 8190.
CHANGEVOL=
If you plan to change the volume serial of the erased volume once the erase completes, you
can use CHANGEVOL= for this task. As a prerequisite for this being a valid operand, you
must also specify CPYVOLID=YES on the main statement.
You can use three types of wildcards for the new volume serials:
* The equivalent character is copied from the original volume serial on the disk. Refer
to Example 2-1 and assume that the original volume serial was SAP001.
Example 2-1 CHANGEVOL= wildcard * usage
CHANGEVOL=ABC***
changes the volume serial name to ABC001.
/Use this to insert a sequence number into the new serial. This wildcard operates in
conjunction with the operand that we describe next, CHANGEVOLNUM=. Refer to
Example 2-2 for some clarification. Assume that CHANGEVOLNUM is set to 555.
Example 2-2 CHANGEVOL= wildcard / usage
The first CHANGEVOL=ABC///
changes the volume serial name to ABC555
The next volume in the row changes the volume serial name to ABC556
&UUU FDRERASE replaces the volume serial name with the device number plus any
other pattern parts. See Example 2-3 as a usage note. Assume that the unit
number is 1ABC and the volume serial is SAP001
Example 2-3 CHANGEVOL= wildcard &UUU usage
CHANGEVOL=XX&UUU
changes the volume serial name to XX1ABC
CHANGEVOL=**&UUU
changes the volume serial name to SA1ABC
CHANGEVOLNUM=
Valid values range from 0 to 999999. The default is 0.
Use this parameter to specify a starting sequence number for the use of one or more /
(slashes), on the CHANGEVOL= operand.
STORAGEGROUP
After the erase, the VTOC has the “SMS-managed” flag turned on. You must add the volume
serial number of the erased disk to your SMS configuration before it can be used as an
SMS-managed volume.
VTOCCYL=
You can use this option to specify a new starting location for the VTOC. This option is only
valid if your have also used CPYVOLID=YES and EMPTYVTOC is specified on the ERASE
statement. VTOCCYL actually specifies the starting cylinder.

Chapter 2. Environment 1 SAP migration experiences using FDRPAS
73
VTOCTRK=
in addition to VTOCCYL, you can also specify a starting track within the starting cylinder for
your VTOC. If you omit this parameter, starting track zero is assumed.
VTOCLOC=
Alternatively to VTOCCYL and VTOCTRK, you can also use VTOCLOC to specify in decimal
form the relative track on the volume where you want to store your VTOC information.
VTOCSIZE=
Valid values are 1 to 9999 tracks.
Use this parameter to specify the new VTOC size in tracks. This parameter can only be used
in conjunction with the other VTOC parameters of the MOUNT statement and must be set if
you use either VTOCCYL or VTOCLOC.
2.6.5 Usage of FDRERASE via ISPF panels
Once you have started the ISPF dialog as described in 2.5, “FDRPAS through ISPF Interface”
on page 58, select option E from the FDR primary option menu panel ABRPRIME to get to
the FDRERASE panel FDRERA1 shown in Figure 2-69.
Figure 2-69 FDRERASE ISPF panel FDRERA1
The line shown in bold type provides a list of valid commands for FDRERASE. One of the
available commands is OP(tions). Enter OP on the command line to check the current
settings, which are shown in Figure 2-70. When you enter the FDRERASE ISPF dialog for the
first time, this screen reflects the default settings of the parameters that we have previously
described. You can either accept the defaults or change any of the parameter setting as
appropriate for your environment.
FDRERA1 ------------------------- FDRERASE ------------------------ Row 1 of 1
COMMAND ===> SCROLL ===> PAGE
No active units
Active QUick ERase SEcure SIm EMpty ABort SUspend REsume OPtions HIstory SOrt
Refresh 0
Command Unit Volume Elapsed Control SMS
Addr Serial Cmp Pass Time Unit Storage
Mask Mask % # Status MM:SS SSID Serial# Group Cyls
------- ---- ------ --- --- -------------- ------- ---- ------- -------- -----
'''
******************************* Bottom of data ********************************

74
IBM TotalStorage Migration Guide for the SAP User
Figure 2-70 FDRERASE ISPF panel FDRERAO
In the upper right part of the panel, More: + indicates, that there is another panel with options
available for you. Press PF8 twice to scroll down to the bottom. The result is shown in
Figure 2-71. We show you this page because you will almost certainly need to change the
Job Statement Information on the second half of the panel. Change the job card to whatever
fits best for your local environment.
Another thing that you must definitely change before you can run your first FDRERASE is the
STEPLIB DD name. Replace DSN=FDRERASE.LOADLIB with the name of the real loadlib.
FDRERAO ------------------------- FDRERASE - Options --------------------------
COMMAND ===> SCROLL ===> PAGE

More: +
Dialog Options:
DISPLAY ===> OFFLINE Display disks: ONLINE, OFFLINE or BOTH

Options:
ACTIVETARGET ===> BYPASS Active volumes or PPRC/SRDF disk: BYPASS or PROCESS
ALREADYERASED ===> BYPASS Disks already erased: BYPASS or PROCESS
CHECKTARGET ===> YES Erase only FDRPAS or empty disks: YES or NO
CONFERASE ===> NO Issue console confirmation WTOR message: YES or NO
CPYVOLID ===> NO Rebuild the VTOC after erase: YES or NO
ERASEPASS ===> Number of times to overwrite each track: 1-8
ERASEPATTERN ===> Hexadecimal value to overwrite each track
LOGMESS ===> YES Issue SYSLOG message at erase completion: YES or NO
MAXTASKS ===> 64 Number of disks to erase concurrently: 1-64
ONLINE ===> BYPASS Disks that are online: BYPASS or VARYOFF
SELTERR ===> YES Abend if selected disk is not processed: YES or NO
VARYON ===> NOAFTER Vary volume online after erase: AFTER or NOAFTER

CHANGEVOL ===> Volume Serial mask to rename volume after erase

. . . . . . . . . . . . . . . . . . . . . . . . . . .

Chapter 2. Environment 1 SAP migration experiences using FDRPAS
75
Figure 2-71 FDRERASE ISPF 2nd Options screen
Once you have set all parameters and JCL statements as needed, you can return to screen
FDRERA1 by pressing the PF3 key.
Simulate an erase
If you want to simulate an erase, you must first specify a unit name or volser for the
simulation. As on many other occasions, you can use the asterisk as wildcard for the
specification. In our case, we specified 990* in the Unit Addr Mask. When you press enter,
the panel shows a list of offline volumes that qualify for the mask. Refer to Figure 2-72 for an
example of this.
Figure 2-72 FDRERASE list of offline volumes for specific unit mask
. . . . . . . . . . . . . . . . . . . . . . . . . . .
FDRERAO-------------------------- FDRERASE - Options --------------------------
COMMAND ===> SCROLL ===> PAGE
END OF DATA
More: -
VTOCCYL ===> New VTOC starting location cylinder: 0-4368
VTOCTRK ===> New VTOC starting location track: 0-14
VTOCSIZE ===> New VTOC size in tracks: 1-9999 (decimal)

Other options:
Maxunits ===> 255 Maximum # of units to generate per jobstep: 1-255
Interval ===> 2 Refresh interval in seconds

Job Statement Information:
===> //XTRTST2A JOB (ACCOUNT),'NAME',NOTIFY=XTRTST2,MSGCLASS=X
===> //*
===> //*
===> //*
===> //*
===> //FDRERASE EXEC PGM=FDRERASE,REGION=0M
===> //STEPLIB DD DISP=SHR,DSN=FDRERASE.LOADLIB
===> //SYSPRINT DD SYSOUT=*
===> //FDRSUMM DD SYSOUT=*
FDRERA1 ------------------------- FDRERASE ----------------------- Row 1 of 10
COMMAND ===> SCROLL ===> PAGE

Active QUick ERase SEcure SIm EMpty ABort SUspend REsume OPtions HIstory SOrt
Refresh 0
Command Unit Volume Elapsed Control SMS
Addr Serial Cmp Pass Time Unit Storage
Mask Mask % # Status MM:SS SSID Serial# Group Cyls
------- ---- ------ --- --- -------------- ------- ---- ------- -------- -----
990E INACTIVE 0805 22222 10017
990D INACTIVE 0805 22222 10017
990C INACTIVE 0805 22222 10017
990B INACTIVE 0805 22222 10017
990A INACTIVE 0805 22222 10017
9909 INACTIVE 0805 22222 10017
9905 3/3 ERASED SECURE 11:02 0805 22222 10017
9902 1/1 ERASED ERASE 3:56 0805 22222 10017
9901 1/1 ERASED QUICK 2:22 0805 22222 10017
9900 INACTIVE 0805 22222 10017
******************************* Bottom of data ********************************

76
IBM TotalStorage Migration Guide for the SAP User
To simulate an erase, you can type SIm on the command line and press Enter. Since all
FDRERASE runs are performed as batch jobs in the background, the dialog asks you which
of the three actions shown in Figure 2-73 you want to take next. We choose option 2 to check
the generated JCL first before we actually submit it.
Figure 2-73 FDRERASE ISPF screen FDRERAJ - Processing options
This option routes you to the ISPF editor, where you can make whatever changes are
necessary or just submit the job. The generated JCL looks like that shown in Figure 2-74. If
you refer back to Figure 2-72, only the volumes which are in status INACTIVE have been
selected by FDRERASE. Those which have been erased previously are ignored.
Figure 2-74 FDRERASE generated JCL for simulation
Either type submit of press PF3 and select option 3 afterwards to finally submit the job. Refer
to “SIMERASE” on page 68 for a functional description of SIMERASE. The job output
provides the information shown in Figure 2-75.
FDRERAJ ------------------------- FDRERASE - Select Processing Option ---------
OPTION ===>

Please select one of the following options or press the END key to cancel

1 - Browse the generated JCL

2 - Edit the generated JCL

3 - Submit the generated JCL
//XTRTST2A JOB (ACCOUNT),'NAME',NOTIFY=XTRTST2,MSGCLASS=X
//FDRERASE EXEC PGM=FDRERASE,REGION=0M
//STEPLIB DD DISP=SHR,DSN=XTRTST2.FDR5441.LOAD
//SYSPRINT DD SYSOUT=*
//FDRSUMM DD SYSOUT=*
//SYSUDUMP DD SYSOUT=*
//SYSIN DD *
SIMERASE TYPE=FULL,
ACTIVETARGET=BYPASS,
ALREADYERASED=BYPASS,
CHECKTARGET=YES,
CONFERASE=NO,
CPYVOLID=NO,
LOGMESS=YES,
MAXTASKS=64,
ONLINE=BYPASS,
SELTERR=YES,
VARYON=NOAFTER
MOUNT ERASEUNIT=(990E,
990D,
990C,
990B,
990A,
9909,
9900)

Chapter 2. Environment 1 SAP migration experiences using FDRPAS
77
Figure 2-75 FDRERASE - SIMERASE messages
Initiate an erase
To actually initiate an erase, you can either enter quick, erase, or secure on the command
line, or you can choose a different erase method for each volume, as shown in Figure 2-76.
Figure 2-76 FRDERASE - initiate different erases at once
The generated JCL ends up in three steps, sorted and packed by erase type: one step for
units 990E, and 990B with option ERASE; a QUICKERASE step for units 990D, 990C, 990A,
and 9909; and a SECUREERASE step for unit 9900.
You can now return to the FDRERASE mail panel to monitor the progress of the different
erase processes as shown in Figure 2-77.
FDRERASE WILL ERASE THE FOLLOWING 7 UNITS:
9900 9909 990A 990B 990C 990D 990E
DEVICE IS ELIGIBLE FOR ERASE UNIT=9900 VOL=SAP001 VOLID=FDR3
DEVICE IS ELIGIBLE FOR ERASE UNIT=9909 VOL=SAP004 VOLID=FDR3
DEVICE IS ELIGIBLE FOR ERASE UNIT=990A VOL=SAP001 VOLID=FDR2
DEVICE IS ELIGIBLE FOR ERASE UNIT=990B VOL=SAP001 VOLID=FDR3
DEVICE IS ELIGIBLE FOR ERASE UNIT=990C VOL=SAP002 VOLID=FDR3
DEVICE IS ELIGIBLE FOR ERASE UNIT=990D VOL=SAP003 VOLID=FDR3
DEVICE IS ELIGIBLE FOR ERASE UNIT=990E VOL=SAP004 VOLID=FDR3
FDR SUCCESSFULLY COMPLETED
FDRERA1 ------------------------- FDRERASE ----------------------- Row 1 of 10
COMMAND ===> SCROLL ===> PAGE

Active QUick ERase SEcure SIm EMpty ABort SUspend REsume OPtions HIstory SOrt
Refresh 0
Command Unit Volume Elapsed Control SMS
Addr Serial Cmp Pass Time Unit Storage
Mask Mask % # Status MM:SS SSID Serial# Group Cyls
------- ---- ------ --- --- -------------- ------- ---- ------- -------- -----
erase 990E INACTIVE 0805 22222 10017
quick 990D INACTIVE 0805 22222 10017
quick 990C INACTIVE 0805 22222 10017
erase 990B INACTIVE 0805 22222 10017
quick 990A INACTIVE 0805 22222 10017
quick 9909 INACTIVE 0805 22222 10017
9905 3/3 ERASED SECURE 11:02 0805 22222 10017
9902 1/1 ERASED ERASE 3:56 0805 22222 10017
9901 1/1 ERASED QUICK 2:22 0805 22222 10017
secure 9900 INACTIVE 0805 22222 10017
******************************* Bottom of data ********************************



. . . . . . . . . . . . . . . . . . . . . . . . . . .

78
IBM TotalStorage Migration Guide for the SAP User
Figure 2-77 FDRERASE - monitor erase progress
FDRERA1 ------------------------- FDRERASE ----------------------- Row 1 of 10
COMMAND ===> SCROLL ===> PAGE

Active QUick ERase SEcure SIm EMpty ABort SUspend REsume OPtions HIstory SOrt
Refresh 0
Command Unit Volume Elapsed Control SMS
Addr Serial Cmp Pass Time Unit Storage
Mask Mask % # Status MM:SS SSID Serial# Group Cyls
------- ---- ------ --- --- -------------- ------- ---- ------- -------- -----
990E 10 1/1 ACTIVE ERASE 2:44 0805 22222 10017
990D 19 1/1 ACTIVE QUICK 2:44 0805 22222 10017
990C 18 1/1 ACTIVE QUICK 2:44 0805 22222 10017
990B 10 1/1 ACTIVE ERASE 2:44 0805 22222 10017
990A 19 1/1 ACTIVE QUICK 2:44 0805 22222 10017
9909 19 1/1 ACTIVE QUICK 2:44 0805 22222 10017
9905 3/3 ERASED SECURE 11:02 0805 22222 10017
9902 1/1 ERASED ERASE 3:56 0805 22222 10017
9901 1/1 ERASED QUICK 2:22 0805 22222 10017
9900 INACTIVE 0805 22222 10017
******************************* Bottom of data ********************************

© Copyright IBM Corp. 2005. All rights reserved.
79
Chapter 3.
Environment 2 SAP migration
experiences
In this chapter we describe three distinct SAP migration scenarios:
The first scenario is a migration from non-IBM storage to an IBM ESS 800 storage
subsystem using IBM’s SAN Volume Controller to migrate the SAP data and Copy
Services (Flashcopy).
The second scenario is a migration of SAP data from non-IBM storage to the IBM ESS
800 using Tivoli Storage Manager (TSM) to back up the data from non-IBM storage and
then restore it to ESS 800.
In the third scenario we use INNOVATION Data Processing Corporation’s product called
FDR/UPSTREAM to move the SAP data off non-IBM storage to the ESS 800.
3

80
IBM TotalStorage Migration Guide for the SAP User
3.1 Environment 2 SAP migration using SVC
This section describes the SAP migration using the IBM SAN Volume Controller (SVC). We
describe how to configure the ESS 800, then the SVC, and then perform the migration.
3.1.1 ESS 800 Open Systems configuration for AIX
The ESS 800 was configured with 15 - 4GB RAID 5 volumes on cluster2, on a single loop,
and across a single disk group. These volumes were to contain the SAP database after
migration from the non-IBM storage. An additional 15 - 4GB RAID 5 volumes were created on
cluster1, also on a single loop, and across a single disk group as well, and contained the
FlashCopy volumes of the SAP data after migration was complete.
When you click the AIX host system icon (aixredbook) or the fibre channel host adapter icon
that the AIX host system is connected to, the ESS Specialist highlights the storage that is
assigned to that host.
Figure 3-1 shows the Open System storage allocation.
Figure 3-1 AIX Open Systems Storage
If you want to view
all
the storage that has been configured on the ESS, click the View All
Storage button on the top right of the ESS Specialist panel. The ESS Specialist will highlight
all storage that has been assigned to all host systems attached to the ESS.
Clicking the Open System Storage button allows you to view the current Open Systems
configuration or to modify the existing Open Systems configuration. You will have the option
to:
Add or modify hosts systems
Configure Host Adapters

Chapter 3. Environment 2 SAP migration experiences
81
Add Volumes
Modify Volumes
Configure Disk Groups
Figure 3-2 shows an example of the Open Systems configuration panel.
Figure 3-2 ESS Specialist Open Systems configuration Panel
Modify host systems
Clicking the Modify Host Systems button allows you to add a new host system or modify an
existing host system entry.
This panel shows the host attributes that were used for this test. The attributes are entered
into the window on left, under Host Attributes; clicking the Add button moves the attributes to
the Host System List window on the right. When all of the host system entries are completed
and added to the Host System List, the ESS is updated by clicking the Perform
Configuration Update button.
The Host Attributes used for this test were:
Attribute name Attribute entry
Nickname aixredbook
Host type IBM RS/6000
Host Attachment Fibre channel Attached
Hostname/ IP Address
Worldwide Port-name 10000000C932A7E6
Fibre-Channel Ports Bay1, Card2, Port1

82
IBM TotalStorage Migration Guide for the SAP User
Figure 3-3 shows the AIX host system configuration used for this test.
Figure 3-3 Modify Host Systems
Configure host adapter
The Configure Host Adapter Port option allows you define, update, or undefine an ESS Host
Adapter port. For hosts attaching to the ESS via fibre channel, the ESS supports three
interconnect topology or network structures:
Arbitrated loop topology
Fibre Channel-Arbitrated Loop (FC-AL) is a ring topology that enables you to interconnect
a set of nodes. ESS supports FC-AL as a “private” loop, it does not support the
fabric-switching in FC-AL. (Use this option to define the Host Adapter when attaching an
AIX host directly to the ESS.)
Switched fabric topology
The ESS supports the switched-fabric topology with the point-to-point protocol. You
should configure the ESS to fibre-channel adapter to operate in point-to-point mode when
connecting to a fabric topology.
Point-to-point topology
The point-to-point topology, or direct connect, allows you to interconnect ports directly.
When using point-to-point topology to directly connect a host to the ESS you must
configure the host adapter to use the FC-AL protocol.
To change an adapter protocol or topology you must first undefine the adapter and then
redefine it.
The configuration used for this test was switched fabric topology with point-to-point protocol.
Figure 3-4 shows the fibre channel host adapter configuration used for this test.

Chapter 3. Environment 2 SAP migration experiences
83
Figure 3-4 ESS Specialist Configure Host Adapter Ports panel
Configure disk group
The Configure Disk Group option is used to create RAID 5 or RAID 10 Arrays on one or more
clusters, adapter pairs, loops, and/or disk groups. This option can also be used to “undefine”
a disk group (delete an Array). As you can see in Figure 3-5, Array1 was configured as a
RAID5 array on Device Adapter Pair1, Cluster2, Loop B. Array1 was used to contain the SAP
database after the migration from non-IBM to the ESS 800. Array 2 was configured on Device
Adapter Pair 2, Cluster1, Loop B, and was used to contain the FlashCopy volumes for Array1.
To configure one or more disk groups, select the disk you want to configure, either one at
time or in a group, then select the attributes for the array. The array attributes are:
Storage type
– Raid5
– Raid10
Track format
– Fixed Block (FB)
When configuring Open Systems storage the track format will always be “Fixed Block.”
Figure 3-5 shows the Open Systems Disk Group configuration used for this test.

84
IBM TotalStorage Migration Guide for the SAP User
Figure 3-5 ESS Specialist Configure Disk Group Panel
Modify volume assignments
The Modify Volume Assignments option allows you to assign or unassign storage volumes to
Open Systems hosts, in this case AIX, by selecting one of the Action radio buttons and then
clicking the Perform Configuration Update button. The list of assigned volumes can be
sorted by any of the columns listed and is always sorted in ascending order.
The sorting choices are:
No sort
First
Second
Third
For example, when you select First for a particular column, the table is sorted first in
ascending order based on the values in that column. If you select Second for another column,
the values in that column pertain if there are two rows in the first column that contain the
same value.
The volume assignment list can also be printed by clicking the Print Table button, which
gives you a separate html screen to print from.
Figure 3-6 shows the Open Systems - Modify Volume Assignments panel with a partial list of
volume assignments used for this test.

Chapter 3. Environment 2 SAP migration experiences
85
Figure 3-6 ESS Specialist Modify Volume Assignment panel
Add volumes
The Add Volumes option allows you to add Open Systems storage volumes to one or more
disk groups. There are two parts to this function. In part 1 you select a host system and host
port, then one or more disk groups. If some volumes are already assigned to a host the ESS
Specialist will highlight them in this panel when you select a host or host port. You can select
a disk group that has space that has not yet been allocated (green). Once the host, host port,
and disk group or groups have been selected, click Next.
From the Add Volumes (2 of 2) panel you can select from the Available Free Space either
RAID 5 or RAID 10 capacity. This free space is based on the Disk Group configurations done
earlier. You may have a choice of one or the other, or both, depending on how the disk
groups are configured. In this case we have a total 420.97GB of RAID 5 free space in a disk
group (array). Once you have selected the appropriate Storage Type from the Available Free
Space window, select the volume size from the list of available volume sizes and the number
of volumes in the Volume Attributes window.
Once the volume size and number of volumes are selected, select Volume Placement from
the following choices:
Place volumes sequentially, starting in the first selected storage area.
Spread volumes across all selected storage areas.
Click ADD to move the selected volumes over to the New Volumes window. You can then
either select more volumes to be created or click Perform Configuration Update to create
the volumes on the selected disk groups.
Figure 3-7 and Figure 3-8 show examples of the Add Volumes panels.

86
IBM TotalStorage Migration Guide for the SAP User
Figure 3-7 ESS Specialist Add Volume (1 of 2) panel
Figure 3-8 ESS Specialist Add Volumes (2 of 2) panel
Table 3-1 shows the ESS volumes that were used for the AIX migration.

Chapter 3. Environment 2 SAP migration experiences
87
Table 3-1
AIX Migration volumes on the ESS800
3.2 FlashCopy process
ESS Copy Services provides a Web-based interface for setting up and managing
Peer-to-Peer Remote Copy (PPRC) and FlashCopy. For this redbook we are only concerned
with FlashCopy.
Figure 3-9 shows the ESS Welcome panel from which Copy Services is invoked.
Volume Location LSS Volume Type Size Storage Type Host Port Host
Nicknames
100-22222
(hdisk24)
Device Adapter Pair 1
Cluster 2, Loop B Array
1, Vol 000
0x11 Open System 004.0 GB RAID-5 Array Fibre Channel ID 00,
LUN 5100
aixredbook
101-22222
(hdisk25)
Device Adapter Pair 1
Cluster 2, Loop B Array
1, Vol 001
0x1 Open System 004.0 GB RAID-5 Array Fibre Channel ID 00,
LUN 5101
aixredbook
102-22222
(hdisk26)
Device Adapter Pair 1
Cluster 2, Loop B Array
1, Vol 002
0x11 Open System 004.0 GB RAID-5 Array Fibre Channel ID 00,
LUN 5102
aixredbook
103-22222
(hdisk27)
Device Adapter Pair 1
Cluster 2, Loop B Array
1, Vol 003
0x11 Open System 004.0 GB RAID-5 Array Fibre Channel ID 00,
LUN 5103
aixredbook
104-22222
(hdisk28)
Device Adapter Pair 1
Cluster 2, Loop B Array
1, Vol 004
0x11 Open System 004.0 GB RAID-5 Array Fibre Channel ID 00,
LUN 5104
aixredbook
105-22222
(hdisk29)
Device Adapter Pair 1
Cluster 2, Loop B Array
1, Vol 005
0x11 Open System 004.0 GB RAID-5 Array Fibre Channel ID 00,
LUN 5105
aixredbook
106-22222
(hdisk30)
Device Adapter Pair 1
Cluster 2, Loop B Array
1, Vol 006
0x11 Open System 004.0 GB RAID-5 Array Fibre Channel ID 00,
LUN 5106
aixredbook
107-22222
(hdisk31)
Device Adapter Pair 1
Cluster 2, Loop B Array
1, Vol 007
0x11 Open System 004.0 GB RAID-5 Array Fibre Channel ID 00,
LUN 5107
aixredbook
Note: A cross reference to the hdisks in AIX have been added to this table in the Volumes column.

88
IBM TotalStorage Migration Guide for the SAP User
Figure 3-9 ESS Specialist Welcome panel
In order to run data copy functions, you need to work with the following components:
Volumes - the basic components in a data-copy task
Logical subsystems
Paths - PPRC only
When you select the Copy Services option from the ESS Welcome panel, the Enterprise
Storage Server: Start Copy Services window opens as shown in Figure 3-10.

Chapter 3. Environment 2 SAP migration experiences
89
Figure 3-10 ESS Start Copy Services panel
This window shows the ESS Storage Server configuration IP configuration and the Available
Actions. Depending on how the ESS is configured, you can start Copy Services from either
ServerA or ServerB. After you make a selection by clicking the appropriate server in the
Action field, the ESS Copy Services main panel is displayed as shown in Figure 3-11.

90
IBM TotalStorage Migration Guide for the SAP User
Figure 3-11 ESS Copy Services main panel
You can either select the volumes individually from a Logical Subsystem (LSS) by clicking the
Volumes button, or select
all
of the volumes in an LSS by clicking the Logical Subsystems
button.
Since not all of the volumes in the source contained data, we chose to select only the
individual volumes that did contain the migrated SAP data. Figure 3-12 shows the ESS Copy
Services Volume selection panel.
.
Figure 3-12 ESS Copy Services - Volume selection panel
From the Copy Services Volumes panel you select the source LSS and volumes from the left
side of the panel and the target LSS and volumes from the right side of the panel. When you
select a source or target LSS the volumes in that LSS are displayed.

