Hi,

sorry, maybe I didn't completely explained my tests.
After you suggestion to update my system version, I did it and today I was
running on a clean updated version without any change to the code.

Starting from this point I have done a deeper analysis of the call park
feature to allow transfer back to parker. I found that, using a different
log level than DEBUG, a segmentation fault happens. Indeed the sipXpark
process crashes.

from core dumps:

*Program received signal SIGSEGV, Segmentation fault.*
*[Switching to Thread 0x7fffcbfff700 (LWP 3741)]*
*CpCall::handleMessage (this=0x7fffbc006940, eventMessage=<value optimized
out>) at /home/admin/sipxecs/sipXcallLib/src/cp/CpCall.cpp:695*
*695                                    OsStatus res =
dtmfEvent->signal((eventData & 0xfffffff0));*


the "signal" method called on dtmfEvent needs a intptr_t parameter:

OsStatus OsQueuedEvent::signal(const *intptr_t* eventData)



So I partially applied the changes used in the patch
http://track.sipfoundry.org/browse/XX-10103 by only changing the type of
the "event" from int to intptr_t. Applying this changes everything works
fine.

Hope this better explains what I did and what I was asking for.
It could be the case to apply this changes to main tree.

Regards
Luciano

2012/6/20 George Niculae <[email protected]>

> On Wed, Jun 20, 2012 at 11:29 AM, Luciano Berardi
> <[email protected]> wrote:
> > Hi,
> >
> > thank you for your replies.
> > Currently my tests are performed in a Snom to Snom configuration. I've
> > rebuilt the system with commit 5b137b5216 of Date:   Tue Jun 19 15:08:50.
> >
> > The problem still persists. I'm having problem to identify where it
> exactly
> > crashes because in DEBUG mode everything works correctly, changing log
> level
> > from INFO on, it gives me just this warning message:
> >
> > "2012-06-20T08:09:31.455664Z":10:MP:WARNING:testpbx.labsip2ser.net:
> pid-25443:037FE700:sipxpark:"_REALTIME_LINUX_AUDIO_THREADS
> > was defined but application does not have ROOT priv."
> >
> > I'm continuing to investigate, any suggestion is really appreciated.
>
> I'm confused, you said that backporting fix for
> http://track.sipfoundry.org/browse/XX-10103 solved the issue, however
> there is no way to backport that one as the code that introduced the
> bug is not in there.
> Have you installed from rpms or built from sources? If from sources
> make sure you're on release-4.4 branch
>
> George
> _______________________________________________
> sipx-users mailing list
> [email protected]
> List Archive: http://list.sipfoundry.org/archive/sipx-users/
>
_______________________________________________
sipx-users mailing list
[email protected]
List Archive: http://list.sipfoundry.org/archive/sipx-users/

Reply via email to