I'm getting the same behaviour as you may see in my post.
I believe this happens to guarantee that only one instance consumes the messages but I'd like to confirm this.
Did you get any new info about it?
Have you had any resolution on this? We're experiencing the same problem in that our queues depend on the DestinationManager which apparently only starts up on the master node.
How are the other nodes in the cluster aware of the various jms-queues if this is the case?
I'm convinced there is a way to resolve this, but I can't find any documentation that will lead me to how our configuration is off.
Yes, I've got a "workaround" on this.
As you can see, this is an expected behaviour since JBoss 3.2.4: http://jboss.org/wiki/Wiki.jsp?page=JBossMQHA
On option to make the service "activate" before the servers synchronizes state (wich takes 1 minute to happen after the master node crashes) is to do a lookup in the desired component, after that, the user will receive a NameNotFoundException but this call will start the activation process and in the subsequent call everything will work fine.
I'm investigating why we get the firs NNFException as the lookup activates the service and we shouldn't get NNFException and if I have news, I can post it here.
Same problem here. Does anyone has a solution for this problem (3.2.5)