I'm having a real rough time getting Active MQ 4.0 to work property when deployed as a network of brokers using Message Groups. The behaviour even without Message Groups seems highly unstable. I don't even know where to start but I'll give it a shot. Basically every time I try to carry out any manual test, like start two brokers, start client and send message, stop one server, start client etc.. I see a misbehaviour.
I've seen things like a single message group being sent to two different servers, the client hanging, servers not responding ... etc
Build the uploaded sample and give it play around with it and you'll see what I mean.
1. Open two DOS boxes and run two servers by calling "startoptimizerPooled.bat" from dist\bin. First, if brokerName is specified in the configuration, it fails when running on JDK 1.4 with a noclassdef error on a jmx class JMXServiceURL. When using JDK 1.5 and specifying a brokerName, the client hangs when a call is made to the server.
2. Start a client by calling "startclient.bat foobar" from dist\bin. Here "foobar" translated to the Message Group ID. Terminate the client midway, and restart it. Repeat and you'll see that the "foobar" message group is received by both servers.
3. Open a third dos box and start another client and send messages as described in step 2, and you'll notice strange behavior again.
4. When running a "network of broker", after a broker is started and has received a message, when another broker is started, an exception is raised immediately. Try this by opening another DOS box and running startoptimizerPooled.bat.
5. Now try starting a broker on a different machine and repeat above test. Different errors like the ones described in issue https://issues.apache.org/activemq/browse/AMQ-696
start showing up.
Too many things don't seem to be working but hopefully once these issues get sorted out, I'll be able to attain Lingo nirvana.
Bug
Closed
Major
many issue with network of brokers
