We are seeing performance issues in TomEE during the peak load. Our
production has 95 PODs with 4 CPU/8GB ram. We have not seen any issues with
the memory, memory pools , CPU utilization etc..and everything is with in
Threshold. One observation from AppDynamics is that the no. of busy threads
in https openssl apr has increased to 150 (from 5) during the peak time and
the application performance has also come down drastically. Currently we
have set maxThread as ‘200’ in the connector configuration. If we increase
the max thread will it solve the issue? We couldn’t take thread dump during
that time from PODs, however we took it after threedays (have not restarted
the PODs) we can see that there are many (more than 100) IO Dispatcher
threads still alive with the following stack trace . with 50% traffic
everything works fine and we have not experienced any performance issue.
This happens when we enabled 100% traffic. Appreciate a quick help on this.
I/O dispatcher 96
threadId:2732 - state:RUNNABLE
stackTrace:
at sun.nio.ch.EPollArrayWrapper.epollWait(Native Method)
at sun.nio.ch.EPollArrayWrapper.poll(EPollArrayWrapper.java:269)
at sun.nio.ch.EPollSelectorImpl.doSelect(EPollSelectorImpl.java:93)
at sun.nio.ch.SelectorImpl.lockAndDoSelect(SelectorImpl.java:86)
- locked sun.nio.ch.Util$3@617b9c99
- locked java.util.Collections$UnmodifiableSet@49e20aac
- locked sun.nio.ch.EPollSelectorImpl@1cebc63e
at sun.nio.ch.SelectorImpl.select(SelectorImpl.java:97)
at
org.apache.http.impl.nio.reactor.AbstractIOReactor.execute(AbstractIOReactor.java:255)
at
org.apache.http.impl.nio.reactor.BaseIOReactor.execute(BaseIOReactor.java:104)
at
org.apache.http.impl.nio.reactor.AbstractMultiworkerIOReactor$Worker.run(AbstractMultiworkerIOReactor.java:588)
at java.lang.Thread.run(Thread.java:748)
Locked synchronizers: count = 0
APR connector settings in server.xml
--
Sent from: http://tomee-openejb.979440.n4.nabble.com/TomEE-Dev-f982480.html
