| |
|
RSS Feed - WebSphere MQ Support
|
RSS Feed - Message Broker Support
|
  |
|
| SYSTEM.CLUSTER.TRANSMIT.QUEUE problem |
« View previous topic :: View next topic » |
| Author |
Message
|
| vmcgloin |
 Posted: Thu Jun 20, 2002 4:45 am Post subject: SYSTEM.CLUSTER.TRANSMIT.QUEUE problem |
 |
|
Knight
Joined: 04 Apr 2002
Posts: 560
Location: Scotland
|
Hi,
Can anyone tell me why messages should be building up on a SYSTEM.CLUSTER.TRANSMIT.QUEUE when the destination queues/qmgrs are available and everything seems to be defined OK?
The cluster has worked fine in the past, only has 1 repository, and does not show any errors in the qmgr logs.
Two routes within the cluster seem to be affected (one is AIX-AIX qmgrs, the other is AIX-OS/390).
The channels end normally, but are not automatically restarted. Over the past few days we have had to restart the Automatic Cluster Sender channels to get the messages moving (even when these already have a running status).
AMQ9545: Disconnect interval expired.
AMQ9001: Channel program ended normally.
I would raise a PMR but the AIX is running with MQ 5.1 CSD05. (OS/390 is running MQ 2.1)
Thanks!
Vicky |
|
| Back to top |
|
 |
| bduncan |
| Posted: Thu Jun 20, 2002 7:09 am Post subject: |
|
|
Padawan
Joined: 11 Apr 2001
Posts: 1554
Location: Silicon Valley
|
Vicky,
Let me make sure I understand. You say that the cluster sender channels end normally, but don't restart. So are you saying that when you manually issue a start channel command on these channels, that the messages start flowing again? Because later in your post you mention that you have to "restart" the channels to get the messages moving, which I interpret as issuing a stop on a running channel, and then a start. This indicates to me that you see messages piling up on the transmission queue even when the status of the sender channel is running. Is this the case? Then I would also like to know how many cluster sender channels are present on each of the queue manager? I'm assuming more than one. Keep in mind that all the cluster sender channels share the same transmission queue. As an experiment, you may want to alter the cluster sender channels such that their DISCINT is 0, so they never timeout. Then see if your problem goes away. This would indicate that as long as the channels are running, everything works, but that something is wrong with the channel initiator or the trigger monitor such that the channels aren't being restarted....
_________________
Brandon Duncan
IBM Certified MQSeries Specialist
MQSeries.net forum moderator |
|
| Back to top |
|
|
| vmcgloin |
| Posted: Thu Jun 20, 2002 7:32 am Post subject: |
|
|
Knight
Joined: 04 Apr 2002
Posts: 560
Location: Scotland
|
Thanks for the reply.
| Quote: |
| This indicates to me that you see messages piling up on the transmission queue even when the status of the sender channel is running. Is this the case? |
Yes, sometimes. I have seen messages piling up both when the relevant channel is inactive and when it is running.
| Quote: |
| how many cluster sender channels are present on each of the queue manager? |
There are 12 qmgrs in the cluster so 11 possible cluster sender channels. Only 8 are generally active. Also we have only seen the problem with 2 of them - looking at the messages on the cluster xmitq we can see where they are heading.
| Quote: |
| As an experiment, you may want to alter the cluster sender channels such that their DISCINT is 0 |
Good idea. We are doing a restart of most of the qmgrs now in the hope that the problem won't reoccur before we upgrade to 5.2 soon. If it does happen again I will try a DISCINT of 0. Since the sender chl's are auto-defined I guess I'll have to set this on the clusrcvr chls.
Cheers |
|
| 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
|
|
|
|
