On 9/20/06, Damien Sandras <[EMAIL PROTECTED]> wrote: > I thought it was my new code.... Apparently not. We'll have to > investigate...
It looks to me like OnReceivedAuthenticationRequired is being called when SIPEndPoint is being destroyed, so that SIPEndPoint's destructor has already been called (but OpalEndPoint's destructor most likely has not run yet), which then results into OpalEndPoint's pure virtual method being called. In this backtrace thread 1 seems to be in the process of destroying the endpoint (with the wait caused by CloseWait() being an explanation why it happens comparable often that ~SIPEndPoint() has already destroyed part of the object, but ~OpalEndPoint() was not called yet), and thread 12 tries to dispatch an received event to the SIPEndPoint object which is already half-destroyed (oops!). Thread 12 probably should not be alive anymore at this point in time, though I neither know why it still is or how to stop it currectly as I have not looked at this part of OPAL to closely :(. Probably it should be stopped form ~SIPEndPoint()? -- Daniel _______________________________________________ Ekiga-devel-list mailing list Ekiga-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/ekiga-devel-list