| Author |
Message
|
| zekehb |
 Posted: Mon Jul 28, 2003 1:13 pm Post subject: Queues defined in repos1 not flowing to repos2 not sure why? |
 |
|
Newbie
Joined: 28 Jul 2003
Posts: 4
|
I have 5 queues defined in qm1 which is a full repository and I have all the connections between it and another qm2 full repository working. but after I defined these new queues they did not flow to the other repository and did not even go when I recycled both qmanagers. I do not understand what is (not) happening. Can anyone shed some light on this for me. (by the way) these qmanagers are both running on MVS host machines. thanks for anything you can provide.  |
|
| Back to top |
|
 |
| bduncan |
| Posted: Mon Jul 28, 2003 4:54 pm Post subject: |
|
|
Padawan
Joined: 11 Apr 2001
Posts: 1554
Location: Silicon Valley
|
When you defined the queues, did you set the QCLUSTER attribute to the name of the cluster?
Just because the queue manager is a member of the cluster doesn't mean that all of its queues automatically are.
_________________
Brandon Duncan
IBM Certified MQSeries Specialist
MQSeries.net forum moderator |
|
| Back to top |
|
|
| zekehb |
| Posted: Thu Jul 31, 2003 6:46 am Post subject: |
|
|
Newbie
Joined: 28 Jul 2003
Posts: 4
|
| Yes I have the cluster name in the definition of the queue. It is defind the same way as I have in a test system and that one seemed to work. I just donot see what I might have forgotten in the production version of this. Any other ideas? |
|
| Back to top |
|
|
| bduncan |
| Posted: Fri Aug 01, 2003 8:44 am Post subject: |
|
|
Padawan
Joined: 11 Apr 2001
Posts: 1554
Location: Silicon Valley
|
Well, you say the messages aren't flowing, but where are they building up? Are they all getting stuck on the SYSTEM.CLUSTER.TRANSMIT.QUEUE? In that case it sounds like you have a communications problem, even though your original post seemed to indicate that the channels were running properly.
Is the queue manager generating FDCs or putting entries in any of the three error logs?
_________________
Brandon Duncan
IBM Certified MQSeries Specialist
MQSeries.net forum moderator |
|
| Back to top |
|
|
| wmqiguy |
| Posted: Thu Aug 14, 2003 11:31 am Post subject: |
|
|
Centurion
Joined: 09 Oct 2002
Posts: 145
Location: Florida
|
You might have to convert my distributed language to mainframe, but here is a shot in the dark. We were having problems with the cluster definitions being propagated to a particular QMGR.
The symptom was that messages were piling up on the SYSTEM.CLUSTER.COMMAND.QUEUE.
Root Cause: The OS was running into resource issues and shutting down the amqrrmfa process. Once we restarted that particular process (more than once), the definitions were eventually propagated succesfully. We have also added this process into our monitoring stack.
I doubt this is your exact problem, but hope it sparks ideas.
Good Luck!  |
|
| Back to top |
|
|
|
|
