You may have to recover your system using hardware that is different from that used when you backed up your database and created disaster recovery plan file. Before restoring the database, update the device configuration file included in the recovery plan file. After restoring the database, update the device configuration on the database.
This section describes a number of such situations in detail. If the hardware environment is different at the recovery site, you must update the device configuration file. TSM uses the device configuration file to access the devices that are needed to read the database backup volumes. The RECOVERY.VOLUMES.REQUIRED stanza in the plan file identifies the volumes that are needed to restore the database.
Ensure that the DEFINE DRIVE and DEFINE LIBRARY commands in the device configuration file are valid for the new hardware configuration. For example, if an automated tape library was used originally and cannot be used at the recovery site, update the device configuration file. Include the DEFINE LIBRARY and DEFINE DRIVE commands that are needed to define the manual drive to be used. In this case, you must manually mount the backup volumes.
Here is an example of an original device configuration file, which describes an automated tape library:
/* Device Configuration */ define devclass auto8mm_class devtype=8mm format=drive mountlimit=2 mountwait=60 mountretention=60 prefix=tsm library=auto8mmlib define library auto8mmlib libtype=scsi device=lb4.0.0.1 define drive auto8mmlib 8mm_tape0 device=mt1.0.0.1 element=82 online=yes define drive auto8mmlib 8mm_tape1 device=mt2.0.0.1 element=83 online=yes /* LIBRARYINVENTORY SCSI AUTO8MMLIB KEV004 1 101*/ /* LIBRARYINVENTORY SCSI AUTO8MMLIB KEV005 3 101*/
Here is an example of the updated device configuration file when a manual library is used at the recovery site:
/* Device Configuration */ define devclass auto8mm_class devtype=8mm format=drive mountlimit=1 mountwait=60 mountretention=60 prefix=tsm library=manual8mm define library manual8mm libtype=manual define drive manual8mm 8mm_tape0 device=mt1.0.0.1
The following changes were made:
After you restore the database, modify the device configuration file in the database. After starting the server, define, update, and delete your library and drive definitions to match your new configuration.
Manually place the database backup volumes in the automated library and note the element numbers where you place them. Then update the comments in the device configuration file to identify the locations of those volumes.
Here is an example of an original device configuration file, which describes an automated tape library:
/* Device Configuration */ define devclass auto8mm_class devtype=8mm format=drive mountlimit=2 mountwait=60 mountretention=60 prefix=tsm library=auto8mmlib define library auto8mmlib libtype=scsi device=lb4.0.0.1 define drive auto8mmlib 8mm_tape0 device=mt1.0.0.1 element=82 online=yes define drive auto8mmlib 8mm_tape1 device=mt2.0.0.1 element=83 online=yes /* LIBRARYINVENTORY SCSI AUTO8MMLIB KEV004 1 101*/ /* LIBRARYINVENTORY SCSI AUTO8MMLIB KEV005 3 101*/
Here is an example of the updated device configuration file when an automated library is used at the recovery site to read a database volume DBBK01:
/* Device Configuration */ define devclass auto8mm_class devtype=8mm format=drive mountlimit=2 mountwait=60 mountretention=60 prefix=tsm library=auto8mmlib define library auto8mmlib libtype=scsi device=lb4.0.0.1 define drive auto8mmlib 8mm_tape0 device=mt1.0.0.1 element=82 online=yes define drive auto8mmlib 8mm_tape1 device=mt2.0.0.1 element=83 online=yes /* LIBRARYINVENTORY SCSI AUTO8MMLIB DBBK01 1 101*/
In this example, database backup volume DBBK01 was placed in element 1 of the automated library. Then a comment is added to the device configuration file to identify the location of the volume. Tivoli Storage Manager needs this informatiion to restore the database restore. Comments that no longer apply at the recovery site are removed.
The RECOVERY.VOLUMES.REQUIRED stanza in the recovery plan file identifies the required copy storage pool volumes. The restored server uses copy storage pool volumes to satisfy requests (for example, from backup/archive clients) and to restore primary storage pool volumes that were destroyed. These volumes must be available to the restored server. After the database is restored, you can handle copy storage pool volumes at the recovery site in three ways: