I have been trying to find out how I can throtte the cocurrent hits to one of our legacy systems (which is already having some performance problems) by controlling the simultaneous threads which poll messages from a JBoss Messaging gateway queue. The situation is we will be receiving close to 5000 messages on a jms queue (not at all high for my system) and we have 2 nodes in cluster - the destination system is not HA and has only single instance and that system wants to control the simultneous requests my system sends concurrently from all 2 nodes (may be only 10 incoming requests concurrently). I was looking at maxThreads attribute on jms-listener but dont see any documentation around it and default count other than InVM transport. Can somebody help me point to any documentation we may have?
I think I found the default property set to 15 - maxThreads on jms-listener maps to maxSession on activation config.
|maxSession||int||The maximum number of jms sessions that are available to concurrently deliver messages to this mdb||No||15|