| Author |
Message
|
| vennela |
 Posted: Mon Oct 07, 2002 8:54 pm Post subject: FDC with probe id XY324192 |
 |
|
Jedi Knight
Joined: 11 Aug 2002
Posts: 4055
Location: Hyderabad, India
|
The QMGR was down and a few FDC's were cut. The QMGR is up now but we are trying to find out what has happened. Has anybody seen an FDC like this
| PIDS :- 5765B73
|
| LVLS :- 520
|
| Product Long Name :- MQSeries for AIX
|
| Vendor :- IBM
|
| Probe Id :- XY324192
|
| Application Name :- MQM
|
| Component :- GetSubpoolsLock
|
| Build Date :- Mar 27 2002
|
| CMVC level :- p520-CSD04G
|
| Build Type :- IKAP - (Production)
|
| UserID :- 00045557 (mqm)
|
| Program Name :- runmqlsr_nd
|
| Process :- 00397054
|
| Thread :- 00059579
|
| QueueManager :- MQPI0002
|
| Major Errorcode :- xecF_E_UNEXPECTED_SYSTEM_RC
|
| Minor Errorcode :- OK
|
| Probe Type :- MSGAMQ6119
|
| Probe Severity :- 2
|
| Probe Description :- AMQ6119: An internal MQSeries error has occurred
('22
|
| - A system call received a parameter that is not val' from semop.)
|
| Arith1 :- 22 16
|
| Comment1 :- '22 - A system call received a parameter that is
not
|
| val' from
|
---
Venny |
|
| Back to top |
|
 |
| kavithadhevi |
| Posted: Tue Oct 08, 2002 4:40 am Post subject: |
|
|
Master
Joined: 14 May 2002
Posts: 201
Location: USA
|
Hi,
Is this in AIX/UNIX Box.
Go thru the FDC, you can see most of them saying OK but wherever you dont see them theres where the problem is. One thing i have got information is something to do with semaphore in AIX, which is the system level configuration. So better place to start with is to look for some semaphore informations.
Hope this helps.
-- Kavitha  |
|
| Back to top |
|
|
| bower5932 |
| Posted: Tue Oct 08, 2002 10:36 am Post subject: |
|
|
Jedi Knight
Joined: 27 Aug 2001
Posts: 3023
Location: Dallas, TX, USA
|
From an internal site:
The failure was caused by an ipcrm command. It appears someone did an
ipcrm and removed The semaphore we are trying to lock when we have
this failure. |
|
| Back to top |
|
|
| vennela |
| Posted: Tue Oct 08, 2002 10:51 am Post subject: |
|
|
Jedi Knight
Joined: 11 Aug 2002
Posts: 4055
Location: Hyderabad, India
|
Bower,
I too saw this. But the MQ Admin says she cleaned the semaphores and shared memories when the QMGR was down and she did the cleaning before she brought it back.
Any way I will wait to see if this happens again.
---
Venny |
|
| Back to top |
|
|
| tapak |
| Posted: Mon Aug 14, 2006 5:21 am Post subject: |
|
|
Centurion
Joined: 26 Oct 2005
Posts: 149
Location: Hartford,CT
|
Hi Venny,
I am seeing similar errors in MQ on Solaris . Do you remember what you did to resolve this error.
This happens every weekend after the queue manager is shutdown for maintenance .
We are cleaning semaphore and shared memory using ipcrm immediately after the queue manager is down .
Thanks , Deepak |
|
| Back to top |
|
|
| mvic |
| Posted: Mon Aug 14, 2006 6:16 am Post subject: |
|
|
Jedi
Joined: 09 Mar 2004
Posts: 2080
|
| tapak wrote: |
| This happens every weekend after the queue manager is shutdown for maintenance. |
May I ask... What kind of maintenance? Why every weekend?
| Quote: |
| We are cleaning semaphore and shared memory using ipcrm immediately after the queue manager is down. |
May I ask, why you do this?
A couple of extra points: are you using MQ 5.2 ? And is it runmqlsr that gives the error like the above? (NB. this is a thread from 2002!)
Supposing it is runmqlsr... the most likely way to avoid these errors is to end your listeners (indeed, all programs accessing MQ) before removing shared memory. |
|
| Back to top |
|
|
| tapak |
| Posted: Mon Aug 14, 2006 6:50 am Post subject: |
|
|
Centurion
Joined: 26 Oct 2005
Posts: 149
Location: Hartford,CT
|
MQ is shut down for weekily backup .
