Re: Question about queue's reusable message
Sam Wilson <Sam.Wilson@...>
Thanks for getting back to me! I've opened Bug 186 (http://bugs.openmama.org/show_bug.cgi?id=186) about this issue. I think I recreated the same leak in the qpid bridge by detaching messages.
Thanks for your help,
From: Gary Molloy [g.molloy@...]
Sent: Monday, March 30, 2015 12:58 PM
To: Sam Wilson; openmama-dev@...
Subject: RE: Question about queue's reusable message
Thanks for your email.
Typical scenarios and use cases that we have encountered, the queue is not usually destroyed that often, so we could very well have missed this during our internally testing.
This area of the code is very 'tricky' and we try to be very careful when looking to make changes here.
Could I ask you to raise a Bugzilla ticket with all your findings and tests and we can look into this for you?
Gary Molloy - SR Labs
Adelaide Exchange | 24-26 Adelaide Street | Belfast | BT2 8GD
From: openmama-dev-bounces@... [mailto:openmama-dev-bounces@...] On Behalf Of Sam Wilson
Sent: 27 March 2015 16:06
Subject: [Openmama-dev] Question about queue's reusable message
I'm tracking down a memory leak in our bridge, and I am trying to understand why mamaMsg_setMsgBuffer doesn't set the owner flag in the message. When the queue destroys the reusable message, it doesn't clean up the payload.
Since there is a comment around line 561 of msg.c
(http://tinyurl.com/msg-c-561) I assume this is intended behaviour. How should we go about cleaning up the payload in this case?
Openmama-dev mailing list