| Author |
Message
|
| skydoor |
 Posted: Tue Jul 24, 2007 2:37 am Post subject: Excessive messages between full repositories |
 |
|
Apprentice
Joined: 24 Jul 2007
Posts: 43
Location: Cape Town
|
I have a cluster with two full repositories (QM1 and QM2) and 8 partial repositories. On each of the full repositories I have a cluster sender channel that points to the other full repository which for this purpose I will call TO.QM1 and TO.QM2.
In the Tivoli Enterprise Portal I noticed that the message count per 5 minutes is very high on both the mentioned cluster sender channels. It is in the region of 50000 messages every 5 minutes. The size is +- 3 KB/sec.
My HBINT is set to 300 and my KAINT is set to AUTO on the cluster sender channels so I do not think that is what is causing the high amount of messages being sent.
This is not a problem at the moment but I would like to know what is causing this high amount of traffic and if this is normal for full repositories.
Mike |
|
| Back to top |
|
 |
| skydoor |
| Posted: Tue Jul 24, 2007 2:46 am Post subject: |
|
|
Apprentice
Joined: 24 Jul 2007
Posts: 43
Location: Cape Town
|
Apologies, version below:
C:\IBM\WebSphere MQ\bin>dspmqver
Name: WebSphere MQ
Version: 6.0.2.0
CMVC level: p600-200-060921
BuildType: IKAP - (Production) |
|
| Back to top |
|
|
| jefflowrey |
| Posted: Tue Jul 24, 2007 2:52 am Post subject: |
|
|
Grand Poobah
Joined: 16 Oct 2002
Posts: 19981
|
You know that those same channels are used to send data between your applications, right? That they aren't dedicated to use for repository updates?
You might be seeing all that traffic for application messages.
_________________
I am *not* the model of the modern major general. |
|
| Back to top |
|
|
| skydoor |
| Posted: Tue Jul 24, 2007 3:35 am Post subject: |
|
|
Apprentice
Joined: 24 Jul 2007
Posts: 43
Location: Cape Town
|
| I thought about that but I was not too sure if that happens. However, in this case, except for 5 alias queue managers (shared in the cluster) on each of the full repositories that I use to communicate with 5 of the partial repositories, I do not use the full repositories as anything other than repositories. I had a look at the message count on the alias queue manager's TO.PARTIALS and it adds up to about 300 messages on each of the full repositories. |
|
| Back to top |
|
|
| skydoor |
| Posted: Wed Aug 15, 2007 7:10 am Post subject: |
|
|
Apprentice
Joined: 24 Jul 2007
Posts: 43
Location: Cape Town
|
We are monitoring it at the moment but I think we have found the source of the excessive messages between the Full Repositories.
We had Alias Queue Managers defined on the Full Repositories for our clustered QM's (partials). They were defined with the same names as the clustered QM's.
As soon as we removed the Alias Queue Managers out of the Full Repositories, the excessive messages stopped. There was some type of conflict between the Alias Queue Managers and the Clustered Queue Managers which caused the excessive messages. |
|
| Back to top |
|
|
| PeterPotkay |
| Posted: Mon Aug 20, 2007 1:59 pm Post subject: |
|
|
Poobah
Joined: 15 May 2001
Posts: 7722
|
I think you had an MQ message (or multiple) rattling back and forth between the 2 QMs as fast as MQ could move it. Because of these particular Aliases and the way you had them defined, you had an endless loop where as soon as the message arrived at QM1, MQ name resolution sent it to QM2, where MQ name resolution sent it to QM1, where name resolution sent it to QM2, etc.
Once removed, the message(s) probably went to the destination queue after logging some serious milage! 
_________________
Peter Potkay
Keep Calm and MQ On |
|
| Back to top |
|
|
|
|