Chapter 3. Environment 2 SAP migration experiences
91
Figure 3-13 shows the source and target LSSs used to FlashCopy the SAP data migrated to
the ESS. LSS11 is the source and LSS12 is the target.
Figure 3-13
SAP Source and Target volumes
In the Volumes panel’s normal mode you can select only one source and target volume. If
you want to select multiple source and target volumes at once, click the Enter Multiple
Selection Mode button at the bottom of the panel.
Once the source and target have been selected the source volumes with be highlighted with a
blue box and the target will be highlighted with a red box. See Figure 3-14.
Figure 3-14 ESS Copy Services Volumes panel

92
IBM TotalStorage Migration Guide for the SAP User
After all of the source and target volumes have been selected, you must set up a Flashcopy
task. A task is created using the Task Wizard, which in invoked by right-clicking the last target
volume in the target list.
Figure 3-15 ESS copy Services Task Wizard
Select “Establish FlashCopy Pair,” then click Next. The next panel, shown in Figure 3-16,
displays all of the FlashCopy copy options.
Of the many choices listed in the copy options window, two are of interest to us at this time:
No background copy
Selecting the “No background copy” option causes FlashCopy to create a logical copy of
the data instead of a physical copy. Since we want to create a complete physical copy of
the data we left this option blank.
Persistent FlashCopy
Select this option if you want the FlashCopy relationship to remain even after the
FlashCopy operation completes. The FlashCopy relationship remains indefinitely and
must be broken with a Withdraw FlashCopy task. This prevents another FlashCopy task
from writing to the target before a withdraw is done.
After the Copy options selections have been made, click Next to continue.

Chapter 3. Environment 2 SAP migration experiences
93
Figure 3-16 Copy Services wizard copy options
The next window displayed by the FlashCopy wizard is the FC Sequence Number. This
option is used to help keep track of FlashCopy tasks should you have more than one, and
also establishes in what sequence the tasks should run. For simplicity, we used 1 as shown in
Figure 3-17.
After entering a sequence number click Next to continue.
Figure 3-17 FC sequence number
The next window displayed is the Define Task window shown in Figure 3-18. Enter a task
name (using
no
spaces or special characters) and an optional task description. You can either
click Save to save this task so it can be run at a later time, and more than once; or click Run
to run this task immediately. We elected to save the task so we could run it later and have the
option to run it more than once.
Note: Only a saved task can be modified, renamed, copied, or run at a later time.

94
IBM TotalStorage Migration Guide for the SAP User
Figure 3-18 FC Define Task window
Once the task has been saved the FlashCopy Volumes panel resets itself. The selections that
were made on this panel prior to invoking the FlashCopy Task Wizard are now gone. This is in
part because we “saved” the task instead of running it, plus it allows us to create more tasks if
desired. See Figure 3-19.
Figure 3-19 ESS Copy Services volumes panel
We now have a FlashCopy task saved that we can run at our convenience. To run this task,
we select the task, then click the Tasks button on the left side of any the Copy Services
panels.
The Tasks panel (Figure 3-20) shows the task we created and at the bottom of the panel there
are options to Run, Modify, Group/Ungroup, and Remove one or more tasks. From this panel
we select Run to establish the FlashCopy pairs.

Chapter 3. Environment 2 SAP migration experiences
95
Figure 3-20 FlashCopy Tasks
We can run the task by clicking the Run button. Running this task takes only a few seconds
and barring any unforeseen problems, it completes successfully. Now the Task State column
in Figure 3-21 shows that the task run was “Successful.”
Figure 3-21 Copy Services Tasks panel - Successful
Figure 3-22 shows what you will see if you now return to the Copy Services Volumes panel.
Note that there are now icons next to the source and target volumes you selected earlier. The
blue lightening bolt indicates a FlashCopy source volume, the red lightning bolt indicates a
target FlashCopy volume. The grey tab or square indicates a volume copy.
Note: The Task State column in the Tasks panel shows the task as “Not running” since it
has not yet been run.

96
IBM TotalStorage Migration Guide for the SAP User
Figure 3-22 Copy Services Volumes panel FlashCopy established
For a list of icons and their meanings click the Legend button at the top of the panel. The
icons legend is shown in Figure 3-23.
Figure 3-23 ESS Copy Services Icons Legend

Chapter 3. Environment 2 SAP migration experiences
97
3.3 Using IBM TotalStorage SAN Volume Controller to migrate
We recommend that you review the material presented in 1.4, “IBM TotalStorage ESS 800
overview” on page 3 and 1.5, “IBM TotalStorage SAN Volume Controller overview” on page 8
before you continue this chapter. In particular, be sure you understand the definition of terms
that apply to the SVC.
3.4 General SVC planning guidelines
The design and planning of the SVC should be done with a top down method; that is, from the
required vdisks to the physical storage arrays. When migrating from one storage device to
another you want the configurations to match. The actual implementation tasks are
completed in the reverse order.
Use the following process to ensure a successful migration installation:
Overall design and sizing
Physical planning
Preparation work on the client hosts - HBAs, device drivers, and SDD
SAN design and definitions: fabric, zones
SVC node and cluster configuration
SVC logical configuration: mdisks, mdisk groups, and vdisks
Back-end storage subsystem configuration
3.4.1 Sizing the SVC
To ensure the SVC is sized appropriately and will perform as required, an analysis of the
current environment should be performed. Things that should be considered include:
Documentation of the number of hosts, the traffic profile activity (read or write, sequential
or random), the performance requirements (I/Os per second), the total storage capacity,
the storage capacity per host, and the host volume sizes.
Define the number of clusters and the number of node pairs, either 1 or 2. Each pair of
nodes (an I/O Group) is the container for the virtual disks, so there may be a need to
ensure certain vdisks are spread across I/O Groups or nodes. For example, you may want
to spread the host system volumes across I/O Groups or nodes.
Determine the disk subsystem configuration including the number of arrays, LUNs, or
mdisks, and the required sizes.
Make sure there is sufficient free space available in the storage subsystem to migrate the
data; for example, enough free space to migrate image mode vdisk data if required.
The performance criteria and a plan for which arrays are to be included into which mdisk
groups.
The initial guidelines suggesting the relationship between the number of SVC nodes and the
number of attached storage systems are shown in Table 3-2. If the storage systems are not
driven to maximum load, you may be able to attach more.
However, it is strongly recommended that you obtain the latest performance guidelines
before completing your design. More information can be found at:
http://www.storage.ibm.com/software/virtualization/index.html

98
IBM TotalStorage Migration Guide for the SAP User
Table 3-2 Storage subsystems per SVC node
The table also shows that two pairs of nodes will approximately double the throughput of a
single pair. However, the SVC, like any disk system, needs to have sufficient disks attached
to deliver its maximum throughput.
3.4.2 Physical planning
There are many factors that have to be taken into account when carrying out the physical
planning of an SVC installation. These are detailed in the SAN Volume Controller: Planning
Guide, GA22-1052 and include the following:
Physical site
Power, cooling, and location requirements for the SVC and the UPSs. An SVC node is
one EIA unit high, a UPS is two EIA units.
The Master Console is two EIA units: one for the server, the other for the keyboard and
monitor.
Will other hardware devices be placed in the rack, for example, DS4000(FAStT), switches,
Ethernet switch, and so on?
The maximum power rating of the rack and input power supply must not be exceeded.
3.4.3 Network and IP addresses
IP addresses will need to be allocated for the SVC cluster, the SVC service point, the Master
Console, and switches. For more information refer to the SVC planning guide.
The back-end storage subsystems will also require IP addresses if they are new to the site.
3.4.4 Naming convention
It is advisable to decide on the naming conventions you will be using to label your back-end
storage, mdisks, mdisk groups, and vdisks. This will make it easier to trace the path from a
SAN File System volume down through the SVC, and finally to the actual LUN and physical
array.
3.5 Client host preparation
This section identifies some of the preparations that need to be taken at the client.
3.5.1 Device drivers
AIX
The most current AIX maintenance levels and device drivers (devices.fcp.disk.ibm2145.rte)
can be found in the Supported Software and Download sections at the Web site:
Number of nodes Should support up to:
DS4300(FAStT600) DS4500(FAStT900) ESS F20 ESS 800
2 2-3 1 2 1
4 7 3 4 2

Chapter 3. Environment 2 SAP migration experiences
99
http://www.ibm.com/storage/support/2145
There are several entries in these sections, so you may have to navigate through several
pages before you find the correct download files.
The method for installing the package is described in the Host Attachment Guide,
SC26-7563.
Check that the SVC drivers have been installed at the correct level by entering the command:
lslpp -L *2145*
Windows 2000
The most current Windows 2000 hardware and software level support for the SVC can also
be found at:
http://www.ibm.com/storage/support/2145
The process to install the correct drivers is detailed in the Host Attachment Guide,
SC26-7563.
Check that the correct driver is installed by selecting My Computer

Manage

Device
Manager. Click the plus (+) sign to expand the menu for SCSI and RAID Controllers. Click
the adapter you want to check and select Properties.
3.5.2 Hardware support
The client host HBAs and HBA drivers supported on the SVC are listed in the Planning Guide,
GA22-1052. Details are also in the supported hardware list, which can be found at:
http://www.ibm.com/storage/support/2145
3.5.3 SDD installation
The multi-path support product SDD must be installed onto the client hosts. The currently
supported version of SDD for the SVC for AIX (rte.tar) and Windows (zip file) is
documented and can be downloaded from the Web site:
http://www.ibm.com/storage/support/2145
Use the AIX command lslpp -L *sdd* to verify the correct version of SDD has been
installed, and lssrc -s sddsrv to ensure it is active.
On the Windows 2000 platform from Start

Programs, select Subsystem Device
Driver

SDD Manager option. Enter the SDD commands, for example, datapath query
device to ensure SDD is operational.
3.6 General SAN planning
The SAN must be designed for high availability. The total number of ports and the bandwidth
needed between the SAN File System client host to the SVC, and the SVC to the back-end
storage, must be considered.
The following general guidelines apply:
On the SAN, the SVC nodes are always connected to SAN switches and nothing else.

100
IBM TotalStorage Migration Guide for the SAP User
The FC ports on each SVC node must be spread evenly across each of the counterpart
SAN fabrics, a redundant dual fabric SAN. The SVC nodes have two FC HBAs, each with
two ports. Therefore, there is a total of four FC ports per node.
The number of paths through the network from the SVC nodes to a client host must not
exceed eight.
The switches supported by the SVC and the required firmware levels are documented in the
supported hardware list at the Web site:
http://www.ibm.com/storage/support/2145
More information on SVC SAN planning also can be found by clicking Configuration
requirements and guidelines at the same Web site.
3.6.1 SAN zones
Create the SAN fabric with the following zones:
SVC Master Console zone
SVC storage zone
SVC host zone
3.6.2 SVC Master Console zone
This zone will include all the SVC node FC ports and the Master Console FC ports. This
isolates the console from the other hosts and storage devices within the SAN. This is shown
in Figure 3-24.

Chapter 3. Environment 2 SAP migration experiences
101
Figure 3-24 SVC Master Console zone

102
IBM TotalStorage Migration Guide for the SAP User
SVC storage zone
The FC switch will need to be zoned to permit all the SVC nodes to see all the storage
controllers that may contain mdisks that it will manage. Any existing storage controllers that
contain data to be migrated into the SVC will also need to be placed into this zone. This is
shown in Figure 3-25.
Figure 3-25 SVC storage zone
SVC host zone
The client hosts will be in a zone that includes the SVC node FC ports. The client hosts’ HBAs
and the back-end storage must not be in the same zone. There may be several host zones
required to separate different platforms, for example, Windows 2000 hosts from AIX hosts; or
even to separate hosts within the same platform, for example, AIX production from
development. This is shown in Figure 3-26.

Chapter 3. Environment 2 SAP migration experiences
103
Figure 3-26 SVC host zone
3.7 SVC planning
This section describes some important SVC planning considerations.
Vdisks
Create a list of the vdisks you plan to define with their required sizes, performance, or
workload characteristics, and name them.
Plan the distribution of critical or high workload vdisks between the different SVC I/O Groups
and nodes in a way that optimizes the I/O load between the hosts and the SVC. Moving
vdisks between I/O Groups is possible, but is disruptive, because following the move, host
systems will need to rediscover the LUs at different WWPNs.
For more explanation on MDS volume limitations, refer to IBM TotalStorage: Introducing the
SAN File System, SG24-7057.
Mdisk groups
The managed disk groups can be designed to include an entire storage controller (ESS or
DS4000(FAStT)) or only part of it; for example, groups of arrays based on the array drive
capacity or performance characteristics.
You will need to map your vdisks to the mdisk groups you have planned to ensure you can
match the required workload characteristics.

104
IBM TotalStorage Migration Guide for the SAP User
The mdisk group extent size needs to be considered because it will determine the maximum
storage that can be supported within the group. Also, vdisks can only be migrated between
mdisk groups with the same extent size.
Mdisks
The mdisks should be named in such a way that the name will assist you to identify the array
within the storage controller where it resides. This will be helpful when tracking problems or
monitoring performance issues.
The LUNs that are seen as mdisks to the SVC cannot be shared with any other hosts directly,
meaning access to the LUNs must be through the SVC only.
Hosts
You will need to define a set of host objects to represent each of the client hosts. Each object
can have multiple WWPNs associated with it, so only one object per server need be defined.
You should decide on a naming standard for your hosts.
3.8 Back-end disk support
The back-end disk subsystems supported by the SVC are identified in the supported
hardware list found at:
http://www.ibm.com/storage/support/2145
3.8.1 ESS
To reduce the likelihood of an mdisk or mdisk group being unavailable, configure the ESS
arrays as RAID-5 or RAID-10.
Configure a full array sized LUN and let the SVC carve up the storage pool through smaller
sized vdisks.
The LUNs have to be masked to the SVC FC ports, so a host must be defined for each SVC
port on each SVC node. The hosts are then assigned to the LUN through the Modify Volume
Assignment panel.
3.8.2 DS4000(FAStT)
To reduce the likelihood of an mdisk or mdisk group being unavailable, configure the
DS4000(FAStT) arrays to provide redundancy. For example, only use RAID-1, RAID-3, or
RAID-5.
As with the ESS, configure a full array sized LUN and let the SVC carve up the storage pool
through smaller sized vdisks. When in doubt about array size, arrays with six to eight member
disks are recommended.
The LUNs have to be masked to the SVC FC ports so a host mapping must be created for
each SVC port on each SVC node. The hosts are then mapped to the required LUNs.

Chapter 3. Environment 2 SAP migration experiences
105
3.9 Sample configuration planning
This section presents a sample configuration plan. Table 3-3 shows a list of all available LUNs
and storage controllers, and describes their usage, names, and mdisk group allocations.
Table 3-3 LUN to mdisk plan
LUN
Serial/Name
Ctl Type
Name
Array
Size
HDD Host Mdisk
name **
Mdisk group Comment
HI_DRIVE_1 DS4300(FAStT
600))
F1
6+P
200GB
36GB SVC1 HIDR1 HIGHGRP1 High performance drives.
Data with high I/O rates
and small response time demands.
HI_DRIVE_2 DS4300(FAStT
600)
F1
6+P
200GB
36GB SVC1 HIDR2 HIGHGRP1 As above.
GEN_DRIVE_1 DS4300(FAStT
600)
F1
6+P
400GB
73GB SVC1 GENDR1 GENERALGRP1 Good performance and capacity.
Fits most data workloads.
GEN_DRIVE_2 DS4300(FAStT
600)
F1
6+P
400GB
73GB SVC1 GENDR2 GENERALGRP1 As above.
LC_DRIVE_1 DS4300(FAStT
600)
F1
6+P
800GB
146GB SVC1 LCDR1 LOWGRP1 Low cost storage.
Data with no performance
needs or default SFS pool.
LC_DRIVE_2 DS4300(FAStT
600)
F1
5+P
680GB
146GB SVC1 LCDR2 LOWGRP1 As above.
600-14830 ESSF20
ESS1
6+P
4GB
73GB W2K2 600 IMAGE_GENLOW Contains existing W2K data.
601-14830 ESSF20
ESS1
6+P
4GB
73GB AIX2 601 IMAGE_GENLOW Contains existing AIX data
602-14830 ESSF20
ESS1
6+P
4GB
73B AIX1 602 IMAGE_HIGH Contains existing AIX data
Note: The mdisk names are prefixed with MDISK_< Controller name> _, for example, MDISK_F1_HIDR1 or MDISK_ESS1_600.

106
IBM TotalStorage Migration Guide for the SAP User
3.9.1 Sample vdisk mapping
Table 3-4 lists all available vdisks and describes their usage, names, and mdisk group
allocations.
Table 3-4 Vdisk plan
Vdisk name ** Size Type Mdisk group Extent size Owner Comments
AIX-VSYSTEM1 16GB Stripe HIGHGRP1 16MB host1
host2
AIX System volume
AIX1_VIDISK1 4GB Image IMAGE_HIGH 16MB SVC1 AIX1 Existing data. Pre SAN File System
format and accessed through SVC vdisk.
AIX1_VIDISK2 4GB Image IMAGE_GEN_LOW 32MB SVC1 AIX1 Existing data. Pre SAN File System
format and accessed through SVC vdisk.
AIX2_VIDISK1 4GB Image IMAGE_GEN_LOW 32MB SVC1 AIX2 Existing data. Pre SAN File System
format and accessed through SVC vdisk.
W2K1_VIDISK1 4GB Image IMAGE_HIGH 16MB SVC1 W2K1 Existing data. Pre SAN File System
format and accessed through SVC vdisk.
W2K2_VIDISK1 4GB Image IMAGE_GEN_LOW 32MB SVC1 W2K2 Existing data. Pre SAN File System
format and accessed through SVC vdisk.
** The vdisk names are prefixed with < Primary Owner> _V, or for image mode disks < Primary Owner> _VI.
For example, SFS1_VSYSTEM1 or AIX1_VIDISK1.

Chapter 3. Environment 2 SAP migration experiences
107
3.10 Configuring the ESS 800 and SVC for migration
When properly connected to the SAN network, the SVC presents multiple initiator ports
(wwpns) to the ESS800 for redundancy (see SVC Configuration Guide SC26 -7543 for more
information). The ESS800 must be configured to use
all
of the available initiator ports,
meaning there should be a Host system entry for each initiator port. If you
do not
use all of the
initiator ports that are presented to the ESS800 from the SVC, the logical drives on the
ESS800 will be seen as mdisks by the SVC, but you will not be able to put these mdisks into
a Managed Disk Group because the mdisks will go offline.
This section describes how to configure the ESS800 and SVC. We begin by modifying the
host systems as shown in Figure 3-27. (For the purpose of this discussion, we assume that
you are familiar with the ESS Specialist interface, so we have not included details for every
button you must click to access the various configuration screens.)
3.10.1 Modify hosts systems
Figure 3-27 Modify host systems
The Host Type should be IBM SAN Volume Controller or RS/6000; the Host Attachment
should be Fibre Channel Attached.
3.10.2 Modify Volume Assignments
In order for the SVC to properly use the ESS 800 volumes, all SVC Host Systems must be
assigned to all volumes that are to become mdisks on the SVC.
The next four figures and accompanying discussion provide a detailed example of how to
configure the ESS800 volumes for SVC.

108
IBM TotalStorage Migration Guide for the SAP User
Select Modify Volume Assignments.
Figure 3-28 Modify Volume Assignments
Select the volumes that are to become SVC mdisks.
Figure 3-29 Modify Volume Assignments #2

Chapter 3. Environment 2 SAP migration experiences
109
Make sure the “Assign selected volumes to target host” radio button is selected and select
all

SVC hosts in the Target Hosts window. Click Perform Configuration Update to update the
ESS800 with the new configuration.
Figure 3-30 Modify Volume Assignments #3
Volume Assignments should look something like Figure 3-31 when complete.
Figure 3-31 Modify Volume Assignments #4
Click Cancel Configuration Update to exit.

110
IBM TotalStorage Migration Guide for the SAP User
3.10.3 Configuring the SAN Volume Controller
This section shows the process of discovering the available mdisks, and creating a Disk
Group and a vdisk to be available to AIX. We begin with the SAN Volume Controller Welcome
panel in Figure 3-32.
Figure 3-32 SVC Welcome panel
Some setup is required prior to configuring storage and host systems on the SVC: you must
create a cluster or clusters and to add nodes to the clusters. Find details for this process in
the redbook IBM TotalStorage SAN Volume Controller and SAN Integration Server,
SG24-6423.
3.10.4 Start the SVC application
To start the SVC application, select Manage Cluster from the list on the left side of the SVC
Welcome panel. When the Viewing Clusters window is displayed, select “Launch the San
Volume Controller application” from the drop-down list and click Go.

Chapter 3. Environment 2 SAP migration experiences
111
Figure 3-33 SVC Viewing Clusters panel
3.10.5 Creating a host
Select Work with Virtual Disks to open the menu list, then select Hosts. Click the Bypass
Filter button.
Figure 3-34 Filtering Hosts panel

112
IBM TotalStorage Migration Guide for the SAP User
Select “Create a Host” from the drop-down list and click Go.
Figure 3-35 Viewing Hosts Panel
Enter the Host Name and select both available ports by clicking the Add button for each port.
Figure 3-36 Creating Hosts Panel

Chapter 3. Environment 2 SAP migration experiences
113
Click OK to create the Host.
Figure 3-37 Creating Hosts panel
If the host has been created successfully, you will see the screen shown in Figure 3-38.
Figure 3-38 Successful creation of host

114
IBM TotalStorage Migration Guide for the SAP User
3.10.6 Creating managed disks
Select Work with Managed Disks

Managed Disks. Click the Bypass Filter button.
Figure 3-39 Filtering Managed Disks panel
If mdisks are not already displayed, select “Discover MDisks” from the drop-down list and
click Go.
Figure 3-40 Viewing Managed Disks panel 1

Chapter 3. Environment 2 SAP migration experiences
115
Mdisks are displayed and show the Mdisk Name, Status, Mode, Capacity in megabytes,
Controller Name, and LUN Number.
Figure 3-41 Viewing Managed Disks panel 2
3.10.7 Creating managed disk groups
Select Work with Managed Disks

Managed Disk Groups. Managed Disk Groups has
the same Filters page. Click the Bypass Filter button.
Figure 3-42 Filtering Managed Disk Groups panel

116
IBM TotalStorage Migration Guide for the SAP User
Select “Create an MDisk Group” from the pull-down list and click Go.
Figure 3-43 Viewing Managed Disk Groups panel
Give the Disk Group a name and select the Mdisks from the MDisk Candidates list that are to
be part of the Disk Group.
Figure 3-44 Creating Managed Disk Groups panel 1

Chapter 3. Environment 2 SAP migration experiences
117
Mdisks move from the Mdisk Candidates list to the Selected Mdisks list.
Figure 3-45 Creating Managed Disk Groups panel 2
Select the Extent Size and click OK. (These Extents will be used later to create the VDisk).
Figure 3-46 Creating Managed Disk Groups panel 3

118
IBM TotalStorage Migration Guide for the SAP User
When the Disk Group is successfully created, the panel shown in Figure 3-47 is displayed.
Figure 3-47 Viewing Managed Disk Group panel
The Managed Disk menu now shows the MDisks as “Managed” and indicates to what MDisk
Group they now belong.
Figure 3-48 Viewing Managed Disks panel
Important: If you add an MDisk to an MDisk group as a managed disk, and there is data
on the MDisk, the data will be lost. Ensure that you create image mode VDisks from the
MDisks that contain data before you start adding any MDisks to groups

Chapter 3. Environment 2 SAP migration experiences
119
3.11 SVC migration
The SVC migration facility allows the mapping of vDisk extents to managed disk extents to be
changed without interrupting a host’s access to that virtual disk. This can be performed for
any vDisk managed by the IBM TotalStorage SVC. This facility can be used to migrate a
virtual disk from one managed disk group to another managed disk group such as the
ESS800.
3.11.1 Migrating a vDisk
From time to time, you may want to migrate vDisks from one set of mDisks to another. Some
reasons that you might want to do this are to retire an old disk subsystem, to have better
balanced performance across your virtualized environment, to migrate from one storage
platform to another, or simply to migrate data into the SVC environment transparently using
image mode.
To perform the migration, use the svctask migratevdisk command. The full syntax of the
command is:
migratevdisk -mDiskgrp name|id [-threads threads] -vDisk name|id
Where:
mDiskgrp is the mDisk group name or ID
threads is the number of threads
vDisk is the vDisk name or ID
To supply these parameters, before you can migrate your vDisk you must know the name of
your VDisk and the name of the MDG to which you want to migrate. To find the names, simply
run the svcinfo lsvdisk and svcinfo lsmdiskgrp commands.
The user can inspect the progress of each migration as it is performed. The CLI command is:
svcinfo lsmigrate
Refer to the redbook IBM TotalStorage SAN Volume Controller and SAN Integration Server,
SG24-6423 for information on how to set up PUTTY for command line interface.
http://www.redbooks.ibm.com/redbooks/pdfs/sg246423.pdf
3.11.2 Migrating a vDisk between managed disk groups
An entire virtual disk can be migrated to a new managed disk group (MDG) using the migrate
virtual disk command. If the vDisk is operating in image mode, it is first converted into a
managed mode vDisk. For the duration of the move, the vDisk is listed as being a member of
the original MDG. For the purposes of configuration, the vDisk moves to the new MDG
instantaneously at the end of the migration.
Assuming we had a vDisk for the non-IBM subsystem called XYZ_VDisk we would use the
following command to migrate the non-IBM VDISK to the ESS800:
Important: After migration is started, it continues to completion unless it is stopped or
suspended by an error condition or the vDisk being migrated is deleted.
Important: In order for migration to work, the source and destination managed disk groups
must have the same extent size.

120
IBM TotalStorage Migration Guide for the SAP User
svctask migratevdisk -mdiskgrp non-IBM_Disk_Group -threads 4 -vdisk non-IBM_VDisk
3.12 Install and configure the TSM Client
Installing the TSM Client
IBM Tivoli Storage Manager (TSM) provides automated, policy-based storage management
for file servers and workstations. Use the following steps to install the TSM Client.
1.Log on as root.
2.Install the software components with the AIX tool smit. Select the fast path smitty
install_package.
3.Select the components to install as shown in Figure 3-49.
Figure 3-49 AIX Smit menu
After the installation is finished the screen shown in Figure 3-50 is displayed.

Chapter 3. Environment 2 SAP migration experiences
121
Figure 3-50 AIX Smit Menu after installation
Configuring the client
Use the following steps to configure the TSM Client:
1.Copy the file /usr/tivoli/tsm/client/ba/bin/dsm.opt.smp to dsm.opt.
2.Edit the dsm.opt file.

