IBM Spectrum Protect

Migrating IBM Spectrum Protect

Servers from One Operating System
to Another
Document Version: 1.03

Daniel Eisenberg, Larry Schuh, Ken Hannigan

 Copyright IBM Corporation 2014, 2016.
US Government Users Restricted Rights Use, duplication or disclosure restricted by GSA ADP Schedule
Contract with IBM Corp.
Contents

Introduction............................................................................................................................................................ 1
Prerequisites..................................................................................................................................................... 1
Database backup and restore...................................................................................................................... 1
DSMSERV EXTRACTDB and DSMSERV INSERTDB................................................................................ 2
Migration Process.................................................................................................................................................. 3
Planning the migration...................................................................................................................................... 3
Preparing the target system for the migration................................................................................................... 4
Preparing the source system for the migration.................................................................................................. 5
Migrating the server using database backup and restore.................................................................................. 7
Migrating the server using database extract and insert utilities......................................................................... 8
Migrate using the network method............................................................................................................... 8
Migrate using the media method.................................................................................................................. 9
Extracting the data to media................................................................................................................... 9
Loading the extracted data into the target database............................................................................. 10
Taking the first steps after migration................................................................................................................ 11
Appendix A. Utilities for server migration.............................................................................................................. 15
DSMSERV EXTRACTDB (Extract data from a server database).................................................................... 15
Syntax........................................................................................................................................................ 15
Parameters................................................................................................................................................ 15
Examples................................................................................................................................................... 16
The manifest file for the data extraction to media...................................................................................... 17
DSMSERV LOADFORMAT (Format a database)............................................................................................ 17
Syntax........................................................................................................................................................ 17
Parameters................................................................................................................................................ 18
DSMSERV INSERTDB (Move a server database into an empty database) .................................................... 19
Requirements for insertion by using media................................................................................................ 19
Syntax........................................................................................................................................................ 20
Parameters................................................................................................................................................ 20
Notices................................................................................................................................................................. 23
Trademarks..................................................................................................................................................... 24

Copyright IBM Corporation 2014, 2016 iii

Introduction

Tip: The product now known as IBM Spectrum Protect was named IBM Tivoli Storage Manager in
releases earlier than Version 7.1.3. To learn more about the rebranding transition, see technote
1963634.
You can migrate an IBM Tivoli Storage Manager or IBM Spectrum Protect server from one operating system
to another. Depending on the version of the server, and the operating systems involved in the migration, different
migration methods are available:
Tivoli Storage Manager V5 servers on IBM z/OS systems can be migrated to AIX or Linux on System
z as part of an upgrade to V6.3 or V7.1.
Restriction: Upgrades from V5 to V8 are not supported.
Tivoli Storage Manager V5 servers on AIX, HP-UX, and Solaris operating systems can be migrated to
Linux x86_64 as part of an upgrade to V6.3.4 or V7.1.
Restriction: Upgrades from V5 to V8 are not supported.
Tivoli Storage Manager V6.3 or V7.1 servers can be migrated by using node replication to move one or
more nodes to another server on any supported operating system.
Any version of a Tivoli Storage Manager or IBM Spectrum Protect server can be migrated by using
export and import operations to move one or more nodes to another server on any supported operating
system.
This paper describes two additional cross-platform migration methods to migrate a server database to another
operating system.
The first relies on database backup and restore operations.
The second uses a variation of the V5 upgrade utilities, which have been adapted to work with an IBM
Spectrum Protect V7.1.5 or later database.
As with most other methods, a limited set of server levels and operating systems support these methods.

Prerequisites
Your system must meet the hardware and software requirements. The different migration methods have different
prerequisites.

Database backup and restore

You can use database backup and restore operations to migrate a Tivoli Storage Manager or IBM Spectrum
Protect server from one IBM AIX, HP-UX, Linux, or Oracle Solaris operating system to a different AIX, HP-UX,
Linux, or Solaris operating system, provided that both systems have the same endianness.

The following table lists operating systems with the same endianness.

Big endian AIX, Solaris SPARC, HP-UX, Linux on System z, and Linux on POWER systems
Little endian Linux x86_64 and Solaris x86_64

The source server must be at one of the following levels:

V6.2.5 or later
V6.3.4 or later

Copyright IBM Corporation 2014, 2016 1

 V7.1.0 or later
In all cases, you will migrate a server to a target system that has the same level of Tivoli Storage Manager or
IBM Spectrum Protect installed. After you migrate the server, you can upgrade it to any level that is supported on
the target system.

DSMSERV EXTRACTDB and DSMSERV INSERTDB

You can use the DSMSERV EXTRACTDB and DSMSERV INSERTDB utilities to migrate a Tivoli Storage
Manager or IBM Spectrum Protect server from IBM AIX to Linux x86_64. The source server must be at level
V7.1.5 or later. Beginning with level V7.1.7, you can also use the DSMSERV EXTRACTDB and DSMSERV
INSERTDB utilities to migrate an IBM Spectrum Protect server from HP-UX, Oracle Solaris, or Linux on POWER
to Linux x86_64. In all cases, the target system must have the same level of server installed as the source.
After you migrate the server, you can upgrade it to any level that is supported on the target system.

