Operator Considerations

When a job is resubmitted to perform a deferred checkpoint/restart (the RESTART parameter is coded on the JOB statement with a checkid operand), the processing is essentially the same as that during an automatic checkpoint/restart after the restart reader has reinterpreted the job. A message is issued to the operator, indicating the virtual storage requirements of the job.

The required virtual storage area can be unavailable for the following reasons:

The REGION size parameter for the step is larger when the job is resubmitted than in the original execution, and the area required is at different addresses than the area available. The parameter REGION=0K causes this type of restart failure.
A new IPL is performed and, because of different IPL options specified by the operator, the area required is larger than the area available.

If these conditions exist, a message is displayed, indicating that virtual storage for the job step to be restarted is unavailable.