Message Consumer becomes passive

[tonytam]

Hi,

   We are using ActiveMQ version 5.2.0.1 version. At services start up  topics and queues are created and consumers are registered to each one of them.

   One connection is used to register consumer for topics and queues with each one of them having unique session.We have problem in some Linux environment

   with respect to consumers.Initially  all the consumers are registered properly to the destinations(which we verified). After sometime say 5 to 10  minutes, consumers are starting to die   (i.e)Number of consumers for some topics and queues(choosen randonly) are becoming zero. Next time on server startup some other consumers are becoming zero and   this behavior is undefined. This problem does not seem to happen in windows platform.Please suggest what may be the root cause and what is the best way to monitor a consumer  for a queue or topic.

Edited by: tonytam on Feb 19, 2009 2:39 AM

[tonytam]

I can see channel becomes inactive and consumers are getting closed in logs

org.apache.activemq.transport.InactivityIOException: Channel was inactive for too long: /127.0.0.1:41214

     at org.apache.activemq.transport.InactivityMonitor$4.run(InactivityMonitor.java:164)

     at java.util.concurrent.ThreadPoolExecutor$Worker.runTask(Unknown Source)

     at java.util.concurrent.ThreadPoolExecutor$Worker.run(Unknown Source)

     at java.lang.Thread.run(Unknown Source)

2009-02-19 12:21:34,236 DEBUG TransportConnection            - Stopping connection: /127.0.0.1:41214

2009-02-19 12:21:34,236 DEBUG TcpTransport                   - Stopping transport tcp:///127.0.0.1:41214

2009-02-19 12:21:34,237 DEBUG TransportConnection            - Stopped transport: /127.0.0.1:41214

2009-02-19 12:21:34,238 DEBUG TransportConnection            - Cleaning up connection resources: /127.0.0.1:41214

2009-02-19 12:21:34,238 DEBUG AbstractRegion                 - Removing consumer: ID:m1.testqa.thinkpassenger.com-33028-1235074593699-0:22:1:1

2009-02-19 12:21:34,238 [127.0.0.1:41214

org.apache.activemq.transport.InactivityIOException: Channel was inactive for too long

[martinmurphy]

Can you try and disable inactivity monitoring and see if that helps:

tcp://localhost:61616?wireformat.maxInactivityDuration=0

The property is described in the documentation here