2 Migrating IBM Spectrum Protect Servers

Migration Process
The migration process consists of several steps:
1. Planning the migration
2. Preparing the source and target systems
3. Moving the server database from the source system to the target system, either by
Backing up the server database on the source system and restoring it on the target system
when the two systems have the same endianness, or by
Extracting the contents of the database on the source system and inserting it into an empty
database on the target system when moving from AIX, HP-UX, Oracle Solaris, or Linux on
POWER to Linux x86_64.
4. Reconfiguring the server on the target system to account for differences between the source and target
systems
Most of the planning, preparation, and reconfiguration steps are required to ensure that the server can continue
to access its storage from the target system. For servers that use tape storage, you might have to install device
drivers, and you must ensure that the target system can access the devices. You might have to update
configurations to account for changes in device names.
Ensuring access to data for servers that primarily use disk storage can be more difficult. Few AIX, HP-UX, Linux,
or Solaris operating systems have a file system in common with a different operating system, so it is usually not
possible to connect an existing disk LUN to the target system and use it as-is. Instead, use one of the following
solutions:
Move all data from disk to tape before the migration
Leave the disk data on the source system, and make the data available to the target system via a
Network File System (NFS) server

Using documentation links: The following procedures include links to supporting information in IBM
Knowledge Center. Most of the links are to Tivoli Storage Manager V7.1.1 topics. These topics are similar to
topics from earlier versions of the server and can be used as-is, even if the server that is being migrated is at
V6.2 or V6.3. However, if you prefer to use documentation that was specifically written for the server version
that is being migrated, go to IBM Knowledge Center:
V6.2: http://www.ibm.com/support/knowledgecenter/SSGSG7_6.2.0/com.ibm.itsm.ic.doc/welcome.html
V6.3: http://www.ibm.com/support/knowledgecenter/SSGSG7_6.3.4/KC_ditamaps/product_welcome.html
V7.1: http://www.ibm.com/support/knowledgecenter/SSGSG7_7.1.0/com.ibm.itsm.ic.doc/welcome.html
V8.1: http://www.ibm.com/support/knowledgecenter/SSEQVQ_8.1.0/tsm/welcome.html

Planning the migration

Plan the migration carefully to minimize server downtime. Thoroughly review this document before you start the
server migration. If you are migrating multiple servers on multiple systems, plan to upgrade one server first in a
test environment.
1. Ensure that each server that you plan to migrate is at V6.2.5 or later, V6.3.4 or later, or V7.1.0 or later. If
the server is being migrated from AIX to Linux x86_64, then it must be at level V7.1.5 or later. If the
server is being migrated from HP-UX, Oracle Solaris, or Linux on POWER to Linux x86_64, then it must
be at level V7.1.7 or later. If the server is not at the appropriate level, upgrade it before you start the
migration process. For instructions about upgrading the server, see the online documentation:
Upgrading from Tivoli Storage Manager V6.1
Upgrading to Tivoli Storage Manager V6.3 or later
Upgrading to Tivoli Storage Manager V7.1
2. Review the hardware and software requirements for running the server on the target system. For more

Copyright IBM Corporation 2014, 2016 3

 information, see Prerequisites and IBM Spectrum Protect Supported Operating Systems.
3. Plan data movement and ensure that the devices that you plan to use are supported on both operating
systems. For information about supported devices, see Supported Devices for AIX, HP-UX, Solaris, and
Windows systems and Supported Devices for Linux systems. Device support can differ, depending on
the operating system, for example:
Device classes with DEVTYPE=GENERICTAPE are not supported on Linux. If the target system
is installed on Linux, all data on generic tape devices must be migrated or backed up to
supported tape devices before migration.
All data in DISK or FILE storage pools must be migrated or backed up to tape devices that are
supported on the target system. Alternatively, you can export the storage pools via an NFS
server so that the data can be read from the target system. For requirements and restrictions on
the use of NFS file systems with the server, see Tivoli Storage Manager server support of NFS
file system. DISK storage pools that use raw logical partitions cannot be exported over NFS,
and must be migrated or backed up to tape.
4. Provision at least as much disk storage for the target system as is used by the source system.
5. If you plan to back up large storage pools to tape, consider starting this process before you migrate the
server. In this way, you might be able to reduce the time that you spend backing up these storage pools
while the server is unavailable.

Preparing the target system for the migration

By completing these steps before you migrate the server, you can reduce the time that the server is unavailable.
1. Install Tivoli Storage Manager on the target system at the same level as on the source system. For
example, if Tivoli Storage Manager V6.2.5 is installed on a source system, install V6.2.5 on the target
system. For information about installing a V6.2 or V6.3 server, see Installing the server. For Tivoli
Storage Manager 7.1, see Installing Tivoli Storage Manager.
2. Install the device drivers for the devices that you plan to use on the target system. Configure the system
to ensure that the devices can be accessed from the target system.
3. Complete the following steps on the target system for each server instance that is running on the source
system:
a. Create a user account (a user ID and group) that will own the server instance. Ensure that this
account has the same user name as on the source system. If the target system will use NFS to
access DISK or FILE storage pool volumes remotely, the account must have the same user ID and
group ID as on the source system.
For instructions about creating a user account, see Creating the user ID and directories for the
server instance.
Change to the user account that you just created.
b. Create the directories that the server requires, starting with the server instance directory. Create the
database, active log, and archive log directories for the server instance by using the same paths as
on the source system. Also, create the log mirror and secondary archive log directories, if these
exist on the source system, by using the same paths as on the source system. For instructions, see
Creating the user ID and directories for the server instance.
c. Configure a new server instance by using the user account and directories that you created in the
previous steps. Take one of the following actions:
To configure the new server instance by using the configuration wizard, follow the instructions
for your operating system:
AIX: Configuring Tivoli Storage Manager using the configuration wizard
HP-UX: Configuring Tivoli Storage Manager using the configuration wizard

