A great tool within Oracle WebLogic Server is the WLDF ( WebLogic Diagnostics Framework). Before 11g deliverd as an extension, but now integrated into the Console, It delivers a framework for monitoring your entire WebLogic Server with monitoring capabilities through MBeans.Out of the box, the WLDF console delivers some basic views, but if you’d like to extend the monitoring capabilities, here’s a way to do this.
First of all, your WebLogic uses watches to monitor all kinds of things to see how you system is performing, but not all watches are running. In this example, we would like to see how the CPU’s are holding under heavy load during testing.
Our Oracle WebLogic runs on a AIX LPAR p690, version 6 64 bits O/S.
First we need to create a watch to pick up notifications about the state of the CPU.
In the Admin Console, section Domain Structure, click on Diagnostic Modules to create a new Collected Metric
We use the ServerRuntime to create a new Metric:
In the next screen, we select an MBean which can monitor some IBM specific things. These MBeans are delivered with the com.ibm.lang.management package which WebLogic uses to monitor O/S CPU, memory, and other things.
In the next screen, we can select our metrics, for this example we choose all
In the next screen select the instance and click on Finish
Next step, we can create our own view in the WLDF Console
In the Admin Console home, click on Monitoring Dashboard
Add a new view and give it a useful name, nex select the Metric Browser tab. In there you can see the metric you’ve created. Be sure that the watch is active.
Select the package, the instance and the various metrics and drag and drop them into you view….et voila!. You’ve created your own metric page!