Catalina logs and application logs should be monitored regularly for any out-of-memory error messages. If possible, additional tools such as VisualVM should be deployed to collect real-time diagnostic data from the system at the exact time corresponding to these errors.

Heap dumps can also be collected using VisualVM, the Support Subsystem, or the jmap command that is available with Java Development Kit (JDK). If you collect heap dump for a server that has significant memory allocated to it, the server may become non-responsive for up to 30 minutes during the collection. You must collect the heap dump when advised by PTC Support. Heap dumps can be sent to PTC for analysis, or PTC support can help in analyzing the data collected locally. If a heap dump is collected at the right time, it is the most definitive method of identifying contributors to high memory consumptions. However, this requires analyzing the collected data, which is not in the scope of this guide.