The user ID is not sufficient because of user ID propagation. We want to segregate scheduler-submitted production work; production work created/submitted by users; and 'other'. Unfortunately, some jobs in the scheduler were 'grandfathered' in without review, and use IEBEDIT or IEBGENER, etc. to write to INTRDR. So it goes like this - job submitted by scheduler with userid APPSCHID (application scheduler ID) "submits" a job via INTRDR. That job also gets the ID of APPSCHID through propagation, and "submits" one or more jobs via INTRDR which also inherit the ID.
So - we need to know the name of the address space that submits the job to know that it came from a batch job and not from the scheduler. This situation also prevents us from using the ID to restrict the job class right now, because they all run under the same ID. Under 2.1, if we can restrict the class via RACF _and_ use WHEN(PROGRAM(scheduler-pgm)) that would solve the problem, I think. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
