Hi David- I have looked into this issue, and I can confirm that I am seeing the same results as you: after an hour or so some of the listeners report as being "stopped".
Looking further into this, I think this is actual correct behaviour. The following seems to be happening. Each queue has multiple sessions each with one listener on it. When there are multiple consumers on a queue then messages are always delivered to the first consumer on the queue unless it is busy, then they will go to the second, the third etc. Therefore if the consumers tend to process them faster than they arrive on the queue then the last consumer on the queue is unlikely to get any to process. I shall run the experiment overnight to make sure, but I don't expect the consumers to stop altogether (if they do that certainly is a problem). There is a task to provide different routing policies (so you can do round robin, random etc) for consumers, but this wouldn't give you any better throughput in your case. JBossMQ has a similar behaviour (see http://wiki.jboss.org/wiki/Wiki.jsp?page=JBossMQReceiverImpl) View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3962581#3962581 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3962581 _______________________________________________ jboss-user mailing list [email protected] https://lists.jboss.org/mailman/listinfo/jboss-user
