The version of WebSphere Applications agent is delivered as part of IBM Cloud Application Performance Management V8.1.3.2.

The following enhancements of this agent are available in this release:

  • Linux on Power Little Endian (pLinux LE) is supported.
  • Added support for IBM WebSphere Application Server traditional V9.
  • Added the Memory Analysis dashboard for the diagnostics mode.
  • Added support to use the Application Health Status data set to create event thresholds for application status monitoring.
  • Simplified the manual configuration of data collector.

Memory Analysis dashboard

The Memory Analysis dashboard is a diagnostics dashboard. The JRE used by your application server must be at an eligible level and the diagnostics mode must be available in the first place. You can use this dashboard to check the heap usage and the garbage collector duration during the history period. It gives all the objects that are currently allocated on the heap ordered by size. Check the heap usage information for each object in a heap dump. This gives an idea about the highest heap occupants.

You can access this dashboard by clicking View Memory Analysis on the JVM GC (history) group widget. Three group widgets are provided on this dashboard:

  • Samples by Object: Displays the object allocation information.
  • Top 100 Instance Count by Request Site: Displays the object allocation information for each request site.
  • Allocation Site (Corresponds to selected row): Displays the allocation sites for the selected object and the percentage of heap size that is allocated for each site.

this image shows the memory analysis dashboard

For more information about data availability on this dashboard, see Enabling memory leak monitoring in the IBM Cloud APM Knowledge Center.

Application status monitoring

The Application Health Status data set is inherited from the ITCAM Agent for WebSphere. Now you can use this data set to create event thresholds to monitor the application status and get alerts on the Application Performance dashboard.

The data collection for this data set is disabled by default and no predefined event thresholds are provided. To monitor the application status, you must first enable the data collection by editing the data collector properties file and then define the situation when you want to get alerts by using the Threshold Manager on the Application Performance dashboard. Detailed instructions can be found at Creating event thresholds to monitor application status in the IBM Cloud APM Knowledge Center.

For example, after you have enabled the data collection for this data set, log in to the Cloud APM console to define an event threshold. To get a warning alert when the application whose name contains DayTrader is stopped, you can create an event threshold like this:

this image shows the data set used to define the threshold, which is application health status.

And you set the conditions like this,

this image shows the condition defined for the threshold.

When the condition is true, you will get the warning alert from the Application Performance dashboard.

Simplified the manual configuration of data collector

The procedure of manual configuration is simplified in this release. For WebSphere Applications Server, you only need to add some JVM arguments and variables for the application server on the WebSphere administrative console. For Liberty, you only need to modify three files for the server.

For detailed instructions, see Manually configure the data collector in the IBM Cloud APM Knowledge Center.

Join The Discussion

Your email address will not be published. Required fields are marked *