Profile Image
Ajay Pandey

Jboss Stopped responding

I'm not able to find root cause, today our production server again stopped responding. We had to restart jboss for immediate fix and captured thread dump.

Please help me to analyze and find the root cause.

 

Thanks



Report URL - https://fastthread.io/my-thread-report.jsp?p=c2hhcmVkLzIwMjEvMDgvMy8tLWhpZ2gtY3B1LXRkdW1wLm91dC0tOC0zMi0xNw==

  • jboss

  • threaddump

  • rootcause

Please Sign In or to post your comment or answer

Profile Image

Mahesh

Hello Ajay,

 

As per my understanding, the application is using a watch service to monitor directories or changes in a file. In the attached fastThread report, I could see that 42 threads that are related to "Linux Watch Service" are consuming high CPU and all threads are having the same stack trace. This may cause your application to become unresponsive. Can you take a look at the following 2 links:

 

https://stackoverflow.com/questions/24739280/watchservice-uses-100-of-cpu-on-centos

https://github.com/linkerd/linkerd/issues/391

 

Here is a stack trace of thread:

stackTrace:
java.lang.Thread.State: RUNNABLE
at sun.nio.fs.LinuxWatchService.poll(Native Method)
at sun.nio.fs.LinuxWatchService.access$600(LinuxWatchService.java:47)
at sun.nio.fs.LinuxWatchService$Poller.run(LinuxWatchService.java:314)
at java.lang.Thread.run(Thread.java:748)
Locked ownable synchronizers:
- None

Profile Image

Ram Lakshmanan

Hello Ajay!

 You had posted similar question on Jul' 30 - JBoss freezing during peak hours .  We have posted our response earlier. Did you had a chance to investigate it, based on our comments? This thread dump analytics is very similar to earlier one. 

Profile Image

Ajay Pandey

Thanks Ram,

Yes I had posted similar trace earlier, but from the trace it is not clear which component is causing this issue. it seems trace is not captured completely.

Got something else on mind? Post Your Question

Not the answer you're looking for? Browse other questions tagged
  • jboss

  • threaddump

  • rootcause