Richard, That's a good question. We do use IEFYS to write some messages to JESYSMSG but the messages that were missing are WTOs that we send to syslog and to JESMSGLG. Until you asked about IEFYS I hadn't noticed that the JESYSMSG messages did appear correctly in the affected jobs. That means the smf exit was being taken, killing one of my theories.
Since my first email we have figured out that all the problem jobs executed in one JES2 initiator. Stopping and restarting that initiator fixed the problem for now. The affected users all happened to be using the same jobclass and were getting the same initiator consistently. The userid seems to be irrelevant after all. We still don't know why the problem occurred, though. There were no obvious errors in the initiator; no syslog message that we could ask the automation group to react to in the future; no logrec entries. Any ideas about what could have happened to the initiator to affect an smf exit's WTOs? Thanks very much, Ruth -----Original Message----- From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On Behalf Of Richard Peurifoy Sent: Friday, January 18, 2008 3:42 PM To: [email protected] Subject: Re: IEFACTRT exit O'Connor, Ruth wrote: > Hello > > We have an intermittent problem with IEFACTRT and are wondering if > anyone else has ever seen this phenomenon. We're running z/OS 1.7 with > Top Secret 9.0 but the problem goes back at least several years and > software levels. > > Like many shops, we use the exit to write WTO messages at the end of > each batch job step. Every once in a while a user reports that the > messages are no longer being written for his batch jobs, even though > the messages were being produced for that user the previous day (or > previous > hour) and the messages are still being produced for other users' jobs. > The messages don't come back for that user's batch jobs until the next > time we IPL the LPAR. > > We have never been able to figure out what causes this problem. The > type 30, subtype 4, smf records are still being cut for the jobs > without the messages. I have looked through the exit code without > finding anything useful. We're wondering if something prevents the > IEFACTRT exit from being called for certain users. > > This week the problem has been reported in our applications > development LPAR. For at least 3 userids (or ACIDs in Top Secret > terminology) the messages stopped appearing early Monday morning. > There is nothing in logrec at that time and I don't see anything strange in syslog, either. > > A search of the archives turned up one thread about a similar problem > at a DR site but there didn't appear to be any resolution. > > Thanks in advance for any suggestions. > > Ruth O'Connor > Senior Systems Software Specialist > Boston University Sorry, sent this to the newsgroup rather than the list the first time. Are these realy WTO's, or are you using the IEFYS interface to write messages to JESYSMSG (where the allocation messages are written). -- Richard ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
