Application in Sun app server hangs after processing a few messages

Appliation server Sun appserver 8.1 JVM - java 1.5.0.11

We have a JMS receiver processing messages deployed in Sun application Server 9.x After processing about 50 odd messages the appserver hangs. Upon restart the 50 odd messages gets processed and hangs again.

No exceptions / error are thrown (We modified the code to catch Throwable and log in severe mode). Hence the bhavior cannot be attributed to code or messages as they get re-processed.

Would deeply appreciate any f/b suggestions I could not load the entire stack, is there a way to a开发者_StackOverflow中文版ttach a file?

Attached is JSTACK o/p

regds, Chiths

Thread t@139: (state = BLOCKED)
 - com.sun.httpservice.spi.HttpService.stop() @bci=0, line=286 (Interpreted frame)
 - com.sun.enterprise.web.HttpServiceWebContainer.stopHttpService() @bci=16, line=1080 (Interpreted frame)
 - com.sun.enterprise.web.HttpServiceWebContainer.stopInstance() @bci=24, line=913 (Interpreted frame)
 - com.sun.enterprise.web.HttpServiceWebContainerLifecycle.onShutdown() @bci=9, line=62 (Interpreted frame)
 - com.sun.enterprise.server.ApplicationServer.onShutdown() @bci=55, line=414 (Interpreted frame)
 - com.sun.enterprise.server.Shutdown.startShutdown(com.sun.enterprise.admin.event.ShutdownEvent) @bci=3, line=41 (Interpreted frame)
 - com.sun.enterprise.admin.event.AdminEventMulticaster.invokeShutdownEventListener(com.sun.enterprise.admin.event.ShutdownEventListener, com.sun.enterprise.admin.event.ShutdownEvent) @bci=7, line=1445 (Interpreted frame)
 - com.sun.enterprise.admin.event.AdminEventMulticaster.handleShutdownEvent(com.sun.enterprise.admin.event.AdminEvent) @bci=56, line=1434 (Interpreted frame)
 - com.sun.enterprise.admin.event.AdminEventMulticaster.processEvent(com.sun.enterprise.admin.event.AdminEvent) @bci=247, line=443 (Interpreted frame)
 - com.sun.enterprise.admin.event.AdminEventMulticaster.multicastEvent(com.sun.enterprise.admin.event.AdminEvent) @bci=37, line=139 (Interpreted frame)
 - com.sun.enterprise.admin.server.core.channel.AdminChannelServer.sendNotification(byte[], com.sun.enterprise.admin.event.AdminEvent) @bci=52, line=90 (Interpreted frame)
 - sun.reflect.NativeMethodAccessorImpl.invoke0(java.lang.reflect.Method, java.lang.Object, java.lang.Object[]) @bci=0 (Compiled frame)
 - sun.reflect.NativeMethodAccessorImpl.invoke(java.lang.Object, java.lang.Object[]) @bci=87, line=39 (Compiled frame)
 - sun.reflect.DelegatingMethodAccessorImpl.invoke(java.lang.Object, java.lang.Object[]) @bci=6, line=25 (Compiled frame)
 - java.lang.reflect.Method.invoke(java.lang.Object, java.lang.Object[]) @bci=111, line=585 (Compiled frame)
 - sun.rmi.server.UnicastServerRef.dispatch(java.rmi.Remote, java.rmi.server.RemoteCall) @bci=246, line=294 (Compiled frame)
 - sun.rmi.transport.Transport$1.run() @bci=23, line=153 (Interpreted frame)
 - java.security.AccessController.doPrivileged(java.security.PrivilegedExceptionAction, java.security.AccessControlContext) @bci=0 (Interpreted frame)
 - sun.rmi.transport.Transport.serviceCall(java.rmi.server.RemoteCall) @bci=163, line=149 (Interpreted frame)
 - sun.rmi.transport.tcp.TCPTransport.handleMessages(sun.rmi.transport.Connection, boolean) @bci=185, line=460 (Compiled frame)
 - sun.rmi.transport.tcp.TCPTransport$ConnectionHandler.run() @bci=685, line=701 (Interpreted frame)
 - java.lang.Thread.run() @bi=11, line=595 (Interpreted frame)

The thread is blocked which is impossible to analyze without context.

Consider attaching with jvisualvm to investigate better. It can also detect a deadlock.