Hi there, first post on the site, I currently have a live application (GT-X7 running on JBoss 3.2.7) which makes use of JGroups for clustering. We have 5 "kernels" involved in the group and both today and yesterday we have experienced a "hang" on the 1st kernel which is the originator of the group – several hours after startup. This hang renders the kernel completely inaccessible, all logging is halted (Garbage Collection halted in the middle of spooling a row), it's as if the kernel has paused itself waiting for something to happen. It would appear in the meantime that the other 4 kernels continue working away. On the first occasion we did not manage to the get the kill -QUIT to output anything before we restarted. Fortunately today on the 2nd occassion we managed to get the kill -QUIT to go through and I will enclose the output below. The reason I come to JGroups is the item at "waiting" on the dump appears to be part of the JGroups setup. - "MessageDispatcher up processing thread" (from google searching) On both occasions I have observed that around 1 minute before the "hung" kernel occurs, another member of the cluster is suspected, removed and then quickly re-introduced to the cluster, from google searching I did see a bit about "simultaneous" kernel exclusion causing a problem, however this doesn't appear "simultaneous". Thanks in advance for any assistance or even advice on areas to focus on – I’m a bit between JVM (Garbage Collection) and JGroups on this one, will be posting questions both here and on the sun site. I’ll post the “dump” first and then snippets from the STD Out Logs which contain the JGroups comings and goings….. Ross View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4130802#... Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&a...