Staging threshhold odd behaviour on C++ broker

[Edlira Nano <ed...@gmail.com>]

Hi everyone,

I am sending a 7M message on a qpid C++ broker, no persistence,  with the
default staging-threshhold value (5M).
Then I consume the message, the queue is properly dequeued, my message is as
I expect it to be, everything seems perfect.

Then I notice that in the .qpidd directory I have 500M of log files that
were created (for a single file of 7M) and that after I consumed
 the message the logs are still there. After deleting the queue, they are
still there. After restarting the broker they are still there.

Is this the wanted behaviour of staging-threshhold? Is this a bug?

I am running the latest version from trunk (qpid broker/client C++ 0.6).

Regards,
Edlira Nano

Re: Staging threshhold odd behaviour on C++ broker

[Gordon Sim <gs...@redhat.com>]

On 03/18/2010 05:07 PM, Edlira Nano wrote:
> Hi everyone,
>
> I am sending a 7M message on a qpid C++ broker, no persistence,  with the
> default staging-threshhold value (5M).
> Then I consume the message, the queue is properly dequeued, my message is as
> I expect it to be, everything seems perfect.
>
> Then I notice that in the .qpidd directory I have 500M of log files that
> were created (for a single file of 7M) and that after I consumed
>   the message the logs are still there. After deleting the queue, they are
> still there. After restarting the broker they are still there.
>
> Is this the wanted behaviour of staging-threshhold? Is this a bug?

Yes, it is a bug. I have created a Jira: 
https://issues.apache.org/jira/browse/QPID-2456. Thanks for highlighting it!

---------------------------------------------------------------------
Apache Qpid - AMQP Messaging Implementation
Project:      http://qpid.apache.org
Use/Interact: mailto:users-subscribe@qpid.apache.org