High Memory Usage is seen on an RA Agent with some unwanted restarts of the Agent.
This issue occurs if the Java memory assigned to the agent is exceeded.
How can this be prevented?
Sub-Component: RA JD Edwards Agent
Cause type: Configuration
Root Cause: Not enough memory allocated for heap size because too many JDE jobs are running simultaneously.
Increase the Heap Space Memory allocation (parameter -xmx in the agent startup command) so it can be as high as possible and at the very least 2GB (-Xmx2048m).
Performing an upgrade of the Agent, both the RA Framework and the RA JD Edwards solution should also help as they may contain some memory improvements that could fix the problem.
Additionally, if you are reaching a limitation of the concurrent amount of RA JDE jobs that can run simultaneously, please reduce the number including some temporization or a maximum number of concurrent jobs at the Agent or Queue level to avoid the issue.