I have a scenario in which, one of the Managed servers (2 managed servers on different machines ) is utilizing more sapce ( physically : the size od the server/data folder is increasing every minute ) for the same application deployed on to the cluster having only 2 servers on two unix boxes.
Using weblogic server WebLogic Server Version: 10.3.0.0
2 manged servers in cluster across 2 different machines
Please provide some hint on the possible reasons that would invite this kind of server behavior. also if possible some suggestions on trouble shoot and solutions.
please find the thread stack for the server which has this issue.
"Main Thread" waiting for lock weblogic.t3.srvr.T3Srvr@100a44b4 WAITING
I checked it more on the diagnostics part, and it is defenitely the diagnostics that creates this issue with one of my managed server which size upto huge 150GB under the server/data/store/diagnostics folder.
but the other managed server diagnostics data size looks fine .
could you please give me some poniter to get rid of this unwanted diagnostics data getting created for the managed servers.
We first need to understand why the diagnostic information is being collected on that particular server.
Please check what is WLDF watch/rule or collection configured for in your environment.
May be there is an issue on that server due to which it is collecting huge data. If I have further details, I might be able to help.
This configuration was present earlier also, but it looks like the logs statrted accumulating recently due to some other changes made to console. There is nothing coming under the WLDF console extension in server admin console.
Other than changing the start script, will it be fine to go the file location ./server/data/store/diagnostics/WLSDIAGNOSTIC**.DAT and delete the older file while the server is running now?
Incase not posisible , an example on the change in startscript and the config.xml will b helpful
I am not really comfortable suggesting you to delete the WLSDIAGNOSTIC**.DAT files for a running server.
If you have restarted the server and the files were from the previous run then you can delete them.
The JDBC connection pool might have been targeted to both the servers. So, then why the collection is high on only one managed server.
Is it possible that there is some issue with the JDBC Connection Pool on that server?
Can we check and try to fix it. I believe that will be the best way forward.