| |
|
RSS Feed - WebSphere MQ Support
|
RSS Feed - Message Broker Support
|
  |
|
| [SOLVED] MQWF not working |
« View previous topic :: View next topic » |
| Author |
Message
|
| Pats21 |
 Posted: Tue Jul 31, 2007 8:26 pm Post subject: [SOLVED] MQWF not working |
 |
|
Disciple
Joined: 08 Sep 2006
Posts: 154
|
Hi All,
We have an issue with MQWF v3.4.
I know that this version is out of support.
However, we are facing an issue with it.
Following is the error message in the fmcsys.log:
*********************************
FMC12240E Execution Server instance(s) stopped with an error
FMC31065E The retry limit for message SDDS has exceeded. The message is stored in the execution server hold queue for later processing.
*********************************
Any idea whats the issue?
Any help appreciated.
Thanks for any help.
Regards,
Patrick
Last edited by Pats21 on Sat Aug 11, 2007 4:29 pm; edited 1 time in total |
|
| Back to top |
|
 |
| Pats21 |
| Posted: Tue Jul 31, 2007 8:45 pm Post subject: |
|
|
Disciple
Joined: 08 Sep 2006
Posts: 154
|
Hi All,
Following is the message that we have browsed for the hold queue.
********************
- System: WEB099
- Message: ExpirationNotifications
- Component: SchedSvr
- Number of failed replays: 0
- Message content: -
NotificationData= {
NumberOfElements=1
-------------------------- element number 1------------
ActInstID=OID(0000000102d7c0010000000000000000),4,OID(0000000104294463000000000
0000000),4
TimeActivated=2006-11-22 17:02:07
NIKind=7 }
********************
Any idea on this.
Pats... |
|
| Back to top |
|
|
| hos |
| Posted: Tue Jul 31, 2007 11:44 pm Post subject: |
|
|
Chevalier
Joined: 03 Feb 2002
Posts: 470
|
Hi,
a (server) message is put onto the Hold Queue when an error prevents its proper execution repeatedly. In most cases the root cause is a database problem, e.g. deadlocks.
You need to find out and fix the problem that prevented the message from proper execution. Check the system log and the error log for further error messages with a matching timestamp. You might need to apply maintenance to your database or start an MQ resource, etc to fix the problem.
Once you have identified and fixed the root cause, replay the messages on the Hold Queue using the administration utility.
There were several fixes in MQWF to prevent messages to be put onto the Hold Queue. So, if possible, apply the latest ServicePac available for MQWF 3.4. |
|
| Back to top |
|
|
| Pats21 |
| Posted: Wed Aug 01, 2007 3:14 am Post subject: |
|
|
Disciple
Joined: 08 Sep 2006
Posts: 154
|
Hi hos,
Thanks for your response.
We have performed DB2 maintenace on the FMC tables
reorg table fmc.CONTAINER;
reorg table fmc.BLOCK_ACT_INST;
reorg table fmc.LOC_W_I_STUBS_SEG;
reorg table fmc.PROCESS_ACT_INST;
reorg table fmc.PROCESS_INST;
reorg table fmc.PROCESS_MODEL_SEG;
reorg table fmc.PROGRAM_ACT_INST;
reorg table fmc.WORK_ITEM;
wanted to ask you a query.
The messages in the hold queue are increasing.
What is the impact of reprocessing these messages?
Are there any side effects?
Do I need to take care of any thing before replaying them?
Also we had 4516 messages in the hold queue. To rectify our problem we had restarted the server. After that there were 28 more msgs added to this queue.
What are these messages for?
FYI, there was a DB2 database lock which had occured which has been rectified.
Thanks,
Pats... |
|
| Back to top |
|
|
| hos |
| Posted: Wed Aug 01, 2007 11:43 pm Post subject: |
|
|
Chevalier
Joined: 03 Feb 2002
Posts: 470
|
Hi Pats,
unfortunately you did not yet hit the important point causing your problems.
You need to find out and fix whatever prevents the messages from being processed!
There is no magic about replaying a message from the Hold Queue. Just replay it and - if the error still persists - it will fail again and be put onto the Hold Queue again. This might give you more insight about the root cause.
Each message on the Hold Queue represents a process instance where MQWF navigation could not proceed. You should be able to monitor the affected process instaces and see them 'hanging' around (e.g. activities remaining in State RUNNING or FINISHED but nothing is going on). In your case it seems to be an activity that never reaches the EXPIRED state. If you 'fix' these hanging process instances manually by using force finish or restart, the related message on the Hold Queue becomes obsolete, because it triggers an event that is no longer valid. Replaying such an obsolete Hold Queue message does not hurt, however. MQWF server just skips (consumes) such an obsolete message.
When you are doing a mixed approach of replaying and manual fixing, the problem is, that you are unable to correlate the messages to the process instances. MQWF 3.6 SP4 has made significant improvements in the usability area of Hold Queue handling. 
Bottom line: in order to get rid of Hold Queue messages you need to find out the problem that is causing them.
GOOD LUCK |
|
| Back to top |
|
|
| Pats21 |
| Posted: Fri Aug 10, 2007 5:23 am Post subject: |
|
|
Disciple
Joined: 08 Sep 2006
Posts: 154
|
Hi All,
First of all ... THANK YOU ALL FOR ALL THE SUPPORT AND HELP ...
Finally we have solved the problem... 
The issue was with the AUDIT_TRAIL table. It had grown to a very large size, hence any subsequent entry into this table was giving problem.
As every flow starts inserting records into this table as soon as it starts any new transactions happening on the flows were not getting logged into the table and very giving problem.
To solve this we took the backup of the Audit_Trail table, renamed it and then created a new table with the same name and structure.
Once we did this, everything worked fine.
hey guys ... do let me know how do we intimate in this forum that the issue is solved or closed.
Regards,
Pats ... |
|
| 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
|
|
|
|
