As soon as I saw the "reply-to" header (a few minutes ago) I started suspecting that each client would use two queues, one for sending requests to and another for receiving replies on. I was looking around for ways to create unique queues and though perhaps I can use the connection information I get back from CONNECT to build a transient queue name.
Am I right in thinking then that a client might receive messages on both queues? -- Visit <http://tggagne.blogspot.com/>,<http://gagne.homedns.org/> or <http://gagne.homedns.org/~tgagne/> for more great reading.
