Michael Broughton wrote: > I have no reason to doubt that the tsx instance I posted is bad, but the > backtrace makes no sense to me (ignoring line numbers). I don't > understand how the code goes from tsx_destroy() to > tsx_on_state_completed_uas() without any intermediate steps... if it is > even possible to make that step. That's not possible indeed. > I was running an old version of pjsip... but I think the last svn up I > did was a week or two ago. So I don't think that explains it. From your core dump, could you "print PJ_VERSION" ? That should give us more info about the version. cheers, -benny > It is possible that there is some other source for this weirdness. I > have also been getting seg faults in two other places, which I have not > been able make sense of either. > -- Benny Prijono http://www.pjsip.org
