Hi Lizette, I have been running with the same settings as Mark for years. He suggested them and they were a good starting point that just worked never had to change them as they had the desired result.
These are my current DIAGxx settings on z/OS 1.10 for the morbidly curious. VSM TRACK CSA(ON) SQA(ON) VSM TRACE GETFREE(OFF) VSM CHECKREGIONLOSS(300K,10M) VSM ALLOWUSERKEYCSA(NO) CBLOC VIRTUAL31(IHALCCA,IHAPCCA) VSM USEZOSV1R9RULES(NO) VSM BESTFITCSA(YES) REUSASID(YES) AUTOIPL SADMP(NONE) MVS(LAST) Working on some further procedure/documentation changes before we incorporate AUTOIPL for SAD currently that is only in our test partition but soon will be rolled out. The INIT restarts around Endevor were determined to be due to heavy VIO activity and went away with subsequent z/OS upgrades where some reordering of system cleanup and checking occurred. See the archives as Jim Mulder's posts on this subject were educational. Just do it. Since I put this in S822 ABENDs just went away as an issue we had to deal with. I was able to eliminate schedule INIT recycles. I see a handful of INIT restarts not enough to justify investigating them further. I also put the dishes into the dishwasher without trying to determine who left the dirty glasses in the sink unless it is really out of hand:-) If you don't have an IEFUSI exit consider implementing one. REGION is one of those things that should not be left to application JCL to manager properly. Best Regards, Sam Knutson, GEICO System z Team Leader mailto:[email protected] (office) 301.986.3574 (cell) 301.996.1318 "Think big, act bold, start simple, grow fast..." -----Original Message----- From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf Of Mark Zelden Sent: Tuesday, September 28, 2010 3:57 PM To: [email protected] Subject: Re: Anyone using the DIAG Parm VSM CHECKREGONLOSS yet? On Tue, 28 Sep 2010 12:34:31 -0400, Lizette Koehler <[email protected]> wrote: >We have few S822 abends. I am looking to turn this on using the IBM >defaults for now. > >Is anyone using this parm? Has it been a good, bad, or indifferent >experience? > Been discussed at various times in the past. Mine is set at VSM CHECKREGIONLOSS(300K,10M) to have a 1 size fits all setting. It's mostly the LPARs that run Endevor that had issues in the past (I haven't look at this in years). I recall Sam Knutson trying to work with Endevor support on the orhpan / fragmented storage left in the initiator. >Has this reduced you S822 abends? > Probably a few, but prior to this enhancement and in the days before we used WLM inits, we recycled JES2 inits a few times a day and there was also automation to recycle an INIT anytime after an abend (that code also was changed to work with WLM inits too). This is one of those no-brainer enhancements. It doesn't hurt to set it. Even a "bad setting" doesn't leave you worse off than you were prior to setting it. At worse case, with values too low, you would recycle the initiator after a single job ran in it and cause some extra overhead. If you set the values too high, it's like you haven't done anything at all and would still see some S822 abends. Mark -- Mark Zelden - Zelden Consulting Services - z/OS, OS/390 and MVS mailto:[email protected] Mark's MVS Utilities: http://home.flash.net/~mzelden/mvsutil.html Systems Programming expert at http://expertanswercenter.techtarget.com/ ==================== This email/fax message is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution of this email/fax is prohibited. If you are not the intended recipient, please destroy all paper and electronic copies of the original message. ---------------------------------------------------------------------- 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