122
IBM TotalStorage Migration Guide for the SAP User
Figure 3-51 TSM - dsm.opt file
3.Edit the DSM.sys file shown in Figure 3-52.
Choose the servername and server TCP/IP address.
Figure 3-52 TSM - dsm.sys
4.Export the following variables:

Chapter 3. Environment 2 SAP migration experiences
123
DSM_CONFIG /usr/tivoli/tsm/client/ba/bin/dsm.opt
DSM_DIR /usr/tivoli/tsm/client/ba/bin
For ease of use copy these files into the user’s .profile as shown in Figure 3-53.
Figure 3-53 Environment Variables
5.Edit the /etc/services:
tc support 1500/tcp
The configuration of the client is now complete.
3.13 Installation of the TSM Server
In this section we describe the installation of the TSM Server in preparation for the migration.
3.13.1 Basic configuration
The basic configuration has the following characteristics:
The server automatically starts in quiet mode after a system reboot.
The default communication method is TCP/IP.
The server options file (dsmserv.opt) contains server configuration options.
Database
The database has one 17MB volume named db.dsm. The size of log, database, and storage
pool volumes is for a new install only.
Recovery log
The recovery log has one 9MB volume named log.dsm.

124
IBM TotalStorage Migration Guide for the SAP User
Disk storage pools
The following disk storage pools are created:
BACKUPPOOL, with one 8MB volume defined, named backup.dsm
ARCHIVEPOOL, with 8MB volume defined, named archive.dsm
SPACEMGPOOL, which has one 8MB volume defined, named spcmgmt.dsm
Server administrator defined
Name: client
Password: client
AIX backup-archive client defined
Name: client
Password: client
Standard policy
When a backup file is no longer associated with a backup copy group, it remains in server
storage for 30 days.
When an archived file is no longer associated with an archived copy group, it remains in
server storage for 363 days.
3.13.2 Installation
Follow these steps to install the TSM Server:
1.Using smit install, choose LICENSE agreement

Preview new

Yes.
Language: en_US
2.Log on as root.
3.Select smit install_update.
4.Select:
Install and Update Software
Install and Update from the LATEST Available Software
5.Select the CD-ROM drive.
6.If you are running with 32-bit hardware, install the following Tivoli Storage Manager
packages:
– tivoli.tsm.server.com
– tivoli.tsm.server.rte
– tivoli.tsm.msg.en_US.server
– tivoli.tsm.license.cert
– tivoli.tsm.license.rte
If you are running with 64-bit hardware, install the following Tivoli Storage Manager
packages:
– tivoli.tsm.server.com
– tivoli.tsm.server.aix5.rte64
– tivoli.tsm.msg.en_US.server
– tivoli.tsm.license.cert
– tivoli.tsm.license.aix5.rte64
When the installation is complete, the screen shown in Figure 3-54 is returned.

Chapter 3. Environment 2 SAP migration experiences
125
Figure 3-54 TSM - Server installation
3.13.3 Configuration of the server
To configure the server, use the administrative Web interface. Java 1.1.6 is a prerequisite for
this.
Configure the dsmserv.opt file:
commethod: http
httpport: 1580
Use the following environment variables:
For ksh
– export DSMSERV_DIR=/usr/tivoli/tsm/server/bin
– export DSMSERV_CONFIG=/usr/tivoli/tsm/server/bin/dsmserv.opt
For csh
– setenv DSMSERV_DIR /usr/tivoli/tsm/server/bin
– setenv DSMSERV_CONFIG=/usr/tivoli/tsm/server/bin/dsmserv.opt
To save this environment, save these entries in the .cshrc or .kshrc file of your $HOME
directory.
Edit the /etc/services file as follows:
tcpport 1500/tcp
Edit the /usr/tivoli/tsm/ file by deleting the comment sign from the following lines:
TCPPort 1500
TCPADMINPort 1500
HTTPPort 1580

126
IBM TotalStorage Migration Guide for the SAP User
3.13.4 Starting and Stopping the Server
To start the server from the /usr/tivoli/tsm/server/bin directory, enter:
./dsmserv
To stop the server, do one of the following:
Enter halt
or
Select Server

Server Status

Operations

Halt Server.
Click Finish.
3.13.5 Starting and stopping an administrative client or the Web interface
ID Name: admin
Password: admin
To start the administrative client command line interface, enter the following command from
the /usr/tivoli/tsm/client/ba/bin/dsmadmc directory:
dsmadmc
To start the administrative Web interface enter the following string in your browser:
http://tsmserv:1580
Changing the administrative client password
To change the password from ADMIN to SUMMER:
update admin admin summer
Select Administrators.
Select the desired administrator’s name.
From Operations, select Update an administrator.
Enter the administrator information and click Finish.
Registering licenses
From the administration command line:
Enter:
register license file=spacemgr.lic numer=10
register license file=mgsyslan.lic number=10
register license file=drm.lic
or
Select Server

License Information.
From Operations, select Register License.
Enter the required information and click Finish.
Start the client
Start the backup archive client graphical user interface by entering the command dsm.

Chapter 3. Environment 2 SAP migration experiences
127
3.13.6 Increasing disk storage pool sizes
To create a 100MB volume named sapbkup01.dsm in BACKUPPOOL, do the following:
Definevolume backup sapbkup01.dsm formatsize=100
or
Expand Server Storage

Storage Pools

Disk Storage Pools.
Select Volumes.
From Operations, select Define a disk storage pool volume.
Enter the required information and click Finish.
Client register
To register the client use the following command:
register node <node-name> <password>
We used the password: itso13sj
The command syntax is:
register node columbia itso13sj
3.13.7 Back up data for migration
We recommend using a graphical screen to manage the backup and restore tasks.
1.Start the client with the command dsm. The login screen is shown in Figure 3-55.
2.Insert a User ID and Password.
Figure 3-55 TSM - Client login screen
The screen shown in Figure 3-56 is displayed.

128
IBM TotalStorage Migration Guide for the SAP User
Figure 3-56 Tivoli Storage Manager
3.Choose Backup files and directories and the tree shown in Figure 3-57 appears.
Figure 3-57 TSM - Client backup tree
4.Expand the Local entry in this tree. Figure 3-58 shows the expanded tree.

Chapter 3. Environment 2 SAP migration experiences
129
Figure 3-58 TSM - Expanded backup tree
5.Select the files belonging to your SAP R/3 environment and click Backup to start the
backup as shown in Figure 3-59.
Figure 3-59 TSM - Selected Backup Files
The progress of the file transfer is displayed in the screen shown in Figure 3-60.

130
IBM TotalStorage Migration Guide for the SAP User
Figure 3-60 TSM - File transfer during backup
6.When the backup is finished the screen shown in Figure 3-61 is displayed.
Figure 3-61 TSM - Backup is finished
3.13.8 Restore to migrate data to the IBM ESS Storage Subsystem
1.Open the Backup and Restore Client.
2.Click Restore as shown in Figure 3-62.

Chapter 3. Environment 2 SAP migration experiences
131
Figure 3-62 TSM - Restore tree
3.14 Environment 2 migration experiences: FDR/UPSTREAM
This section describes the architecture, features, installation, and handling for the
INNOVATION Data Processing Corporation product, FDR/UPSTREAM Reservoir. The
product is a useful tool to move data such as SAP R/3 applications from a vendor storage
subsystem to an IBM ESS in an AIX environment.
The product FDR/UPSTREAM Reservoir is a high-speed storage management and disaster
recovery tool for open systems data stored on LAN servers, workstations, and storage
subsystems. Almost all of the popular server and workstation operating systems in use today
are supported by FDR/UPSTREAM Reservoir.
FDR/UPSTREAM Reservoir offers several options for high-speed and flexible backups.
These backups, which can be taken at either a volume level or an incremental/file level, are
directed to a storage server system for safe, reliable storage and retention. When required,
individual files and directories can be stored.
FDR/UPSTREAM Reservoir has been designed to back up and restore data as quickly and
as efficiently possible. It offers the choice of transmission of data using the TCP/IP protocol in
a high performance network connection such as GB-Ethernet.
FDR/UPSTREAM Reservoir has an integrated tool for centralized administration and control
monitoring. The tool enables the automation of operations and keeps administrators informed
about activities. An event notification can send e-mail to inform administrators about traps
and errors.
In the lab, we ran a scenario that used the FDR/UPSTREAM Reservoir application to move
data from a vendor storage subsystem to an IBM ESS storage subsystem. The application
server was an IBM pSeries system installed with AIX. In conjunction with the IBM DB2 UDB

132
IBM TotalStorage Migration Guide for the SAP User
Database Version 7, we ran the application SAP R/3 4.6C. The disks where the data
belonged were on a vendor storage subsystem.
3.14.1 Supported client environments
The following clients are supported:
Windows NT/2000
Windows XP
Windows Server 2003
Novell NetWare
Sun Solaris
Sun X86
HP-UX
Tru64 UNIX
Intel® Linux
zLinux
OS/390 UNIX System Services (USS) VMware
OS/2®
3.14.2 Supported databases
The following databases are supported:
IBM DB2 UDB
Oracle
SQL Server
Domino® and Lotus® Notes®
MS Exchange
SQL Back-Track Interface for Sybase, Informix®, ORACLE and IBM DB2 UDB
Novell GroupWise
Backint Interface for ORACLE with SAP and SAPDB with SAP
3.14.3 Client-server application
The product FDR/UPSTREAM Reservoir is a client-server application.
In our environment the server component resides on an Intel platform based system using
Microsoft Windows 2000.
The client component is installed on an IBM pSeries system with AIX.
FDR/UPSTREAM Reservoir Repository (Server)
This component resides on the Windows side. The Windows system can be Microsoft
Windows 2000/XP or Server 2003 with a minimum of 512 MB memory. The component is
responsible for the storage and retention of backup. It is also where the main configuration
control files reside, together with the backup recording mechanism and history files. It can be
controlled form a Java interface. The Java interface provides an easy to use GUI that fully
utilizes the FDR/UPSTREAM Reservoir storage management functions especially for
restores, and can be used from a PC or an X terminal.
FDR/UPSTREAM Reservoir (Client)
This component resides on the IBM pSeries system installed with AIX and is responsible for
accessing AIX files and the IBM DB2 Universal Database™ during the backup and restore. A
full set of GUI panels is provided to allow for the creation of individual backup and restore

Chapter 3. Environment 2 SAP migration experiences
133
tasks. During backup and restore, data is transferred between the two components across
one or more connections.
3.14.4 FDR/UPSTREAM Reservoir server components
FDR/UPSTREAM service
The FDR/UPSTREAM service (USSERVER.EXE) is a safe and secure process that starts up
when your system comes up and listens for requests from remote systems. When requests
are received they are processed by an instance of the UPSTREAM task program
(USTASK.EXE).
Configuration file
The UPSTREAM service program is controlled by a configuration file (USSERVER.CFG)
which defines overall system information like the work path, listening ports, and so forth.
Backup Profile
The Backup Profile is used to describe and govern the capabilities enabled and storage rules
to be used by FDR/UPSTREAM when processing product function requests. For each
Backup Profile, there is a configuration file (<profile>.PRF) which contains information like
backup retention, tape pools, security, and so forth. The Backup Profile is the entity used to
define backup rules. Backup Profiles can be combined into Profile Sets to allow multiple
Backup Profiles to be written on a single tape. Backup Profiles can be combined into Groups
to allow a number of profiles to be migrated or copied to tape in a single request.
Logging file
The FDR/UPSTREAM logging file contains all the messages that have been issued by the
FDR/UPSTREAM components while it has been operating. This file is named usserver.log,
by default it is located in the directory that FDR/UPSTREAM is started from. Logging is
performed on both the client and reservoir sides.
Message file
This file contains all the FDR/UPSTREAM system messages that it can issue. The messages
have a specified format and can be modified by the customer, if so desired, to include
additional information. This information will then be displayed when the system message is
issued and logged to the logging file.
Database files
FDR/UPSTREAM maintains databases so that it can provide quick inquiry and restore
facilities. These database files are stored in a subdirectory Backups, under the defined work
path, by backup profile name. Each backup consists of three index database files.
Disk backups
Disk backups are stored in the same directory as the databases unless the
DISKBACKUPATH is specified in the profile that is used to create this backup.
Tape set files
Groups of tapes are managed in a tape set file, which is in a subdirectory Tapes under the
defined work path, by backup pool. Each tape set is named using the version date and type of
the first backup in the set.

134
IBM TotalStorage Migration Guide for the SAP User
Registered Name Facility
The FDR/UPSTREAM Registered Name Facility allows for the assignment of a logical name
to the FDR/UPSTREAM client process. This specification takes the place of the actual IP
address and TCP port number in function requests. This makes the request independent of
the actual network address and allows these values to dynamically change, for instance when
using DHCP or a similar scheme. This facility is enabled by either a FDR/UPSTREAM client
configuration file setting or via a manual entry of the required information via the UPSTREAM
Director Interface. It can also be used for automatic upgrade with silent installation.
Centralized control (USTBatch)
A command line program, USTBatch allows all UPSTREAM facilities to be requested directly
from the Reservoir machine with full logging and result checking on both sides. For
centralized requests, USTBatch supports a number of very useful features. These include
pre-authorized security which keeps you from having to store user names and passwords.
Also, USTBatch requests initiate all communications from the Reservoir machine, which
avoids many firewall issues.
Backup copy and migration (USCopy)
A command line program, USCopy allows you to copy backups to tape for off-site storage, or
migrate backups from disk or other tapes to tape to more completely utilize the tape systems
you have on site.
3.14.5 FDR/UPSTREAM client-based components
UPSTREAM Director
Written as a Java application, the UPSTREAM Director is a management interface that
provides a user friendly Windows-like graphical user interface that can run on any platform
that provides a Java Virtual Machine and TCP/IP support covering almost every substantial
operating system in use today. The FDR/UPSTREAM Director can: specify backups and
restores; check the status of all Client FDR/UPSTREAM systems;, manage your UPSTREAM
stored backups; define, modify and delete UPSTREAM stored backup profiles; and more.
This is the recommended administrative tool. It can also be used by a browser.
End User Restore Interface
The End User Restore Interface of FDR/UPSTREAM is a separate program, written as a
Java application, that provides a Windows Explorer-like interface for the selection and
monitoring of restores for administrators, help-desk operators, and most other end users.
This facility greatly reduces training and administrative configuration tasks when you wish to
allow for user-specified restores. This is recommended only for end users.
Client User Interface
The Client User Interface of FDR/UPSTREAM is a separate program that runs on a
supported remote system. This interface is intended primarily to control and administer
FDR/UPSTREAM functions on supported open systems platforms.
Centralized scheduler (the UPSTREAM Dispatcher)
A centralized scheduler (the UPSTREAM Dispatcher) is built into the UPSTREAM Client
software and is configured and controlled with the UPSTREAM Director. However, the
scheduler knows about the Reservoir server software and will use USTBatch. This enables all
of the features of USTBatch, including security pre-authorization and centralized
communications requests for fire walls.

Chapter 3. Environment 2 SAP migration experiences
135
3.14.6 Installation
This section describes the installation of the FDR/UPSTREAM Reservoir server component,
the FDR/UPSTREAM client component, and the FDR/UPSTREAM Director.
During the installation of the FDR/UPSTREAM client component we show how to configure a
client to get a connection to the assigned FDR/UPSTREAM Reservoir server.
Installing the FDR/UPSTREAM server component
The installation of FDR/UPSTREAM Reservoir consists of running the UPSTREAM
installation program. This program:
Creates a directory for FDR/UPSTREAM Reservoir
Installs the UPSTREAM Windows Media Label Library module (USLABEL.DLL)
Sets up the UPSTREAM Storage Reservoir service
Detailed steps for performing the installation follow.
1.To begin the installation, run the Setup.exe program in the root directory of the
FDR/UPSTREAM CD. The installation program performs a number of steps that are
cumbersome to perform manually, so you will want to use the installation program to
install FDR/UPSTREAM Reservoir. The FDR/UPSTREAM Installation Wizard appears as
shown in Figure 3-63.
Figure 3-63 FDR/UPSTREAM Installation Wizard Welcome Screen

136
IBM TotalStorage Migration Guide for the SAP User
2.Click Next. The window shown in Figure 3-64 is displayed.
Figure 3-64 FDR/UPSTREAM Installation Wizard Customer Information Screen
3.Insert the appropriate User Name and Company Name; click Next.
4.The screen shown in Figure 3-65 is displayed. Select the components you want to install.
These include:
– FDR/UPSTREAM Reservoir
This is the client version on a machine where you will be running the UPSTREAM
Windows Reservoir.
– ULTra
The UPSTREAM LAN Transport facility used for backing up workstations.
– FDR/UPSTREAM Director
The UPSTREAM Director is used to direct client and Reservoir versions of
UPSTREAM and should be installed on the Storage Reservoir machine.
– SAR or Reservoir
There are two check boxes:
• Main UPSTREAM Reservoir Component
This component is checked and can not be unchecked.
• UPSTREAM Reservoir History Reporting
The UPSTREAM Reservoir can generate history records in XML format.
First deselect all components you do not want to install. Then select SAR or Reservoir to
install the server components. After you choose the items you want to install, click Next.

Chapter 3. Environment 2 SAP migration experiences
137
Figure 3-65 FDR/UPSTREAM Installation Wizard - Select Features screen
5.The screen shown in Figure 3-66 is the last one that appears during the installation of the
FDR/UPSTREAM Reservoir Client. Click Install and the installation starts.
Figure 3-66 FDR/UPSTREAM Wizard Installation Program
6.After the installation is finished there will be a new entry on your PC in the path Start


Program

Upstream History Reports

Reservoir History Reports.
When you select this item the window shown in Figure 3-67 is displayed.

138
IBM TotalStorage Migration Guide for the SAP User
Figure 3-67 FDR/UPSTREAM History Reports screen
Information about all your backups is displayed in this window. Details about these reports is
beyond the scope of this book; see the pertinent product documentation for more details.
Our scenario in this chapter involves migrating data from a vendor storage subsystem to IBM
ESS, so we used only one backup to move our data to the new disks.
Installing the client component
The FDR/UPSTREAM Reservoir is distributed on the UPSTREAM CD in a separate directory
for each operating system type.
To install the client:
1.Log in as the user root.
2.Mount the UPSTREAM CD in your AIX machine,
3.Switch to the directory /<CD-ROM mount point>/upstream/aix.
4.Run the install script: ./install
5.After you execute the install script the default installation directory /usr/lpp/fdrupstream is
displayed. Press Enter to accept this installation directory.
6.Press y. The software is now installed.
FDR/UPSTREAM user interface
The FDR/UPSTREAM user interface is a character-mode interface, so it can be used from
dumb terminals or telnet.
Configure the client components
Use the following steps to configure the client component.
1.Switch to directory /usr/lpp/fdrupstream.
2.Start the configurator with the command ./uscfg.

Chapter 3. Environment 2 SAP migration experiences
139
Figure 3-68 FDR/UPSTREAM Client Configurator
3.The configuration dialog shown in Figure 3-68 is displayed.
Configure the following parameters:
– Connection Type
– Host TCP/IP Address
– Host TCP/IP Port
– Registered Name (recommended)
The values we used were:
– Connection TYPE: TCP/IP
– Host TCP/IP: 192.168.150.18
– HOST TCP/IP Port: 1974
– Registered Name: AIXRED
4.On the next screen press y to save the configuration.
5.Leave the FDR/UPSTREAM configurator by pressing ESC and X.
Installing the director component
Install the FDR/UPSTREAM Director on your local workstation to manage and monitor your
backup and restore activities.
Use the same procedure to install the director component that you used to install the server
component, but with a few different selections.
1.On the FDR/UPSTREAM Installation Wizard Select Feature screen, unselect all
components, then select the item FDR/UPSTREAM Director as shown in Figure 3-69.
Click Next.

140
IBM TotalStorage Migration Guide for the SAP User
Figure 3-69 FDR/UPSTREAM Wizard - Selecting the Director feature
2.The window shown in Figure 3-70 is displayed. You can specify a disk drive and folder into
which the FDR/UPSTREAM Director should be installed or you can leave the setting at the
default. If you want to change the drive and the folder, select a new one, otherwise
continue the installation by clicking Next.
Figure 3-70 FDR/UPSTREAM Wizard - Choose Destination Location screen

Chapter 3. Environment 2 SAP migration experiences
141
3.The window shown in Figure 3-71 is displayed. Enter the IP address for the server in this
screen.
Figure 3-71 FDR/UPSTREAM IP-Address field
4.In the next window you have to customize the communication. Enter the TCP/IP address
from your server and enter the FDR/UPSTREAM server port number. Figure 3-72 shows
this window.
Figure 3-72 FDR/UPSTREAM Configuration Popup
5.After you have specified the parameters for FDR/UPSTREAM Server TCP/IP Address or
DNS Name, FDR/UPSTREAM Server Port Number, and Local Port Number, click Save.
6.The next window, shown in Figure 3-73, prompts you to log in to your server.

142
IBM TotalStorage Migration Guide for the SAP User
Figure 3-73 FDR/UPSTREAM Login
7.After you insert a chosen profile name and a storage server address, click OK to connect
to the server.
3.15 Configuring FDR/UPSTREAM Reservoir
You must configure FDR/UPSTREAM Reservoir before you perform for your first backup.
The following sections describe how to:
Configure Removable Storage Manager if you intend to perform tape backups
Configure FDR/UPSTREAM Reservoir overall parameters
Configure FDR/UPSTREAM Reservoir backup profile specific parameters
3.15.1 Removable Storage Manager
Microsoft Windows2000 and above includes a feature called Removable Storage Manager
(RSM) that can be used to manage the data that is stored in your system. RSM provides a
single interface for managing the data that is stored in a stand-alone drive and in storage
libraries, and it allocates media to applications.
FDR/UPSTREAM Reservoir uses Removable Storage Manager because it gives access to
virtually all tape drives and library systems available today.

Chapter 3. Environment 2 SAP migration experiences
143
If you will be using FDR/UPSTREAM Reservoir with tapes, you need to understand and
prepare RSM for use with FDR/UPSTREAM Reservoir. Otherwise, you can skip to
“FDR/UPSTREAM Reservoir overall parameter configuration” on page 145.
Removable Storage Manager is accessed as a computer management snap-in. From the
Windows Start menu, select Settings

Control Panel

Administrative Tools


Computer Management. Expand the Storage tree and than the Removable Storage
sub-tree. If you are running a non-server version of Windows, highlight Removable Storage,
pull down the View menu, and select Full for a complete list of the components in Removable
Storage.
Creating a pool
A media pool is a logical collection of cartridges that share some common attributes. A media
pool contains media of only one type, but media in the media pool can reside in more than
one library. Each cartridge is in a media. Both sides of a two-sided cartridge are always in the
same pool.
Media pools can be used hierarchically. A media pool can be used to hold other media pools,
or it can be used to hold cartridges. An application that needs to group media of several types
into one collection can create one application media pool for the whole collection and then
additional media pools within the original pool - one for each media type. The
free pool

contains a media pool for each media type. There are two classes of media types: system
and application.
System media pools
are created by Removable Storage for its own use and
include the free pools, import pools, and unrecognized pools.
Application media pools
are
created by applications to group media. Grouping media is especially important if several
applications share the libraries that are attached to a system.
Each application that uses cartridges managed by Removable Storage, including
FDR/UPSTREAM Reservoir, uses one or more application pools. You must create at least
one application pool. Use the following steps to do this:
1.Highlight the Media Pools subtree in Removable Storage, pull down the Action menu,
and select Create Media Pool. The resulting screen is shown in Figure 3-74.

144
IBM TotalStorage Migration Guide for the SAP User
Figure 3-74 Windows2000 - Removable Storage Manager
2.Insert a name for your pool and any text to help you remember this definition. In the Media
Information section select the media type you wish to use for FDR/UPSTREAM Reservoir
backups. The Media Information section contains a field where you can build a hierarchy
of pools. The items are not used for your backup.
3.Do not select any items in the Allocation/Deallocation policy field.
Adding tapes
FDR/UPSTREAM Reservoir uses a mainframe-compatible tape labeling system. This system
requires a standard naming convention for tape labels, particularly that the name consist of
up to 6 numbers and letters. The USLabel.dll module installed with FDR/UPSTREAM
Reservoir will allow these tapes to be recognized by RSM.
If you have new tapes without bar codes, put the new media in the drive. For some drives, for
the media to be recognized by RSM, you may have to highlight the device, pull down the
Action menu and select Inject. When RSM sees the media, it will appear in a tree in
Computer Management/Media Pools/Unrecognized for the device you have. Using the
mouse, highlight the media, pull down the Action menu and select Free or Prepare. This will
cause RSM to write an MTF tape label on the media and move it to the Media Pools/Free
pool for the device you have.
Highlighting that free pool displays the list of media in it. Highlight all media you wish to use
for FDR/UPSTREAM Reservoir and drag it over to the FDR/UPSTREAM Reservoir pool. It
can now be used by FDR/UPSTREAM Reservoir.
If you have existing tapes which meet the name label standard, you can use them with
FDR/UPSTREAM Reservoir. In Windows Computer Management, located the tapes.
Deallocate them if they are assigned to an application. Drag them with the mouse to the
FDR/UPSTREAM Reservoir pool.

Chapter 3. Environment 2 SAP migration experiences
145
If the tapes do not meet the FDR/UPSTREAM Reservoir name label standards, you must use
RSM to rename the tapes; otherwise, FDR/UPSTREAM Reservoir will rename the tapes
using the RSM sequence number.
3.15.2 FDR/UPSTREAM Reservoir overall parameter configuration
The FDR/UPSTREAM Reservoir overall parameters are maintained in the file usserver.cfg in
the FDR/UPSTREAM Reservoir default directory. This file is created and modified with the
FDR/UPSTREAM Reservoir Server Configuration (asservercfg.exe)
This configuration is optional. The FDR/UPSTREAM Reservoir will run without a configuration
file, so you can get your service up quickly by skipping this step and proceeding directly to
startup of the FDR/UPSTREAM Windows Reservoir service. This will be done normally when
your systems restarts.
3.15.3 Starting FDR/UPSTREAM Reservoir service
To start FDR/UPSTREAM Reservoir for the first time, or manually later on, use the following
steps:
1.From the Windows Start button, select Settings

Control Panel

Administrative
Tools

Computer management.
2.Expand Services and Applications and highlight Services. In the right-hand pane,
select USSERVER. Right-click the item and select Start.
3.When the FDR/UPSTREAM Windows Reservoir is running you may see a text pane which
displays the current functions being performed. An example is shown in Figure 3-75.
Figure 3-75 FDR/UPSTREAM Windows Reservoir Services
Keyboard or mouse input is not recognized in this pane - it is for display only.
3.16 Start the client service
Before you select your files or file systems to back up, the FDR/UPSTREAM service has to
be started.

