| Author |
Message
|
| Segs |
 Posted: Thu Jun 06, 2002 6:33 am Post subject: Monitoring Execution Groups |
 |
|
Voyager
Joined: 04 Oct 2001
Posts: 78
Location: Zurich Financial Services
|
Not sure if this is the correct forum to post this question, if not please advise.
When I was a boy we used Tivoli to monitor MQSI, we used wildcards to look for processes like * DataFlowEngine * <execution group name>
However, this I'm told by the Sys Management people is not possible under Candle. Does anyone have any thoughts or ideas on this using Candle? The execution groups are on a UNIX box running AIX 4.33 ml8
Thanks |
|
| Back to top |
|
 |
| meekings |
| Posted: Thu Jun 06, 2002 8:46 am Post subject: |
|
|
Voyager
Joined: 28 Jun 2001
Posts: 86
Location: UK, South West
|
| I believe the broker monitors its own execution groups, and restarts them if they die. You can check this by kill'ing one of them - you should see entries in the log that the broker is restarting the EG. Is that what you wanted to do? |
|
| Back to top |
|
|
| Segs |
| Posted: Thu Jun 06, 2002 9:07 am Post subject: |
|
|
Voyager
Joined: 04 Oct 2001
Posts: 78
Location: Zurich Financial Services
|
| Thanks, what I'm after is an alert if they can't be restarted by the broker, does anyone monitor execution groups? |
|
| Back to top |
|
|
| kirani |
| Posted: Thu Jun 06, 2002 10:30 am Post subject: |
|
|
Jedi Knight
Joined: 05 Sep 2001
Posts: 3779
Location: Torrance, CA, USA
|
Introduction and Planning manual says this,
| Quote: |
WebSphere MQ Integrator brokers publish event messages, using fixed topics, in
response to configuration changes, state changes, and user actions such as
subscription registrations.
A system management agent can subscribe to these topics, or to a subset of these
topics, to receive the detailed information about activity and state changes in the
WebSphere MQ Integrator broker domain. The event messages have a fixed
structure, defined in XML (Extensible Markup Language).
|
Never monitored ExecutionGroups so far. You can write an application, which parses these XML messages and then sends required data to Candle. I don't know anything about Candle so not sure whether this makes sense or not 
_________________
Kiran
IBM Cert. Solution Designer & System Administrator - WBIMB V5
IBM Cert. Solutions Expert - WMQI
IBM Cert. Specialist - WMQI, MQSeries
IBM Cert. Developer - MQSeries
|
|
| Back to top |
|
|
| mrodel |
| Posted: Fri Jun 14, 2002 1:13 am Post subject: Monitoring |
|
|
Newbie
Joined: 13 Jun 2002
Posts: 5
|
Hi
Unless Patrol for MQSI V2.1 is not released we assure that flows (think this is more important than execution groups) are running with monitoring the throughput on queue basis. means that we monitor via Patrol for MQSeries if the input and output queues of a flow are regularely processed.
Regards
Marisa |
|
| Back to top |
|
|
| dhaksr |
| Posted: Fri Jun 14, 2002 8:39 am Post subject: |
|
|
Apprentice
Joined: 20 Feb 2002
Posts: 37
|
In addition to what mrodel has suggested; you can monitor the
cpu utilization to ensure that broker is not looping. An extended
usage of 100% CPU and non-decreasing Queue depth indicates
a looping problem within broker. |
|
| Back to top |
|
|
| Paul D |
| Posted: Wed Jul 17, 2002 5:28 am Post subject: |
|
|
Master
Joined: 16 May 2001
Posts: 200
Location: Green Bay Packer Country
|
| The Event messages that you can use are in the Problem Determination Guide: Chapter 9 - Sources of Diagnostic Information. The configuration changes events might have what you are looking for. Subscribe to these and make a flow that processes them in the manner needed. I'm taking these and using then in some homegrown monitoring to reflect the status of my message flows on a web page. |
|
| Back to top |
|
|
|
|
