Re: LPAR Frozen with HIGH CPU

[Marcy Cortes]

Oh, yeah,  I had that one during share in march 2005.  Lots of NDMBKs and things fell apart.   If you don't have a dump to know for sure, get current on your RSU level , and then learn how to do one or how to get on the phone with IBM to get one before IPLing next time :)   Now you can tell management the plan!   Marcy Cortes “This message may contain confidential and/or privileged information.  If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose, or take any action based on this message or any information herein.  If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message.  Thank you for your cooperation."   From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of David Pitner Sent: Thursday, September 14, 2006 2:34 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: [IBMVM] LPAR Frozen with HIGH CPU Greetings, We had this exact same problem on our z/VM 5.1 (RSU 0501) LPAR earlier this week. I have had this issue happen a few times, but this was the first time I was able to obtain a dump from the PSW RESTART. I did open an incident with IBM support. I have received a reply from them to apply a few PTFs. They reported that the dump showed some issues with processing NDMBKs. I’m hesitant to post the APAR numbers here in case they don’t fit your situation. If you need them, I can pass them on. I have not applied them yet. I’m considering just doing an RSU upgrade, but I do not know if these APARs are in the latest RSU yet. I hope this information helps!   -Dave.   From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of William Boyer Sent: Tuesday, September 12, 2006 1:58 PM To: IBMVM@LISTSERV.UARK.EDU Subject: LPAR Frozen with HIGH CPU   Last Friday we had a situation where our production LPAR – running z/VM 5.1 with two z/OS guests – hung or rather ran away with the CPUs.  From the test LPAR, we only have two LPARs, I could see using the Performance Toolkit that the production LPAR was running at 97% + of all three CPU’s. The test LPAR, which only has one CPU assigned to it was using the remaining percentage.  When I tried to logon to the production z/VM, I got no response until I got a message the External Security Manager was unavailable.  We run RACF/VM on the system.  Luckily, I had a userid logged on via the ICC however; when I typed a command in I got an immediate NOT ACCEPTED.  I then tried a PA1 to get a CP READ that did work although it was slow.  I was then able to issue an IND commands but could not really tell which processes were consuming the CPUs.  I say processes plural because the only userid with more then one CPU defined is the one z/OS and it only has two CPUs defined.  The LPAR has 3 CPUs total assigned to it.  I tried to force some of the users I saw in the CPU queues but they just hung with a LOGOFF/FORCE pending.   We re-activated the LPAR and everything came up normally.  There was not trace of what happened in dumps or anomalies in any logs either z/VM or z/OS.  Anyone every have something like this occur?  If it occurs again does anyone have some ideas of what to try to figure out what is happening?   Thanks.   William L. Boyer Senior Systems Programmer   ViPSÒ, an Emdeon Company One West Pennsylvania Avenue Baltimore, MD  21204 Office:  410.832.8300 ext. 8419 Fax:     410.832.8327   This message is confidential, intended only for the named recipient(s) and may contain information that is privileged or exempt from disclosure under applicable law.  If you are not the intended recipient(s), you are notified that the dissemination, distribution, or copying of this message is strictly prohibited.  If you receive this message in error or are not the named recipient(s), please notify the sender at either the fax address or telephone number above and delete this message.  Thank you.