And the cleanup is done to make sure that ,there wont be an issue with MQStart .
We are using MQ Ver5.2 . How can I check if runmqlsr is throwing this error .
Here is the FDC . It looks like the client program is throwing the error.
+-----------------------------------------------------------------------------+
| |
| MQSeries First Failure Symptom Report |
| ===================================== |
| |
| Date/Time :- Saturday August 12 12:03:46 EDT 2006 |
| Host Name :- racstst08 (SunOS 5. |
| PIDS :- 5765B75 |
| LVLS :- 520 |
| Product Long Name :- MQSeries for Sun Solaris 2 (Sparc) |
| Vendor :- IBM |
| Probe Id :- XY324192 |
| Application Name :- MQM |
| Component :- GetSubpoolsLock |
| Build Date :- Apr 2 2002 |
| CMVC level :- p520-CSD04G |
| Build Type :- IKAP - (Production) |
| UserID :- useriID (modified) |
| Program Name :- programName (modified) |
| Process :- 00003505 |
| Thread :- 00000002 |
| Major Errorcode :- xecF_E_UNEXPECTED_SYSTEM_RC |
| Minor Errorcode :- OK |
| Probe Type :- MSGAMQ6119 |
| Probe Severity :- 2 |
| Probe Description :- AMQ6119: An internal MQSeries error has occurred ('22 |
| - Invalid argument' from semop.) |
| Arith1 :- 22 16 |
| Comment1 :- '22 - Invalid argument' from semop. |
| |
| |
+-----------------------------------------------------------------------------+
Last edited by tapak on Mon Aug 14, 2006 7:31 am; edited 1 time in total |
|
| Back to top |
|
|
| mvic |
| Posted: Mon Aug 14, 2006 7:29 am Post subject: |
|
|
Jedi
Joined: 09 Mar 2004
Posts: 2080
|
| tapak wrote: |
MQ is shut down for weekily backup .
And the cleanup is done to make sure that ,there wont be an issue with MQStart . |
The cleanup is causing the problem, I think.
| Quote: |
| We are using MQ Ver5.2 . How can I check if runmqlsr is throwing this error . |
It's the following text in the FDC header, which would say "runmqlsr" or "runmqlsr_nd" if it was the listener. In your case it's an app called msBLTrade:
| Quote: |
| Program Name :- msBLTrade |
I'd suggest you (a) ensure msBLTrade (and all other apps) ends before attempting to remove any MQ shared memory segments, and (b) unless you're installing a CSD or other fix, you might choose not to remove shared memory and semaphores etc. It should only be necessary to touch this stuff if you're installing a CSD or other fix. |
|
| Back to top |
|
|
| tapak |
| Posted: Mon Aug 14, 2006 7:39 am Post subject: |
|
|
Centurion
Joined: 26 Oct 2005
Posts: 149
Location: Hartford,CT
|
Is there a CSD for MQ52 which fix this . ?
Does MQ53 has the fix for this ? May be upgrading to 53 is a better option .
If the fix has to be done at the client end , I guess we need to stop the client application before mq shutdown or add FAIL_IFQUIESCING in the openOptions . Am I right ? |
|
| Back to top |
|
|
| mvic |
| Posted: Mon Aug 14, 2006 7:44 am Post subject: |
|
|
Jedi
Joined: 09 Mar 2004
Posts: 2080
|
| tapak wrote: |
| Is there a CSD for MQ52 which fix this . ? |
The problem is that shared memory is being removed which an app needs. If you choose not to remove the shared memory, this FDC will not get written.
| Quote: |
| Does MQ53 has the fix for this ? May be upgrading to 53 is a better option . |
Upgrading to MQ v6 would be the best idea. If you remove shared memory being used by an app at v6, it will still complain - or at least it should.
| Quote: |
| If the fix has to be done at the client end , I guess we need to stop the client application before mq shutdown or add FAIL_IFQUIESCING in the openOptions . Am I right ? |
Choosing not to remove shared memory, unless installing CSDs or fixes, should be sufficient to prevent the FDCs appearing. |
|
| Back to top |
|
|
| tapak |
| Posted: Mon Aug 14, 2006 9:34 am Post subject: |
|
|
Centurion
Joined: 26 Oct 2005
Posts: 149
Location: Hartford,CT
|
Thanks Vic for the detailed expanation . Your inputs were very much valuable .
Thanks < Deepak |
|
| Back to top |
|
|
|
|