146
IBM TotalStorage Migration Guide for the SAP User
To start this service automatically after each boot, insert the following line into the inittab file:
usdaemon:2:wait:/usr/lpp/fdrupstream/usdaemon
parameter=/usr/lpp/fdrupstream/rmtparm.dat
configfile=/usr/lpp/fdrupstream/upstream.cfg > /dev/console 2>&1
To start the service for the first time, bring up the AIX window as the user root and switch to
the directory /usr/lpp/fdrupstream. Start the client with the command ./uscmd.
After the client starts the window shown in Figure 3-76 is displayed.
Figure 3-76 AIX Window Messages
The command to bring up the config file is found in the same directory. If you wish to change
any parameters, issue the command ./uscfg.
In our environment we used the default parameters.
3.17 Backup
After you have successfully installed the FDR/UPSTREAM Reservoir server, client, and
Director, you can use the UPSTREAM Director to start the backup.
Use the following steps to start the UPSTREAM Director:
1.From the Windows Start menu, select Start

Programs

FDR_UPSTREAM


UPSTREAM Director. The window shown in Figure 3-77 is displayed.
Note: Enter this as a single line.
Note: Be sure to stop the application SAP R/3 and stop the IBM DB2 UDB database
before the backup starts.
Note: If you wish to back up the database online, use the FDR/UPSTREAM IBM DB2
Database Agent.

Chapter 3. Environment 2 SAP migration experiences
147
Figure 3-77 FDR/UPSTREAM Director Login
2.The banner screen prompts you to sign in. You need to enter a Backup Profile name in the
Profile field. Any value up to 8 characters will generally work unless security is enabled, in
which case you need to enter a valid User Name/Password on the system you are
connecting to.
In FDR/UPSTREAM Reservoir, backup profiles define characteristics about how backup
files and tapes are created, maintained, and released. Each server, database, or other
entity that you are backing up needs its own profile. For a real-world environment where
backups run repeatedly, you need to verify that the settings here meet your needs. This is
particularly true for those settings concerning releasing tapes and disk resources, and
those involved with combining multiple backups on single tapes.
The FDR/UPSTREAM Reservoir backup profile definition parameters are maintained in
the file <Backup Profile>.prf in the FDR/UPSTREAM WorkPath directory.
3.In the Storage Server field, enter the <IP Address>:<Port Number> of the server system,
which is assign for you. Click OK, as shown in Figure 3-78.

148
IBM TotalStorage Migration Guide for the SAP User
Figure 3-78 FDR/UPSTREAM Director Login definitions
4.To start a backup, choose the Backup task label shown in Figure 3-79.
Figure 3-79 FDR/UPSTREAM Backup Window
5.At the bottom of the screen, select your client in the Target field to specify where the files
should be backed up.
6.To save your files which are located on a vendor storage subsystem, choose the item
Type Non-merge. Specify whether to save your data on tape or on disk (we specified
tape). Then select all files which belong to the SAP system.
In the pictured example we selected the files which belong to an SAP R/3 system called
P11. To store these files choose the following path in the FDR/UPSTREAM Backup menu:
–/db2
Note: For the first time in a production environment, specify the backup Type as First
time full, if you plan to create other incremental backups for the same files.

Chapter 3. Environment 2 SAP migration experiences
149
–/sapmnt/P11
–/usr/sap/P11
–/db2/db2as (it will be used in IBM DB2 UDB Version 7)
–/usr/sap/trans (if the file system is physically installed on your machine)
7.If you have other file systems to move from the vendor storage subsystem, like file system
/usr/sap/trans if it is physically install on the vendor disk, or other file systems you need,
select them from the backup window.
8.The screen shown in Figure 3-80 lists the files you have included in the backup as well as
the parameters you have specified.
Figure 3-80 FDR/UPSTREAM Backup Window
In our environment we used the following parameters:
Profile name: AIXROOTV
Target: AIXRED
Type: Non-merge
Storage: Tape
9.After you choose the files or file systems to back up, click the Start Backup button.

150
IBM TotalStorage Migration Guide for the SAP User
10.A screen showing the progress of your backup opens. An example is shown in
Figure 3-81.
Figure 3-81 FDR/UPSTREAM Backup Status
11.After your backup is finished the screen shown in Figure 3-82 appears.
Figure 3-82 FDR/UPSTREAM Backup completed message
3.18 Restore the data to IBM ESS
In this section we discuss how to migrate data in the form of a restore from a vendor storage
subsystem to an IBM Enterprise Storage Subsystem.
Installing AIX logical volumes and file systems on IBM ESS
We show you two scenarios for this:
In the first scenario we migrate data from the vendor storage subsystem to the IBM ESS
without changing the application server.
In the second scenario we migrate data from the vendor storage subsystem to the IBM
ESS
including
changing the application server.

Chapter 3. Environment 2 SAP migration experiences
151
First scenario
Perform the following steps to migrate data from the vendor storage subsystem to the IBM
ESS without changing the application server.
1.Connect the IBM pSeries machine physically to the IBM Enterprise Storage Subsystem.
2.Install the required device driver.
3.Make sure to have access to all disks on the IBM ESS.
4.Unmount all file systems on the vendor storage subsystem that belong to the SAP R/3
system and contain the associated data.
5.Deactivate the volume group that belongs to the SAP R/3 file systems on the vendor
storage subsystem.
6.Export the volume group that belongs to the SAP R/3 file systems on the vendor storage
subsystem.
If you want to use the same volume group name on the IBM ESS storage subsystem, you
have to remove this volume group first. Otherwise, you have to create a volume group with
another name for your disks to belong to the IBM ESS.
7.Create a new volume group on the disk that belongs to the IBM ESS.
8.Create new logical volumes of the same size you used for your logical volumes on the
vendor subsystem.
9.Create new file systems and mount points.
10.Mount these file systems.
11.Make sure you have access to all file systems needed for your SAP R/3 system.
12.Make sure all file systems have the same size as defined on the disks belong to the
vendor subsystem.
Second scenario
Perform the following steps to migrate data from the vendor storage subsystem to the IBM
ESS
including
changing the application server.
1.Connect the new IBM pSeries machine physically to the IBM ESS.
2.Install AIX and the AIX-prerequisites for SAP R/3 on this IBM pSeries machine.
3.Create groups and users with the same names and parameters you used on the previous
machine.
4.Insert the required ports in the /etc/services.
5.Enable the asynchronous I/O.
6.Adjust the kernel parameters described in the SAP R/3 OS Dependencies documentation
and the appropriate OSS-Notes.
7.Install the required device driver.
8.Make sure to have access to all disks on the IBM ESS.
9.Create a volume group on the disks with belongs to the IBM ESS.
10.Create new logical volumes of the same size you used for your logical volumes on the
vendor subsystem.
11.Create new file systems.
12.Mount these file systems.
13.Make sure you have all file systems needed for your SAP R/3 system.

152
IBM TotalStorage Migration Guide for the SAP User
14.Make sure all file systems are the same size as defined on the disks belong to the vendor
subsystem.
15.Implement the SAP R/3 OS dependencies with the same parameters, users, and groups
you used on you previous machine.
3.18.1 Use FDR/UPSTREAM Reservoir for restoring
To work with your FDR/UPSTREAM tool, first open the monitor as described in 3.17,
“Backup” on page 146.
The FDR/UPSTREAM tool enables you to perform the following tasks:
Restore
Backup
Profiles
Physical Restore
Physical Backup
SOS Management
Host Reports
Run Job
Dispatcher
ReGen
Use the FDR/UPSTREAM Director to execute your restore task. To start the
FDR/UPSTREAM Director, on your PC select Start

Program

FDR/UPSTREAM


UPSTREAM Director.
After you select the task Restore you will find the backup profile you created previously.
Figure 3-83 shows the window that displays the selected profile.
Figure 3-83 FDR/UPSTREAM Monitor

Chapter 3. Environment 2 SAP migration experiences
153
After you select your profile and open it, you will see which files were backed up. In an SAP
R/3 environment you have to restore the whole file system structure that belongs to your SAP
R/3 system.
Figure 3-84 shows the contents of this structure.
Figure 3-84 Backup files from profile AIXRED
To restore files you have to select them first. Figure 3-85 shows which files you have to select
to restore your SAP R/3 system.
Figure 3-85 Selected directories to restore
If you need other files systems or files, select them also.

154
IBM TotalStorage Migration Guide for the SAP User
In our scenario we installed the file system /usr/sap/trans physically on the vendor storage
subsystem, so we have to restore it, too.
If you want to migrate the file system /usr/sap/trans to the IBM ESS, select it in the restore
window; otherwise, ignore it.
Table 3-5 is an overview of a SAP R/3 installation and the associated files systems to restore.
Table 3-5 AIX File system overview
3.18.2 Monitor your tasks
After you start a restore operation you can monitor the restore status. The status screen
contains the following information:
Start time
Estimated time remaining
Restored file
File byte count
File size
Average character/second
Percent file complete
Percent total complete
Figure 3-86 shows an example of this screen when the restore is started and is waiting for a
tape.
File system 1024-blocks Used Free Mounted on
/dev/lvtrans 32760 10412 317268/usr/sap/trans
/dev/lvsapmntP11 802816 392344 410472/sapmnt/P11
/dev/lvusrP11 450560 132940 317620/usr/sap/P11
/dev/lv dbudb 917504 39812 877692/db2/P11
/dev/lvlog 2310144 1384488 925656/db2/P11/log_dir
/dev/lvretrieve 409600 12908 396692/db2/P11/log_retrieve
/dev/lvarchive 1007616 31680 975936/db2/P11/log_archive
/dev/lvdump 131072 5092 125930/db2/P11/db2dump
/dev/lvdata1 6021120 4653448 1345236/db2/P11/sapdata1
/dev/lvdata2 3817472 3308060 509412/db2/P11/sapdata2
/dev/lvdata3 3416064 2860004 556060/db2/P11/sapdata3
/dev/lvdata4 2015232 1664880 350352/db2/P11/sapdata4
/dev/lvdata5 4014080 3579436 434644/db2/P11/sapdata5
/dev/lvdata6 2260992 2022940 238052/db2/P11/sapdata6
/dev/lvdbas 32768 4548 28220/db2/db2as

Chapter 3. Environment 2 SAP migration experiences
155
Figure 3-86 Restore Status - Sample screen 1
Figure 3-87 shows the status screen while the restore operation is running.
Figure 3-87 Restore Status - Sample screen 2
Figure 3-88 shows a different set of details and a different presentation, but it is also a restore
status screen.

156
IBM TotalStorage Migration Guide for the SAP User
Figure 3-88 Restore Status Screen in Details
After the restore is finished all file systems are available on the IBM Enterprise Storage
Subsystem.
Before you start your SAP R/3 system, check your system by enter the df command on your
SAP R/3 application server. You should see all the file systems used by the rootvg and the
SAP R/3 volume group. Switch to the <sid>adm user and choose the R3trans -x command
to check the connection between application server and database. If all works fine your
migration is finished and you can start your SAP R/3 system.
3.18.3 Summary of the backup process
First install your server component on a Windows-based PC. The installation is performed by
an installation wizard. After you select the item SAR or Reservoir in the FDR/UPSTREAM
installation wizard the installation runs.
To install the client component on your AIX machine mount the CD, switch to the
recommended directory and start the installation by typing ./install.
To monitor your backup or restore tasks install the FDR/UPSTREAM Director on any PC you
want. The procedure is nearly the same to install a server component. Select Installation in
your wizard FDR/UPSTREAM and start the installation.
After you install all components, start the FDR/UPSTREAM Director, select backup, select
the files you want to save and start the backup.

© Copyright IBM Corp. 2005. All rights reserved.
157
Chapter 4.
SAP operational activities after
migration to ESS
Database administration is an import activity in an SAP environment. This chapter describes
the daily activities for an SAP database administrator.
There are different tasks required depending on what databases are used in your
environment. In this chapter, we consider only the activities for managing an SAP landscape
with IBM DB2 Universal Database Version 7 from the point of view of database
administration.
There is a variety of ways to monitor these activities, including implementing tools such as
Tivoli or other vendor products, or using the SAP alert monitor.
The main database activities for an administrator are:
Back up the database
Archive inactive logs
Back up tablespaces
Restore databases
Restore tablespaces
Restore tables
Update statistics
Reorganize tables
Analyze performance of tables and indexes
Monitor space statistics
Check for database errors
Monitor alerts
Monitor deadlocks
Monitor lock waits
4
Note: System management tools are helpful to administer the IT environment and many
are available in the marketplace, but it is beyond the scope of this book to describe these
tools in detail.

158
IBM TotalStorage Migration Guide for the SAP User
4.1 Tools to manage and monitor your database tasks
Some database administration tasks and activities, such as the database backup, have to be
performed regularly. You can schedule and coordinate most such tasks using the:
SAP DBA Planning Calendar
IBM DB2 UDB Control Center
IBM DB2 UDB Command Line (CLP)
In this section we present an overview of how to work with the SAP R/3 DBA Planning
Calendar and the IBM DB2 UDB Control Center.
We do not show how to work with scripts because this is an individual task for each
administrator. We only show the command syntax for specific tasks and activities.
4.1.1 SAP R/3 DBA Planning Calendar
We recommend using the SAP R/3 DBA Planning Calendar to monitor and manage the main
tasks in your SAP R/3 environment.
The main functions of the DBA Planning Calendar are to define the start times and
parameters for database activities.
To access the planning calendar, use the following path in the SAP GUI main menu: Tools


CCMS

DBA

Administration

DBA Planning calendar. You also can switch directly
to the transaction DB13. After you start the transaction the window shown in Figure 4-1
opens.
Note: We recommend using the SAP R/3 DBA Planning Calendar, where possible, to start
tasks for daily or weekly activities.

Chapter 4. SAP operational activities after migration to ESS
159
Figure 4-1 DBA Planning Calendar
To schedule actions in the planning calendar, choose the date by double-clicking the day’s
header bar. The system displays the actions supported by the calendar for your database
platform. Select the action you want to schedule. Enter the following parameters for the
actions you want to:
Starttime
Period
Calendar
After you insert all parameters press Enter to insert the desired action into the DBA Planning
calendar.
Figure 4-2 shows an example of the pop-up list of activities that is returned.

160
IBM TotalStorage Migration Guide for the SAP User
Figure 4-2 DBA Planning calendar with pop-up list of activities
4.1.2 IBM DB2 UDB Control Center
The DB2 UDB Control Center is another helpful tool to manage database activities. You can
use this tool to monitor and manage your activities in an SAP R/3 environment for the
database only without any SAP R/3 facilities.
Before you can use this tool, it must be installed on your local PC. After the installation is
finished start the tool by using the following path on your PC: Start

Programs

IBM
DB2

Control Center. The next step is to catalog your database if it is not done yet.
After you start your control center a graphical user interface is displayed. Figure 4-3 shows
how this looked in our test environment.

Chapter 4. SAP operational activities after migration to ESS
161
Figure 4-3 DB2 Control Center
We installed an SAP R/3 system with the SID P11. The control center shows the tree
expanded below our database P11.
Choose the database you want to work with by left-clicking its name. The pop-up list that
results is shown in Figure 4-4.

162
IBM TotalStorage Migration Guide for the SAP User
Figure 4-4 DB2 Control Center with the options for a chosen database
From this list you can select the tasks to manage or monitor your activities.
Note: In the following sections we discuss miscellaneous transactions to manage and to
monitor your database health. Some of these transactions have nearly the same results
but different screens. It is up to you to choose the transactions that are useful and helpful
for your work as an SAP R/3 database administrator in your particular environment.

Chapter 4. SAP operational activities after migration to ESS
163
4.2 Online backup of the database
In this section we show how to create an online backup using each of the three previously
introduced methods: the SAP R/3 DBA Planning Calendar, the IBM DB2 UDB Control Center,
and from an IBM DB2 command line.
4.2.1 Use SAP R/3 DBA Planning Calendar
To create a full online backup with the SAP R/3 DBA Planning Calendar, open the transaction
as described in 4.1.1, “SAP R/3 DBA Planning Calendar” on page 158. The calendar shown
in Figure 4-5 is displayed. Use the following steps to schedule your backup:
1.Choose what day you want to start your backup by double-clicking within the selected day.
2.Select where to save your backup. You have three options:
– In TSM
– In a device
– In a vendor library
A small pop-up window will appear where you type in your choice.
Figure 4-5 DBA Planning Calender - Full backup
3.In the next window select your backup type as shown in Figure 4-6.

164
IBM TotalStorage Migration Guide for the SAP User
Figure 4-6 DBA Planning Calendar - Backup Type
4.After you select your preferred Backup type one of the following windows appears,
depending on the choice you made in Step 2.
a.Backup into a device:
b.Backup into TSM:
c.Backup into a vendor library:
5.When you have completed all necessary entries, click the green check mark button to
start the backup.
4.2.2 Use the IBM DB2 UDB Control Center
In this section we show how to perform the same backup task using the DB2 Control Center.
1.Start the DB2 Control Center on your PC by selecting Start

Programs

IBM DB2


Control Center. The window shown in Figure 4-7 appears.

Chapter 4. SAP operational activities after migration to ESS
165
Figure 4-7 DB2 Control Center
2.Select the database you want to back up by right-clicking the name. The window shown in
Figure 4-8 appears.
Figure 4-8 DB2 Control Center - Backup
3.You can select Database or Database Using Wizard. If you are not familiar with the input
for the backup window use the Database Using Wizard, otherwise use Database.
We show how to use the input for the parameters without the wizard in Figure 4-9.

166
IBM TotalStorage Migration Guide for the SAP User
Figure 4-9 DB2 Control Center - Database Backup
4.Insert a path where you want to save your backup in the Specify directories field.
5.Click the Options tab and choose Online as shown in Figure 4-10.
Figure 4-10 DB2 Control Center - Backup Online
6.After you insert all necessary entries, click the Backup Now button to start the backup.
4.2.3 Use the IBM DB2 command line
If you need specific parameters for your backup use the DB2 command line interface.
Figure 4-11 and Figure 4-12 show the command syntax for the backup command.

Chapter 4. SAP operational activities after migration to ESS
167
Figure 4-11 DB2 Command Syntax - Backup Database (1)
Figure 4-12 DB2 Command Syntax - Backup Database (2)
Choose the database name and where you want to save your data to, then start your backup
from the DB2 command line shown in Figure 4-11 and Figure 4-12.
An example of a backup command in our database follows:
backup database P11 online to <device>

WITH

num-buffers

BUFFERS

BUFFER

buffers-size

PARALLELISM

n

WITHOUT

PROMPTING

168
IBM TotalStorage Migration Guide for the SAP User
4.3 Offline backup of the database
In this section we show how to create a database backup offline.
We recommend using the SAP R/3 DBA Planning Calendar for this task, but we also show
two other possibilities for creating a backup offline (IBM DB2 Control Center and IBM DB2
UDB Command Line).
4.3.1 Use SAP R/3 DBA Planning Calendar
To use the SAP R/3 Planning Calendar for an offline backup, begin by selecting transaction
DB13 as described previously. Use the same steps we described for an online backup, but in
step 3, instead of specifying Online (shown in Figure 4-6) select Offline in the Backup Type
window shown in Figure 4-13.
Figure 4-13 DBA planning Calendar - Backup Offline
There are no more differences in the SAP R/3 backup menus between online and offline
backup. Continue with your entries as describe in 4.2.1, “Use SAP R/3 DBA Planning
Calendar” on page 163.
4.3.2 Use IBM DB2 UDB Control Center
Next we show how to use the IBM DB2 Control Center to back up a database offline.
Start your DB2 Control Center as described in 4.2.2, “Use the IBM DB2 UDB Control Center”
on page 164 and select Backup

Database.
Click the Options tab and select Offline in the “Process while” field as shown in Figure 4-14.

Chapter 4. SAP operational activities after migration to ESS
169
Figure 4-14 DB2 Control Center - Backup Offline
Click the Backup Now button to start the backup.
4.3.3 Use the IBM DB2 UDB command line
In this section we describe how to back up an IBM DB2 Universal Database for AIX from the
command line.
We recommend using the SAP R/3 tools and transactions, but if you need to use specific
parameters, or if your tools are broken or otherwise unavailable, it is good to know how to run
the backup from a command line.
For DB2 you can use the integrated DB2 command backup. Use the following steps to create
a backup using the DB2 command line processor:
1.Log in as db2 user.
2.The db2 command starts the command line processor (CLP). The CLP is used to execute
database utilities and SQL statements, and to provide online help.
3.Issue the backup command, along with any desired options as shown in Figure 4-15.
A sample backup command using the syntax shown in Figure 4-15 is the following:
backup database P11 to <device>

170
IBM TotalStorage Migration Guide for the SAP User
Figure 4-15 Command Syntax for backup database
4.4 Archive inactive logs
To get a consistent database after a recovery procedure it is imported to save the database
logs.
4.4.1 Use SAP R/3 DBA calendar
We recommend using the DBA Planning Calendar to archive inactive logs for your database.
From the SAP GUI main menu choose Tools

CCMS

DBA

Administration

DBA
Planning calendar.
You also can switch directly to the transaction db13.
Select a day by double-clicking; a pop-up list of tasks appears. You can choose a task to
archive inactive logs into ADSM or onto device as shown in Figure 4-16.
Command Syntax

BACKUP

DATABASE
database-alias


DB

ONLINE

tablespace-name


TABLESPACE

(

)

,
dev

USE TSM

dir

TO

,

OPEN

num-sessions


SESSIONS

LOAD

library-name


OPEN

num-sessions


SESSIONS

USER

USING

username


password


Chapter 4. SAP operational activities after migration to ESS
171
Figure 4-16 SAP R/3 DBA Planning Calendar - Archive Logs
4.5 Back up tablespaces
It is not possible to use the SAP R/3 DBA Planning Calendar to back up a tablespace.
There are two other methods you can use to back up tablespaces:
IBM DB2 UDB Control Center
IBM DB2 command line
4.5.1 Use the IBM DB2 UDB Control Center
To use the DB2 Control Center open it as described previously.
Select the database where the tablespace belongs as shown in Figure 4-17.

172
IBM TotalStorage Migration Guide for the SAP User
Figure 4-17 DB2 Control Center - Backup tablespace
The next window (Figure 4-18) shows the details about the tablespace backup.
Figure 4-18 DB2 Control Center - Tablespace Backup Details
Select a tablespace by right-clicking its name. Choose a device or directory to save it to.
Click the Options tab and choose either online or offline. Click the Backup Now button to
start the backup of a tablespace.

Chapter 4. SAP operational activities after migration to ESS
173
4.5.2 Use the IBM DB2 UDB Command Line
You can create a tablespace backup using the same database backup command described
previously. Use this command with the following option:
backup database <database-alias> tablespace (tablespace-name)
Figure 4-19 shows the command syntax for a tablespace backup.
Figure 4-19 Command Syntax to back up a tablespace
Command Syntax

BACKUP

DATABASE
database-alias


DB

ONLINE

tablespace-name


TABLESPACE

(

)

,
dev

USE TSM

dir

TO

,

OPEN

num-sessions


SESSIONS

LOAD

library-name


OPEN

num-sessions


SESSIONS

USER

USING

username


password


174
IBM TotalStorage Migration Guide for the SAP User
4.6 Restore databases
Sometimes the requirement is to restore a database. This could happen, for example, in a
test environment to restore a previous version after running tests or migration scenarios.
We show two different possibilities to restore your data using:
IBM DB2 UDB Control Center
IBM DB2 UDB Command Line
4.6.1 Use the IBM DB2 Control Center
To use the DB2 Control Center open it as described in 4.1.2, “IBM DB2 UDB Control Center”
on page 160.
Select the database you want to restore by right-clicking its name.
The next window, shown in Figure 4-20, provides details about your last restore.
Figure 4-20 IBM DB2 UDB Control Center - Restore Database
Select the database you want to restore and click OK to start the restore.
4.6.2 Use the IBM DB2 command line
You can use the command line to restore a database.
Log in as db2 user and execute the db2 command. At the db2 prompt use the restore
command with the appropriate options shown in Figure 4-21 and Figure 4-22.

Chapter 4. SAP operational activities after migration to ESS
175
Figure 4-21 Restore command syntax
c
Figure 4-22 Restore command syntax (continued)
Command Syntax

RESTORE

DATABASE

source-database-alias


DB
restore-options

CONTINUE

ABORT
restore-options:

USER

USING

username


password


ONLINE

TABLESPACE ONLINE

tablespace-name


TABLESPACE

(

)

,

HISTORY FILE

ONLINE

device


USE TSM

directory


FROM

,

OPEN

num-sessions


SESSIONS

LOAD

shared-library


OPEN

num-sessions


SESSIONS

TAKEN AT

date-time

TO

target-directory

INTO

target-database-
alias

NEWLOGPATH

directory

WITH

num-buffers

BUFFER

buffer-size

BUFFERS
DLREPORT

filename

REPLACE EXISTING

REDIRECT

PARALLELISM

WITHOUT ROLLING FORWARD

WITHOUT DATALINK

WITHOUT PROMPTING

176
IBM TotalStorage Migration Guide for the SAP User
4.7 Restore tablespaces
We recommend using the same method to restore tablespaces that you used to restore a
database.
4.7.1 Use the IBM DB2 UDB Control Center
Open the IBM DB2 Control Center.
Select the database you want to restore the tablespace to. Select the item tablespace by
left-clicking it. Select a tablespace from the menu by right-clicking it (Figure 4-23); then
choose Restore.
Figure 4-23 IBM DB2 Control Center - Restore Tablespace
After you select the tablespace the window shown in Figure 4-24 is displayed.

Chapter 4. SAP operational activities after migration to ESS
177
Figure 4-24 IBM DB2 Control Center - Restore Tablespace
Click the Options tab and select Offline or Online backup. When all entries are complete,
click OK to start the restore.
4.7.2 Use the IBM DB2 UDB command line
You can use the command line to restore a tablespace. The command reference shows the
appropriate options to restore a tablespace. In our environment we worked with DB2
Universal Database Version 7. The command syntax a to restore a tablespace is shown in
Figure 4-25 and Figure 4-26.
A sample command to restore a tablespace is the following:
restore database P11 tablespace <tablespace name> from <device>

178
IBM TotalStorage Migration Guide for the SAP User
Figure 4-25 Restore Tablespace command syntax
Figure 4-26 Restore Tablespace command syntax (continued)
Command Syntax

RESTORE

DATABASE

source-database-alias


DB
restore-options

CONTINUE

ABORT
restore-options:

USER

USING

username


password


ONLINE

TABLESPACE ONLINE

tablespace-name


TABLESPACE

(

)

,

HISTORY FILE

ONLINE

device


USE TSM

directory


FROM

,

OPEN

num-sessions


SESSIONS

LOAD

shared-library


OPEN

num-sessions


SESSIONS

TAKEN AT

date-time

TO

target-directory

INTO

target-database-
alias

NEWLOGPATH

directory

WITH

num-buffers

BUFFER

buffer-size

BUFFERS
DLREPORT

filename

REPLACE EXISTING

REDIRECT

PARALLELISM

WITHOUT ROLLING FORWARD

WITHOUT DATALINK

WITHOUT PROMPTING

Chapter 4. SAP operational activities after migration to ESS
179
4.8 Other daily activities
Some other activities an SAP R/3 database administrator performs to manage and monitor
the database are:
Update table statistics
Reorganize tables
Analyze performance
Monitor space statistics
Check for database errors and alerts
Monitor deadlocks and waits
We discuss these tasks in the following sections.
4.8.1 Maintain statistics
The database optimizer decides how to run database queries in the most efficient way. To
maintain your statistics use the transaction DB20.
The window shown in Figure 4-27 appears.
Figure 4-27 DB20 - Maintain Statistics - Initial Screen
We used “table t000” only as an example of how the transaction works and how a screen
looks. It is only a small table with the contents of created clients in an SAP R/3.
The window shown in Figure 4-28 appears.

180
IBM TotalStorage Migration Guide for the SAP User
Figure 4-28 DB20 - Maintain Statistics
After you specify the table of interest, the next window shows detailed statistics about the
table as shown in Figure 4-29.
Figure 4-29 DB20 - Maintain Statistics Details

Chapter 4. SAP operational activities after migration to ESS
181
4.8.2 Update table statistics
We recommend using the SAP DBA Planning Calendar to update the database statistics.
Choose the transaction DB13 to obtain a schedule overview. Select a day to execute an
action; a pop-up task list appears. The last two tasks in this list are useful to plan the tasks for
updating the database statistics. You can choose:
Upd. Statistics + Reorgcheck all tables
Select this task to update the statistics for all tables. This task may take several hours
depending on the size of the database. We recommend scheduling this job to run once a
week.
Updates Statistics + Reorgcheck(DBSTATC)
This task operates on a subset of tables entered in the DBSTATC table. You should
schedule this job to run daily.
After you fill out the entries for start time, period, and calendar, an entry in the DBA planning
calendar appears.
4.8.3 Reorganize tables
We recommend that you periodically reorganize tables to achieve the best performance from
your system. We show how to plan a reorganization task and which method we used to
reorganize tables.
There are two options for reorganizing the tables. They are:
Reorganize flagged tables and update statistics
A selection of tables (maximum 100) requiring reorganization is displayed. Information on
runtime and size is given for each table; this information is used to calculate an
approximate time required for the reorganization.
Reorganize tables in tablespaces
This task reorganizes all tables of a specific tablespace using primary indexes. The table
statistics are updated.
Figure 4-30 shows the specific entry in the DBA Planning Calendar.
Note: While the optimizer’s statistics are being updated, the performance of your SAP R/3
system is impacted. Therefore, we suggest that you run the update of the statistics at night
or during weekends.

182
IBM TotalStorage Migration Guide for the SAP User
Figure 4-30 DBA Planning Calendar with the task selection menu
4.8.4 Status of activities and tasks
You can monitor your activities using the same tool you choose the activities with. Open the
DBA Planning Calendar and c
lick the header bar for a particular day.
Unsuccessful or interrupted actions are shown in red in the DBA Planning Calendar and, if
there are only unsuccessful or interrupted actions, then the header bar for that day is also
highlighted in red.
In the scheduling overview, you can see if any logs were written for an action.
To survey the action logs or background job log for a particular action, select Action logs or
Job logs as appropriate. All logs written on the day you have selected are listed. Some
actions do
not
write logs.
The timestamp is used to assign logs to scheduled actions. An action log is assigned to the
action which has the same type and the closest corresponding timestamp. In some cases, for
example, if no background work process is available, the action is delayed and even
postponed until after the next scheduling time. Unfortunately this means that the action log is
then assigned to the next scheduling time and the original scheduling time log is incorrect.
This is the case if the logs for the previous schedules are displayed for the next schedule of
the same type.
The screen shown in Figure 4-31 appears if you choose the header bar for a particular day.

Chapter 4. SAP operational activities after migration to ESS
183
Figure 4-31 DBA Planning Calendar with an action log entry
4.8.5 Database performance: Tables and indexes
As the database administrator you have to monitor the size of your database. If the database
does not have enough space to save data it will become unavailable, so you have to monitor
the growth of tables and indexes regularly.
To access informations about the growth of your database in CCMS, select Control


Monitoring

Performance Menu

Database

Tables/Indexes, or use transaction
DB02.
The window shown in Figure 4-32 appears.

184
IBM TotalStorage Migration Guide for the SAP User
Figure 4-32 DB Performance: Tables and Indexes screen
In the Tablespaces field, click the Detailed analysis button. Detailed information is displayed
about each tablespace, such as used size and free size, as shown in Figure 4-33.
Note: Implement a plan for archiving SAP R/3 data right from the start to prevent SAP R/3
systems from growing too large too quickly.

Chapter 4. SAP operational activities after migration to ESS
185
Figure 4-33 DB02 - Tablespaces
If you click the Graphics button at the top of the screen, a graphical representation of the data
is displayed, as shown in Figure 4-34.
Figure 4-34 DB02 - Tablespace Analysis

186
IBM TotalStorage Migration Guide for the SAP User
4.8.6 Overview of backup status
To get an overview of your backup status select CCMS

Control

Monitoring


Performance Menu

Database

Activity, or specify transaction DB12.
The window shown in Figure 4-35 appears.
Figure 4-35 Overview of Backup Status screen
Two separate areas in this screen monitor backups as follows:
Database backup information from the recovery history file
DB2 log file information
To monitor the database backup information click Last successful database backup or
Overview of all database backups available if this information is included in your statistics.
To monitor your DB2 log file information click Log directory status.
To use the interface between IBM DB2 and SAP R/3 to archive or restore log files, we
recommend that you activate the user exit mechanism by setting the database configuration
parameters USEREXIT and LOGRETAIN. When the user exit program is to be invoked, the
database manager passes control to the executable db2uext2, which can store and retrieve
log files and manage the location of archived log files. Using a user exit program to archive
and retrieve files enables a database to perform roll forward recovery.
4.8.7 Database Alert Monitor
One of the tools available to monitor your system is the Database Alert Monitor. This tool can
monitor any or all of the following aspects of your system:
Performance

Chapter 4. SAP operational activities after migration to ESS
187
Backup
Memory
DB-configuration
System check
To invoke the monitor, select CCMS

Control

Monitoring

Performance Menu


Database

Alert monitor; or specify transaction DB16.
If the systems generates alerts they will be displayed in window shown in Figure 4-36.
Figure 4-36 Database Alert Monitor
Specify the object you want to monitor in the Current selection field.
Before you use the alert monitor we recommend that you customize this tool. Choose the
DB6 Alert Configuration to access the configuration menu shown in Figure 4-37.

188
IBM TotalStorage Migration Guide for the SAP User
Figure 4-37 DB16 - Alert Monitor configuration
It is in your administrative discretion to set the thresholds for objects you want to monitor. In
the next section we show how to customize an alert for an “archive stuck” warning.
After you have customized your alert monitor you will get a global overview about your SAP
R/3 system’s health.
4.8.8 Database check
Choose the transaction DB16B to check a lot of tasks in conjunction with a database check.
The window shown in Figure 4-38 appears when you start the transaction.

Chapter 4. SAP operational activities after migration to ESS
189
Figure 4-38 Database Check
To configure a specific task, switch to the configuration menu by clicking the Configure
Check button in the menu at the top of this screen. The window shown in Figure 4-39
appears.

190
IBM TotalStorage Migration Guide for the SAP User
Figure 4-39 DB16B - Database Check Configuration menu
Using the parameter ARCHIVE_STUCK as our example, we now show you how to configure
a single task. Double-click the row ARCHIVE_STUCK to select it. The window shown in
Figure 4-40 appears.
Figure 4-40 DB16B - Check Parameter for ARCHIVE_STUCK

Chapter 4. SAP operational activities after migration to ESS
191
Table 4-1 and Figure 4-41 identify and illustrate some of the possibilities for customizing the
alert of an archive stuck problem in your database. It is up to you to set the thresholds for
alerts. In our configuration we used the default parameters.
Table 4-1 Overview of Activities and appropriate Parameters
Figure 4-41 DB16B - Selecting Menu for the condition
After you choose Database Check the window shown in Figure 4-42 is displayed. Click Yes
to start the database check. If any errors are detect the overview shows you these entries.
Figure 4-42 DB16B - Database Check
Actv Yes No
Condition Error Warning Exception
Repetion Period Seconds Hours Days
Corrective measure Report Transaction Program Editor Hotline
DB16B - Overview of Activities
Actv
Condition
Repetion
Period
Corrective
measure
Equal to
Not Equal to
Small than
Less than/Equal to
Greater than(older)
Greater Than/Equal to
Greater than/Smaller
Kilo
Mega
Giga
Percentage
Seconds
Hours
Days
Rate
<No Unit>

192
IBM TotalStorage Migration Guide for the SAP User
4.8.9 Database Alert Monitor/System Check
In your SAP R/3 system you can configure a monitor to get warnings or alerts if defined tasks
or activities you enabled and configured fail to execute properly. SAP supplies interfaces to
tasks or activities you can use for monitoring your SAP R/3 system.
Figure 4-43 shows the first screen of the SAP Alert Monitor. Use the transaction DB20 to
access this monitor.
Figure 4-43 DB20 - Alert Monitor
Before you start using the Alert Monitor, we recommend that you configure the tool for your
own requirements. In this section we briefly explain the configuration process, but a detailed
discussion is beyond the scope of this book. The Alert Monitor is a powerful tool, and
customization requires you to apply experience and judgement. See the pertinent product
documentation for more information.
In the following paragraphs we provide an overview of two tools available for configuring
database system monitoring: general alert monitoring and DB2 UDB-specific configuration.
Configuration using general alert monitoring
You can change scheduling data by entering how often the data collection program should
run.
A useful feature of DB-specific alert monitoring is the central, automated notification function
that informs you of an alert by e-mail. If you want to be notified by e-mail as soon as an alert
is raised, you have to define yourself as a recipient of these generated e-mails.
To do this:
1.Choose Tools

CCMS

Configuration

Alert Monitor, or call transaction RZ21.
The Monitoring: Characteristics and Methods screen appears.

Chapter 4. SAP operational activities after migration to ESS
193
2.Click Method Definition

Display Overview.
3.Scroll through the list until you find CCMS_DB6_WATCHDOG.
4.Select this method and choose List

Selected Entries

Edit.
5.Select Parameter.
6.Select Method Definitions

Display

Change.
7.Select the task Parameters.
8.In the SENDER line of the Parameter value column, enter a valid user for your SAP
System.
9.In the RECIPIENT line of the Parameter value column, enter a valid user of your SAP
System who should be notified in the event of an alert.
10.Save your changes.
Not all alerts automatically generate an e-mail notification; you must configure those that you
want to do so.
DB2 UDB-specific configuration
Choose Tools

CCMS

DB administration

Check

Configuration, or call
transaction DB17.
You can specify additional parameters, for example, assignment of logged values for given
alerts.
The initial screen of the database-specific configuration provides you with an overview of all
the configuration entries. You can display details, sort entries, and make selections using the
list boxes. In addition, an entry can be activated or deactivated by selection of a line in the
active column. If you deactivate an entry, there is no further notification of corresponding
alerts.
In the Detail display, you can configure additional parameters, such as:
Threshold - This subscreen is divided as follows:
– Normal operation
– Warning: Limited operation, with reduced performance, for example
– Error: Normal operation is endangered if the error is not corrected
The following fields can be changed depending on operation status:
Relational operators
You can specify how the defined comparison value should be compared with the current
given value. In addition to the relational operators, you can enter whether a value should
lie inside of or outside of a range of values. A full colon (:) must separate the two values.
You can also specify whether or not discrete values are within a set of explicit values.
Semicolons (;) must separate such values.
Comparison value
You can specify a value, a list of values, or a value range depending on the operator. This
value will later be compared with the current measured value.
Comparison value (unit of measurement)
You can specify the unit of measurement of the comparison value. This is important for
time values, which are normally calculated internally in seconds, to be correctly converted
before comparison.

194
IBM TotalStorage Migration Guide for the SAP User
Automatic e-mail notification check box
Values do not need to be entered for every operation status. However, you must make sure
that the sum of comparison values must cover every possible value. If this is not the case, a
special alert is triggered with the message:
There is no configuration entry for the logged value.
The following tabs display additional information:
General (transaction RZ21)
Here you view the scheduling data from the basic alert monitor configuration. The values
are displayed here for completeness. It is not possible to make changes in this
transaction. You can make changes using the general maintenance function in the alert
monitor.
Administration
This screen displays the identity of the user that made the last changes and whether this
entry is currently active.
After you have made your changes, save them. Changes take effect immediately.
Figure 4-44 is an example of how a monitor looks. We used the default SAP monitor for this
example.
Figure 4-44 RZ20 - Alert Monitor Tree
4.8.10 Database performance monitor
To monitor the performance of your SAP R/3 system use the transaction DB6COCKPIT or
ST03N. In this section we briefly mention the tasks that are part of performance and tuning; a
discussion of performance issues is beyond the scope of this book.

Chapter 4. SAP operational activities after migration to ESS
195
When you start the transaction db6cockpit the screen shown in Figure 4-45 is displayed.
Figure 4-45 DB2 Performance Overview
A lot of tasks to manage performance are selectable from this screen. Performance should be
considered in the context of average response times across your entire SAP R/3 system.
Consult the appropriate product documentation for additional information.
4.8.11 Database snapshot analysis
To monitor your database use the transaction DB6DBS. This task is the same as the
performance transaction describe in the previous section.
In the following sections we analyze some representative database snapshots that indicate
some common situations:
deadlock
and
wait
.
The Database Snapshot window shown in Figure 4-46 is displayed when you start the
transaction.

196
IBM TotalStorage Migration Guide for the SAP User
Figure 4-46 Database Snapshot
4.8.12 Deadlocks analysis
The DB2 UDB database event monitor records the occurrence of deadlocks. These are
situations where two or more applications lock each other, each waiting for the others to
release the lock. Deadlock situations are detected automatically by DB2 UDB and resolved
after a specified time period. This time period is determined by database configuration
parameter
DLCHKTIME
. Information on the SAP work processes involved, the database
manager agent processed, and detailed lock information on the objects involved in the
deadlock is provided. For technical reasons, two event monitors are used. These are then
switched on and off alternately, so that one monitor is always active. The only reliable
information is that recorded by the monitor that has just been switched off. This information is
displayed when the function is called.
You can switch between the event monitors using the
Change Event Monitor
function. When
you switch between event monitors, displayed information is lost. For this reason, it is
essential to save information on deadlocks. To do this, choose Save to a local file to
download the displayed information to your computer.
To monitor deadlocks choose transaction DB6DLK or ST04 and select the
Locks_Deadlocks tab. The field Deadlocks Detected gives you a hint about how many
deadlocks have occurred.
When you start transaction ST04 the window shown in Figure 4-47 appears.

Chapter 4. SAP operational activities after migration to ESS
197
Figure 4-47 ST04 - Database Snapshot for deadlocks
The window shown in Figure 4-48 appears after you start the transaction DB6DLK. THe
information presented is an analysis of the specified deadlock.
Figure 4-48 DB2 Deadlocks

198
IBM TotalStorage Migration Guide for the SAP User
4.8.13 Exclusive Lock Waits analysis
To monitor Lock Wait situations select transaction DB6EX or ST04.
The window shown in Figure 4-49 is displayed when you start transaction DB6EX.
Figure 4-49 Analyses Lock Waits
Transaction ST04 was described in the previous section. Choose the same tab, and the field
“Waits” gives you an overview about wait situations in your SAP R/3 system.
4.8.14 Monitor scheduled jobs
To monitor results of scheduled jobs start the transaction SM37, or select Tools

CCMS


Jobs

Maintenance.
This monitor gives you an overview about tasks which are scheduled, released, ready, active,
finished, or canceled. Choose a job name or a user name, or both; select the start condition
for your monitoring; and execute this transaction. The screen shown in Figure 4-50 is
displayed.

Chapter 4. SAP operational activities after migration to ESS
199
Figure 4-50 Job Selection / Overview
Select a job about which you want to know details. A job log like the one shown in Figure 4-51
shows details of what happened during execution of this job.
Figure 4-51 SM37 - Job Log Details

200
IBM TotalStorage Migration Guide for the SAP User
4.8.15 DB2 UDB diag log
The database log is written as a file on disk in the path /db2/<SID>/sqllib/db2dump. The file is
called db2diag.log.
You can access this file by switching to it using the SAP GUI. The path to this transaction is
CCMS

Control/Monitoring

Performance Menu

Database

Activity. Click the
Diagnostics tab, then double-click the task DB2 UDB diag log or choose transaction ST04.
The window shown in Figure 4-52 appears.
Figure 4-52 Database Alert Monitor/DB2 UDB Alert Log
The entries in this log give general information about the status, warnings, or problems with
objects in the database. If a problem exists you get tips or hints about how to handle it.
4.8.16 Transactions in an SAP R/3 environment with IBM DB2 UDB on zOS
This section describes operational activities from the point of view of a database administrator
in an SAP R/3 environment with the IBM DB2 UDB on zOS.
The purpose is to introduce a few specific SAP R/3 transactions in a zOS environment. There
are two main transactions:
DB2D - Deadlock Monitor
DB2T - Time Monitor
Deadlocks
Deadlock situations are described in 4.8.12, “Deadlocks analysis” on page 196. The deadlock
monitor shows you tasks where processes are in wait situations.

Chapter 4. SAP operational activities after migration to ESS
201
Select the following path to choose the deadlock monitor: CCMS

Control/Monitoring


Performance Menu

Database

Exclusive Lock waits, or specify transaction DB2D.
The window shown in Figure 4-53 is displayed.
Figure 4-53 DB2D - Deadlock monitor
Select a day and time for which you want to monitor your deadlock situation.
After you make the necessary entries and start the execution, the results are displayed on a
screen like the one shown in Figure 4-54.

202
IBM TotalStorage Migration Guide for the SAP User
Figure 4-54 DB2D - Deadlocks
In our system there were no deadlock situations. If there had been any, the details would be
included on this screen.
Time out
Truncation of a transaction could be due to a time out situation. In an SAP R/3 environment
with the database IBM DB2 UDB on zOS the time parameter is set to 600 seconds before
transactions will get a truncation.
To monitor time out situations choose transaction DB2T as shown in Figure 4-55.

Chapter 4. SAP operational activities after migration to ESS
203
Figure 4-55 DB2T - Time Out
Select a day and time for which you want to monitor the time out situations in your system
and execute the transaction.
Figure 4-56 DB2T - Time Out Monitor details

204
IBM TotalStorage Migration Guide for the SAP User
If you have no time outs during the selected period the window looks like that shown in
Figure 4-56. In our environment no applications were running, so it was not possible to create
a time out.
If time outs did occur, details are displayed so that you can analyze the situation.
4.8.17 Summary of regular database activities and tasks
Table 4-2 presents a summary of important regular database administrative activities and
tasks.
Table 4-2 Summary of activities and tasks
Activity Task Transaction Frequency Monitor
Update statistics DB13 Schedule once Monitor weekly
Reorganization DB13 Schedule weekly Monitor weekly
Database Space DB02 Monitor weekly
Backup status DB12 Schedule once Monitor daily
Database
Performance
DB6COCKPIT
ST03N
Monitor weekly
Database Log ST04 Monitor daily

© Copyright IBM Corp. 2005. All rights reserved.
205
Appendix A.
INNOVATION Data Processing
Corporation product information
This appendix contains an overview of the INNOVATION Data Processing Corporation’s
products used in some of the migration scenarios. The INNOVATION Data Processing
Corporations. products described here are FDRPAS, FDRERASE and FDR/UPSTREAM. It
also contains a section on how to read the syntax for INNOVATION products/
A
Note: The product information contained in this appendix is copyrighted property of
INNOVATION Data Processing Corporation.

206
IBM TotalStorage Migration Guide for the SAP User
FDRPAS
The following charts describe the INNOVATION Data Processing product called FDRPAS.
©Copyright 2004 Innovation Data Processing
INNOVATION FDRPAS™
A new tool for Non-Disruptive
OS/390® and z/OS™ DASD
Hardware Install and Relocation

Appendix A. INNOVATION Data Processing Corporation product information
207
What are you doing here?
Exploring a familiar good news, bad news situation:

Good news!
Customers are facing explosive data growth…
they need newer, bigger, faster disk storage systems.

Bad news or an opportunity?
Customers want to start using the newer, bigger, faster storage
systems, without any interruption in business processing…
Opportunity!…

Mainframe customers can begin using their new OS/390® and
z/OS™ DASD disk with absolutely no disruption to their ongoing
business.
What are your options?
Customer typically have three options to choose
from for non-disruptive migrations…
contract with;

INNOVATION Data Processing to license FDRPAS

EMC, IBM, Hitachi or StorageTek professional
services to employ FDRPAS

A vendor’s business partner to employ FDRPAS.

208
IBM TotalStorage Migration Guide for the SAP User
Why do customers pick Professional Services?

Need to migrate data QUICKLY

No staff, time nor inclination to learn new software

Professional Services staff has years of experience
How did the hardware vendors come to chose FDRPAS™ ?

Long-term relationship with INNOVATION

Found FDRPAS™ quick to install, easy to use and fast

FDRPAS™ is kept up to date with hardware
enhancements

Some customers have experience with FDRPAS and
prefer to use it
What is FDRPAS™ ?

FDRPAS™ is part of INNOVATION's Suite of
Non-Disruptive Storage Management Solutions…
– To move disk volumes while still in use!
– During the production day
– Whenever convenient
– Grouping Volumes

Staging – Consolidating - Dispersing

Transparent to system and users
– Puts an exact copy of a volume on a new device
– Running in real time handling data and updates

Swaps all I/O operations to the new device

Appendix A. INNOVATION Data Processing Corporation product information
209
When to use FDRPAS™ ?

ESCON / FICON Mix

Distribute Volumes Across Multiple Storage Systems

LSS Configuration PAV Constraints

Incompatible FICON / FICON mix

No Toleration for Any Interruption

Time Constraint to Start

Quick In and Out
How can you insure no cust omer disruption…
INNOVATION brings you

FDRPAS™ …, to non-di sruptively replace any
vendor’s disk….. , any time,.
When customers buy new IBM mainframe disk
storage…

Replace all old dis k storage using
FDRPAS™


210
IBM TotalStorage Migration Guide for the SAP User
FDRPAS™

– Storage Hardware Support

Storage Hardware Independent…
– Supporting all vendors –
EMC, HDS IBM, STK
– Moving Any-to-Any
like hardware device

Across & Between…
– Multiple Systems Updating DASD
– 2105 and 3990 controllers
– FICON, ESCON and/or Parallel
– Channel Extender Distances
– Smaller to Larger
– Mirrored Disk, Raid-S, Raid 10, Raid 5
– PPRC Cache Fast Write
– PAV, non-PAV
– SRDF, Consistency Group
FDRPAS™

– System Software Support

All Current z/OS and OS/390 Operating Systems
– 31 bit and 64 bit mode

Moves Active Application Datasets and Databases
– SMS Managed and Non-SMS Managed
– CICS - USS WebSphere – DB2

Moves Active System Volumes
– PLPA and Common Page volumes
– Active Coupling Facility Dataset volumes
– JES Spool / JES Checkpoint
– JES3 volumes / MVS disks under JES3
– Master and User Catalog volumes

Appendix A. INNOVATION Data Processing Corporation product information
211
FDRPAS™

– System Software Support

All Current z/OS and OS/390 Operating Systems
– 31 bit and 64 bit mode

Moves Active Application Datasets and Databases
– SMS Managed and Non-SMS Managed
– CICS - USS WebSphere – DB2

Moves Active System Volumes
– PLPA and Common Page volumes
– Active Coupling Facility Dataset volumes
– JES Spool / JES Checkpoint
– JES3 volumes / MVS disks under JES3
– Master and User Catalog volumes
FDRPAS™

Customer Benefits

No Disruption to Business Operations

Fast, Safe and Secure

New z/OS and OS/390 Storage Available Sooner
…during normal business hours, customer installation non-disruptively
moved over 500 volumes to a new z/OS storage system in less than
3.5 hours concurrently swapping as many as 40 volumes at a time.

212
IBM TotalStorage Migration Guide for the SAP User
8
8
2
2
1
1
How Does FDRPAS™ Work?
Offline
Target
Device
Online
Source
Volume
Monitor
Task
Monitor
Task
Swap
Task
Swap
Task
Swap
Task
Swap
Task
Intercept
Intercept
Quiesce
Quiesce
Offline
Target
Volume
Online
Source
Device
Original volume online
and target offline
START
7
7
7
7
3
3
4
4
5
5
6
6
3
3
Target now online and
original volume offline
Offline
Source
Volume
Online
Target
Device
END
Installation

Easy-to-install and implement
– IEBCOPY load of libraries
– No SMP/E
– No TCP/IP or VTAM required
– No coupling facility required
– No command dataset
– No separate disk communication files
– No IPL required to activate

Appendix A. INNOVATION Data Processing Corporation product information
213
How can you insure no customer disruption…
INNOVATION brings you

FDRPAS™ …, to non-disruptively replace any
vendor’s disk….. , any time,.
When customers buy new EMC, Hitachi, IBM or
StorageTek mainframe disk storage…

Replace all old disk storage using
FDRPAS™

What is FDRPAS™ ?

