First , my suggestion to take a thread dump did not mean that I was volunteering to analyze your thread dump for you. Second, there are too many stack traces for jstack interspersed with the jbossas stack traces, which makes this hard to read. In addition, the threads that jstack cannot interpret could be the ones causing the problem, You should have sent a SIGINT to jbossas instead. See http://java.sun.com/developer/technicalArticles/Programming/Stacktrace/. Third, scanning through what you posted, nothing jumps out at me (see the previous paragraph). The "socket accept" threads and "object wait" threads are normal - these usually represent pooled threads waiting for things to do. You can usually ignore these, they are expected. Examples of both of these follow: socket accept example: anonymous wrote : 0xb7fac5fe __libc_accept + 0x3e | 0x8be41229 ???????? | 0xb1153898 * java.net.PlainSocketImpl.socketAccept(java.net.SocketImpl) bci:0 (Interpreted fram | e) | 0xb114db6b * java.net.PlainSocketImpl.accept(java.net.SocketImpl) bci:7 line:384 (Interpreted f | rame) | 0xb114db6b * java.net.ServerSocket.implAccept(java.net.Socket) bci:50 line:450 (Interpreted fra | me) | 0xb114db6b * java.net.ServerSocket.accept() bci:48 line:421 (Interpreted frame) | 0xb114da94 * sun.rmi.transport.tcp.TCPTransport.run() bci:59 line:340 (Interpreted frame) | 0xb114de71 * java.lang.Thread.run() bci:11 line:595 (Interpreted frame) object wait example: anonymous wrote : 0xb1153898 * java.lang.Object.wait(long) bci:0 (Interpreted frame) | 0xb114db6b * java.lang.Object.wait() bci:2 line:474 (Interpreted frame) | 0xb114db6b * org.jboss.resource.connectionmanager.PoolFiller.run() bci:86 line:87 (Interpreted | frame) | 0xb114de71 * java.lang.Thread.run() bci:11 line:595 (Interpreted frame) View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4146606#... Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&a...