If I am not wrong, you need to add some Bean for this..Managed Bean..I think
the experts will help you more on this..
Actually, just google on this ...
Regards,
jd
On Tue, Jul 13, 2010 at 6:08 PM, Jari Juslin <zds(a)iki.fi> wrote:
Hello,
everybody.
I have a problem with JBoss leaking memory over repeated re-deployments
of same ear file. Obviously the ear file changes slightly between the
re-deploys, but name is the same, so it should replace the old one
entirely on each re-reploy.
Over time, after dozen or so hot-re-deploys, JBoss stops to
OutOfMemoryError at deploy time. To trace the issue I captured full
memory dump and analyzed it with Ecliple Memory Analysis Tool. Here is a
screenshot from the data it shows:
<
http://zds.iki.fi/zds/temp/jboss_leak_0.png>
And the main leak point as text:
One instance of "org.jboss.mx.server.MBeanServerImpl" loaded by
"org.jboss.classloader.spi.base.BaseClassLoader @ 0x7ffe9d4baa50"
occupies 83 098 176 (21,06%) bytes. The memory is accumulated in one
instance of
"EDU.oswego.cs.dl.util.concurrent.ConcurrentReaderHashMap$Entry[]"
loaded by "org.jboss.bootstrap.NoAnnotationURLClassLoader @
0x7ffe9c604bb8".Keywords
org.jboss.bootstrap.NoAnnotationURLClassLoader @ 0x7ffe9c604bb8
org.jboss.classloader.spi.base.BaseClassLoader @ 0x7ffe9d4baa50
org.jboss.mx.server.MBeanServerImpl
EDU.oswego.cs.dl.util.concurrent.ConcurrentReaderHashMap$Entry[]
Looking further down the object tree, I get drown to sea of interlinked
HashMaps. It would seem the memory leak, if it exists, is about having
lots and lots of internal bookkeeping information gathered in small
pieces to interlinked network of object references.
So, the question is, what can I do to this?
-Jari
_______________________________________________
jboss-user mailing list
jboss-user(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/jboss-user