OPS0025: Metaspace is out of memory
Pega Predictive Diagnostic Cloud™ (PDC) generates the OPS0025 event when an application cannot allocate an object in the Java Metaspace because the Metaspace ran out of memory.
Reason for the event
When an application cannot allocate an object in the Java Metaspace because the Metaspace has run out of memory, Pega Platform™ generates a java.lang.OutOfMemoryError exception. PDC receives this information, generates an OPS0025 event, and creates a unique Metaspace is out of memory case for each node on which the exception occurs. To learn more about cases, see Cases in Pega Predictive Diagnostic Cloud.
Example message text
Metaspace is out of memory on node <node ID>, which is a <node_type> node.
Recommended next steps
To resolve a Metaspace out of memory issue, perform the following steps:
- If you are a Pega Cloud® Services client, make a request to restart the node by selecting New request in My Support Portal, specifying that this is a Cloud Change request for an OutOfMemoryError exception. For the latest documentation on how to make requests, see My Support Portal: New Design, Streamlined Features. If you have an on-premises system, contact your application support team.
Pega Cloud Services clients can restart the Pega web app tier in their dev/test and staging environments themselves, by clicking the self-service My Pega Cloud navigation icon in their My Support Portal. For more information, see Self-service: Restarting the Pega web app tier of your Pega Cloud environment.
- While the restart is in progress, gather node information in PDC to help diagnose the issue.
- On the landing page, select the system on which the issue occurred.
- On the Health tab, select the node on which the issue occurred.
- On the System Info tab, expand the Node Info section.
- Capture and save all the information that is displayed in the Node Info section.
- Contact Pega Support and provide the information that you gathered in step 2. For an on-premises system, contact your application support team. Java memory tuning and Metaspace monitoring require expert assistance.