| |
|
RSS Feed - WebSphere MQ Support
|
RSS Feed - Message Broker Support
|
| WBIM deploy timeout loop??? |
« View previous topic :: View next topic » |
| Author |
Message
|
| klabran |
 Posted: Thu Feb 23, 2006 9:45 am Post subject: |
 |
|
Master
Joined: 19 Feb 2004
Posts: 259
Location: Flagstaff AZ
|
well installing FP9 for MQ and FP 8 for DB2 8.1 didn't change anything.
Since this was a training server I deleted the cfgmgr and broker and DB2 DB's and then recreated everything and it is working again as expected with a simple flow.
However, I then deployed my known good flow from my two other servers and I received the same results as before.
I was able to correct this by shutting down the QM and the brkr and cfgmgr services, restarting QM and deleting my messages in the system. queues used for deployment.
I then realized that my flow that was blowing up has a sendmail node in it. I added a send mail node to the simple flow that was working and voila it blows up. Remove the node and it deploys. I researched what this could be and discovered that the sendmail plugin lil file that was on this server was the one in the debug folder and not the one in the nt folder of the download. I replaced it and now everything is working! 
I am a little suprised that I never received a message in the IDE or in the event log that tipped me off to something being wrong with the sendmail node..... (I was only getting timeout errors in the IDE and in the event viewer)
Ah well. I am still a happy camper now that I know what was wrong.
Thanks for everyone's help!
Kevin |
|
| Back to top |
|
 |
| rmah |
| Posted: Mon Apr 21, 2008 5:08 pm Post subject: |
|
|
Centurion
Joined: 04 May 2007
Posts: 142
|
| mqmatt wrote: |
OK - look at the depths of the following queues:
a) SYSTEM.BROKER.ADMIN.QUEUE
b) SYSTEM.BROKER.ADMIN.REPLY
c) SYSTEM.BROKER.EXECUTIONGROUP.QUEUE
A message on (a) means that the broker isn't running; on (b) means that the Config Manager isn't running; on (c) suggests a problem with the execution group.
-Matt |
What if there's a message on SYSTEM.BROKER.ADMIN.QUEUE, has 1 Open Input Count, 0 Open Output Count, and the message just sits there for the longest time?
I issued a deploy command to remove a message flow from an exec group, and I noticed this problem.
I get this message from the CMP API:
(4/21/08 6:00:03 PM) DeployResult.getCompletionCode() = initiated
(4/21/08 6:00:03 PM) DeployResult.getLogEntries() : BIP1520I: The Configuration Manager has initiated a deployment operation.
(4/21/08 6:00:03 PM) DeployResult.getCompletionCodeForBroker(<mqbroker_01>) = initiated
(4/21/08 6:00:03 PM)
(4/21/08 6:00:03 PM) <---- cmp.exerciser.ClassTesterForMessageFlowProxy.testDeleteMsgFlow
And it just hangs for the longest time. It's been 30 mins now, and the message flow is still there.
Eventually, the message will be processed and the flow will be deleted. I think. This is based on the fact that earlier, when I restarted broker, there were 19 messages sitting on the system.broker.admin.queue (there are 19 exec groups). None of the exec groups started on the fly. After an hour, the 19 message began to be dequeued, and the exec groups started. Do you think this is a system resource issue? I have 4GB of RAM allocated to our Linux VM. Sigh. 
Thanks 
_________________
MQ 6.0.2.3
Broker 6.0.0.7
for Linux |
|
| Back to top |
|
|
| mqmatt |
| Posted: Tue Apr 22, 2008 2:26 am Post subject: |
|
|
Grand Master
Joined: 04 Aug 2004
Posts: 1213
Location: Hursley, UK
|
This means that your broker is running really, really slowly; the deployment messages are simply taking a long time to process.
Use a system monitor to check to see if any one execution process is hogging your CPU. It may also be a good idea to schedule deployments when the system is not heavily loaded (if you can). |
|
| Back to top |
|
|
|
|
|
|
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
|
|
|
|
