[rabbitmq-discuss] How to debug queue memory leak

Max E. Kuznecov syhpoon at gmail.com
Fri Jan 25 14:59:08 GMT 2013


In my application I have a pool of temporary queues, used for RPC-like 
Queues are created with flags durable and exclusive.

I'm observing a constant growth in queues memory as the time passes despite 
the fact rabbitmqctl shows zeros in messages column.
The output looks like this:

rabbitmqctl list_queues memory name messages consumers | sort -rn | head
10772176        amq.gen-qnA00cF1K0o2vq1KyO9_yQ  0       1
10772104        amq.gen-os12dLIRtGJYqEwG7xpZOA  0       1
10771888        amq.gen-YJkYswabMY0eUQi057xdHA  1       0
10771888        amq.gen-X8_I9wVBb6H2bHb_yt5zrw  0       0
10771888        amq.gen-wLAXnQjySs9QgVlaZOlXaw  0       0

And this number just continues to grow until memory high watermark is hit.
Maybe I'm missing something but how is this possible? Message rate is about 
50 msg/s, it is not so high.
Is it possible somehow to find out why is memory not being freed?

rabbitmq version 3.0.1 on linux x86_64 box with erlang client API.

Thank you.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rabbitmq.com/pipermail/rabbitmq-discuss/attachments/20130125/9af43d2b/attachment.htm>

More information about the rabbitmq-discuss mailing list