| Author |
Message
|
| neeha_kashyap120 |
 Posted: Mon Jan 10, 2011 9:45 am Post subject: Cluster issue |
 |
|
Apprentice
Joined: 05 Mar 2009
Posts: 31
|
Hi All
We have a MQ cluster with 4 qmgrs in it. Each of the qmgrs is on a different machine. 2 of the qmgrs are in NorthAmerica and two in Europe. MQ Broker is also on a different NorthAmerica server which has a local instance of all 4 Cluster target queues.
When I send messages to the target queues, messages are going in a round robin fashion only to the two NorthAmerican servers, none to EU servers. The two full repositories of the cluster are one in NA and one in EU. All the queues are NOTFIXED. No netwrok priority set anywhere. Any idea why this is happening.
Does it have something to do with the servers being in Europe?
Thanks
Neeha
_________________
Neeharika Kashyap
IBM Certified System Administrator - WebSphere MQ V6.0
Trying to be IBM Certified System Administrator - WebSphere Message Broker V6.1 and IBM Certified Solution Developer - WebSphere Message Broker V6.1 |
|
| Back to top |
|
 |
| Vitor |
| Posted: Mon Jan 10, 2011 10:01 am Post subject: Re: Cluster issue |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
| neeha_kashyap120 wrote: |
| Does it have something to do with the servers being in Europe? |
Not unless your version of WMQ can read passports! 
I think you could get an entirely North American or European based cluster to misbehave in that what. Use normal cluster problem determination.
(Note also that the queue being NOTFIXED is only a default that an applicaiton can override).
I'd start by checking that all the channels are running & the FRs can see all 4 queues.
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
| bruce2359 |
| Posted: Mon Jan 10, 2011 10:21 am Post subject: |
|
|
Poobah
Joined: 05 Jan 2008
Posts: 9469
Location: US: west coast, almost. Otherwise, enroute.
|
| Quote: |
| Use normal cluster problem determination. |
...as described in the WMQ Clusters manual.
_________________
I like deadlines. I like to wave as they pass by.
ב''ה
Lex Orandi, Lex Credendi, Lex Vivendi. As we Worship, So we Believe, So we Live. |
|
| Back to top |
|
|
| PeterPotkay |
| Posted: Mon Jan 10, 2011 11:20 am Post subject: |
|
|
Poobah
Joined: 15 May 2001
Posts: 7722
|
Assuming everything is otherwise set up correctly...
Lots of posts here on why the volume of App B's messages in a cluster can affect App A's load balancing.
_________________
Peter Potkay
Keep Calm and MQ On |
|
| Back to top |
|
|
| neeha_kashyap120 |
| Posted: Mon Jan 10, 2011 1:24 pm Post subject: |
|
|
Apprentice
Joined: 05 Mar 2009
Posts: 31
|
I checked the setup, everythign seems to be good. I have the same setup in a different environment which works.
I think I will have to take a closer looks and investigate what is happening. Now I know its not the issue of servers being in a different netwrok,
Thanks for the help
Neeha
_________________
Neeharika Kashyap
IBM Certified System Administrator - WebSphere MQ V6.0
Trying to be IBM Certified System Administrator - WebSphere Message Broker V6.1 and IBM Certified Solution Developer - WebSphere Message Broker V6.1 |
|
| Back to top |
|
|
| Vitor |
| Posted: Mon Jan 10, 2011 1:29 pm Post subject: |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
| neeha_kashyap120 wrote: |
| I think I will have to take a closer looks and investigate what is happening. |
Good plan.
| neeha_kashyap120 wrote: |
| Now I know its not the issue of servers being in a different netwrok, |
You said different continents; nothing was said about them not being on the same network! Do you mean that this cluster actually spans 2 TCP/IP networks? If so, this is a good place to start looking; I repeat my advice about checking all the channels are running. If cross-network resolution has failed this will cause problems.
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
| bruce2359 |
| Posted: Mon Jan 10, 2011 1:56 pm Post subject: |
|
|
Poobah
Joined: 05 Jan 2008
Posts: 9469
Location: US: west coast, almost. Otherwise, enroute.
|
| Quote: |
| I checked the setup, everythign seems to be good. |
How exactly did you check the setup?
_________________
I like deadlines. I like to wave as they pass by.
ב''ה
Lex Orandi, Lex Credendi, Lex Vivendi. As we Worship, So we Believe, So we Live. |
|
| Back to top |
|
|
| exerk |
| Posted: Mon Jan 10, 2011 4:11 pm Post subject: |
|
|
Jedi Council
Joined: 02 Nov 2006
Posts: 6339
|
And was it ever working? At a basic level, i.e. the FRs could see each other and the 2 PRs?
_________________
It's puzzling, I don't think I've ever seen anything quite like this before...and it's hard to soar like an eagle when you're surrounded by turkeys. |
|
| Back to top |
|
|
| mvic |
| Posted: Mon Jan 10, 2011 4:30 pm Post subject: Re: Cluster issue |
|
|
Jedi
Joined: 09 Mar 2004
Posts: 2080
|
| neeha_kashyap120 wrote: |
| All the queues are NOTFIXED. |
Besides this, how are the apps opening the queues? They can ignore the "defaults" on the queue definitions, as Vitor has said.
| Quote: |
| No netwrok priority set anywhere. |
Cluster workload balancing goes further than this.
The only way I know to see the variables at work is to use MQ trace (strmqtrc -m QMNAME -t all -t detail as per the MQ info center) and look at the dumped details during the workload balancing logic in the amqzlaa0 processes. |
|
| Back to top |
|
|
| raj429 |
| Posted: Tue Jan 11, 2011 11:20 pm Post subject: |
|
|
Acolyte
Joined: 04 Jul 2009
Posts: 53
|
I will suggest to check..
if the channels are running fine or not?
are the other problematic queues put disabled?
what are the values of CLWLUSEQ values on queue?
are the 4 instaces available in cluster?
etc |
|
| Back to top |
|
|
| John89011 |
| Posted: Wed Jan 12, 2011 10:27 pm Post subject: |
|
|
Voyager
Joined: 15 Apr 2009
Posts: 94
|
| Also, what verision of MQ do you have? |
|
| Back to top |
|
|
|
|
