Confluence site suddenly stops working. There is/are [150] thread(s) Value may be stuck

Question

I have a confluence standalone version running on Windows 2012 and SQL in AWS.
The confluence website stops working every day or so.

After a reboot of the instance, the Confluence starts working again.  But then work for just day or so before it needs a reboot.

Just before it stops working I get the below error.

Is there some kind of setting I can tweak to make it running without getting stuck?

I compare a number of logs when it got stuck and found that when there is 150 threads and value may be stuck message the confluence stop working.

There is/are [150] thread(s) in total that is monitored by this Valve and may be stuck.

15-Jul-2019 16:33:52.678 WARNING [ContainerBackgroundProcessor[StandardEngine[Standalone]]] org.apache.catalina.valves.StuckThreadDetectionValve.notifyStuckThreadDetected Thread [https-jsse-nio2-443-exec-148] (id=[2175]) has been active for [66,752] milliseconds (since [7/15/19 4:32 PM]) to serve the same request for [https://help.pexxxx.com.au/rest/quickreload/latest/1477398?since=1563171193340&_=1563172367298] and may be stuck (configured threshold for this StuckThreadDetectionValve is [60] seconds). There is/are [150] thread(s) in total that are monitored by this Valve and may be stuck.
java.lang.Throwable
at com.hascode.confluence.plugin.filter.ApplicatorFilter.doFilter(ApplicatorFilter.java:48)
at com.atlassian.plugin.servlet.filter.DelegatingPluginFilter.doFilter(DelegatingPluginFilter.java:64)
at com.atlassian.plugin.servlet.filter.IteratingFilterChain.doFilter(IteratingFilterChain.java:37)
at com.atlassian.plugin.servlet.filter.DelegatingPluginFilter$1.doFilter(DelegatingPluginFilter.java:58)
at

charleslcso · Answer

My Confluence 7.3.1 has only 1 thread, and it still stuck.

[Catalina-utility-1] org.apache.catalina.valves.StuckThreadDetectionValve.notifyStuckThreadDetected Thread [http-nio-8090-exec-12] (id=[221]) has been active for [60,689] milliseconds (since [5/2/20 7:41 PM]) to serve the same request for [xxx] and may be stuck (configured threshold for this StuckThreadDetectionValve is [60] seconds). There is/are [1] thread(s) in total that are monitored by this Valve and may be stuck.

Stephen Sifers · Answer

Hello,

Thank you for providing what infrastructure you’re running on along with that steps help to resolve the issue. Additionally, the logs you sent are helpful. The type of error, however, is not always clear in what could be the cause. There are a few items which may cause this type of timeout and/or performance issue:

* Check CPU utilization on both your Confluence server and SQL instance.
* Ensuring your Memory utilization is not over-allocated
* Ensuring disks are not queuing and are adequate for your instances

Additional resources to review would be: Performance Tuning

Regards,
Stephen SIfers