4 Migrating IBM Spectrum Protect Servers

 Linux: Configuring Tivoli Storage Manager using the configuration wizard
Solaris: Configuring Tivoli Storage Manager using the configuration wizard
To configure the new server instance manually, follow the instructions for your operating system:
AIX: Configuring the server instance manually
HP-UX: Configuring the server instance manually
Linux: Configuring the server instance manually
Solaris: Configuring the server instance manually
d. To prepare the new server instance for the migration, clean up the temporary database and other
files that will interfere with the migration process.
i. Use the DSMSERV REMOVEDB utility to remove the temporary database. Issue the following
command:
# dsmserv removedb TSMDB1
For more information about using this command, see DSMSERV REMOVEDB (Remove a
database).
ii. Manually delete the archived log volumes in the archive log directory. The DSMSERV
REMOVEDB utility does not remove these volumes, and you cannot format a new database
using the DSMSERV LOADFORMAT utility while they exist.
iii. Manually delete the SSL key database, exported certificates and password stash files that were
created when the instance was created. These files are associated with the temporary server
database, and cannot be used with the migrated database.
cert.kdb
cert256.arm
cert.arm
cert.crl
cert.rdb
cert.sth
e. If using the DSMSERV EXTRACTDB and DSMSERV INSERTDB utilities to migrate from AIX, HP-
UX, Oracle Solaris, or Linux on POWER to Linux x86_64 then use the DSMSERV LOADFORMAT
utility to create a new, empty database into which the source database contents will be inserted. For
more information about using this utility, see DSMSERV LOADFORMAT (Format a database).
4. If the target system will use NFS to access DISK and FILE storage pool volumes remotely, export the
file systems from the source system and mount them on the target system. Ensure that all volumes can
be accessed by using the same path as on the source system.
5. Copy the following configuration files from the source server to the target server:
The server options file, dsmserv.opt
The device configuration file
The volume history file
The server password file, dsmserv.pwd. Note that this file is only created by servers at level V7.1.3
or higher

Preparing the source system for the migration

When you are ready to begin the migration, complete the following steps on the source system:
1. Disable sessions.

Migrating IBM Spectrum Protect Servers 5

 2. Back up or migrate data stored on DISK and FILE devices. This step is not required if all of the source
server's DISK and FILE storage pools are to be accessed remotely via NFS.
3. If the target system runs on a Linux operating system, migrate data that is stored on GENERICTAPE
devices.
4. If the source server's backup set volumes will not be accessed remotely via NFS, move backup sets
that are stored on FILE devices.
5. To prevent potential issues during and after the migration, make the following adjustments to settings on
your server and clients. These adjustments must be completed so that you can revert to using the
original server after the migration, if a problem occurs.
a. For each sequential-access storage pool on tape, use the UPDATE STGPOOL command to set the
REUSEDELAY parameter. Set the parameter to the number of days during which you want to be
able to revert to the original server, if necessary. For example, if you want to be able to revert to the
original server for up to 30 days after the migration, set the REUSEDELAY parameter to 31 days:
update stgpool sequential_access_storage_pool reusedelay=31
where sequential_access_storage_pool specifies the name of the storage pool. For more
information about setting the REUSEDELAY parameter, see UPDATE STGPOOL (Update a storage
pool).
b. For each copy storage pool on tape, use the QUERY STGPOOL command to determine the value
of the RECLAIM parameter. Note the value for future reference. If you must revert to the source
server at any time during the migration process, it is useful to know the value of the RECLAIM
parameter so that you can restore the setting. Issue the following command:
query stgpool sequential_access_storage_pool format=detailed
where sequential_access_storage_pool specifies the name of the copy storage pool. In the system
output, the value of the RECLAIM parameter can be found in the Reclamation Threshold field. For
more information about determining the value of the RECLAIM parameter, see QUERY STGPOOL
(Query storage pools).
c. For each copy storage pool on tape, use the UPDATE STGPOOL command to set the RECLAIM
parameter to 100, meaning 100%. For example, issue the following command:
update stgpool copy_storage_pool reclaim=100
where copy_storage_pool specifies the name of the copy storage pool.
d. For each tape, DISK, or FILE volume that was used by the server on the source system, specify the
read-only access mode. Issue the following administrative command:
update volume volume_name access=readonly
where volume_name specifies the name of the tape, DISK, or FILE volume. For more information
about specifying read-only access mode, see UPDATE VOLUME (Change a storage pool volume).
e. For each client that plays an essential role in your system, verify that the value for the
schedlogretention client option is set to retain the client schedule log for a sufficient time.
The client schedule log can be useful if you must revert the system to the original server. If the
retention period for the schedule log is too short, the schedule log information might be deleted too
soon.
For example, to prune the log every 45 days and save the log entries, add the following option:
schedlogretention 45 S
Add the schedlogretention client option to the dsm.sys file within the server stanza for the server
that is being migrated. For more information about pruning the log, see Schedlogretention.
6. Determine whether any server processes are running. Either cancel them or allow them to finish. Use

6 Migrating IBM Spectrum Protect Servers

 the following commands:
query process
cancel process process_number
where process_number specifies the number of the process. Allow time for the processes to be
stopped. Some processes might take a considerable amount of time to stop.
7. Create a summary of database contents.

