HALDB partition reorganization numbers

IMS assigns and maintains a reorganization number for each partition to ensure the integrity of data across database reorganizations.

IMS also uses the reorganization number in the HALDB self-healing pointer process after reorganizations of HALDB partitions that use either logical relationships or secondary indexes.

The reorganization number is stored in the following places in each partition:

Attention: If the reorganization number of a partition becomes corrupted, future reorganizations or modifications of the partitions in the HALDB database might produce duplicate segment ILKs and data will be lost.

Reorganization numbers can become corrupt if the HALDB reorganization number verification function is not enabled and either a reorganization fails to increment the reorganization number of a partition correctly or a segment that has a low reorganization number in its EPS is moved into a partition and lowers the reorganization number of the destination partition.

A corrupt reorganization number is difficult to detect. If you do not use logical relationships or secondary indexes, a corrupt reorganization number does not cause any immediate problems. However, if you later add either logical relationships or secondary indexes to a HALDB database that has a corrupt reorganization number, you are likely to lose data.

To ensure the consistency of partition reorganization numbers, enable the HALDB reorganization number verification function. The HALDB reorganization number verification function records the reorganization number of each partition in the RECON data set and ensures that reorganization numbers are always incremented properly. When enabled, the HALDB reorganization number verification function applies to all HALDB databases that are recorded in the RECON data set.

To enable the HALDB reorganization number verification function:

When you enable the HALDB reorganization number verification function, the reorganization numbers for all HALDB partitions in the RECON data set are reset to zero. Accessing a record in each partition updates the RECON data set with the current reorganization number that is stored in each partition.