Yeah, unfortunately without a proper thread dump it is hard to tell where the problem is.
Can you not simulate this in a test or staging environment, by loading one server (which
runs the TcpCacheServer) artificially (maybe even give it less memory so it has to perform
GC more often) and try and lock up the other servers to get a thread dump?
In general though, the TcpCacheServer/TcpDelegatingCacheLoader code is due for an
overhaul:
https://jira.jboss.org/jira/browse/JBCACHE-1262
https://jira.jboss.org/jira/browse/JBCACHE-789
https://jira.jboss.org/jira/browse/JBCACHE-1259
View the original post :
http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4196821#...
Reply to the post :
http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&a...