FDRPAS™ is part of INNOVATION's Suite of
Non-Disruptive Storage Management Solutions…
– To move disk volumes while still in use!
– During the production day
– Whenever convenient
– Grouping Volumes

Staging – Consolidating - Dispersing

Transparent to system and users
– Puts an exact copy of a volume on a new device
– Running in real time handling data and updates

Swaps all I/O operations to the new device

214
IBM TotalStorage Migration Guide for the SAP User
When to use FDRPAS™ ?

ESCON / FICON Mix

Distribute Volumes Across Multiple Storage Systems

LSS Configuration PAV Constraints

Incompatible FICON / FICON mix

No Toleration for Any Interruption

Time Constraint to Start

Quick In and Out
FDRPAS™

– Storage Hardware Support

Storage Hardware Independent…
– Supporting all vendors –
EMC, HDS IBM, STK
– Moving Any-to-Any
like hardware device

Across & Between…
– Multiple Systems Updating DASD
– 2105 and 3990 controllers
– FICON, ESCON and/or Parallel
– Channel Extender Distances
– Smaller to Larger
– Mirrored Disk, Raid-S, Raid 10, Raid 5
– PPRC Cache Fast Write
– PAV, non-PAV
– SRDF, Consistency Group

Appendix A. INNOVATION Data Processing Corporation product information
215
FDRPAS™

– System Software Support

All Current z/OS and OS/390 Operating Systems
– 31 bit and 64 bit mode

Moves Active Application Datasets and Databases
– SMS Managed and Non-SMS Managed
– CICS - USS WebSphere – DB2

Moves Active System Volumes
– PLPA and Common Page volumes
– Active Coupling Facility Dataset volumes
– JES Spool / JES Checkpoint
– JES3 volumes / MVS disks under JES3
– Master and User Catalog volumes
FDRPAS™ Restrictions
Customers who are not familiar with FDRPAS often ask…
- Can we copy JES2 data sets? - What about CICS?
- Logger files? - - Will FDRPAS™ handle DB2 database data
sets? - What about “RES” packs?

The simple answer is…
FDRPAS™ will move everything except volumes containing
“active” local page or swap data sets, which you can
non-disruptively relocate to a new volume without requiring an
IPL, using the “PAGEDEL REPLACE” z/OS console command.

216
IBM TotalStorage Migration Guide for the SAP User
FDRPAS™

Customer Benefits

No Disruption to Business Operations

Fast, Safe and Secure

New z/OS and OS/390 Storage Available Sooner
…during normal business hours, customer installation non-disruptively
moved over 500 volumes to a new z/OS storage system in less than
3.5 hours concurrently swapping as many as 40 volumes at a time.
8
8
2
2
1
1
How Does FDRPAS™ Work?
Offline
Target
Device
Online
Source
Volume
Monitor
Task
Monitor
Task
Swap
Task
Swap
Task
Swap
Task
Swap
Task
Intercept
Intercept
Quiesce
Quiesce
Offline
Target
Volume
Online
Source
Device
Original volume online
and target offline
START
7
7
7
7
3
3
4
4
5
5
6
6
3
3
Target now online and
original volume offline
Offline
Source
Volume
Online
Target
Device
END

Appendix A. INNOVATION Data Processing Corporation product information
217
How Does FDRPAS Work?
FDRPAS has two distinct type system tasks
Online
Source
Volume
Monitor
Task
Monitor
Task
Swap
Task
Swap
Task
Swap
Task
Swap
Task
Swap task
…running on system with
most update activity or only system
Monitor task
…Run on all systems
Sample Monitor Job
//MONITOR EXEC PGM=FDRPAS,REGION=0M
//STEPLIB DD DISP=SHR,DSN=FDRPAS.loadlib
//SYSPRINT DD SYSOUT=*
//SYSUDUMP DD SYSOUT=*
//SYSIN DD *
MONITOR TYPE=SWAP,DURATION=120
MOUNT SWAPUNIT=(21C*,…,…,31B*)

218
IBM TotalStorage Migration Guide for the SAP User
FDR303 CARD IMAGE -- MONITOR TYPE=SWAP,DURATION=120
FDR303 CARD IMAGE -- MOUNT SWAPUNIT=21C*
FDR235 FDRPAS IS MONITORING THE FOLLOWING 16 UNITS:
FDR235 21C0 21C1 21C2 21C3 21C4 21C5 21C6 21C7 21C8 21C9 21CA 21CB 21CC 21CD 21CE 21CF
Monitor Main Task
Control
Statements
16 offline
target devices
Sample Monitor Output
2
2
How Does FDRPAS Work?
Start a Swap task… Monitor task joins in swap
Online
Source
Volume
Monitor
Task
Monitor
Task
Swap
Task
Swap
Task
Offline
Target
Device

Appendix A. INNOVATION Data Processing Corporation product information
219
Figure 4-57
Starting an FDRPAS™ SWAP

FDRPAS™ tasks can be run as batch jobs:
–//SWAP EXEC PGM=FDRPAS,REGION=0M
//STEPLIB DD DISP=SHR,DSN=FDRPAS.loadlib
//SYSPRINT DD SYSOUT=*
//SYSUDUMP DD SYSOUT=*
//SYSIN DD *
SWAP TYPE=FULL,MAXTASKS=3
MOUNT VOL=TSO010,SWAPUNIT=226F
MOUNT VOL=TSO014,SWAPUNIT=226E
MOUNT VOL=TSO017,SWAPUNIT=226B
–//SWAP EXEC PASPROC
//SYSIN DD *
SWAP TYPE=FULL
MOUNT VOL=DB001A,SWAPUNIT=074E
Installation

Easy-to-install and implement
– IEBCOPY load of libraries
– No SMP/E
– No TCP/IP or VTAM required
– No coupling facility required
– No command dataset
– No separate disk communication files
– No IPL required to activate

220
IBM TotalStorage Migration Guide for the SAP User
Figure 4-58
A
dditional SWAP Options

#SYSTEMS=nnn

LARGERSIZE=OK

CHECKTARGET=YES

PACEDELAY=nnnnn

PACING=DYNAMIC

SWAPIOERR=RETRY

EXCLUDE CPUID=cpuid

MIN#SYSTEMS=nnn
F D R P A S ™ C o n t r o l s

N o t h i n g t o s e t u p a h e a d o f t i m e
– N o p a r a m e t e r f il e s
– N o c o m m u n i c a t i o n f i l e s

S i m p l e F D R - s t y l e c o n t r o l s t a t e m e n t s
– S W A P T Y P E = F U L L
M O U N T V O L = A B C 1 2 3,S W A P U N I T = 0 3 A 2
– M O N I T O R T Y P E = S W A P
M O U N T S W A P U N I T = 0 3 *

Appendix A. INNOVATION Data Processing Corporation product information
221
Figure 4-59
Sample Swap Output
FDR303 CARD IMAGE -- SWAP TYPE=FULL
FDR303 CARD IMAGE -- MOUNT VOL=TSO010,SWAPUNIT=226F
FDR233 SYSA (SERIAL# 0109419672) ACKNOWLEDGES THE SWAP OF VOL=TSO010
FDR233 SYSC (SERIAL# 0132429672) ACKNOWLEDGES THE SWAP OF VOL=TSO010 AND HAS JOINED
FDR233 SYSJ (SERIAL# 0209419672) ACKNOWLEDGES THE SWAP OF VOL=TSO010 AND HAS JOINED
FDR233 SYSZ (SERIAL# 0054502064) ACKNOWLEDGES THE SWAP OF VOL=TSO010 AND HAS JOINED
FDR233 SYSE (SERIAL# 0309419672) ACKNOWLEDGES THE SWAP OF VOL=TSO010 AND HAS JOINED
FDR233 SYSD (SERIAL# 0146279672) ACKNOWLEDGES THE SWAP OF VOL=TSO010 AMD HAS JOINED
FDR233 SYSH (SERIAL# 0270039672) ACKNOWLEDGES THE SWAP OF VOL=TSO010 AND HAS JOINED
FDR233 SYSI (SERIAL# 0032429672) ACKNOWLEDGES THE SWAP OF VOL=TSO010 AND HAS JOINED
FDR233 SYSB (SERIAL# 0145399672) ACKNOWLEDGES THE SWAP OF VOL=TSO010 AND HAS JOINED
FDR233 SYSK (SERIAL# 0432429672) ACKNOWLEDGES THE SWAP OF VOL=TSO010 AND HAS JOINED
FDRW66 SWAP OF VOL=TSO010 TO UNIT=226F STARTED ON 10 SYSTEMS
(SYSA SYSC SYSJ SYSZ SYSE SYSD SYSH SYSI SYSB SYSK)
Control
Statements
This is the
system the
swap task is
running on
Monitor tasks
on 9 other
systems join
in the swap
The expected
10 systems
have joined
so the swap
begins
1
1
How Does FDRPAS Work?
Start a Monitor task on every sharing system
Offline
Target
Device
Online
Source
Volume
Monitor
Task
Monitor
Task

222
IBM TotalStorage Migration Guide for the SAP User
FDR303 CARD IMAGE -- MONITOR TYPE=SWAP,DURATION=120
FDR303 CARD IMAGE -- MOUNT SWAPUNIT=21C*
FDR235 FDRPAS IS MONITORING THE FOLLOWING 16 UNITS:
FDR235 21C0 21C1 21C2 21C3 21C4 21C5 21C6 21C7 21C8 21C9 21CA 21CB 21CC 21CD 21CE 21CF
FDR238 FDRPAS STARTED MONITOR JOIN TASK FOR UNIT=21C2
Monitor Main Task
Swap detected,
subtask started
Sample Monitor Output
FDR303 CARD IMAGE -- MONITOR TYPE=JOIN,NOS/MOUNT SU=21C1
FDR303 CARD IMAGE -- MOUNT SU=21C1
FDR235 FDRPAS IS MONITORING THE FOLLOWING 1 UNITS:
FDR235 21C1
Monitor Subtask
Internally
created Control
Statements
Single target
for swap
Sample Monitor Output

Appendix A. INNOVATION Data Processing Corporation product information
223
How Does FDRPAS Work?
Intercepts installed:
Offline
Target
Device
Online
Source
Volume
Monitor
Task
Monitor
Task
Swap
Task
Swap
Task
Intercept
Intercept
3
3
3
3
FDR236 SYSA ACTIVATED I/O INTERCEPTS ON UNIT=2121
FDR007 STARTING TIME OF FULL VOL SWAP - 10.17.59 -UNIT=3390,IN=D#TSO010,OUTPUT=TAPE1
I/O Intercept
installed on this
system
Swap process
has begin
Sample Swap Output

224
IBM TotalStorage Migration Guide for the SAP User
How Does FDRPAS Work?
Tracks copied:
Offline
Target
Device
Online
Source
Volume
Monitor
Task
Monitor
Task
Swap
Task
Swap
Task
Swap
Task
Swap
Task
Intercept
Intercept
4
4
FDR236 SYSA ACTIVATED I/O INTERCEPTS ON UNIT=2121
FDR007 STARTING TIME OF FULL VOL SWAP - 10.17.59 -UNIT=3390,IN=D#TSO010,OUTPUT=TAPE1
FDR158 DATA SET IS ACTIVE DSN=GEORGE.ISPF.PROFILE
FDR158 DATA SET IS ACTIVE DSN=GEORGE.ISPF.CLIST
FDR158 DATA SET IS ACTIVE DSN=WALTER.TEST.PGMLIB
FDR158 DATA SET IS ACTIVE DSN=PHILLIP.JCL.CNTL
FDR158 DATA SET IS ACTIVE DSN=WALTER.INPUT.PDSE
These datasets
are active at
beginning of
swap
Sample Swap Output

Appendix A. INNOVATION Data Processing Corporation product information
225
How Does FDRPAS Work?
update list collected from each monitor task:
Offline
Target
Device
Online
Source
Volume
Monitor
Task
Monitor
Task
Swap
Task
Swap
Task
Swap
Task
Swap
Task
Intercept
Intercept
5
5
How Does FDRPAS Work?
Updated tracks copied,
Offline
Target
Device
Online
Source
Volume
Monitor
Task
Monitor
Task
Swap
Task
Swap
Task
Swap
Task
Swap
Task
Intercept
Intercept
6
6

226
IBM TotalStorage Migration Guide for the SAP User
FDR239 182 TRACKS UPDATED BY SYSA
FDR239 8 TRACKS UPDATED BY SYSJ
FDR239 7 TRACKS UPDATED BY SYSH
FDR239 191 TOTAL UNIQUE TRACKS UPDATED IN PASS 1 - RE-COPYING UPDATED TRACKS
Updated tracks
above threshold
in pass 1
Sample Swap Output
How Does FDRPAS Work?
I/O quiesced:
Offline
Target
Device
Online
Source
Volume
Monitor
Task
Monitor
Task
Swap
Task
Swap
Task
Intercept
Intercept
7
7
7
7
Quiesce
Quiesce

Appendix A. INNOVATION Data Processing Corporation product information
227
FDR239 182 TRACKS UPDATED BY SYSA
FDR239 8 TRACKS UPDATED BY SYSJ
FDR239 7 TRACKS UPDATED BY SYSH
FDR239 191 TOTAL UNIQUE TRACKS UPDATED IN PASS 1 - RE-COPYING UPDATED TRACKS
FDR239 22 TRACKS UPDATED BY SYSA
FDR239 22 TOTAL UNIQUE TRACKS UPDATED IN PASS 2 - RE-COPYING UPDATED TRACKS
FDR236 SYSA DE-ACTIVATED I/O INTERCEPTS ON UNIT=2121 204 TRACKS UPDATED
Updated tracks
above threshold
in pass 1
Updated tracks
below threshold in
pass 2
Final tracks
copied and I/O
intercept
de-activated
Sample Swap Output
How Does FDRPAS Work?
Volume Swapped on every system:
Offline
Target
Device
Online
Source
Volume
Monitor
Task
Monitor
Task
Swap
Task
Swap
Task
Quiesce
8
8
8
8
Quiesce
Online
Target
Device
Offline
Source
Volume

228
IBM TotalStorage Migration Guide for the SAP User
FDR241 FDRPAS SUCCESSFULLY COMPLETED SWAP OF VOL=TSO010 TO UNIT=226F ON SYSA
FDR241 FDRPAS SUCCESSFULLY COMPLETED SWAP OF VOL=TSO010 TO UNIT=226F ON SYSC
FDR241 FDRPAS SUCCESSFULLY COMPLETED SWAP OF VOL=TSO010 TO UNIT=226F ON SYSJ
FDR241 FDRPAS SUCCESSFULLY COMPLETED SWAP OF VOL=TSO010 TO UNIT=226F ON SYSZ
FDR241 FDRPAS SUCCESSFULLY COMPLETED SWAP OF VOL=TSO010 TO UNIT=226F ON SYSE
FDR241 FDRPAS SUCCESSFULLY COMPLETED SWAP OF VOL=TSO010 TO UNIT=226F ON SYSD
FDR241 FDRPAS SUCCESSFULLY COMPLETED SWAP OF VOL=TSO010 TO UNIT=226F ON SYSH
FDR241 FDRPAS SUCCESSFULLY COMPLETED SWAP OF VOL=TSO010 TO UNIT=226F ON SYSI
FDR241 FDRPAS SUCCESSFULLY COMPLETED SWAP OF VOL=TSO010 TO UNIT=226F ON SYSB
FDR241 FDRPAS SUCCESSFULLY COMPLETED SWAP OF VOL=TSO010 TO UNIT=226F ON SYSK
Each system calls
swap service
Sample Swap Output
FDR303 CARD IMAGE -- MONITOR TYPE=SWAP,DURATION=120
FDR303 CARD IMAGE -- MOUNT SWAPUNIT=21C*
FDR235 FDRPAS IS MONITORING THE FOLLOWING 16 UNITS:
FDR235 21C0 21C1 21C2 21C3 21C4 21C5 21C6 21C7 21C8 21C9 21CA 21CB 21CC 21CD 21CE 21CF
FDR238 FDRPAS STARTED MONITOR JOIN TASK FOR UNIT=21C2
FDR241 FDRPAS SUCCESSFULLY COMPLETED SWAP OF VOL=RVA7D2 TO UNIT=21C2 ON CPUB
FDR238 FDRPAS STARTED MONITOR JOIN TASK FOR UNIT=21C1
FDR241 FDRPAS SUCCESSFULLY COMPLETED SWAP OF VOL=RVA7D1 TO UNIT=21C1 ON CPUB
FDR238 FDRPAS STARTED MONITOR JOIN TASK FOR UNIT=21C7
FDR241 FDRPAS SUCCESSFULLY COMPLETED SWAP OF VOL=RVA7D7 TO UNIT=21C7 ON CPUB
FDR999 FDR SUCCESSFULLY COMPLETED
Monitor Main Task
Swap completed,
subtask ended
Sample Monitor Output

Appendix A. INNOVATION Data Processing Corporation product information
229
How Does FDRPAS Work?
Volume is moved,
and all I/O now goes to the new device.
Original volume now offline
and
Target online
Offline
Source
Volume
Online
Target
Device
END
FDR122 OPERATION STATISTICS FOR 3390 VOLUME..................TSO010
FDR122 CYLINDERS ON VOLUME..............3,339
FDR122 DATASETS PROCESSED.................274
FDR122 BYTES READ FROM DASD.....1,467,108,993
FDR122 DASD TRACKS SWAPPED.............29,098
FDR122 UPDATED TRACKS RECOPIED............213
FDR122 DASD EXCPS.......................1,991
FDR122 TARGET DASD EXCPS................2,080
FDR122 CPU TIME (SECONDS)...............2.297
FDR122 ELAPSED TIME (MINUTES).............7.2
FDR122 SWAP TIME . .......................6.5
FDR999 FDR SUCCESSFULLY COMPLETED
FDR007 ENDING TIME OF FULL VOL SWAP - 10.24.24 - UNIT=3390,IN=D#TSO010,OUTPUT=TAPE1
The swap
is complete
Sample Swap Output

230
IBM TotalStorage Migration Guide for the SAP User
FDRPAS™ Security

Controlled by FACILITY class profiles
– FDRPAS.SWAP
– FDRPAS.SWAPDUMP
– FDRPAS.SWAPBUILDIX

If not defined, all operations are allowed.

Optionally checks for DASDVOL authority to the volume
Other ways of Starting FDRPAS™

As started tasks:
– S PASPROC.MON2,PARM='MONITOR TYPE=SWAP/MOUNT SU=226*‘

In a Sysplex, monitor tasks can be started globally
on all systems:
– ROUTE T=0,*OTHER,S PASPROC.MON2,PARM='MONITOR TYPE=SWAP/MOUNT SU=2*‘

FDRPAS™ ISPF panels

Appendix A. INNOVATION Data Processing Corporation product information
231
ISPF Primary Option Menu
Option ===>
A
0 Settings Terminal and user parameters User ID . : DF
1 View Display source data or listings Time. . . : 09:05
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 . : ISR
6 Command Enter TSO or Workstation commands TSO logon : V48ISPF
7 Dialog Test Perform dialog testing TSO prefix: DF
8 LM Facility Library administrator functions System ID : OS24
9 IBM Products IBM program development products MVS acct. : **NONE**
10 SCLM SW Configuration Library Manager Release . : ISPF 4.8
11 Workplace ISPF Object/Action Workplace
A FDR/ABR FDR/ABR DASD Management Functions
ISPF Main Menu
---------- FDR TOTAL DASD MANAGEMENT SYSTEM -- FDR PRIMARY OPTIONS MENU ------
OPTION ===>
P
1 REPORTS - ABR REPORTING FUNCTIONS
2 RESTORE - ABR DATA SET RESTORE
3 ARCHIVE - ABR DATA SET ARCHIVE OR SUPERSCRATCH
4 BACKUP - ABR DATA SET BACKUP
5 REMOTE Q - ABR REMOTE QUEUE UTILITY FUNCTIONS
C COMPAKTOR - COMPAKTOR MAP AND SIMULATION REPORTS
R RELEASE - COMPAKTOR RELEASE
I INSTALL - INSTALLATION AND MAINTENANCE OF FDR AND OPTIONAL PRODUCTS
J JCL PARMS - SPECIFY FDR JCL AND SYSOUT DEFAULTS FOR SUBMITTED JOBS
K FORMAT - MODIFY FORMAT OF GENERATED REPORTS
M MESSAGES - FDR MESSAGES AND CODES QUERY FACILITY
P PLUG & SWAP - FDRPAS™ PLUG & SWAP
Q QUERY - FDR/ABR STATISTICS QUERY
S SRS - SEARCH, REPORT, SERVICES DIALOG
T FDRTSEL - BACKUP FILE MANAGEMENT UTILITY
ISPF ABR Main Menu

232
IBM TotalStorage Migration Guide for the SAP User
Simple ISPF panel for all operations:
---------------------------- FDRPAS™ Plug & Swap ------------- Row 1 to 1 of 1
COMMAND ===> SCROLL ===> PAGE
Active COnfirm SWap DUmp Msg SUspend REsume ABort OPtions HIstory SOrt
Panel: 1 of 9
Command Volume Unit Swap to REfresh 0
Serial Addr Offline
Mask Mask Unit Status
------- ------ ---- ------- ----------------------------------------------
'''
Pass: % Tracks to copy: Copied: Updated:
Source - Reserve: Level: Pace: Type: Cyls:
Target - Reserve: Level: Pace: Type: Cyls:
Storgrp: SSID: CU Serial#:
ISPF Interface
Options panel sets options for swaps:
---------------------------- FDRPAS™ Plug & Swap - Options --------------------
COMMAND ===> SCROLL ===> PAGE
Options for SWAP command: Options for SWAPDUMP command:
#SYSTEMS ===> #SYSTEMS ===>
MIN#SYSTEMS ===> MIN#SYSTEMS ===>
CONFIRMSWAP ===> NO (yes no) CONFIRMSPLIT ===> NO (yes no)
CONFMESS ===> NO (yes no) CONFMESS ===> NO (yes no)
LOGMESS ===> YES (yes no) LOGMESS ===> NO (yes no)
PACEDELAY ===> 0 1/100 seconds PACEDELAY ===> 0 1/100 seconds
SWAPDELAY ===> 15 seconds SWAPDELAY ===> 15 seconds
CHECKTARGET ===> NO CHECKTARGET ===> NO
LARGERSIZE ===> NO (ok no) LARGERSIZE ===> NO (ok no)
PACING ===> STATIC (dynam static) PACING ===> STATIC (dynam static)
Other options:
Max Start ===> 10 maximum number of simultaneous Swap/Swapdump to start
Maxvols ===> 1 maximum number of volumes to generate per process
Maxtasks ===> 1 maximum number of concurrent volumes to process
Interval ===> 2 refresh interval in seconds
Saved in ISPF profile
ISPF Interface

Appendix A. INNOVATION Data Processing Corporation product information
233
Press ENTER to display active swaps:
Command Volume Unit Swap to REfresh 0
Serial Addr Offline
Mask Mask Unit Status
------- ------ ---- ------- ----------------------------------------------
TSO001 07C1 2C31 ACTIVE
Pass: 1 75 % Tracks to copy: 9554 Copied: 7245 Updated: 1091
Source - Reserve: 0 Level: 1 Pace: 0 Type: 3390-3 Cyls: 3339
Target - Reserve: 1 Level: 1 Pace: 0 Type: 3390-3 Cyls: 3339
Storgrp: SSID: 0310 CU Serial#: 14710
SMS123 07C2 2C32 ACTIVE
Pass: 3 95 % Tracks to copy: 1000 Copied: 950 Updated: 0
Source - Reserve: 0 Level: 1 Pace: 0 Type: 3390-3 Cyls: 3339
Target - Reserve: 1 Level: 1 Pace: 5: Type: 3390-3 Cyls: 3339
Storgrp: SGWORK SSID: 0311 CU Serial#: 14710
ISPF Interface
Alternate display formats, more volumes per screen:
Command Volume Unit Swap to Refresh 0
Serial Addr Offline
Mask Mask Unit Status
------- ------ ---- ------- ----------------------------------------------
TSO001 07C1 2C31 ACTIVE
Pass: 1 75 % Tracks to copy: 9554 Copied: 7245 Updated: 1091
TSO002 07C2 2C32 ACTIVE
Pass: 1 25 % Tracks to copy: 4976 Copied: 1244 Updated: 10
TSO003 07C3 2C33 SYNCHRONIZING
Pass: 1 0 % Tracks to copy: 5512 Copied: 0 Updated: 0
TSO003 07C4 2C34 SWAPPED
Pass: 0 % Tracks to copy: Copied: Updated:
TSO003 07C5 2C35 SYNCHRONIZING
Pass: 0 % Tracks to copy: 7729 Copied: 0 Updated: 0
ISPF Interface

234
IBM TotalStorage Migration Guide for the SAP User
Alternate display formats, more volumes per screen:
Command Volume Unit Swap Refresh 0
Serial Addr to Comp
Mask Mask Unit % Status
------- ------ ---- ---- ---- ----------------------------------------------
TSO001 07C1 2C31 75 ACTIVE
TSO002 07C2 2C32 25 ACTIVE
TSO003 07C3 2C33 0 SYNCHRONIZING
TSO004 07C4 2C34 SWAPPED
TSO005 07C5 2C35 47 ACTIVE
TSO006 07C6 2C36 0 SYNCHRONIZING
TSO007 07C7 2C37 SWAPPED
TSO008 07C8 2C38 17 ACTIVE
TSO009 07C9 2C39 0 SYNCHRONIZING
ISPF Interface
Select volumes by various criteria:
---------------------------- FDRPAS™ Plug & Swap ------------- Row 1 to 1 of 1
COMMAND ===> SCROLL ===> PAGE
Active COnfirm SWap DUmp Msg SUspend REsume ABort OPtions HIstory SOrt
Panel: 1 of 9
Command Volume Unit Swap to REfresh 0
Serial Addr Offline
Mask Mask Unit Status
------- ------ ---- ------- ----------------------------------------------
'''
Pass: % Tracks to copy: Copied: Updated:
Source - Reserve: Level: Pace: Type: Cyls:
Target - Reserve: Level: Pace: Type: Cyls:
Storgrp: SSID: CU Serial#:
TSO* 3A2*
SGPROD1 0300 013507SGPROD1 0300 013507
ISPF Interface

Appendix A. INNOVATION Data Processing Corporation product information
235
Initiate swaps:
---------------------------- FDRPAS™ Plug & Swap ------------- Row 1 to 1 of
1
COMMAND ===>
SWAP
SCROLL ===> PAGE
Active COnfirm SWap DUmp Msg SUspend REsume ABort OPtions HIstory SOrt
Command Volume Unit Swap to REfresh 0
Serial Addr Offline
Mask Mask Unit Status
------- ------ ---- ------- ----------------------------------------------
SH20C0 20C0 21c0 INACTIVE
Pass: % Tracks to copy: Copied: Updated:
SH20C2 20C2 21c2 INACTIVE
Pass: % Tracks to copy: Copied: Updated:
SH20C5 20C5 21c5 INACTIVE
Pass: % Tracks to copy: Copied: Updated:
Starts FDRPAS™ swap task
as a started task using PASPROC
ISPF Interface
Terminate active swaps:
Command Volume Unit Swap to REfresh 0
Serial Addr Offline
Mask Mask Unit Status
------- ------ ---- ------- ----------------------------------------------
abort
SH20CC 20CC 21CC ACTIVE (MAIN)
Pass: 1 89 % Tracks to copy: 1246 Copied: 1110 Updated: 2504
Source - Reserve: 0 Level: 1 Pace: 0 Type: 3390-9 Cyls: 10017
Target - Reserve: 1 Level: 1 Pace: 0 Type: 3390-9 Cyls: 10017
Storgrp: SSID: 0310 CU Serial#: 14710
Terminates swap with no harm
to the volume or jobs using it.
ISPF Interface

