>>>>> "JA" == Jim Archer <[EMAIL PROTECTED]> writes:

JA> Were are battling this even as I type...  It seem that if jBoss
JA> hits the operating systems pid (thread) limit for a process, it
JA> throws out of memory exceptions. We just ran some load test
JA> scripts and when we get too many threads, pop goes the weasil.

Yeah.  I've encountered this problem when I build a Unified Messaging
platform on a GNU/Linux architecture.  The voice switch needed a
number of threads per timeslot on the primary rate ISDN cards.  This
means a magnitude of 30xthread_num x 4 (ports) x num_of_cards.  Not
sure what the default process size was.  It's either 256 or 512.  One
can change it by editing a header file and recompiling the kernel
(forgot which file though.  bummer).

I wonder why jboss needs so many threads though.  Can one tweak this
at startup?

JA> I'm looking at two possibilities... Catch the out of memory
JA> exceptions in my code and then sleep for a bit and try again (not
JA> a great olution) or throw more hardware / operating system at it.


JA> Maybe both.

JA> Does anyone know what jBoss should do when the thread limit is
JA> hit?

Hmmm, I don't know why it needs so many threads.  There is very little
activity on the box (dedicated JBoss devel box).  Maybe one should
look at the thread management issue.  I've noticed that on deploy
error and redeploy the number of threads increases drastically
(measured this very primitively via watch 'ps ax | wc -l').

Hope this helps.
-- 
  Nicolai P Guba    http://www.gnu.org         http://www.frontwire.com
                    mailto:[EMAIL PROTECTED]     mailto:[EMAIL PROTECTED]
                    GSM: +44 (0)7909 960 751   DDI: +44 (0)20 7368 9708

_______________________________________________
JBoss-user mailing list
[EMAIL PROTECTED]
http://lists.sourceforge.net/lists/listinfo/jboss-user

Reply via email to