Migrating the server using database backup and restore

Complete the following steps:
1. On the source system, back up the server database to tape. Issue the following command:
backup db type=full devclass=tape_device
where tape_device specifies the name of a tape device class.
With Tivoli Storage Manager V7.1.1, you can compress database backups. Do not use compression for
this backup because compressed database backups cannot be restored on a different operating
system. If you have Tivoli Storage Manager V7.1.1 installed, issue the following command to back up
the server database:
backup db type=full devclass=tape_device compress=no
For more information about backing up the database, see BACKUP DB (Back up the database).
2. Unmount any tapes that are mounted. Use the following commands:
query mount
dismount volume volume_name
where volume_name specifies the name of the volume. For more information about unmounting tapes,
see QUERY MOUNT (Display information on mounted sequential access volumes) and DISMOUNT
VOLUME (Dismount a volume by volume name).
3. Back up the volume history and device configuration files to another directory. Issue the following
commands:
backup devconfig filenames=devconfig_file_name
backup volhistory filenames=volhist_file_name
where devconfig_file_name specifies the file in which to store device configuration information, and
volhist_file_name specifies the file in which to store volume history information.
For more information about using the commands, see BACKUP DEVCONFIG (Create backup copies of
device configuration information) and BACKUP VOLHISTORY (Save sequential volume history
information).
4. Stop the server by issuing the following command:
halt
5. If you did not complete the steps in Preparing the target system for the migration, complete them now.
6. Move copies of the volume history file, device configuration file, and server options file to the target
system.
7. On the target system, edit the device configuration file to update the device path information for any
devices that are required to restore the database. For each relevant DEFINE PATH entry in the device
configuration file, update the DEVICE parameter to the corresponding path on the target system.
8. Restore the server database by using the following command:

Migrating IBM Spectrum Protect Servers 7

 # dsmserv restore db
For more information about restoring the database, see DSMSERV RESTORE DB (Restore a database
to its most current state).

Migrating the server using database extract and insert utilities

The database extract and insert utilities are used to extract the contents of the source Tivoli Storage Manager
server's database and insert the data into an empty database on the target system. There are two utilities
DSMSERV EXTRACTDB
DSMSERV INSERTDB
Both utilities are based on the Tivoli Storage Manager V5 Upgrade Utilities. Starting with V7.1.5, the utilities
were enhanced to work with data from a version 7.1.5 DB2 database.
The database extract and insert utilities support the following two methods of operation.
Network method
The database contents are extracted from the source database and transmitted over a network to the
target server, where they are simultaneously inserted into the target database.
Media method
The database contents are extracted from the source database and stored on tape or FILE media. The
contents are inserted into the target database at a later time.

Migrate using the network method

Move the database by starting the insertion process for the target server to accept the server database. Then,
start the extraction process for the source server to extract and send the database.
Before you begin
Ensure that the source and target servers are not running.
Procedure
To move the server database over a network, complete the following steps:
1. Verify that there is a good network connection between the two systems.
2. Start the insertion process on the source server to accept the database. Use the DSMSERV INSERTDB
command. To monitor the process, direct the output of the process to a file. For example, to start the
server, allow the default time of 60 minutes for the source server to contact the source server, and direct
the process output to the insert.out file, use this command:
nohup /opt/tivoli/tsm/server/bin/dsmserv insertdb \
sesswait=60 >insert.out 2>&1 &
For more information about inserting data, see DSMSERV INSERTDB (Move a server database into an
empty database).
The server starts and waits up to 60 minutes to be contacted by the original server. Some time might
pass during which no messages are issued. During this time, DB2 operations are running in the
background.
Optional: To verify that operations are continuing as expected, monitor the processor and I/O usage for
the server process and the corresponding DB2 process.
3. Monitor the output of the DSMSERV INSERTDB process. Verify that the DSMSERV INSERTDB
process issues the following message before you continue to the next step:
ANR1336I INSERTDB: Ready for connections from the source server

8 Migrating IBM Spectrum Protect Servers

 Issue the following command to monitor the process output in the insert.out file:
tail -f insert.out
4. Start the data extraction from the original server by using the DSMSERV EXTRACTDB command. Issue
the command from the source server's instance directory. Specify the TCP/IP address and port for the
target server. Direct the output of the process to a file for monitoring. For example, enter the following
command on one line:
nohup /opt/tivoli/tsm/server/bin/dsmserv extractdb \
hladdress=9.11.25.124 lladdress=1500 >extract.out 2>&1 &
For more information about extracting data, see DSMSERV EXTRACTDB (Extract data from a server
database).
5. Monitor the processes for errors and warning messages, and for items that might require attention.
From the instance directory for the server that you are upgrading, issue the following command to
monitor the extraction process:
tail -f extract.out
The length of time that the process runs depends on the size of the database, the hardware, and the
network.
6. Examine the process outputs for the extraction and insertion processes to find the messages that
indicate the success or failure of the operations.

Process Success message Failure message

Extraction ANR3402I EXTRACTDB: Process ANR3414E EXTRACTDB: Process
1, database extract, has 1, database extract, has
completed. completed with errors.
Insertion ANR1395I INSERTDB: Process 1, ANR1396E INSERTDB: Process 1,
database insert, has database insert, has
completed. completed with errors.

Migrate using the media method

Extracting the data to media

Use the DSMSERV EXTRACTDB utility to extract data from the source server database and store it on a tape
device.
Procedure
1. Log in by using the server's instance id on the source system.
2. Ensure that the storage device is available. The server database and the device configuration file must
contain a valid device class definition for the device.
3. From the instance directory for the server that you are migrating, issue the command to start the
extraction. Direct the output of the process to a file for monitoring. For example, issue the following
command on one line:
nohup /opt/tivoli/tsm/server/bin/dsmserv extractdb \
devclass=tape manifest=./manifest.txt >extract.out 2>&1 &
For more information about extracting data, see DSMSERV EXTRACTDB (Extract data from a server
database).
Tip: Messages that are issued during the extract operation are not saved in the server activity log. Direct
the output of the utility to a file, as shown in the examples, to record the messages.

