I'm trying to figure out this behavior and what I may be doing wrong.
Topic is running on 5.7.0 broker
Multi-threaded consumer on a topic using this connection
tcp\://10.2.100.209\:61616?keepAlive=true,jms.prefetchPolicy.all=0
I'm testing my app under 2 different machines so I have 2 different topic
queues that contain the same messages.
I should mention both these machines are WAN to the broker so 100ms+ round
trip times are quite possible.even over 1 second at times.
I'm doing CLIENT_ACKNOWLEDGE. Receive msg, process it, then acknowledge it.
My situation is that Dispatched Queue Size increments once in a while and
then never decrements below a minimum value EVER.
If I stop and restart the client that queue size stays the same.
But.I have received all the messages and acknowledged them (though like I
said that might've taken a second or more).
Right now both clients show receiving and successfully acknowledging 134
messages and the subscribers tab shows this for one client:
Dispatched Queue Size = 2
Dispatched Counter = 134
Enqueued Counter = 134
Dequeue Counter = 132
And the other client has 0 queue size and 134 Dequeue as it should
I've tried setting the redelivery policy with the thought that 0 means don't
redeliver but I get the same behavior.
ActiveMQConnectionFactory tcf = new
ActiveMQConnectionFactory(BDQueueConnectString);
RedeliveryPolicy policy = new RedeliveryPolicy();
policy.setInitialRedeliveryDelay(10*1000L); // may not need
this if 0 redeliveries works
policy.setMaximumRedeliveries(0);
tcf.setRedeliveryPolicy(policy);
I even tried this:
tcf.setUseRetroactiveConsumer(true);
But other than sending me 1000 messages from the past the Dispatched Queue
Size still stayed at 2.
#1 How do I receive/acknowledge/delete those 2 messages that have been
dispatched again?
#2 How do I stop this behavior where my ack's aren't being processed
correctly apparently.
Michael Black