baOn 05/27/2011 12:43 AM, sujith wrote:
Hi
Please help me. I have configured qpid(version .10) in 2 node cluster using
openais. When I started to send message one node always get down after few
hours or some times after a day.
This is the log when qpidd went down
...
2011-05-26 21:08:44 critical Error decoding events, may indicate a broker
version mismatch: St9bad_alloc
The bad_alloc means you are running out of memory. This could be because
a) your consumers are slower than your producers and you're building up a
backlog of messages on queues.
b) there's a bug in the broker that's using excessive memory.
Don't reply to this mail, instead create a JIRA at
https://issues.apache.org/jira
so we can track the discussion about this issue.
Things to try (include the results in the JIRA):
Add these qpidd options (command line or config file):
--log-enable=info+ --log-enable=debug+:cluster
Attach the full logs from both brokers to the JIRA
Check your queue depths coming up to the crash. You can run "qpid-stat -q"
periodically to check.
Give details of what your clients are doing up to the crash.
Thanks!
Alan.
---------------------------------------------------------------------
Apache Qpid - AMQP Messaging Implementation
Project: http://qpid.apache.org
Use/Interact: mailto:[email protected]
