Persistent TCP connections to UniMRCP cause memory to grow due to "leaked" messages

[GoogleCodeExporter]

What steps will reproduce the problem?
1. Establish a persistent MRCPv1 connection over TCP 
2. Every message received by UniMRCP will use memory allocated from the
pool associated with the connection, so they can not be freed.
3. Once the connection is dropped the pool is released, freeing the
"leaked" memory.

What version of the product are you using? On what operating system?
v0.8.0 on RedHat linux ES 3 update 8

Please provide any additional information below.
Some media servers seem to try to establish a heartbeat connection to
monitor the resource.  This causes a memory issue if the heartbeat
connection remains up for a long enough time. 

Original issue reported on code.google.com by [email protected] on 10 Feb 2010 at 6:52

[GoogleCodeExporter]

Original comment by [email protected] on 10 Feb 2010 at 7:12

• Changed state: Accepted
• Added labels: Type-Defect, Priority-Medium, OpSys-All, Component-Server, Usability

[GoogleCodeExporter]

Hi Arsen,

Do you have plan a new mechanism to resolve this issue ? I read the code to 
find a 
solution.

on loading test, the issue becomes blocking.

Regards
Anthony

Original comment by [email protected] on 5 May 2010 at 1:33

[GoogleCodeExporter]

Hi Anthony,

Sorry, but there is no trivial or simple fix to apply.

I basically raised a few ideas on how the issue could be fixed here.
http://groups.google.com/group/unimrcp/browse_frm/thread/417c93756df64840/7500ae
cd4627ffac?lnk=gst&q=persistent+TCP+conn#7500aecd4627ffac

But I'm afraid this will go nowhere, unless someone covers the development time.

Regards,
Arsen

Original comment by [email protected] on 5 May 2010 at 4:42