Hi, I just wanted to give a head's up about a problem we had. Before I begin let me just say I am not blaming CA in any way. Although I do wish their "IBM APARs" hiper notification would have mentioned MIM 11.5 and 11.6 and I probably would have not run into this issue. I am not blaming IBM either since there was a HIPER APAR and notification via ASAP (if you signed up for that). The HIPER APAR in our case just hadn't gone through internal testing yet and MIM 11.6 was a sysres roll / cycle ahead of the IBM PTF.
The problem I had is an IBM bug but it was not listed in the $IBMAPAR member that CA provided with MIM 11.6 when I ordered it (this member details IBM APARs that are known to cause problems when running MIM). As previously mentioned, it was documented in the MIM hiper notices that are distributed periodically, but MIM 11.6 is not mentioned in that document so I missed it since I was already working on 11.6 and was in the process of rolling it out and there document says it's for MIM 11.5. The problem is described by OA22795 / UA38252. The APAR closed on 11/28/2007 and the official PTF became available on 12/12/2008. I ordered and downloaded MIM 11.6 and all MIM PTFs at the time on 01/09/2008. But since MIM 11.6 is still at base level and that level was cut prior to the IBM APAR being confirmed, the $IBMAPAR member did not have the warning. So here is how the problem bit us: After we rolled out MIM 11.6, all batch jobs were getting a dispatching priority of x'FF'. We actually ran for a couple of weeks on some development LPARs and never noticed because they had 3 or 4 engines on a z9 EC processor. As soon as this hit production LPARs that run at or near 100% busy at peak hours (and one penalty box LPAR with only 2 processors), this of course caused a *HUGE* problem since batch jobs were getting the highest possible priority in the system. We saw JES2 checkpoint lockouts, MIM control file lockouts, ThruPut Manager control file lockouts, HSC CDS lockouts, etc. All the LPARs involved were z/OS 1.8 run MII/MIA/MIC and use WLM controlled inits. Only WLM controlled INITs are involved with IBM APAR OA22795. We also have some other monoplex LPARs that only use MIA and those were not affected. The IBM fix hits the nucleus, so we backed off to MIM 11.5 since an IPL was out of the question. I have since tested 11.6 on my sandbox LPAR again with OA22795 applied and the problem is fixed. BTW, a work around seemed to be a quiesce and resume of the batch job. But I also had to move some of these jobs into a "KILLIT" service class with a resource cap of 1 until I backed off MIM to 11.5. Mark -- Mark Zelden Sr. Software and Systems Architect - z/OS Team Lead Zurich North America / Farmers Insurance Group - ZFUS G-ITO mailto:[EMAIL PROTECTED] z/OS Systems Programming expert at http://expertanswercenter.techtarget.com/ Mark's MVS Utilities: http://home.flash.net/~mzelden/mvsutil.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
