Hi Jesse, Just to add to Martin's reply that the most frequent cause of OOM on the broker is unconsumed messages backing up there.
What size is your heap ? Are you consuming the messages published ? Regards, Marnie On Fri, Feb 5, 2010 at 1:31 PM, Jesse W. Hathaway <[email protected]>wrote: > I posted this question to the users mailing list, but received no > replies, perhaps the dev subscribers can provide some insight. > > I have been running into an issue where unpredictably, my java > qpid-broker runs out of memory. I tried upgrading to 0.6 to see if it > resolved the issue, but it seems to still be present. > > In an attempt to debug the problem I connected the QPID JMX Management > console to the broker. What I noticed is that when only my synapse > process was running the ActiveConsumerCount did not increase. Yet, when > my jruby JMS client was running the ActiveConsumerCount consistently > increased without ever decreasing. > > Questions: > > 1. What does the ActiveConsumerCount represent? > > 2. Is it possible that the increasing of the ActiveConsumerCount is > causing the broker to exhaust its memory? > > 3. What might be the reason my jruby JMS process is causing this value > to increase? > > thanks, Jesse > > -- > There is no expedient to which man will not > resort to avoid the real labor of thinking. > - Sir Joshua Reynolds > > --------------------------------------------------------------------- > Apache Qpid - AMQP Messaging Implementation > Project: http://qpid.apache.org > Use/Interact: mailto:[email protected] > >