Migrating IBM Spectrum Protect Servers 9

 4. Monitor the process for errors and warning messages, and for items that might require attention. A
message near the end of the process output indicates the success or failure of the operation:
Success message:
ANR3402I EXTRACTDB: Process 1, database extract, has completed.
Failure message:
ANR3414E EXTRACTDB: Process 1, database extract, has completed with
errors.
For example, from the instance directory for the server that you are upgrading, issue the following
command to monitor the process:
tail -f extract.out
The length of time that the process runs depends on the database size. The time is approximately the
length of time that is required for a full database backup.

Loading the extracted data into the target database

After you format an empty database by using the DSMSERV LOADFORMAT utility, load the data that you
extracted from the original server database.
Before you begin
Ensure that the following requirements are met before you begin to load the data:
The manifest file from the DSMSERV EXTRACTDB operation must be available.
The server options file must contain an entry for the device configuration file.
The device configuration file must have information about the device class that is specified in the
manifest file.
The media that contains the extracted database must be available to the target server. The device must
be physically attached to the system. The permissions must be set to grant access to the media for the
instance user ID.
Procedure
Complete the following steps:
1. Verify that the target server can access the extracted data. The tape drive that is used for the extracted
data must be physically attached to the target system.
2. Ensure that the instance user ID has ownership or read/write permission for the manifest file that was
created by the extraction process.
3. Log in with the instance user ID on the system where you installed the target server.
4. Copy the manifest file that was created by the extraction process to the target system.
5. On the target server, complete the following steps:
a. Verify that the server options file from the source server includes the DEVCONFIG option, and that
the option specifies the full path of the device configuration file.
b. Verify that the device configuration file from the source server is available in the location that is
specified by the DEVCONFIG option.
c. Verify that the permissions on the device configuration file allow read access for the instance user
ID.
6. Verify that the contents of the device configuration file are correct. The device class that was used for
the extraction step is recorded in the manifest file, and that device class must exist and be valid on the

10 Migrating IBM Spectrum Protect Servers

 target system. Verify entries for tape. For example, the device names might have changed.
7. Verify the contents of the manifest file and edit the file if necessary:
a. Ensure that the device names in the manifest file are valid for the target system. Device names for
the same device might be different on source and target systems.
b. Ensure that the manifest file contains a list of volumes to be used when the extracted data is loaded
into the new database.
8. To load an extracted server database into the prepared, empty target database, issue the DSMSERV
INSERTDB command. Direct the output of the process to a file for monitoring. For example, enter the
following command on one line:
nohup /opt/tivoli/tsm/server/bin/dsmserv insertdb \
manifest=./manifest.txt >insert.out 2>&1 &
For more information about loading an extracted database into a new database, see DSMSERV
INSERTDB (Move a server database into an empty database).
9. Monitor the process for error messages, warning messages, and any items that might require your
attention. The system displays interim statistics about the process of loading the database. However,
there might be periods when no messages are issued. During this time, DB2 operations are running in
the background. The length of time that the process runs depends on the size of the database. For more
information, see Example: Estimating the upgrade time based on the database size.
Optional: Verify that the database is being loaded by monitoring the processor and I/O usage for the
server process and the corresponding DB2 process. For example, issue the following command to
monitor the process:
tail -f insert.out
A message in the output of the DSMSERV INSERTDB command indicates the status of the operation:
Success message:
ANR1395I INSERTDB: Process 1, database insert, has completed.
Failure message:
ANR1396E INSERTDB: Process 1, database insert, has completed with
errors.
10. After the data is loaded into the database, remove or check out from the library the tape that holds the
extracted data. Prevent the tape from being reused until you are sure that you do not need to run the
database-loading operation again.

Taking the first steps after migration

1. Verify the migration results:
a. Start the server. Monitor the messages that the server issues as it starts. Watch for error and
warning messages.
b. Set ownership or permissions to the device special files to ensure that the storage devices of the
original server that are supported by the target system are accessible to the server. If the server
storage area network (SAN) discovery function is enabled, the server should update device
information automatically. Compare the device names on the target system with the device
parameters of the path definitions on the server. To see the server path definitions, issue the
following command:
query path format=detailed
Update server device path definitions manually as necessary. For more information about this task,
see QUERY PATH (Display a path definition), UPDATE PATH (Change a path), and

