[OpenSIPS-Users] memory leak with TLS?

Zappasodi Daniele Daniele.Zappasodi at seltatel.it
Thu Oct 11 10:04:41 CEST 2012


Hi,
I think that there is a memory leak with TLS. 
I have experienced it in my calls stress test, but it is possible to reproduce it easily with sipp with a simple scenario: 
you have only to send REGISTERs in multi socket mode (-t ln). 
With this test I can observe the internal memory that sometimes increases (both for "TCP main" and the "TCP receivers" processes). 
After some hours appears the messages:

WARNING:core:fm_malloc: Not enough free memory, will atempt defragmenation
WARNING:core:fm_malloc: Not enough free memory, will atempt defragmenation
ERROR:core:tcpconn_new: shared memory allocation failure

and Opensips is no longer able to accept new connections.

I wait some time (even hours) in order to free the allocated memory and then I start again the same test, but after some minutes the error appears again. 

Note that:
the Opensips config file is very simple, it performs only the "save" and then exits, without authentication or any other actions.
I have tried to compile with the DBG_QM_MALLOC option, but I can't see any difference compared to a normal situation.
I have tried also Opensips 1.8.1, but the memory leak is present also in this version.


Thanks,
Daniele



**********************************************************************
The information in this message is confidential and may be legally
privileged. It is intended solely for the addressee. Access to this 
message
by anyone else is unauthorized. If you are not the intended 
recipient, any
disclosure, copying, or distribution of the message, or any action 
or
omission taken by you in reliance on it, is prohibited and may be 
unlawful.
Please immediately contact the sender if you have received this 
message in error.


**********************************************************************
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opensips.org/pipermail/users/attachments/20121011/9659cb0d/attachment.htm>


More information about the Users mailing list