User development, A new message was posted in the thread "Client hangs when get JMS connection factory": http://community.jboss.org/message/520960#520960 Author : Rakesh Naidu Profile : http://community.jboss.org/people/GoldenHawk Message: -------------------------------------------------------------- Similarly, I've managed to simulate the error as below:   1. Start a client application on a remote computer 2. Abruptly kill this client application (in this case,using the Task Manager) 3. Check the SERVER.LOG generated for this period The moment the application is killed, the marshalling exception appears on the SERVER.LOG. I actually kept the SERVER.LOG scrolling for this experiment & the exception appears exactly when the application is killed. I think this closely resembles the action of disconnecting the network point, where from the server's view, the Client has 'disappeared'. When I check the NonDurableMessage count at this very moment, I can see a spike in the number but it lasts for a few seconds only before returning to 0. I assume all the messages were consumed or dealt with properly by JBREM.   Under a mass-scale environment like the one I'm in (with an excess of 100 nodes to 1 server), its difficult to pin-point which client triggered this exception or how it happened. All we have in the end of the day is the exception in the SERVER.LOG & a whole bunch of unconsumed messages under NonDurableMessages. Is it possible to track back which node caused this issue? At the moment, all I can do is restart the server every time my memory runs low due to the message pile-up. -------------------------------------------------------------- To reply to this message visit the message page: http://community.jboss.org/message/520960#520960