Red Hat Bugzilla – Bug 1005163
Collection interval change for Message Count metrics on the fly does not work properly
Last modified: 2013-09-06 06:41:07 EDT
Description of problem:
Collection interval change for "Message Count (avg) per Minute" on the fly will break monitoring of this metric and it's value will drop to 0 and it will stay 0 until the RHQ Agent is restarted.
The same behaviour is observed for "Message Count (avg)" - so it's value will drop to 0 and this will be resolved with RHQ Agent restart.
However, if I change the collection period for "Message Count", it will be properly collected but values for "Message Count (avg)" and "Message Count (avg) per Minute" will drop to 0 and will stay 0 until the RHQ Agent is restarted.
Version-Release number of selected component (if applicable):
Steps to Reproduce:
1. Install and set up a SOA-P instance.
2. Start the SOA-P instance.
3. Start JBoss ON Server and RHQ Agent
4. Add a new SOA-P instance to RHQ inventory - RHQ server/agent must have SOA-P 5 plug-in installed.
5. Make sure that the Principal/Credentials are set properly in the Connection Settings page for SOA-P instance
6. Execute $SOA-P_HOME/samples/quickstarts/helloworld sample.
7. Run "ant deploy"
8. Once the JBoss ESB deployment is done and discovered - navigate to SimpleListener Service: JBossAS Servers -> SOA-P instance -> JBoss ESB -> Deployments -> Quickstart_helloworld.esb -> Services -> SimpleListener Service -> Monitoring -> Schedules and set Message Count, Message Count (avg) and Message Count (avg) per Minute to 30 seconds;
9. Run "ant runtest"
10. Wait at least 30 seconds+.
11. Repeat the steps 9. and 10. several times.
12. From left navigation tree, expand and select SimpleListener Service (as above).
13. From right pain, select "Monitoring" tab -> "Tables" and see metrics.
14. From right pain, select "Monitoring" tab -> "Graphs" and see metrics.
15. Metrics are properly updated.
14. Navigate to Monitoring -> Schedules page for "Message Count (avg) per Minute" and change this value to something else (for instance 1 minute);
Value of the "Message Count (avg) per Minute" will drop to 0 and it will stay there.
Value of the "Message Count (avg) per Minute" is collected properly after collection period is changed.
As mentioned above - it seems that RHQ Agent restart can be used as a workaround.