Currently I am running one queue with a 15 message per second rate on average, those failover queues are created and deleted at pretty much the same rate. The producers are producing at one message per connection/session. This seems to create a new queue bound to the amq.failover exchange for each message.. I'm thinking I may be doing something wrong in this regard, but I don't see another obvious way.
Etrik --- On Tue, 6/16/09, Carl Trieloff <[email protected]> wrote: > From: Carl Trieloff <[email protected]> > Subject: Re: Qpid C++ Broker memory use > To: [email protected] > Date: Tuesday, June 16, 2009, 12:32 PM > > that is normal, and those queues should always how a low > message count. > it is just updates for changes in cluster membership. > > Carl. > > Nominal wrote: > > This did fix several issues I was observing with qpid > and the bdbstore. However I am still seeing a gradual > increase in memory over time. Doing some more poking > around with qpid-tool I am seeing what appears to be the > same behavior for failover communication. > > > > From what I can gather the cluster is doing its > communication via temporary queues on the amq.failover > exchange. > > > > 1862 16:06:07 > 16:06:07 103.amq.failovera99cf518-6516-4719-b84e-8e4e368d5ed0 > > 1863 16:06:07 > 16:06:07 103.amq.failoverb768b9bf-c37e-4eac-b1ff-a7e463b9f9d1 > > 1864 16:06:07 > 16:06:07 103.amq.failover65a0742c-7a1c-4502-bb07-a406a29eb0d7 > > 1865 16:06:07 > 16:06:07 103.amq.failoverae655da7-ae98-4160-b699-fc9f980ce6ae > > 1866 16:06:07 > 16:06:07 103.amq.failovere5a8537d-d553-4ab8-ad3d-f8a42c9c0a82 > > 1867 16:06:07 > 16:06:07 103.amq.failovera6ca1249-d0e6-4504-b87f-092f3cad5365 > > 1868 16:06:07 > 16:06:08 103.amq.failover1ed66c6a-a3c0-4519-9d27-7c5d4e0d816a > > 1869 16:06:08 > 16:06:08 103.amq.failover1e0b11f3-e858-47fe-a929-24f3ac55ebb0 > > 1870 16:06:08 > 16:06:08 103.amq.failovera738967f-c013-429d-b108-f092119af7bd > > 1871 16:06:08 > 16:06:08 103.amq.failover11861a49-f753-407d-87c5-42cf08d87e84 > > > > > > This is what I'm observing. Is this normal > behavior? > > > > Etrik > > > > --- On Mon, 6/15/09, Gordon Sim <[email protected]> > wrote: > > > > > >> From: Gordon Sim <[email protected]> > >> Subject: Re: Qpid C++ Broker memory use > >> To: [email protected] > >> Date: Monday, June 15, 2009, 6:49 AM > >> > > = > > > >> This is due to a bug[1] that has been fixed on > trunk[2], > >> but did not make it in time for the 0.5 release. > If there > >> are active clients when you add a new node to the > cluster, > >> their temporary queues become 'non-temporary' on > the new > >> node and don't get cleaned up automatically when > the > >> original owning session ends. > >> > >> [1] https://issues.apache.org/jira/browse/QPID-1873 > >> [2] http://svn.apache.org/viewvc?view=rev&revision=779183 > >> > >> > >> > > > > > > > > > > > --------------------------------------------------------------------- > > Apache Qpid - AMQP Messaging Implementation > > Project: http://qpid.apache.org > > Use/Interact: mailto:[email protected] > > > > > > --------------------------------------------------------------------- Apache Qpid - AMQP Messaging Implementation Project: http://qpid.apache.org Use/Interact: mailto:[email protected]
