JDBC monitoring is a big part of FusionReactor but currently when you install FusionReactor, it does not monitor JDBC connections by default. The installer does not help you to create wrapped data sources, nor does it inform you that you have to configure a wrapped datasource to be able to see JDBC metric data. You must do this manually.
Follow the confluence documentation here:
When you get to step 3 of the above documentation, please enter the wrapped url and new driver class instead (with your server details)
After you have done that, you must follow the steps below.
1. Ensure the J2EE engine can locate, load and use the unwrapped data source.
2. Locate the two new split jars within your /FusionReactor/etc/lib folder. These are named fusionreactor-core.jar and fusionreactor-jdbc.jar.
3. Locate the existing installed fusionreactor.jar file, for example confluence3.5.7/lib/fusionreactor.jar
4. Stop your J2EE (e.g. Tomcat Server) engine.
5. Make a copy of the installed ...confluence3.5.7/lib/fusionreactor.jar and store this safely, in case a rollback is required.
6. Remove the installed ...confluence3.5.7/lib/fusionreactor.jar file and replace it with the fusionreactor-core.jar file.
7. Copy the fusionreactor-jdbc.jar to the same location as the jar containing your unwrapped JDBC driver for example: C:/confluence3.5.7/confluence/WEB-INF/lib
8. Restart the engine and attempt to verify the datasource.
Now when starting Tomcat, you should see Fusion Reactor JDBC: Driver loaded if successful and a load of database queries output to console. The reason for this is Fusion Reactor is started later and doesn't catch the queries until it starts.
|Last Updated:||09/Aug/11 1:01 PM|
Comments are closed.