Well, there's nothing like being embarsed to get my lazy but workin.
I've allready commited some several changes that should reduce the number of
thread being wasted by JBossMQ. The change will have eliminated all the 38
threads here:
--> 38: org.jbossmq.SpyConnection$1.run(SpyConnection.java:99)
It should also significantly reduce the number of threads at :
--> 122: org.jbossmq.Mutex.waitLock
And I have a feeling that I just took care of these too. There is a new
"INVM" IL (invocation layer) that get stored in JNDI using the
NonSerializable factory. If the ASF stuff is configured to use the
connection factorys of the INVM IL, no extra threads will be generated.
--> 39:
org.jbossmq.distributed.server.DistributedJMSServerOIL.run(DistributedJMSSer
verOIL.java:89
--> 38:
org.jbossmq.distributed.server.ConnectionReceiverOIL.run(ConnectionReceiverO
IL.java:110)
The changes have been commited to the JBossMQ module.. I start working on
integrating with the the rest of the jboss modules so we can test to see If
what I did really did help.
Regards,
Hiram
----- Original Message -----
From: "danch" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Thursday, June 21, 2001 10:57 PM
Subject: Re: [JBoss-dev] JMS thread usage
> Chris' nightly test runs have only been failing with these OutOfMemory
> exceptions for the last couple of nights. Either it was a recent change
> that caused this or a recently added test that brought it into the open
>
> -danch
>
> Hiram Chirino wrote:
>
> > I'm embarased.
> >
> > I'll look into it right away. Since the JMS client and server stuff is
> > running in the same VM when using ASF I should be able to get rid of the
> > socket calls and thus free up the threads that are polling the sockets.
> >
> > I'll also see if I can get rid of the SpySession Threads when using the
> > ASF..
> >
> > Regards,
> > Hiram
> > ----- Original Message -----
> > From: "Scott M Stark" <[EMAIL PROTECTED]>
> > To: "JBoss Dev" <[EMAIL PROTECTED]>
> > Sent: Thursday, June 21, 2001 8:04 PM
> > Subject: [JBoss-dev] JMS thread usage
> >
> >
> >
> >> I have been running the jbosstest suite off and on throughout the day
> >
> > against
> >
> >> a snapshot of main on linux to look into the issue people have reported
> >
> > concerning
> >
> >> running out of processes on linux. I am seeing a steady increase in
> >
> > threads from
> >
> >> 32 for the baseline just after startup to 229 after one interation of
the
> >
> > test suite to
> >
> >> 329 after about 10 iterations of the test suite.
> >>
> >> Clearly were leaking threads and it could be that the test code is not
> >
> > cleaning up
> >
> >> correctly, but I am seeing a large portion of the threads associated
with
> >
> > JMS and I
> >
> >> question why this should be. Out of the 329 threads there are 266 JMS
> >
> > associated
> >
> >> threads.
>
>
> _______________________________________________
> Jboss-development mailing list
> [EMAIL PROTECTED]
> http://lists.sourceforge.net/lists/listinfo/jboss-development
>
_______________________________________________
Jboss-development mailing list
[EMAIL PROTECTED]
http://lists.sourceforge.net/lists/listinfo/jboss-development
