For the past week, there has been an increase in CPU usage (95% - 99%) on the instance where we host

[Jose Roberto Nava Morales]

We have a Jira v8.13.4, and for over a week, CPU usage has been reaching up to 99% on our dedicated m6a.4xlarge (16 CPU and  64 GB RAM) instance. These spikes persist, and CPU usage only returns to normal after restarting the service. We haven't found anything unusual in the logs.

While reviewing the logs, we couldn't find any reference to this excessively high consumption, nor the cause behind it.

top

[Nic Brough -Adaptavist-]

Welcome to the Atlassian Community!

This is symptomatic of a memory leak (processes getting allocated memory, but then not releasing it), not performance.  The CPU usage goes up because as less and less space being available for running processes means the CPU has to swap more and more frequently, and eventually, it stumbles to a halt.

You need to be digging through the garbage collection logs and getting thread dumps so you can see what long-running threads you have consuming your memory.

[Dan Breyen]

Hi @Jose Roberto Nava Morales welcome to the community.  I found these articles on Atlassian's site regarding performance on the server: troubleshoot-performance-issues server-optimization performance-troubleshooting-tools 

On Atlassian's jira site, I couldn't find anything specific to this issue. You could check with support, (support.atlassian.com/contact) as I may be missing something.  

I'm not sure if something in the latest code could help you.  Might be another thing to check.