A couple of questions about threading model. First, it seems like (on the client-side) there's a thread created per session. Is this true?
Second, it seems like a session is required per thread. Is that true as well? What I'm trying to accomplish is to implement a long-polling http service. My original approach is to implement an async WCF service that uses the NMS async deliver pattern creating multiple consumers against a single session. Then, when I get a callback from NMS, I then do the plumbing work to finish off the async WCF call. The problem that I'm running into, is that with this approach, I'm not reliably getting the callback. I suspect that it's because one session is required per thread. Using a session per thread may not be possible because NMS's thread creation overhead. Answers to the above questions and suggestions on an implementation approach would be *GREATLY* appreciated. -- View this message in context: http://activemq.2283324.n4.nabble.com/NMS-Threading-model-question-tp3087991p3087991.html Sent from the ActiveMQ - User mailing list archive at Nabble.com.
