Adding another fact:
At about the time I switched over to debug level logging this was present on
the env page
freenet.interfaces.PublicNIOInterface$ConnectionShell 219
This is at least 10 times as many ConnectionShellers when compared to what I
am used to see. All the other thread consumers looked normal though. I dont
know if this is causing the CPU usage problem or if the high CPU usage is
causing it. It is abnormal for my node though.
/N
----- Original Message -----
From: "Niklas Bergh" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Tuesday, August 19, 2003 8:56 AM
Subject: Re: [freenet-dev] System hogging with 5026 & 5027
> Hmm.. I forgot to say that this is with unstable CVS of 14 hours back..
>
> /N
>
> ----- Original Message -----
> From: "Niklas Bergh" <[EMAIL PROTECTED]>
> To: <[EMAIL PROTECTED]>
> Sent: Tuesday, August 19, 2003 8:49 AM
> Subject: Re: [freenet-dev] System hogging with 5026 & 5027
>
>
> > Try switch to loglevel debug and compare your results to this..
> >
> > When my node was using 100% CPU (and had been doing so for, say, the
last
> > 8+ hours) I switched over to debug level logging for about a minute or
> so..
> >
> > Here are two messages from two different parts of the log (30 seconds
> > inbetween them as you can see)
> > Aug 19, 2003 8:12:23 AM (freenet.ConnectionHandler, QThread-33368,
DEBUG):
> > Terminating ([EMAIL PROTECTED] for
> > null,null,accumulator:(null)) (not trailing)(reading)
java.lang.Exception:
> > debug
> > <..snip..>
> > Aug 19, 2003 8:12:54 AM (freenet.ConnectionHandler, QThread-34575,
DEBUG):
> > Terminating ([EMAIL PROTECTED] for
> > null,null,accumulator:(null)) (not trailing)(reading)
java.lang.Exception:
> > debug
> >
> > Between those two there are exactly 16368 more instances of 'Terminating
> > ([EMAIL PROTECTED]' (done by an assortment of different
> > QThreads). Am I interpreting the log right when I believe that my node
has
> > tried to establish about 16k connections in the last 30 seconds? Or
maybe
> it
> > is only so that my node has tried to Terminate the *same* (1c17c3a)
> > ConnectionHandler 16k times during those seconds?
> >
> > Attached to all of those log messages above there is this stack trace:
> > at freenet.ConnectionHandler.logDEBUG(ConnectionHandler.java:287)
> > at freenet.ConnectionHandler.terminate(ConnectionHandler.java:2603)
> > at freenet.OpenConnectionManager.put(OpenConnectionManager.java:125)
> > at freenet.ConnectionHandler.registerOCM(ConnectionHandler.java:328)
> > at
> >
>
freenet.interfaces.FreenetConnectionRunner.handle(FreenetConnectionRunner.ja
> > va:133)
> > at
> >
>
freenet.interfaces.PublicNIOInterface$ConnectionShell.run(PublicNIOInterface
> > .java:147)
> > at freenet.thread.QThreadFactory$QThread.run(QThreadFactory.java:214)
> >
> > I have had some trouble figuring out exactly when the problems start
> > occurring, my best estimation up until now has been; a while after
> restart..
> > sometimes one hour, sometimes more. A thought that occurred to me right
> > now.. can it be so that my node runs just fine until it hits its
> > connectionlimit? There are no connection termination going on in the
node
> > until that point...
> >
> > /N
> >
> > _______________________________________________
> > devl mailing list
> > [EMAIL PROTECTED]
> > http://hawk.freenetproject.org:8080/cgi-bin/mailman/listinfo/devl
> >
>
> _______________________________________________
> devl mailing list
> [EMAIL PROTECTED]
> http://hawk.freenetproject.org:8080/cgi-bin/mailman/listinfo/devl
>
_______________________________________________
devl mailing list
[EMAIL PROTECTED]
http://hawk.freenetproject.org:8080/cgi-bin/mailman/listinfo/devl
