Backup methods for the IMS catalog

Because the IMS catalog is a HALDB PHIDAM database, you can use the IMS image copy utilities to back up the partition data sets of the main database and the secondary index.

Like the recovery of other HALDB databases, the recovery of the IMS catalog does not use image copies of the primary index data set or the indirect list data set (ILDS). Instead, these data sets are recovered by rebuilding them by using the HALDB Index/ILDS Rebuild utility (DFSPREC0).

If the IMS management of ACBs is enabled, the IMS catalog also uses additional system data sets to manage the ACBs:
  • One more IMS directory data sets that contain the ACBs that are active in the IMS system
  • A bootstrap data set
  • A staging data set that stores new or changed ACBs that are not yet active in the IMS system.

The names of these data sets are derived by IMS from the high-level qualifier of the IMS catalog.

These additional system data sets are PDSE data sets and cannot be backed up by using the IMS image copy utilities. Instead, you can rebuild them from the recovered IMS catalog database data sets by using the IMS Catalog Directory Recovery utility (DFS3RU00), which recreates the data sets and loads the active ACBs into the IMS directory. Rebuilding the IMS catalog PDSEs by using this utility ensures that they match the current IMS catalog.

Alternatively, you can backup the ACBs in the IMS directory data sets to an ACBLIB data set by using the IMS Catalog Library Builder utility (DFS3LU00). During recovery, you rebuild the IMS directory from the backup ACBLIB by running the IMS Catalog Populate utility (DFS3PU00) with MANAGEDACBS=SETUP specified on the SYSINP DD statement.

You can also make backup copies of the additional IMS catalog data sets by using your preferred method of copying PDSE data sets. If you use backup copies of the IMS directory that are not created by IMS, you are responsible for ensuring that the copies match the current or recovered IMS catalog database.

Regardless of the method you use to back up the IMS directory, the active ACBs in the backup copy of the IMS directory must match exactly the active ACBs that are in the corresponding image copy of the IMS catalog database data sets that you would use for recovery.

When the IMS image copy utilities are used and the IMS catalog is managed by DBRC, the image copies and the required log data sets are managed by DBRC.

If the IMS catalog is shared between IMS systems, the IMS Catalog must be added to a change accumulation (CA) group so that the logs from all of the IMS systems can be merged before a recovery.