Migrating IBM Spectrum Protect Servers 11

 SANDISCOVERY.
c. Rerun commands to get a summary of information in the database. Compare the summary with the
results for the same commands before the migration.
d. Update the network address, and if necessary the port number, that is used by backup-archive
clients, administrative clients, storage agents, library client servers, and other servers for
communicating with the migrated server. Use the SET SERVERHLADDRESS command and, if
necessary, the SET SERVERLLADDRESS command.
Alternatively, consider whether you can use the network address that was used by the source
system as the address of the target system. You might also be able to update the domain name
service (DNS) to point to the target system instead of the source system. Consult your network
administrator.
e. Verify that you can connect to the server by using an administrative client as you did for the earlier
version of the server.
f. Back up several client nodes and verify that the backups work as expected. Ensure that the clients
could connect to the target server, and that the data was written to the appropriate storage devices.
g. Verify that operations such as LAN-free data movement and library sharing work correctly. In some
cases, it might be necessary to reset the server verification keys of both the library manager and
library client servers by using the UPDATE SERVER command and specifying the
FORCESYNC=YES parameter.
After you are satisfied that the server is operating as expected and you do not have to revert to the
source system, return any settings that you changed to prepare for the migration back to their original
values.
2. Register licenses.
3. If you did not use the instance configuration wizard to configure the server, ensure that you have
completed the steps to manually configure the system for database backups. Then, perform a full
database backup.
4. Back up the volume history file.
5. If you must change the host name of the server, follow the instructions in Changing the host name for
the Tivoli Storage Manager server.
6. Update DISK, FILE,and directory-container storage pools that use NFS mounted file systems from the
source system to use local non-NFS storage for new data. Define replacement DISK volumes on non-
NFS storage for each random access storage pool.
7. If DISK or FILE data was migrated to tape before the migration, migrate the data from tape back to DISK
or FILE devices.
8. Redefine backup sets as necessary by using the DEFINE BACKUPSET command.
9. Configure the server for communications by using Secure Socket Layer (SSL) and Transport Layer
Security (TLS) following the instructions at Setting up Secure Sockets Layer and Transport Layer
Security.
You might be able to use the same certificates on the target server as were used on the source
server if the target server address is covered by the fields in the source server certificate.
For third-party certificates, if the Certificate Authority (CA) certificates for the new target server
already appear in partner key databases, only the target certificates must be replaced. If the new
certificates uses a CA whose certificates are not in partner key databases, distribute the certificates
to clients and other partner systems so that the new target certificates can be trusted.
When using self-signed certificates, distribute the newly generated target certificates to clients and
other partner systems for inclusion in their key databases.

12 Migrating IBM Spectrum Protect Servers

You can now upgrade the server to any level that is supported on the target system. For more information, see
the appropriate topic:
Upgrading from Tivoli Storage Manager Version 6.2 to 6.3 or later
Upgrading from Tivoli Storage Manager V6.2 or V6.3 to V7.1.1

Migrating IBM Spectrum Protect Servers 13

Appendix A. Utilities for server migration
Utilities are available to migrate a server from AIX, HP-UX, Oracle Solaris, or Linux on POWER to Linux x86_64.

DSMSERV EXTRACTDB (Extract data from a server database)

Use the DSMSERV EXTRACTDB utility to extract data from a version 7.1.5 or later server database. The data
can be inserted into a target server database either later or at the same time as the extraction process.

Syntax
>>-DSMSERV---+---------+--+-------------------+--EXTRACTDB------>
'- -quiet-' '- -o -options_file-'

>--+-| A: Extract to media |------------------------+----------><

'-| B: Extract and transmit by using a network |-'

A: Extract to media

|--DEVclass--=--device_class_name--MANifest--=--file_name------->

.-SCRatch--=--Yes-----.
>--+---------------------+-------------------------------------->
'-SCRatch--=--+-Yes-+-'
'-No--'

>--+---------------------------------------+--------------------|
| .-,-----------. |
| (1) V | |
'-------VOLumenames--=----volume_name-+-'

B: Extract and transmit by using a network

|--HLAddress--=--ip_address--LLAddress--=--tcp_port-------------|

Notes:
1. You must specify VOLUMENAMES if you specify SCRATCH=NO.

Parameters
-quiet

Copyright IBM Corporation 2014, 2016 15

 Specifies that messages to the console are suppressed. This parameter is optional.
-o options_file
Specifies an options file to use. This parameter is optional.
DEVclass
Specifies a sequential device class to use for storing the extracted data. This parameter is required if
you want to extract the database to media.
Restriction: You cannot use a device class with a device type of NAS or CENTERA.
MANifest
Specifies the location of the manifest file. Use a fully qualified file name, or place in a local directory. For
example: ./manifest.txt
This parameter is required if you want to extract the database to media.
SCRatch
Specifies whether scratch volumes can be used to store the data. This parameter is optional. The
default value is YES. Possible values are:
Yes
Specifies that scratch volumes can be used.
If you specify SCRATCH=YES and also specify volumes with the VOLUMENAMES parameter,
the specified volumes are used first. Scratch volumes are used only if the specified volumes do
not have enough space to contain the extracted data.
If the device type for the device class is FILE, the names for the scratch volumes are generated
based on a time stamp.
No
Specifies that scratch volumes cannot be used. If you specify this value, you must also specify
volumes by using the VOLUMENAMES parameter.
If the volumes that you specify with the VOLUMENAMES parameter do not have enough space
to contain the extracted data, the process fails.
VOLumenames
Specifies the volumes to use to store the extracted database. To specify multiple volumes, separate the
names with commas and without intervening spaces.
HLAddress
Specifies either the numeric IP address or the domain name of the target server. This parameter is
required if you want to extract and simultaneously transmit the data to the target server by using the
network.
LLAddress
Specifies the low-level address of the target server. This address is the same as the value that is
specified with the TCPPORT server option for the target server. This parameter is required if you want to
extract and simultaneously transmit the data to the target server by using the network.

Examples
Extract and transmit the data to a server by using the network:

dsmserv extractdb hladdress=xyz.company.com lladdress=1555

16 Migrating IBM Spectrum Protect Servers

Extract and store the data on media:

dsmserv extractdb devclass=tapeclass manifest=keepthis.txt

The manifest file for the data extraction to media

During the data extraction to media, the DSMSERV EXTRACTDB utility creates a manifest file. The manifest file
contains information about the volumes and device classes that are used for storing the extracted data.
The manifest file is required by the data insertion process when you are using media to move the database to
the target server. You specify the name of the manifest file when you run the DSMSERV INSERTDB utility.
In the manifest file, the device configuration section contains information about the configuration settings of the
source server device. You might need to edit settings, such as device names and drive element numbers, before
you use the manifest file with the DSMSERV INSERTDB utility, especially if the utility is run on a different
system.
A manifest file has content that is similar to the following example:
* Version 2 manifest file for EXTRACTDB
DEVCLASS=3584L2
STREAMS=1
VOLUMENAMES000=L12345
* Device Configuration Information
DEFINE DEVCLASS 3584L2 DEVTYPE=LTO FORMAT=DRIVE MOUNTLIMIT=DRIVES MOUNTWAIT=60
MOUNTRETENTION=60 PREFIX=ADSM LIBRARY=3584L2 WORM=NO DRIVEENCRYPTION=ALLOW
SET SERVERNAME MOKSHA
DEFINE LIBRARY 3584L2 LIBTYPE=SCSI SERIAL="00000002267304AB" SHARED=YES
AUTOLABEL=YES RESETDRIVE=YES
DEFINE DRIVE 3584L2 3584L2_DRV1 ELEMENT=265 ONLINE=Yes WWN="500507630F18BA09"
SERIAL="0007859130"
DEFINE PATH MOKSHA 3584L2 SRCTYPE=SERVER DESTTYPE=LIBRARY DEVICE=/dev/IBMchanger5
ONLINE=YES
DEFINE PATH MOKSHA 3584L2_DRV1 SRCTYPE=SERVER DESTTYPE=DRIVE LIBRARY=3584L2
DEVICE=/dev/IBMtape20 ONLINE=YES

DSMSERV LOADFORMAT (Format a database)

Use the DSMSERV LOADFORMAT utility to format an empty database in preparation for inserting an extracted
database into the empty database.

Syntax
>>-DSMSERV -+------------------+--+-------------------+--------->
| | | |
'- -u--user_name-' '- -i--instance_dir-'

Migrating IBM Spectrum Protect Servers 17

>--+-------------------+--+------------+--+---------+----------->
'- -o--options_file-' '- -noexpire-' '- -quiet-'

.-,---------.
V |
>--LOADFORMAT--+-DBDir--=----directory-+-+---------------------->
'-DBFile--=--file---------'

.-ACTIVELOGSize--=--16384-----.
>--+-----------------------------+------------------------------>
'-ACTIVELOGSize--=--megabytes-'

>--ACTIVELOGDirectory--=--directory----------------------------->

>--ARCHLogdirectory--=--directory------------------------------->

>--+----------------------------------------+------------------->
'-ARCHFailoverlogdirectory--=--directory-'

>--+----------------------------------+------------------------><
'-MIRRorlogdirectory--=--directory-'

Parameters
-u user_name
Specifies a user name to switch to before you initialize the server. This parameter is optional.
-i instance_dir
Specifies an instance directory to use. This directory becomes the current working directory of the
server. This parameter is optional.
-o options_file
Specifies an options file to use. This parameter is optional.
-noexpire
Specifies that expiration processing is suppressed when starting. This parameter is optional.
-quiet
Specifies that messages to the console are suppressed. This parameter is optional.
DBDir
Specifies the relative path names of one or more directories that are used to store database objects.
Directory names must be separated by commas but without spaces. You can specify up to 128 directory
names. You must specify either the DBDIR or the DBFILE parameter.
DBFile

18 Migrating IBM Spectrum Protect Servers

 Specifies the name of a file that contains the relative path names of one or more directories that are
used to store database objects. Each directory name must be on a separate line in the file. You can
specify up to 128 directory names. You must specify either the DBDIR or the DBFILE parameter.
ACTIVELOGSize
Specifies the size of the active log in megabytes. This parameter is optional. The minimum value is 2048
MB (2 GB); the maximum is 131,072 MB (128 GB). If you specify an odd number, the value is rounded
up to the next even number. The default is 16384 MB.
ACTIVELOGDirectory (Required)
Specifies the directory in which the Tivoli Storage Manager server writes and stores active log files.
There is only one active log location. The name must be a fully qualified directory name. The directory
must exist, it must be empty, and it must be accessible by the user ID of the database manager. The
maximum number of characters is 175.
ARCHLogdirectory (Required)
Specifies the directory for the archive log files. The name must be a fully qualified directory name. The
maximum number of characters is 175.
ARCHFailoverlogdirectory
Specifies the directory to be used as an alternative storage location if the ARCHLOGDIRECTORY
directory is full. This parameter is optional. The maximum number of characters is 175.
MIRRorlogdirectory
Specifies the directory in which the server mirrors the active log (those files in the
ACTIVELOGDIRECTORY directory). This parameter is optional. The directory must be a fully qualified
directory name. The maximum number of characters is 175.
Example: Format a database

dsmserv loadformat dbdir=/tsmdb001 activelogsize=8192

activelogdirectory=/activelog archlogdirectory=/archlog
archfailoverlogdirectory=/archfaillog mirrorlogdirectory=/mirrorlog

DSMSERV INSERTDB (Move a server database into an empty database)

Use the DSMSERV INSERTDB utility to move a server database into a new database. The database can be
extracted from the original server and inserted into a new database on the new server by using a network
connection between the two servers. The database can also be inserted from media that contains the extracted
database.
Before you use the DSMSERV INSERTDB utility, complete the planning and preparation tasks, such as backing
up the database and saving configuration information. Ensure that you meet all requirements before you move
the server database.

Requirements for insertion by using media

Before you run the utility to insert the server database into an empty database, ensure that your system meets
the following requirements.
The manifest file from the DSMSERV EXTRACTDB operation must be available.
The media that contains the extracted database must be available to the target server. Also, the
permissions must be set to grant access to the media for the user ID that owns the target server
instance.

