| Author |
Message
|
| Ward |
 Posted: Thu Feb 05, 2004 9:56 am Post subject: SYSTEM.BROKER.ADMIN.QUEUE not serviced |
 |
|
Voyager
Joined: 27 Jun 2001
Posts: 98
Location: Europe
|
Hi,
We have some problems with a WMQI2.1 broker on AIX. The symptoms are :
* Non of the execution groups are doing any work (processes exist though)
* Some execution groups are twice in the process list (ps -ef)
* Core files created and process numbers of execution groups change all the time
* MQSI log shows following error messages:
Feb 5 17:43:12 eaia1 WMQIv210[169528]: (BRMQI001T.IP_140_SOLR_Aux)[3342]BIP2228E: Severe error: /build/S210_P/src/CommonServices/Unix/ImbNamedMutex.cpp 1367 semop Abend File: /var/wmqi/errors/BRMQI001T.IP_140_SOLR_Aux.169528.3342.abend action: abort
* The SYSTEM.BROKER.ADMIN.QUEUE is not serviced
* root puts following messages in the SYSTEM.BROKER.ADMIN.QUEUE :
<Broker uuid="bcc370ec-f000-0000-0080-be2518b3d47a" version="1">
<Report recursive="no">
<AllExecutionGroups/>
</Report>
</Broker>
Since the SYSTEM.BROKER.ADMIN.QUEUE is not serviced I cannot (un)deploy anymore.
I have already stopped the broker; cleared out all SYSTEM.BROKER.* queues and re-cycled the broker database... but nothing seems to help.
Any ideas ?
Regards,
Ward. |
|
| Back to top |
|
 |
| kirani |
| Posted: Thu Feb 05, 2004 11:32 pm Post subject: |
|
|
Jedi Knight
Joined: 05 Sep 2001
Posts: 3779
Location: Torrance, CA, USA
|
It appears that your DataflowEngine process is abending constantly, so it will restart with a different process-id.
Did you try to capture Broker service trace to see what's going wrong? I'd recommend that you clear any outstanding deploy record from Configmgr database.
Was this working before? What CSD level you are at? Did anything change in the system? Did you open a PMR for this problem?
_________________
Kiran
IBM Cert. Solution Designer & System Administrator - WBIMB V5
IBM Cert. Solutions Expert - WMQI
IBM Cert. Specialist - WMQI, MQSeries
IBM Cert. Developer - MQSeries
|
|
| Back to top |
|
|
| Ward |
| Posted: Fri Feb 06, 2004 1:29 am Post subject: |
|
|
Voyager
Joined: 27 Jun 2001
Posts: 98
Location: Europe
|
Thanks for your reply...
ALL DataflowEngine processes are indeed constantly abending.
When trying an mqsichangetrace, the command hangs... so, no I don't have a Broker service trace.
We're at CSD 04.
This is an acceptance test system... rather then going through the possible lengthly procedure of a PMR we will rebuild the environment if no solution found within the first couple of hours...
Is there a way of dropping the broker and recreating it without having to recreate the config manager if I can't deploy the topology after removing the broker?
cheers,
Ward. |
|
| Back to top |
|
|
| Ward |
| Posted: Fri Feb 06, 2004 6:40 am Post subject: Delete broker does not help ! |
|
|
Voyager
Joined: 27 Jun 2001
Posts: 98
Location: Europe
|
Hi,
I deleted the broker, then I removed all references of that broker in the configmanager database. After that I recreated and started the broker.
So far all went well. But then I add the broker to the topology and deployed that topology to my broker. At this point I have the same problem with my broker as before:
Again the DataflowEngine process aborts and restarts all the time (only this time there is only one DataflowEngine of the default execution group)
Again the SYSTEM.BROKER.ADMIN.QUEUE fills with messages from root containing :
<Broker uuid="bcc370ec-f000-0000-0080-be2518b3d47a" version="1">
<Report recursive="no">
<AllExecutionGroups/>
</Report>
</Broker>
Any thoughts ?
Regards,
Ward. |
|
| Back to top |
|
|
| vmcgloin |
| Posted: Fri Feb 06, 2004 7:11 am Post subject: |
|
|
Knight
Joined: 04 Apr 2002
Posts: 560
Location: Scotland
|
Does the content of the abend file(s) in /var/wmqi/errors/ give any clues to the problem?
Does it match anything in a later CSD? For example, there is something similar in CSD05...
IY40186 - SEMOP FAILURE IN IMBNAMEDMUTEX CAUSES DATAFLOWENGINE A
Cheers,
Vicky |
|
| Back to top |
|
|
| Ward |
| Posted: Fri Feb 06, 2004 8:04 am Post subject: |
|
|
Voyager
Joined: 27 Jun 2001
Posts: 98
Location: Europe
|
Bingo !
Upgraded to CSD06 and this solved the problem.
Thanks Vicky.
Regards,
Ward. |
|
| Back to top |
|
|
|
|
