We met OpsCenter agent out of memory issue before and increased -Xmx in datastax-agent-env.sh to 2048M according to Datastax doc.

It has been running well for past few month, recently two agents has out of memory issue again, during running cleanup outdated backup files.

Could we increase -Xmx again? Maybe change it to 3072M?


1 Answer 1


A 2GB heap for an OpsCenter agent is already double the default (1024 MB) and is pretty large. In older versions of OpsCenter, agents were configured with only a 128 MB heap (OPSC-12313).

Yes, you can increase the heap size provided there is sufficient available RAM left on the nodes but this is not going to solve the underlying problem -- you're just hiding in temporarily.

Without knowing anything about your cluster, my best guess is that there are either (a) too many keyspaces, (b) too many tables, or (c) both.

My recommendation is that you analyse the heap dump to determine the root cause so you can fix it for good. If you're not sure how to do that or need assistance then log a support ticket. Cheers!

  • Thanks, we have about 10 keyspaces and less than 100 tables totally, but our backup retention policy is 6 months, lots of small files saved in NFS volume, I guess maybe be related since out of memory happened during cleanup outdated backup files. I will try to analyse the heap dump.
    – kevin
    Sep 2, 2022 at 9:05

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.