236
IBM TotalStorage Migration Guide for the SAP User
Suspends all copy I/O but
continues to monitor updates
Suspend and resume swaps:
Command Volume Unit Swap to REfresh 0
Serial Addr Offline
Mask Mask Unit Status
------- ------ ---- ------- ----------------------------------------------
suspend
SH20CC 20CC 21CC ACTIVE (MAIN)
Pass: 1 89 % Tracks to copy: 1246 Copied: 1110 Updated: 2504
Source - Reserve: 0 Level: 1 Pace: 0 Type: 3390-9 Cyls: 10017
Target - Reserve: 1 Level: 1 Pace: 0 Type: 3390-9 Cyls: 10017
Storgrp: SSID: 0310 CU Serial#: 14710
ISPF Interface
Display swap history:
Command Volume Unit Swapped System Date Time
Serial Addr to Unit
------- ------ ---- ------- -------- ---------- --------
SH20CC 20CC 21CC CPUA 02/05/2002 10:18:43
SH20CC 20CC 21CC CPUB 02/05/2002 10:18:44
SH20CC 20CC 21CC CPUC 02/05/2002 10:18:42
SH20CE 20CE 21CE CPUA 02/05/2002 10:22:12
SH20CE 20CE 21CE CPUB 02/05/2002 10:22:14
SH20CE 20CE 21CE CPUC 02/05/2002 10:22:13
ISPF Interface

Appendix A. INNOVATION Data Processing Corporation product information
237
E-MAIL Notification

Sample FDREMAIL input:
– MAILSERVER MAIL.MYCOMPANY.COM
FROM:<DILBERT@MYCOMPANY.COM>
TO:<DOGBERT@MYCOMPANY.COM>
TO:<CATBERT@MYCOMPANY.COM>
CC:<DILBERT@MYCOMPANY.COM>
SUBJECT: SWAP FAILED ON &SYSTEMS &EMAILDATE &EMAILTIME &MSG
&ALLMSG
EMAIL:
TO:Pager<9735551212@vtext.com>
SUBJECT: FDRPAS™ ERROR &MSG

Sends e-mail to multiple recipients with all FDRPAS™ messages
from failure

Sends brief e-mail to pager
The “PAS Method” (Pre-Engagement )
FDRPAS Data Migration process… questionnaire/checklist
• Prepare a “from/to” list or spreadsheet that shows the volser of the
current (source) volume and the device address of the new (target)
volume
• “familiarize” yourself with the your hardware and software
• Current DASD hardware and DASD type moving to.
• Hardware features i.e. PAV, copy facilities (Timefinder™, FlashCopy™)
PPRC, SRDF and so on.
• Operating system(s) releases and maintenance level(s)
• Other Hardware and Software Special Considerations
Old UCB
VOL New UC
B
1000 D0008E 3000
1001 D00063 3001
1002 D00011 3002

238
IBM TotalStorage Migration Guide for the SAP User
“PAS Method” ( On site set up)

I nstall FDRPAS software (APF authorize)

Run FDRPAS™ License Program Report

Run FDRPAS™ HealthCheck jobs (Fast CPK and FDREPORT)

Run FDRPAS™ HealthCheck jobs (Fast CPK and FDREPORT)

Create FDRPAS™SWAP and MOUNT control statements

Use SI MSWAP to validate your FDRPAS j ob control statements.
CONTROLLER SSID STORGRP UNIT VOLSER % ALLOC # CYLS SIZE(GB) TOTAL(GB) # 3390-3
---------- ---- -------- ---- ------ ------- ------ -------- ----------- --------
*UNKNOWN* 18 28.647 10
EMC 01336 56 98.371 34
IBM 14710 12 225.302 79
IBM 22935 18 194.225 68
IBM 70941 44 117.933 41
IBM 91468 24 15.136 5
EMC 00938 10 48.458 17
EMC 01889 4 200.515 70
HTC 45278 19 63.764 22
IBM 00000 84 5.676 2
CPU SERIAL SYSTEM NAME % ALLOC TOTAL(GB) # 3390-3
---------- ----------- ------- ----------- --------
0209417060 CPUB 21 998.032 351
FDRPAS™ License Output

Appendix A. INNOVATION Data Processing Corporation product information
239
“PAS Method” ( On site Execution)

NOW you are ready for some fun!

Start a MONI TOR job on each system/LPAR that “knows” about
the “old” DASD

Begin submitting SWAP jobs (HI NT: Start out “low and slow”)

Submit one job with 1 volume
- Watch channel activity if possible during the migration to get
a feel for how this system will respond
– Gradually build up
- Balance MAXTASKS with number of concurrent SWAP j obs
- Generally only submit SWAP jobs from one system
FDRPAS™ Performance

FDRPAS™
– SWAP Task uses very little CPU time - MONITOR even less

Many factors impact elapse time
– ** Number of volumes concurrently swapping **
– Number of volumes, % full, disk speed, channel type
– Other system and application I/O, set up and checking
Raw speed…
.concurrently swap 30 3390-9 vols in 15min (17.4GB/m)
.concurrently swap 30 3390-3 vols in 8min (10.9GB/m)
Project elapse time…
. 2.3TB /18 hours
. 1.5TB /11.5 hours

240
IBM TotalStorage Migration Guide for the SAP User
When to use FDRPAS™ ?
SUMMARY…

ESCON / FICON Mix

Distribute Volumes Across Multiple DMX Systems

LSS Configuration PAV Constraints

Incompatible FICON / FICON mix

No Toleration for Any Interruption

Time Constraint to Start

Quick In and Out

Appendix A. INNOVATION Data Processing Corporation product information
241
FDR/UPSTREAM Overview
What follows is a set of charts that provide an overview of FDR/UPSTREAM a product of
INNOVATION Data Processing Corporation.
UPSTREAM Reservoir
The Next Generation of FDR/UPSTREAM Reservoir:
A Non-Mainframe Backup Solution
Simplified Backup Storage Management with UPSTREAM Reservoir
SHARE New York City August 16, 2004 Session 1916
Patrick Fitzsimmons and Bob Perper INNOVATION Data Processing

242
IBM TotalStorage Migration Guide for the SAP User
The Next Generation FDR/UPSTREAM…
UPSTREAM Reservoir
UPSTREAM Reservoir extends the power of UPSTREAM
so it can be used either in mixed MVS and Windows and
UNIX environments or entirely non-mainframe
environments.
UPSTREAM Reservoir Backup server is a Windows
System–Windows 2000/XP or Server 2003
Fairly powerful machine with minimum 512MB
Tape - requires Windows Removable Storage Manager
(RSM) capable tape drives or tape library
TCPIP
Client
UPSTREAM Reservoir Heritage
FDR/UPSTREAM MVS
A z/OS or OS/390 (MVS)
server-based solution,
for enterprise-wide high
speed backup/restore,
hierarchical storage
management and disaster
recovery of distributed NAS
and SAN attached storage on
open system platforms.
UPSTREAM
MVS

Appendix A. INNOVATION Data Processing Corporation product information
243
Windows Reservoir

244
IBM TotalStorage Migration Guide for the SAP User
UPSTREAM Reservoir
Can use any tape drive or library that supports
Windows Removable Storage Manager (RSM) such
as 3490, 9840, Magstar, DLT, LTO, AIT, Ultrium,
Jaguar 3592
UPSTREAM provides the ability to backup to disk
or tape
Option for backup to disk, staged from disk and
automatically moved
Can provide tape stacking for efficient tape usage
Windows Removable Storage Manager
An API that is called by UPSTREAM to make any
supported media accessible
Microsoft Management Console Providing Interface
capabilities
I.E. Injecting/Ejecting, Mount/Dismount
RSM Database
Almost all libraries and stand-alone tape drives
are supported
Supports sophisticated features such as bar-code
readers

Appendix A. INNOVATION Data Processing Corporation product information
245
UPSTREAM Director
Director-JAVA & browser GUI
UPSTREAM management interface
Ad-hoc backups and restores
Run jobs, reports, examine logs
Dispatcher scheduler
UPSTREAM console
List of all operations in progress
Request for any resources (i.e. tape mounts)
UPSTREAM Director

246
IBM TotalStorage Migration Guide for the SAP User
UPSTREAM Director
Ad-hoc restores
Explorer-like interface
Directory tree of backups and files
within backups
Add or exclude files
All without mounting any tapes
Director Operations
Notifies operators when operations require
user intervention (like mounting a tape)
Allows administrators to view, or terminate
running operations
Multiple operator consoles available even in
the browser interface
You can monitor it from home

Appendix A. INNOVATION Data Processing Corporation product information
247
UPSTREAM Dispatcher
Dispatcher is the UPSTREAM Centralized
scheduling system
Integrated to UPSTREAM client and the Director
Only One communication pipe needed
No security issues as no passwords to
be stored
Event notification facility
Easy to use as it’s wizard driven
You can also use external scheduler ie cron, NT
scheduler, MVS scheduler etc
Dispatcher Setup

248
IBM TotalStorage Migration Guide for the SAP User
Dispatcher Setup
Wizard driven setup
Specify the combination of system and
backup profile
Specify what is to be backed up
Specify when it is to be backed up
and that is pretty much it
Dispatcher Maintenance
Very easy to:
Add new systems and profiles to the schedule
Run a job NOW
Make a change to affect one, several or
ALL systems (reporting, excludes,
compression, etc.).
Suspend a schedule for a day for a company
holiday, scheduled down-time, etc.

Appendix A. INNOVATION Data Processing Corporation product information
249
Dispatcher Maintenance
Very easy to:
Add new systems and profiles to the schedule
Run a job NOW
Make a change to affect one, several or
ALL systems (reporting, excludes,
compression, etc.).
Suspend a schedule for a day for a company
holiday, scheduled down-time, etc.
Monitoring…Event Notification
The ability of sending Email and SNMP Trap notifications that
can contain logs or reports when unforeseen events occur
Highly recommended for Dispatcher controlled backups
STEP 1:Define the notification type (Email or SNMP) and the
administrator’s destination in a notification target file.
STEP 2:Then the types of possible actions or specific events are
specified, such as backup or restore.
STEP 3:Rule for when the notification is to be sent. Example the
success or failure of any UPSTREAM operation and what
type of informational attachments, UPSTREAM log,
report file, or any user defined file

250
IBM TotalStorage Migration Guide for the SAP User
Automatic Upgrades
UPSTREAM maintains
an inventory of
installed versions
When an administrator
wishes to upgrade one
or more machines, use
the Director, select the
machine and request the
automatic update.
Silent Install
Allows UPSTREAM to be pushed out to
remote locations where there is not a user
available to respond to installation prompts
Allows administrator selective feature installs
Includes a silent upgrade script to ease
upgrades

Appendix A. INNOVATION Data Processing Corporation product information
251
Supported Client Environments
Windows NT/2000
Windows XP
Windows Server 2003
Novell NetWare
IBM AIX
Sun Solaris
Sun X86
HP-UX
Tru64 UNIX
Intel Linux
zLinux
OS/390 UNIX
OS/2
Any system that can be
NFS mounted can be
backed up/restored
End-to-End Support…
Online Data Base Agents
Oracle
SQL Server
Domino & Lotus Notes
Sybase Informix UDB
MS Exchange
SQL Back-Track Interface
Novell GroupWise
Database Support

252
IBM TotalStorage Migration Guide for the SAP User
Multiple Interfaces
Director-JAVA & browser GUI
UPSTREAM management interface
UPSTREAM open systems console
End-User Restore
Client Interface (UPSTREAM Local)
Telnet compatible interface for UNIX
Advanced Data Reduction methods
Logical Volume Full Merge Backup
Logical File Granularity
Incremental Backup Processing
Data Compression
Exclude/Include
On-Line Database Agents
Auto Migrate inactive data

Appendix A. INNOVATION Data Processing Corporation product information
253
Forward Merge Backup
Creates a full backup WITHOUT transmitting any
unchanged data
Full Merge is constructed with Incremental Backup +
carry unchanged data from previous full = Full Server
Image on 1 tape or tape set
Helps overcome many limitations such as line speed
and backup window
Supports backups to both disk and Tape
Uses 1 backup profile per server
Forward Merge Backup
Full merge utilizes tapes efficiently as multiple
backups are stored on a single tape.
When the full is performed, if a file was
backed up on an incremental and has not been
changed at the time of the full, UPSTREAM
will not recopy the file, but will note it on the
tape and in the database, saving tape and
improving performance.

254
IBM TotalStorage Migration Guide for the SAP User
Support for your SAN
Client begins the backup or restore
The Reservoir machine mounts the tape
Clients directly access SAN attached tape drives
without going through the network
Supports Fibre or SCSI
All that is required is that the tape drive be visible to
both the Reservoir and client systems.
Will work on ANY vendors SAN

Appendix A. INNOVATION Data Processing Corporation product information
255
Profile Sets
A Profile Set is a single definition for a number of backup profiles.
Profile Sets let you define the policies of a backup profile a single
time, add Backup Profiles to that Profile Set definition and then
store all of the tape backups for that profile set on a single tape.
This simplifies configuration and allows you to be able to take full
advantage of high capacity tape drives.
The backups that comprise a Profile Set follows merge
backup rules.
First incremental backup after a full for any profile in the profile
set begins a new tape.
Full backup immediately after a full backup for any profile in the
profile set will start a new tape. Otherwise all backups are
appended to the existing tape.

256
IBM TotalStorage Migration Guide for the SAP User

Appendix A. INNOVATION Data Processing Corporation product information
257
Tape and Disk Management
Tape and Disk management
policy rules are defined by
parameters when you create
the rules for the Profile.
The destination whether
backups go to disk or tape is
setup when you define the
backup job.
Tape and Disk Management
Retention Type parameter:defines retention schemes for tape
and Disk .
Count of Backups:use to control when backups to be deleted,
when the count exceeds the value you specify here.
Count of Days:use to delete backup when it has exceeded the
number of days specified. Incrementals and fulls are treated as
independent backups
Count of Tape Sets:use to delete all backups contained on the
oldest tape sets which exceed the specified count. If there are
disk backups, only disk backups older than the oldest tape set is
deleted.
Count of Full Backups:use to delete all backups which are
older than the number of fulls in the s
p
ecified count.

258
IBM TotalStorage Migration Guide for the SAP User
Tape Usage & Management
For efficient tape utilization, UPSTREAM can stack
more than one backup on a tape. Specify
NEWTAPEINCR N, UPSTREAM will append
incrementals after the first incremental onto the same
tape set.
Specify NEWTAPEFULL N,UPSTREAM will
append the full onto the same tape set as the
incrementals. The first incremental after a full always
starts a new tape.
Tape Usage & Management
Whenever a backup is expired the index files for the backup
are deleted, if the backup is to disk, then the backup file is
deleted. If the backup is to tape, UPSTREAM will delete the
index files for any expired backups.
Tape is not scratched until all of the backups on the tape have
been deleted.
Can recover any backups which are stored on a non-deleted
tape using the Regen facility.
UPSTREAM will NEVER delete your last full backup–
administrator will have to manually delete that backup if
desired.

Appendix A. INNOVATION Data Processing Corporation product information
259
Copying & Migrating Backups to Tape
UPSTREAM utility to copy or migrate backup to a
new tape. Use this when you want to:
Store multiple backups for multiple profiles on a single
tape or tape set and do not wish to use Profile Sets.
Run a large number of backups at the same time but
you do not have enough tapes. Run backups to disk and
use migration option to migrate them to a single tape
set.
Reclaim unused space…collect the active backups from
multiple tapes that also have deleted backups on tape to
a new tape reclaiming space.
Copying & Migrating Backups to Tape
Make a backup (vault) copy of a disk or tape
backup for off-site storage. Use the copy
option to copy the backups onto a new tape set
and then regen them into UPSTREAM at the
off-site location.
Move data from one media type to another
i.e. to take advantage of a new tape
drive technology.

260
IBM TotalStorage Migration Guide for the SAP User
Vault Tapes
Vaults are another copy of backup made after
backup is completed
Create separate tape pool–Vault
Create separate Vault profile
Vault tape name numeric # 1 thru 999
Able to construct different retention schemes
Able to use different tape media types
View contents of vault tape thru Director interface
Vault tape are automatically cataloged for
quick recovery
Vault Tapes
Vaults are another copy of backup made after
backup is completed
Create separate tape pool–Vault
Create separate Vault profile
Vault tape name numeric # 1 thru 999
Able to construct different retention schemes
Able to use different tape media types
View contents of vault tape thru Director interface
Vault tape are automatically cataloged for
quick recovery

Appendix A. INNOVATION Data Processing Corporation product information
261
Tape Interchange
UPSTREAM’s Reservoir provides flexibility
by allowing logically interchangeable tapes
Tapes generated with FDR/UPSTREAM MVS
can be imported and read on the UPSTREAM
Reservoir and visa versa as long as media is
compatible.
Example 3590 tapes on the UPSTREAM
Reservoir system can be imported and read on
the FDR/UPSTREAM mainframe system.
Reporting
UPSTREAM can generate detailed
information in machine manageable, XML
(Extensible Markup Language) form.
Records can be imported into a standard
database package for additional handling. This
allows the user to define their own reports in
their own format, without having to rely on a
limited fixed internal report generated by
UPSTREAM

262
IBM TotalStorage Migration Guide for the SAP User
Reporting
To allow simplified viewing of UPSTREAM
Reservoir functions, UPSTREAM can generate
overall information in human readable form, one
line per function. The file name is summary.rpt
TASK ADDRESS PROFILE FUNCTION START END ELAPSED File Ct BYTES
---- ------- ------- -------- ----- --- ------- ----- -----
1042 192.168.75.36 BOBMIG1 Backup 12:26:52 12:26:54 00:00:02 78 174,080
1043 192.168.75.36 BOBMIG1 Backup 12:27:28 12:27:29 00:00:01 78 174,080
1055 127.0.0.1 BOBNT1 Backup 13:08:07 13:13:23 00:05:16 329 149,788,672
1056 192.168.150.136 JIM51A Backup 13:21:17 13:23:34 00:02:17 480 13,021,184

Appendix A. INNOVATION Data Processing Corporation product information
263
Disaster Recovery
To the rescue…
Reliable restores for systems, applications and files when
things go wrong and one or more servers are gone...
Disaster Recovery
Reliable restores of data files and
systems databases
NDS
Active Directory
System State
Windows XP Automated System Recovery
UPSTREAM Rescuer

264
IBM TotalStorage Migration Guide for the SAP User
Security
Reservoir provides 3 levels using the Microsoft
security already defined on your system
Checks users/passwords against the security database in
the Reservoir machine
Verifies users have access to a given set of directories
which allows you to tailor specific users to access
specific sets of data
Client implements local security
New Features
VSS support-Support for Microsoft VSS
(Volume Shadowcopy Service)
Backups of volume snapshots allowing backups
of data consistent open files as well as support
of any compatible VSS writer.
Support for ACSLS. This will allow support
for drives and libraries not controlled with
RSM, such as STK Powerhorn

Appendix A. INNOVATION Data Processing Corporation product information
265
Summary: The UPSTREAM Benefits
Straight forward and Simple-to-Use
Enterprise-wide backup solution
Peace of mind
Provides a safety net with fast, reliable
protection for new applications
End-to-End Business Continuance Protection
Non-Disruptive Database Protection
04192-04388

266
IBM TotalStorage Migration Guide for the SAP User
FDRERASE
Here are the overview charts for FDRERASE.
Erasing Data From MVS
OS/390 and z/OS Disks
With
FDRERASE
FDRERASE

Appendix A. INNOVATION Data Processing Corporation product information
267
Requirements that security of certain data be maintained
throughout its life cycle.
• US Government legislationHealth Insurance Portability and
Accountability Act (HIPAA), Sarbanes-Oxley Act, Gramm-
Leach-Bliley Act (GLB), National Security Agency Guidelines.
• National Government regulations such as the European Data
Protection Directive 95/46 or Australian Government IT
Security Manual ACSI 33
• Industry guidelines
• Cor
p
orate
g
overnance rules for the deletion of data.
Erasing Data...
Why is it necessary?
• DASD migration.
• De-commissioning, selling off, or scrapping disks.
• Leaving a DR test site.
• Moving data around (FDRPAS).
• Re-using DASD Volumes in-house.
Erasing Data...
When is it necessary?

268
IBM TotalStorage Migration Guide for the SAP User
Technique
Consequence?
• Deleting datasets Secure enough?
• ICKDSF Init (Minimal) Secure enough?
• ICKDSF Init (Medial) Fast enough?
How is data typically erased ?
Erasing Data...
Deleting datasets –
Is it secure enough? NO!
VTOC
Deleting a dataset only
removes the entry from the
VTOC ….
The actual data remains on disk
and can still be accessed..
Erasing Data...
VTOC Entry
Dataset
VTOC

Appendix A. INNOVATION Data Processing Corporation product information
269
VTOC
Dataset #1
Re-Labelling a disk with
ICKDSF only re-writes the
VTOC
The actual datasets remain on
disk and can still be
accessed..
Dataset #2
Dataset #3
Dataset #4
Erasing Data...
ICKDSF Minimal INIT –
Is it secure enough? NO!
VTOC
ICKDSF Medial INIT –
Is it Fast enough? NO!
An ICKDSF Medial Init of a single
3390-3 takes approximately 5-8 mins.
INIT runs as a serial operation
- so to
“ERASE” multiple volumes, you must
run multiple individual steps/ jobs.
Erasing Data...
ICKDSF
INIT

270
IBM TotalStorage Migration Guide for the SAP User
• A fast, easy-to-use, effective tool to erase data.
• A complimentary cost option to FDRPAS.
• Or available as a separately licensed product.
FDRERASE
The Solution...
FDRERASE
Safe and Secure
Data Erasure
Secure:Complies with the U.S. Government
National Computer Security Center (NCSC)
guidelines for erasing computer disks.
Safe:By default, operates against OFFLINE
disks.

Appendix A. INNOVATION Data Processing Corporation product information
271
FDRERASE
– ‘QUICK’ERASE
– ‘ERASE’
– ‘SECURE’ERASE
Three levels of data erasure;
advancing levels of security
• More secure than ‘QUICK’ERASE
• Re-Writes tracks with binary zeros.
• Meets NCSC standard for “clearing” data.
• Appropriate for less sensitive data when
selling off, de-commissioning, or scrapping
disks and upon leaving an unsecured D/R site.
FDRERASE
‘ERASE’

272
IBM TotalStorage Migration Guide for the SAP User
• The most secure erasure of data
• Re-writes tracks multiple times
varying a
two’s-compliment data pattern.
• Meets NCSC standard for “purging” data.
• Appropriate for your most sensitive
data.
FDRERASE
‘SECURE’ERASE
FDRERASE
Additional Functions
– EMPTYVTOC
Empties the VTOC
– ‘SIM’ERASE
Validates control
statements and lists target volumes.
– LABEL, MOUNT, VARY Utility controls

Appendix A. INNOVATION Data Processing Corporation product information
273
– Simple control statements.
– Initiation via JCL.
– Can erase multiple disks in one step.
– Operates against offline
volumes, by default.
FDRERASE
Simple & Safe
FDRERASE
Examples...
Test/verify FDRERASE statements before execution for real.
(Example #1
Use SIMERASE)
Remove data from a range of disks before they are re-used within
the organisation. (Example #2
Use QUICKERASE)
Securely erase all data from a set of disks before decommissioning
and sending them back to the manufacturer (or after a DR test).
(Example #3
Use SECUREERASE)

274
IBM TotalStorage Migration Guide for the SAP User
Example JCL #1 -
SIMERASE
//FDRERASE EXEC PGM=FDRERASE,REGION=0M
//STEPLIB DD DISP=SHR,DSN=fdrerase.loadlib
//SYSPRINT DD SYSOUT=*
//FDRSUMM DD SYSOUT=*
//SYSIN DD *
SIMERASE TYPE=FULL,ALREADYERASED=PROCESS
MOUNT ERASEUNIT=07C*
This SIMERASE
command validates FDRERASE control statements.
It lists the offline disk volumes (in the range 07C0-07CF) that will be
erased when SIMERASE is changed to an erase command...
Example #1Output -
SIMERASE
FDR001 FDR ERASE VOLUMES - FDRERASE - INNOVATION DATA PROCESSING PAGE 1
FDR303 CARD IMAGE -- SIMERASE TYPE=FULL,ALREADYERASED=PROCESS 00410001
FDR303 CARD IMAGE -- MOUNT ERASEUNIT=07C* 00420000 12.13.55
FDR234 ERASE BYPASSED VOL=RVA7CF - UNIT=07CF REASON=8 - TARGET ERASEUNIT IS NOT OFFLINE OR NOT CONNECTED
FDR234 ERASE BYPASSED VOL=RVA7CE - UNIT=07CE REASON=8 - TARGET ERASEUNIT IS NOT OFFLINE OR NOT CONNECTED
FDR234 ERASE BYPASSED VOL=RVA7CD - UNIT=07CD REASON=8 - TARGET ERASEUNIT IS NOT OFFLINE OR NOT CONNECTED
FDR234 ERASE BYPASSED VOL=RVA7CC - UNIT=07CC REASON=8 - TARGET ERASEUNIT IS NOT OFFLINE OR NOT CONNECTED
FDR234 ERASE BYPASSED VOL=RVA7C9 - UNIT=07C9 REASON=8 - TARGET ERASEUNIT IS NOT OFFLINE OR NOT CONNECTED
FDR234 ERASE BYPASSED VOL=RVA7C8 - UNIT=07C8 REASON=8 - TARGET ERASEUNIT IS NOT OFFLINE OR NOT CONNECTED
FDR234 ERASE BYPASSED VOL=RVA7C7 - UNIT=07C7 REASON=8 - TARGET ERASEUNIT IS NOT OFFLINE OR NOT CONNECTED
FDR234 ERASE BYPASSED VOL=RVA7C6 - UNIT=07C6 REASON=8 - TARGET ERASEUNIT IS NOT OFFLINE OR NOT CONNECTED
FDR234 ERASE BYPASSED VOL=RVA7C3 - UNIT=07C3 REASON=8 - TARGET ERASEUNIT IS NOT OFFLINE OR NOT CONNECTED
FDR234 ERASE BYPASSED VOL=RVA7C2 - UNIT=07C2 REASON=8 - TARGET ERASEUNIT IS NOT OFFLINE OR NOT CONNECTED
FDR234 ERASE BYPASSED VOL=RVA7C0 - UNIT=07C0 REASON=8 - TARGET ERASEUNIT IS NOT OFFLINE OR NOT CONNECTED
FDR235 FDRERASE WILL ERASE THE FOLLOWING 5 UNITS:
FDR235 07C1 07C4 07C5 07CA 07CB
FDR270 DEVICE IS ALREADY ERASED UNIT=07C1 VOL=RVA7C1 OVERRIDDEN
FDR270 DEVICE IS ELIGIBLE FOR ERASE UNIT=07C1 VOL=RVA7C1 VOLID=FDR5
FDR270 DEVICE IS ELIGIBLE FOR ERASE UNIT=07C4 VOL=RVA7C4 VOLID=VOL1
FDR270 DEVICE IS ELIGIBLE FOR ERASE UNIT=07C5 VOL=RVA7C5 VOLID=VOL1
FDR270 DEVICE IS ALREADY ERASED UNIT=07CA VOL=RVA7CA OVERRIDDEN
FDR270 DEVICE IS ELIGIBLE FOR ERASE UNIT=07CA VOL=RVA7CA VOLID=FDR5
FDR255 ERASE TARGET UNIT 07CB CONTAINS DATA SETS FIRST DSN=ICF1.ERASE008.KSDS.DATA
FDR270 DEVICE IS NOT ELIGIBLE FOR ERASE UNIT=07CB VOL=RVA7CB VOLID=VOL1
FDR998&& FDR COMPLETED WITH ERRORS
Here is the output for the previous SIMERASE command.
The FDR234 messages show the volumes that are bypassed
because they are not offline.
The FDR270 messages show volumes already erased and other eligible volumes…
the parm
ALREADYERASED=PROCESS
, allows volumes to be re-erased.
The FDR255 message identifies a volume that is not empty (has files on it) and is bypassed.

