100% CPU usage from netty epoll

[Alan Willemesn]

I am running the Web Call Server (5.0.2940) locally and running into a problem where CPU usage stays at 100% even if there are no connections.
I run into this issue running on both ubuntu 16.04 (with OpenJDK Runtime Environment (build 1.8.0_151-8u151-b12-0ubuntu0.16.04.2-b12) and CentOS Linux release 7.4.1708 (with OpenJDK Runtime Environment (build 1.8.0_161-b14))

I used jstack to find the threads using up the cpu and found that it is netty causing the problem.

"pool-39-thread-1" #159 prio=5 os_prio=0 tid=0x00007f422c2d0e00 nid=0xa6d2 runnable [0x00007f421af4c000]
java.lang.Thread.State: RUNNABLE
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 <0x00000000c0348d78> (a sun.nio.ch.Util$3)
- locked <0x00000000c0348d60> (a java.util.Collections$UnmodifiableSet)
- locked <0x00000000c08c6d90> (a sun.nio.ch.EPollSelectorImpl)
at sun.nio.ch.SelectorImpl.select(SelectorImpl.java:97)
at org.jboss.netty.channel.socket.nio.SelectorUtil.select(Unknown Source)
at org.jboss.netty.channel.socket.nio.AbstractNioWorker.run(Unknown Source)
at org.jboss.netty.channel.socket.nio.NioWorker.run(Unknown Source)
at org.jboss.netty.util.ThreadRenamingRunnable.run(Unknown Source)
at org.jboss.netty.util.internal.DeadLockProofWorker$1.run(Unknown Source)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)​

It looks like it is a known issue with older versions of netty.
Is there a workaround for this issue?

 

[Max]

Is there way to reproduce the issue?
You can also try Oracle JDK, latest 1.8.x
Here you can find installation instructions:
https://flashphoner.com/docs/wcs5/w...eparing_for_installation-jdk_installation.htm