This may be related to my other post (http://www.jboss.org/modules/bb/index.html?module=bb&op=viewtopic&t= ), but I'm not sure, so I'm posting it separately:
Whenever I attempt to shut down a clustered instance of JBoss 3.2.2RC1 that is running on my HP/UX server using kill , it hangs when it gets to the point of shutting down the HANamingService. In order to fully stop the JBoss instance, I have to resort to kill -9 . When I turn up the logging to DEBUG level, it shows that JBoss is hanging when attempting to stop the AutomaticDiscovery component of the HANamingService.
What would cause this to hang? Looking at the source code for HANamingService, the only call that could be blocking would be the socket.leaveGroup(group) call in the stop() method of the AutomaticDiscovery inner class. What would cause leaveGroup() to hang?
this has never been a problem for me, I suspect you have a JVM issue, please check with HP.