Appendix A. INNOVATION Data Processing Corporation product information
275
Example JCL #2 -
QUICKERASE
//FDRERASE EXEC PGM=FDRERASE,REGION=0M
//STEPLIB DD DISP=SHR,DSN=fdrerase.loadlib
//SYSPRINT DD SYSOUT=*
//FDRSUMM DD SYSOUT=*
//SYSIN DD *
QUICKERASE TYPE=FULL
MOUNT ERASEUNIT=(07C*,07D*,07E*,07F*)
QUICKERASE command is used to erase disks in the range of 07C0
through 07FF….
Example JCL #3 -
SECUREERASE
//FDRERASE EXEC PGM=FDRERASE,REGION=0M
//STEPLIB DD DISP=SHR,DSN=fdrerase.loadlib
//SYSPRINT DD SYSOUT=*
//FDRSUMM DD SYSOUT=*
//SYSIN DD *
SECUREERASE TYPE=FULL
MOUNT ERASEUNIT=(07C5,07D2,07D3,07DA,
07E5,07F4)
SECUREERASE will “purge” all data from the specified disks. Each
track on each target disk will be overwritten 3 times
with twos
complement patterns.

276
IBM TotalStorage Migration Guide for the SAP User
“Before & After” -
ERASE
FDR520 PRINTING FROM DEVICE TYPE 3390-3 TO OUTPUT DDNAME TAPE1
FDR521 CYLINDER 00003 TRACK 0000 ------------------ RECORD ZERO 0003000000000008 -- DATA 0003000000000000 ------
FDR522 COUNT FIELD 0003000001002350
000000 4E4EE4E2 C5D9D4D6 C44DD3C6 F5F0F3F5 F65D4040 40404040 40404040 40404040 *..USERMOD.LF50356. *
000020 615C4040 40404040 40404040 40404040 40404040 40404040 40404040 40404040 *.. *

FDR520 PRINTING FROM DEVICE TYPE 3390-3 TO OUTPUT DDNAME TAPE1
FDR521 CYLINDER 00003 TRACK 0000 ------------------ RECORD ZERO 0003000000000008 -- DATA 0000000000000000 ------
FDR522 COUNT FIELD 000300000100DD58
000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 *...............................*
LINE 000020-00DD40 SAME AS ABOVE

FDR520 PRINTING FROM DEVICE TYPE 3390-3 TO OUTPUT DDNAME TAPE1
FDR521 CYLINDER 00003 TRACK 0000 ------------------ RECORD ZERO 0003000000000008 -- DATA 0003000000000000 ------
FDR522 COUNT FIELD 0003000001002350
000000 4E4EE4E2 C5D9D4D6 C44DD3C6 F5F0F3F5 F65D4040 40404040 40404040 40404040 *..USERMOD.LF50356. *
000020 615C4040 40404040 40404040 40404040 40404040 40404040 40404040 40404040 *.. *

FDR520 PRINTING FROM DEVICE TYPE 3390-3 TO OUTPUT DDNAME TAPE1
FDR521 CYLINDER 00003 TRACK 0000 ------------------ RECORD ZERO 0003000000000008 -- DATA 0000000000000000 ------
FDR522 COUNT FIELD 000300000100DD58
000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 *...............................*
LINE 000020-00DD40 SAME AS ABOVE

AFTER DATA SET DELETE and ICKDSF INIT… but
BEFORE FDRERASE
AFTER
AFTER DATA SET DELETE and ICKDSF INIT… but
BEFORE FDRERASE
AFTER
After ERASE, the data is gone: its now all zeros
The printout still shows the data on the track…
After ERASE, the data is gone: its now all zeros
The printout still shows the data on the track…
Output & performance -
ERASE
FDR001 FDR ERASE VOLUMES - FDRERASE - INNOVATION DATA PROCESSING
FDR303 CARD IMAGE -- ERASE TYPE=FULL,ALREADYERASED=PROCESS
FDR303 CARD IMAGE -- MOUNT ERASEUNIT=21C5
FDR235 FDRERASE WILL ERASE THE FOLLOWING 1 UNITS:
FDR235 21C3
FDR270 DEVICE IS ALREADY ERASED UNIT=21C5 VOL= OVERRIDDEN
FDR272 ERASE STARTED PASS 1 PATTERN=00
FDR273 ERASE HARDENED DATA TO UNIT=21C5 IN 2 SECS
FDR272 ERASE ENDED PASS 1
FDR241 FDRERASE SUCCESSFULLY COMPLETED ERASE OF VOL= ON UNIT=21C5
FDR122 OPERATION STATISTICS FOR 3390 VOLUME.................. 21C5
FDR122 CYLINDERS ON VOLUME..............3,339
FDR122 DATASETS PROCESSED...................0
FDR122 BYTES READ FROM DASD.................0
FDR122 DASD TRACKS ERASED..............50,085
FDR122 UPDATED TRACKS RECOPIED..............0
FDR122 DASD EXCPS.......................3,345
FDR122 TARGET DASD EXCPS....................0
FDR122 CPU TIME (SECONDS)...............0.254
FDR122 ELAPSED TIME (MINUTES).............2.8
FDR122 ERASE TIME.........................2.8
FDR999 FDR SUCCESSFULLY COMPLETED
An ERASE function is quite fast. In this example, overwriting each track once with binary
zeros, the ERASE of a 3390-3 volume took just 2.8 minutes..

Appendix A. INNOVATION Data Processing Corporation product information
277
Output & performance -
SECUREERASE
FDR001 FDR ERASE VOLUMES - FDRERASE - INNOVATION DATA PROCESSING
FDR303 CARD IMAGE -- SECUREERASE TYPE=FULL,ALREADYERASED=PROCESS
FDR303 CARD IMAGE -- MOUNT ERASEUNIT=21C3
FDR235 FDRERASE WILL ERASE THE FOLLOWING 1 UNITS: 21C3
FDR270 DEVICE IS ALREADY ERASED UNIT=21C3 VOL=SH21C3 OVERRIDDEN
FDR272 ERASE STARTED PASS 1 PATTERN=FA
FDR273 ERASE HARDENED DATA TO UNIT=21C3 IN 2 SECS
FDR272 ERASE ENDED PASS 1
FDR272 ERASE STARTED PASS 2 PATTERN=05
FDR273 ERASE HARDENED DATA TO UNIT=21C3 IN 2 SECS
FDR272 ERASE ENDED PASS 2
FDR272 ERASE STARTED PASS 3 PATTERN=4B
FDR273 ERASE HARDENED DATA TO UNIT=21C3 IN 2 SECS
FDR272 ERASE ENDED PASS 3
FDR241 FDRERASE SUCCESSFULLY COMPLETED ERASE OF VOL=SH21C3 ON UNIT=21C3
FDR122 OPERATION STATISTICS FOR 3390 VOLUME..................SH21C3
FDR122 CYLINDERS ON VOLUME..............3,339
FDR122 DATASETS PROCESSED...................0
FDR122 BYTES READ FROM DASD.................0
FDR122 DASD TRACKS ERASED.............150,255
FDR122 UPDATED TRACKS RECOPIED..............0
FDR122 DASD EXCPS......................10,027
FDR122 TARGET DASD EXCPS....................0
FDR122 CPU TIME (SECONDS)...............3.820
FDR122 ELAPSED TIME (MINUTES).............7.7
FDR122 ERASE TIME.........................7.7
FDR999 FDR SUCCESSFULLY COMPLETED
SECUREERASE writes a non-zero record, multiple times, to every track. It also waits at the end of each pass to be
sure that the data is hardened (written) to disk before continuing; this may take some seconds. In the following
example, the erasure of a 3390-3 volume took 7.7 minutes, overwriting each track 3 times. The FDR272 messages
show the 3 passes and the data pattern used for each one.
(Because of the 3 passes, the DASD TRACKS ERASED in the FDR122 message is 3 times the number of tracks on the volume)
Because of the 3 passes, the
FDR!22 message for DASD
TRACKS ERASED (150,255) is 3
times the number of tracks on the
volume.
– Initiate FDRERASE tasks via a menu
– Control the function(s)
• Suspend/Resume etc
– View History records
FDRERASE
ISPF Interface

278
IBM TotalStorage Migration Guide for the SAP User
--------------------------------- FDRERASE ------------------- Row 1 to 9 of 9
COMMAND ===> SCROLL ===> PAGE
Active QUick ERase SEcure SIm EMpty ABort SUspend REsume OPtions HIstory SOrt
Refresh 0
Command Unit Volume Elapsed Control SMS
Addr Serial Cmp Pass Time Unit Storage
Mask Mask % # Status MM:SS SSID Serial#Group Cyls
------- ---- ------ --- --- -------------- ------- ---- ------- -------- -----
erase 21CD INACTIVE 0311 3390 1113
quick 21CC INACTIVE 0311 3390 1113
Bring up a list of volumes and then issue the appropriate FDRERASE command...
Command Unit Volume Elapsed Control SMS
Addr Serial Cmp Pass Time Unit Storage
Mask Mask % # Status MM:SS SSID Serial#Group Cyls
------- ---- ------ --- --- -------------- ------- ---- ------- -------- -----
suspend 21C0 12 1/1 ACTIVE QUICK 0:19 0310 10017
resume 21C3 12 1/1 SUSPEND QUICK 0:15 0313 10017
abort 21C5 33 1/1 ACTIVE QUICK 0:19 0315 10017
Once started, you can also suspend, resume and abort FDRERASE tasks...
FDRERASE
ISPF Interface...
Some Words Of Caution...
• FDRERASE is designed to erase data!!!
• It is your
responsibility to ensure that the
right
disks are erased, understand your
requirements and use appropriate procedures.
– Use the checklist in FDRERASE User Guide
– Always use SIMERASE first.
– Be careful when overriding the defaults.
– Watch out for Multi-CPU environments...
– Use your security system (e.g. RACF) to protect
access to the FDRERASE load library.

Appendix A. INNOVATION Data Processing Corporation product information
279
Multi-CPU Environments #1
• By default, FDRERASE operates on disks which
are offline
to the system on which it is executing.
• However, except on most EMC subsystems,
FDRERASE cannot tell if a disk may be online to
other
systems
• On each of those other systems, use the following
console command to see if the volume is online
D U,,,uuuu,nn
Multi-CPU Environments #2
ROUTE *all,d u,,,1c0,2
CPUA RESPONSES ----------------------------
IEE457I 14.02.46 UNIT STATUS
UNIT TYPE STATUS VOLSER VOLSTATE
01C0 3390 OFFLINE /RSDNT
01C1 3390 OFFLINE /RSDNT
CPUB RESPONSES ----------------------------
IEE457I 14.02.46 UNIT STATUS
UNIT TYPE STATUS VOLSER VOLSTATE
01C0 3390 O SYM000 PRIV/RSDNT
01C1 3390 OFFLINE /RSDNT
CPUC RESPONSES ----------------------------
IEE457I 14.02.46 UNIT STATUS
UNIT TYPE STATUS VOLSER VOLSTATE
01C0 3390 O SYM000 PRIV/RSDNT
01C1 3390 OFFLINE /RSDNT
If your systems are in a sysplex, use ROUTE *ALL,DS
D U,,,uuuu,nn to execute across all systems...
This example shows
that 01C1 is offline
to all systems, but
01C0 is online (O)
to 2 out of 3
systems...

280
IBM TotalStorage Migration Guide for the SAP User
State Agency:
Over 1000 (3390-3) volumes images containing financial, medical, motor
vehicle and sensitive personal information.
Disaster Recovery Test in a 48 hour slot at an outside site.
ICKDSF min INIT leaving data on the disks…
Not acceptable to their auditors.
ICKDSF Medial INIT to erase the disks takes 12 hours...
Inefficient process creating an “additional expense”.
Employing QUICKERASE, the agency recovers over 80% of that time,giving
them an additional 10 hours of productive DR testing...
FDR QUICKERASE
- 612 (3390-3) volumes in 83 minutes.
- 332 (3390-3) volumes in 43 minutes.
- 21 (3390-9) volumes in 9 minutes.
FDRERASE
User Experience
• FDRERASE offers several types of data
erasure with various levels of security.
• Choose the level of erase that best suits
your needs.
FDRERASE
Summary

Appendix A. INNOVATION Data Processing Corporation product information
281
Syntax For FDR Products
How to read the syntax diagrams
The following rules apply to the syntax diagrams that are used in this book:
Read the syntax diagrams from left to right, from top to bottom, following the path of the
line. Refer to Figure 4-60 for an explanation about how the beginning, continuation and
end of a statement looks like.
Figure 4-60 Start/Continuation/End of syntax
Required items appear on the horizontal line, that is the main path. Refer to Figure 4-61
for an example.
Figure 4-61 Required item
Optional items appear below the main path. Figure 4-62 shows you how this looks like.
Figure 4-62 Optional statements/operands
If you can choose from two or more items, they appear vertically, in a stack. If you must
choose on of the items, one item of the stack appears on the main path. An example is in
Figure 4-63.
Figure 4-63 Choose from several required items
If choosing one of the items is optional, the entire stack appears as shown below in
Figure 4-64.
Figure 4-64 List of optional items

The

symbol indicates the beginning of a statement.

The

symbol indicates the statement syntax is continued on the next
line

The

symbol indicates that a statement is continued from the previous line.

The

symbol indicates the end of a statement.

required_item

required_item

optional_item


required_item

required_choice2
required_choice1

required_item

optional_item

282
IBM TotalStorage Migration Guide for the SAP User
An arrow returning to the left as shown in Figure 4-65, indicates an item that can be
repeated.
Figure 4-65 Repeatable operand
If the repeat arrow contains a comma as in Figure 4-66, you must separate repeated items
with a comma.
Figure 4-66 Repeatable item separated by comma.

re
q
uired
_
item
re
p
eatable
_
item

required_item

,
repeatable_item

© Copyright IBM Corp. 2005. All rights reserved.
283
Related publications
The publications listed in this section are considered particularly suitable for a more detailed
discussion of the topics covered in this redbook.
IBM Redbooks
For information on ordering these publications, see “How to get IBM Redbooks” on page 284.
Note that some of the documents referenced here may be available in softcopy only.
IBM TotalStorage Expert Hands-On Usage Guide, SG24-6102
IBM TotalStorage Enterprise Storage Server Model 800, SG24-6424
IBM TotalStorage Enterprise Storage Server: Implementing the ESS in Your Environment,
SG24-5420
IBM ESS and IBM DB2 UDB Working Together, SG24-6262
DB2 UDB Backup and Recovery with ESS Copy Services, SG24-6557
IBM TotalStorage SAN Volume Controller and SAN Integration Server , SG24-6423
Volume Migration Using SAN Volume Controller, TIPS0400
IBM Tivoli Storage Manager: LAN/WAN Backup; ServerFree Backup; LANFree Backup;
and Split-Mirror Backup - What Does It All Mean?, TIPS011
DB2 UDB for z/OS V8: Through the Looking Glass and What SAP Found There,
SG24-7088
Other publications
These publications are also relevant as further information sources:
SAN Volume Controller V1.2.0 - Planning Guide, GA22-1052-02
IBM TotalStorage Enterprise Storage Server Introduction and Planning Guide, GA26-7444
IBM TotalStorage Enterprise Storage Server, Command-Line Interfaces User’s Guide
Online resources
These Web sites and URLs are also relevant as further information sources:
IBM publications:
http://www.ibm.com/Support/publications/
Enterprise Storage Systems information:
http://www.ibm.com/servers/storage/disk/ess/ess800
IBM Redbooks:
http://www.Redbooks.ibm.com
Innovation Data Processing:
http://www.innovationdp.fdr.com

284
IBM TotalStorage Migration Guide for the SAP User
How to get IBM Redbooks
You can search for, view, or download Redbooks, Redpapers, Hints and Tips, draft
publications and Additional materials, as well as order hardcopy Redbooks or CD-ROMs, at
this Web site:
ibm.com/redbooks
Help from IBM
IBM Support and downloads
ibm.com/support
IBM Global Services
ibm.com/services

© Copyright IBM Corp. 2005. All rights reserved.
285
Index
Numerics
26388
Head 1
7.4 FDRPAS SWAP via Batch job 35
A
ACTIVE SWAP 63
Advanced Copy Services Functions 6
AIX 169
AIX maintenance levels 98
Alert Monitor 192
alternate path 9
ARCHIVE_STUCK 190
arrays 97
Authorizing FDRPAS 30
Authorizing the FDRPAS ISPF PROGRAM 30
B
back-end 13
Back-end disk support 104
backend storage 10
backup 187
backup status 186
balancing 9
Benefits of the ESS 800 6
block aggregation 11
block level virtualization 14
BM DB2 2
Book 2
boss node 12
C
CHANGEVOL= 72
CHANGEVOLNUM= 72
CKD 65
Client host preparation 98
CLP 169
Cluster 12
cluster 10
Command Line 163
Concurrent Swap 54
CONFERASE= 69
configuration activity 13
configuration commands 13
configuration node 13
configure 10
CONFIRMSPLIT=YES 58
CONFIRMSWAP=YES 58
Control Center 168, 171
CPU serial numbers 34
CPYVOLID= 69
D
Database 165
database 2
DB2 168
DB2 command backup 169
DB2 Control Center 168
DB2 UDB 200
DB2 UDB Command Line 173
DB2 Universal Database 169
DB6EX 198
DBA Planning Calendar 158
DB-configuration 187
Deadlocks 200
destage data 10
Diag Log 200
E
EMC Symmetrix 38
emulates 38
Environment 1 2
Environment 2 2
ERASE 65
ERASEPASS=n 70
ERASEPATTERN= 70
ERASEUNIT= 71
ESS 104
ESS 800 2–3
ESS Copy Services 7
ESS Specialist 8
Extensive Connectivity 6
extent 13
extent size 104
F
fail-over 12
FBA 65
FDRERASE 2, 15, 65, 68
FDRPAS 2, 16
FDRPAS batch job 18
FDRPAS INSTALL ISPF 25
FDRPAS tasks 21
FDRW68 53, 60
Fixed Block Architecture 65
FlashCopy 7
H
hardware elements 9
Hardware support 99
Heterogeneous Support 5
host objects 104

286
IBM TotalStorage Migration Guide for the SAP User
I
I 17
I/O capability 12
I/O Group 9, 12–13
I/O intercepts 19
I/O operation 19
IBM 2105 38
IBM 3990-3/RVA 20
IBM RVA 38
IBM3990-3 38
INACTIVE 61, 64
in-band 9
independent pairs 12
Install FDRPAS 24
internal disk 10
IP addresses 98
isolate 10
ISPF Interfac 58
ISPF panels 73
L
Lock Wait 198
logical disk images 11
logical unit 13
LU 13
LUN 13
M
manage 10
Managed Disk Group 13
managed disk groups 11, 103
Managed Disks 11
managed disks 9
management 12
mapping 12
Master Console 10, 14
MAXTASKS= 70
MDG 11
mdisk 13
Mdisk groups 103
mdisks 11, 13
MDS 10
MDS to SVC Zone 103
memory 187
metadata servers 10
migrate 2
monitor 186
MOUNT statement 71
multi-path support 99
multi-pathing 10
Multiple System 20
multiple systems 20
N
named 104
naming conventions 98
naming standard 104
NCSC 66
NCSC requirements 66
nodes 98
non-preferred 9
O
Objective 2
ONLINE= 70
optimize 103
Options 62, 166
Overview 1, 3
P
path failover 11
Performance 7
performance 186, 194
physical planning 98
physical storage 11
Preferred node 14
preferred path 9
preferred SDD path 14
purging 66
Q
QUICKERASE 65
R
RDAC 10
Redbooks Web site 284
Contact us x
redundancy 104
Redundant Disk Array Controller 10
redundant paths 11
redundant SAN 10
Resiliency Family 4
restore a database 174
restore tablespaces 176
RETRY 54
round-robin manner 11
S
SAN 9
SAN appliance 14
SAN Volume Controller 8
SAN Volume Controller Overview 8
SAP 1
SAP database 2
SAP Migration 2
SAP R/3 158
SAP R/3 DBA 163
SAP R/3 environment 200
SAP R/3 system 192
Scalability 7
scalable 9
scales well 12
SCSI bus 13
SDD 10, 99
SDD installation 99

Index
287
SECUREERASE 65
SELTERR= 71
server engines 14
service 10
Simulate 75
Single System 20
source device 18
SQL 169
Storage Area Network (SAN) 7
Storage virtualization and the SVC 9
StorageTek SVA 38
StorageTek SVA/Eclipse 20
Subsystem Device Driver 10
successful installation 97
Summary 1
Supported 16
SVC cluster members 12
SVC compatibility 10
SVC configuration 12
SVC host zone 102
SVC logical components 11
SVC Master Console zone 100
SVC multi-pathing 10
SVC node 10, 14
SVC physical components 9
SVC storage zone 102
Swap scenario 57
Swap scenario IV 58
Swap tasks 21
SYNCHRONIZING 60
syntax diagram 37
syntax diagram p 38
SYSIN 36
SYSPRINT 36
System Check 192
SYSUDUMP 36
T
tables 181
Tablespaces 184
tablespaces 171
time out 202
TORAGEGROUP 72
Total Cost of Ownership 7
TotalStorage ESS 800 2
Transactions 200
TSM 163–164
TSO EXEC 58
U
UDB 168
Uninterruptible Power Supplies 10
UPS 10
V
VARYON= 71
vdisk 14
vdisks 11, 97, 103
virtual disk 14
virtual disks 9, 11
virtualization 9
virtualization function 12
voltage regulation 10
volume label 19–20
volume management 11
VTOC, VTOC index 19
VTOCCYL= 72
VTOCLOC= 73
VTOCSIZE= 73
VTOCTRK= 73
VVDS 19
W
Wizard 165
workload 103
WWPNs 104
Z
z/OS 2, 65
z/OS system images 20
zoned 11
zoning 10

288
IBM TotalStorage Migration Guide for the SAP User

(0.5” spine)
0.475”<->0.873”
250 <-> 459 pages
IBM TotalStorage Migration Guide for the SAP User IBM TotalStorage Migration Guide for
the SAP User
IBM TotalStorage Migration Guide
for the SAP User
IBM TotalStorage Migration Guide for the SAP User

IBM TotalStorage Migration Guide
for the SAP User
IBM TotalStorage Migration Guide
for the SAP User


®
SG24-6400-00 ISBN 0738491659
INTERNATIONAL
TECHNICAL
SUPPORT
ORGANIZATION
BUILDING TECHNICAL
INFORMATION BASED ON
PRACTICAL EXPERIENCE
IBM Redbooks are developed
by the IBM International
Technical Support
Organization. Experts from
IBM, Customers and Partners
from around the world create
timely technical information
based on realistic scenarios.
Specific recommendations
are provided to help you
implement IT solutions more
effectively in your
environment.
For more information:
ibm.com/redbooks
IBM TotalStorage
Migration Guide for the
SAP UserUser
Learn about SAP
operational tasks after
migrating to IBM ESS
Become familiar with
tools like FDRPAS to
migrate easily
Learn how to use the
IBM SAN Volume
Controller to migrate
This IBM Redbook demonstrates the relative ease with which SAP
users can migrate to the IBM TotalStorage Enterprise Storage
Server (ESS). It describes how this can be accomplished in
numerous ways, across a variety of hardware and software
platforms.
The book offers migration scenarios for customers who want to
migrate from OEM storage subsystems to the ESS, and it covers
a variety of migration tools and methodologies you can use to
perform these migrations.
In some cases, the use of software migration tools (such as
FDRPAS) from third party vendors is described, while in other
cases the use of the IBM SAN Volume Controller (SVC) to migrate
users in an open systems environment is detailed, demonstrating
the added value of IBM's preferred migration tools, including the
SVC and ISV partner tools.
Back cover
First page image
Due to scheduled site updates, ip.com and all related websites (including Prior Art publishing as well as Pantros IP) will be unavailable on Sunday, September 21st, starting at 2:00PM EDT and lasting for up to 4 hours.

We thank you for your patience and apologize in advance for any inconvenience this may cause.
We are pleased to offer a download of this document free of charge.
Files available for download:
  • a representative PDF of the primary file (contains all the relevant information for most users)
To obtain the file, please enter the "captcha" below and click the Download button.
Avoid entering CAPTCHAs! Sign In or Create a Free Account.

Challenge image
  • Please enter letters and numbers only; no spaces.
  • Cannot read this one? Click the image.
  • Difficulty with captchas? Contact us with the URL of this page and we will email it to you.