| Author |
Message
|
| Arlene |
 Posted: Fri Nov 02, 2007 3:35 am Post subject: SYSTEM.CLUSTER.TRANSMIT.QUEUE filling up rapidly |
 |
|
Newbie
Joined: 17 May 2006
Posts: 6
Location: UK
|
Hi,
The SYSTEM.CLUSTER.TRANSMIT.QUEUE on my mainframe queue manager is filling up rapidly. This queue manager is part of a cluster, yesterday one of my colleagues stopped the listeners for 2 of the other queue managers within the cluster as he didn't want the queues there accessed any more. The problem seems to be that the m/f queue manager is constantly 'retrying' to start the cluster sender channels attached to the 2 removed yesterday.
My colleague on the Unix side has issued a qmgr suspend for the 2 he removed yesterday.
I still have messages sitting on the S.C.T.Q tho from yesterday when he first made this change, I'm making an assumption that it's the retries on the 2 CLUSSDR channels that is causing rapid fill of this queue...... can someone please offer some assistance as to how I can stop this happening, and how I can remove the msgs from yesterday when the change was first made.
We are running MQ v6.0 |
|
| Back to top |
|
 |
| Gaya3 |
| Posted: Fri Nov 02, 2007 4:32 am Post subject: |
|
|
Jedi
Joined: 12 Sep 2006
Posts: 2493
Location: Boston, US
|
|
| Back to top |
|
|
| Arlene |
| Posted: Fri Nov 02, 2007 4:53 am Post subject: |
|
|
Newbie
Joined: 17 May 2006
Posts: 6
Location: UK
|
Gayathri,
Thanks for this - the only problem I then have is that my m/f qmgr is still 'retrying' to send messages on CLUSSDRA channels to.pps1 and to.pps2, the corresponding receiver channels are on qmgrs which have been suspended from the cluster, how do I stop this 'retry' status |
|
| Back to top |
|
|
| PeterPotkay |
| Posted: Fri Nov 02, 2007 9:46 am Post subject: |
|
|
Poobah
Joined: 15 May 2001
Posts: 7722
|
Well. I posted this reply to your question the list serve:
| Quote: |
Channels don't know if a QM is suspended or not. They don't care. They were probably running when he stopped the listeners, they went into retry as designed and there they will stay until retry attempts are exhausted or you manually stop them or the listeners are restarted. |
But here you provide more info. If you want those 2 QMs out of the cluster they should have been removed properly from the cluster. Stopping their listeners as a first step prevented that.
Can you restart the listeners, let the channels start talking again and the follow the correct procedures for unclustering them?
_________________
Peter Potkay
Keep Calm and MQ On |
|
| Back to top |
|
|
| Arlene |
| Posted: Wed Nov 07, 2007 7:38 am Post subject: |
|
|
Newbie
Joined: 17 May 2006
Posts: 6
Location: UK
|
Peter,
The listeners were re-started and I provided the correct instructions for removal from the cluster.
Thanks for your help
Arlene |
|
| Back to top |
|
|
| PeterPotkay |
| Posted: Wed Nov 07, 2007 11:17 am Post subject: |
|
|
Poobah
Joined: 15 May 2001
Posts: 7722
|
_________________
Peter Potkay
Keep Calm and MQ On |
|
| Back to top |
|
|
|
|
