Confluence Crashes Due to CodeCache in JVM Becoming Full

Still need help?

The Atlassian Community is here for you.

Ask the community

Symptoms

The following appears in the atlassian-confluence.log:

Caused by: java.lang.VirtualMachineError: out of space in CodeCache for adapters

Diagnosis

The instance will have crashed, locked up or frozen, but once restarted the instance will continue to perform normally until the next freeze/hang/crash.

Cause

The reason this occurs isn't clear, but the problem is that the CodeCache of the JVM becomes full and as a result the JVM locks up and the instance will crash.

Resolution

  • Add the following argument to the Java startup options by following these instructions.

    -XX:+UseCodeCacheFlushing
  • Restart the application to initialise the new configuration.

Last modified on Nov 2, 2018

Was this helpful?

Yes
No
Provide feedback about this article
Powered by Confluence and Scroll Viewport.