Migrating IBM Spectrum Protect Servers 19

Syntax
>>-DSMSERV -+----------------+--+-------------------+----------->
'- -u--user_name-' '- -i--instance_dir-'

>--+-------------------+--+------------+--+---------+----------->
'- -o--options_file-' '- -noexpire-' '- -quiet-'

>--INSERTDB--+-| A: Insert from media |-----+------------------->

'-| B: Insert over a network |-'

.-PREview--=--No------.
>--+---------------------+-------------------------------------><
'-PREview--=--+-Yes-+-'
'-No--'

A: Insert from media

|--+--------------------------------+--MANifest--=--file_name --|
'-DEVclass--=--device_class_name-'

B: Insert over a network

.-SESSWait--=--60------.
|--+----------------------+-------------------------------------|
'-SESSWait--=--minutes-'

Parameters
-u user_name
Specifies a user name to switch to before you initialize the server. This parameter is optional.
-i instance_dir
Specifies an instance directory to use. This directory becomes the current working directory of the
server. This parameter is optional.
-o options_file
Specifies an options file to use. This parameter is optional.
-noexpire

20 Migrating IBM Spectrum Protect Servers

 Specifies that expiration processing is suppressed when starting. This parameter is optional.
-quiet
Specifies that messages to the console are suppressed. This parameter is optional.
DEVclass
Specifies a sequential-access device class. You can specify any device class except for the DISK device
class. The definition for the device class must exist in either the manifest file or the device configuration
file.
This parameter is optional and is used only when the database that you want to insert into the empty
V7.1.1 database was extracted to media. If the database is on media and you do not specify a device
class, the device class that is identified in the manifest file is used.
Restriction: You cannot use a device class with a device type of NAS or CENTERA.
MANifest
Specifies the location of the manifest file. Use a fully qualified file name, or place in a local directory. For
example: ./manifest.txt
This parameter is required when the database that you want to insert into the empty V7.1.1 database
was extracted to media.
SESSWait
Specifies the number of minutes that the V7.1.1server waits to be contacted by the original server. The
default value is 60 minutes.
Use this parameter only if the data that is inserted into the empty V7.1.1 database is transmitted from
the source server with a network connection.
PREview
Specifies whether to preview the insertion operation. This parameter is optional. The default value is
NO.
Use the PREVIEW=YES parameter to test a database. When you use this parameter, the operation
includes all steps of the process, except for the actual insertion of data into the new database. When
you preview the insertion operation, you can quickly verify that the source database is readable. You can
also identify any data constraint violations that might prevent an upgraded database from being put into
production.

Migrating IBM Spectrum Protect Servers 21

Notices
This information was developed for products and services offered in the US. This material might be available
from IBM in other languages. However, you may be required to own a copy of the product or product version in
that language in order to access it.
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 grant you any license to these patents. You can send license inquiries, in
writing, to:
IBM Director of Licensing
IBM Corporation
North Castle Drive, MD-NC119
Armonk, NY 10504-1785
US
For license inquiries regarding double-byte character set (DBCS) information, contact the IBM Intellectual
Property Department in your country or send inquiries, in writing, to:
Intellectual Property Licensing
Legal and Intellectual Property Law
IBM Japan Ltd.
19-21, Nihonbashi-Hakozakicho, Chuo-ku
Tokyo 103-8510, Japan
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 jurisdictions 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 websites are provided for convenience only and do not in any
manner serve as an endorsement of those websites. The materials at those websites are not part of the
materials for this IBM product and use of those websites is at your own risk.
IBM may use or distribute any of the information you provide in any way it believes appropriate without incurring
any obligation to you.
Licensees of this program who wish to have information about it for the purpose of enabling: (i) the exchange of
information between independently created programs and other programs (including this one) and (ii) the mutual
use of the information which has been exchanged, should contact:
IBM Director of Licensing
IBM Corporation
North Castle Drive, MD-NC119
Armonk, NY 10504-1785
US
Such information may be available, subject to appropriate terms and conditions, including in some cases,

Copyright IBM Corporation 2014, 2016 23

payment of a fee.
The licensed program described in this document and all licensed material available for it are provided by IBM
under terms of the IBM Customer Agreement, IBM International Program License Agreement or any equivalent
agreement between us.
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 actual people or business enterprises is entirely coincidental.
COPYRIGHT LICENSE:
This information contains sample application programs in source language, which illustrate programming
techniques on various operating platforms. You may copy, modify, and distribute these sample programs in any
form without payment to IBM, for the purposes of developing, using, marketing or distributing application
programs conforming to the application programming interface for the operating platform for which the sample
programs are written. These examples have not been thoroughly tested under all conditions. IBM, therefore,
cannot guarantee or imply reliability, serviceability, or function of these programs. The sample programs are
provided "AS IS", without warranty of any kind. IBM shall not be liable for any damages arising out of your use of
the sample programs.

Trademarks
IBM, the IBM logo, and ibm.com are trademarks or registered trademarks of International Business Machines
Corp., registered in many jurisdictions worldwide. Other product and service names might be trademarks of IBM
or other companies. A current list of IBM trademarks is available on the Web at "Copyright and trademark
information" at http://www.ibm.com/legal/copytrade.shtml.

Linux is a registered trademark of Linus Torvalds in the United States, other countries, or both.
Microsoft and Windows are trademarks of Microsoft Corporation in the United States, other countries, or both.
UNIX is a registered trademark of The Open Group in the United States and other countries.

24 Migrating IBM Spectrum Protect Servers