mina-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Emmanuel Lecharny (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (DIRMINA-1111) 100% CPU (epoll bug) on 2.1.x, Linux only
Date Fri, 17 May 2019 16:12:00 GMT

    [ https://issues.apache.org/jira/browse/DIRMINA-1111?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16842296#comment-16842296
] 

Emmanuel Lecharny commented on DIRMINA-1111:
--------------------------------------------

Also we would need the complete stack trace.  The part that is shown might be a red-herring:
just because a thread is marked RUNNABLE does not mean it's actually doing something. When
the call is to a native method, the JVM can't know if the method is doing something, so by
default tag it as RUNNABLE.

> 100% CPU (epoll bug) on 2.1.x, Linux only
> -----------------------------------------
>
>                 Key: DIRMINA-1111
>                 URL: https://issues.apache.org/jira/browse/DIRMINA-1111
>             Project: MINA
>          Issue Type: Improvement
>    Affects Versions: 2.1.2
>            Reporter: Guus der Kinderen
>            Priority: Major
>
> We're getting reports [reports|https://discourse.igniterealtime.org/t/openfire-4-3-2-cpu-goes-at-100-after-a-few-hours-on-linux/85119/13]
of a bug that causes 100% CPU usage on Linux (the problem does not occur on Windows). 
> This problem occurs in 2.1.2. but does _not_ occur in 2.0.21.
> Is this a regression of the epoll selector bug in DIRMINA-678 ?
> A stack trace of one of the spinning threads:
> {code}"NioProcessor-3" #55 prio=5 os_prio=0 tid=0x00007f0408002000 nid=0x2a6a runnable
[0x00007f0476dee000]
>    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 <0x00000004c486b748> (a sun.nio.ch.Util$3)
> 	- locked <0x00000004c486b738> (a java.util.Collections$UnmodifiableSet)
> 	- locked <0x00000004c420ccb0> (a sun.nio.ch.EPollSelectorImpl)
> 	at sun.nio.ch.SelectorImpl.select(SelectorImpl.java:97)
> 	at org.apache.mina.transport.socket.nio.NioProcessor.select(NioProcessor.java:112)
> 	at org.apache.mina.core.polling.AbstractPollingIoProcessor$Processor.run(AbstractPollingIoProcessor.java:616)
> 	at org.apache.mina.util.NamePreservingRunnable.run(NamePreservingRunnable.java:64)
> 	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
> 	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
> 	at java.lang.Thread.run(Thread.java:748)
>    Locked ownable synchronizers:
> 	- <0x00000004c4d03530> (a java.util.concurrent.ThreadPoolExecutor$Worker){code}
> More info is available at https://discourse.igniterealtime.org/t/openfire-4-3-2-cpu-goes-at-100-after-a-few-hours-on-linux/85119/13



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Mime
View raw message