| |
|
RSS Feed - WebSphere MQ Support
|
RSS Feed - Message Broker Support
|
  |
|
| Queue manager abending for lack of storage on z/OS |
« View previous topic :: View next topic » |
| Author |
Message
|
| tleichen |
 Posted: Thu Apr 02, 2009 1:05 pm Post subject: Queue manager abending for lack of storage on z/OS |
 |
|
Yatiri
Joined: 11 Apr 2005
Posts: 663
Location: Center of the USA
|
Since installing z/OS 1.10 last week, there have been issues on our queue manager. We really did not see the problem occur until the queue manager came crashing down on a weekend. Auto-ops restarted the task successfully, so it was not readily apparent that there were any serious issues..
The problem looks like some kind of memory leak, which is ultimately draining away all the space available to the task. This seems to cause the coupling facility to fail, more often than not, prior to the queue manager shutting down. At first this looked like just a coupling facility problem. But after looking at the logs more, it looks more like the initial problem was the lack of private extended space to the task, which is what is gradually ebbing away over its lifecycle. So, the coupling facility failure looks like more of a symptom rather than a cause.
Are there changes in the region size that MQ needs to run in with z/OS 1.10 compared to earlier versions and how should it be reconfigured?
_________________
IBM Certified MQSeries Specialist
IBM Certified MQSeries Developer |
|
| Back to top |
|
 |
| bruce2359 |
| Posted: Thu Apr 02, 2009 1:31 pm Post subject: |
|
|
Poobah
Joined: 05 Jan 2008
Posts: 9469
Location: US: west coast, almost. Otherwise, enroute.
|
Any details would help here. What abend code when the qmgr came crashing down? What did you (they) do in response?
A qmgr instance needs enough virtual storage? How much is allocated in the // EXEC statement in the startup proc?
//stepname EXEC PGM=CSQYASCP,REGION=0M,MEMLIMIT=2G is a nice place to start.
z/OS 1.10? Are you upgrading any time soon?
_________________
I like deadlines. I like to wave as they pass by.
ב''ה
Lex Orandi, Lex Credendi, Lex Vivendi. As we Worship, So we Believe, So we Live. |
|
| Back to top |
|
|
| ctefehinoz |
| Posted: Thu Apr 02, 2009 6:18 pm Post subject: |
|
|
Apprentice
Joined: 27 Oct 2003
Posts: 29
Location: Australia
|
tleichen,
Bit worrying since my shop is on Z/OS 1.09 at the moment.
Do the CSQY220I messages from the MSTR indicate runaway usage by MQ? Can you access the pre Z/OS1.10 syslog's to check prior usage?
Ctefehinoz |
|
| Back to top |
|
|
| gbaddeley |
| Posted: Mon Apr 06, 2009 1:33 am Post subject: Re: Queue manager abending for lack of storage on z/OS |
|
|
Jedi Knight
Joined: 25 Mar 2003
Posts: 2538
Location: Melbourne, Australia
|
| tleichen wrote: |
| ...The problem looks like some kind of memory leak, which is ultimately draining away all the space available to the task. This seems to cause the coupling facility to fail, more often than not, prior to the queue manager shutting down. At first this looked like just a coupling facility problem. But after looking at the logs more, it looks more like the initial problem was the lack of private extended space to the task, which is what is gradually ebbing away over its lifecycle. So, the coupling facility failure looks like more of a symptom rather than a cause. |
This seems to be quite serious. Are your systems programmers and performance management people investigating it? They should be able to identify which area of memory is growing. Have you reported the problem to IBM?
_________________
Glenn |
|
| Back to top |
|
|
|
|
|
| |
|
Page 1 of 1 |
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
|
|
|
