We have a job that runs dozens of times per day, copying a circular log file to a GDG backup. On occasion we get something like the following:
IEC331I 042-006(040B0425),INIT ,MICR01PR,SCRT,IGG0CLH0 IEC331I VOL,BK0068,NAME,PROD.XMTIN.MICR.MICRDLY.BKP.G1514V00 The issue does not cause the job itself to fail in any way, or even set a high return code for the job. What seems to be occurring is when z/OS rolls in a new generation of a GDG, and that GDG already has reached it's maximum number of generations, it deletes the oldest generation. Except in this case that delete process fails. The new generation is still rolled in, but the old one remains out there indefinitely. This does not cause any immediate issues, but (if I am recalling correctly), once the G#### name rolls back around (from 9999, I assume, back to 0000 or 0001, and then to the specific number -- 1514 in this example) z/OS attempts to roll in a generation with this same name. Only its already cataloged, so that fails. We currently have our AO software look for IEC331I and when it sees it it sends an email to the development group, who then has to create a job to have the dataset deleted, which is then run by the operators. Rather a pain. I'm wondering if there's a way to actually automate the process to delete the dataset. Has anyone done this? Thanks, Frank ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
