Active output jobs and server restarts

If you boot an enterprise server region that shares a catalog with another enterprise server region that is currently running, the boot process flushes the restarted server's active jobs to the Complete queue. As a result, these jobs disappear from each region's Active queue. The flush process only flushes jobs that have the same region name as the newly started region.

When a job is moved from Active to Complete as a result of a server restart, the Completion Code as viewed in the JES Complete queue is set to RFLS. This indicates that the job was flushed as a result of a restart.

To modify this behavior, there is an environment variable that you can set to disable it.

If the environment variable ES_JES_DISABLE_RESTART_FLUSH is set to any value, then a region's jobs are not moved from the Active queue to the Complete queue when the region is restarted.

The recommended way to use this environment variable is to:

Set the ES_JES_DISABLE_RESTART_FLUSH environment variable for all online servers that share the Batch server's catalog.
Do not set the ES_JES_DISABLE_RESTART_FLUSH environment variable for all the Batch server.

This prevents the Online servers from purging the JES Active queue when they are cycled.

Note: When running in a PAC or with the MFDBFH database locking for multiple regions, active jobs are not flushed at region start-up. This is to avoid incorrectly removing jobs that are currently active in another region with the same name as the one starting up. [2]