After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 558587 - segfault when call connects
segfault when call connects
Status: RESOLVED NOTGNOME
Product: ekiga
Classification: Applications
Component: general
3.0.x
Other All
: Normal critical
: ---
Assigned To: Ekiga maintainers
Ekiga maintainers
Depends on:
Blocks:
 
 
Reported: 2008-10-30 20:20 UTC by Peter Robinson
Modified: 2008-11-17 21:36 UTC
See Also:
GNOME target: ---
GNOME version: 2.23/2.24

Attachments
ekiga -d5 -u4 (62.11 KB, text/plain)
2008-10-30 20:21 UTC, Peter Robinson 		Details
Ekiga crash backtrace (16.40 KB, text/plain)
2008-11-13 15:04 UTC, Peter Robinson 		Details
ekiga crash using avahi instead of SIP (15.44 KB, text/plain)
2008-11-13 15:43 UTC, Peter Robinson 		Details

Description Peter Robinson 2008-10-30 20:20:02 UTC 
Steps to reproduce:
Upstreaming this from Redhat BZ # 469132

Description of problem:

Ekiga seg faults when answering incoming call

Version-Release number of selected component (if applicable):

F10-Snap3-i386  Ekiga 3.0.1

How reproducible:


Steps to Reproduce:
1.Configure ekiga with sip account (fedoratalk or ekiga)
2.Dial extension from landline or sip call from another machine
3.After incoming call pop-up is displayed click Accept to answer call

Actual results:

Ekiga closes with a segmentation fault.

Expected results:

Ekiga should answer the call

Additional info:

Attempted this on an XO and a rawhide testing PC with the same results so it is
not specific to the XO

Stack trace:

+ Trace 208935

Thread 1 (Thread 0xb7fd8770 (LWP 25328))

  • #0 IA__g_type_check_instance_cast
    at gtype.c line 3723
  • #1 closed_cb
    at gui/main.cpp line 3644
  • #2 IA__g_cclosure_marshal_VOID__UINT
    at gmarshal.c line 251
  • #3 IA__g_closure_invoke
    at gclosure.c line 767
  • #4 signal_emit_unlocked_R
    at gsignal.c line 3244
  • #5 IA__g_signal_emit_valist
  • #6 IA__g_signal_emit
    at gsignal.c line 3034
  • #7 _close_signal_handler
    at notification.c line 687
  • #8 notify_marshal_VOID__UINT_UINT
    at notify-marshal.c line 81
  • #9 marshal_dbus_message_to_g_marshaller
    at dbus-gproxy.c line 1665
  • #10 IA__g_closure_invoke
    at gclosure.c line 767
  • #11 signal_emit_unlocked_R
    at gsignal.c line 3244
  • #12 IA__g_signal_emit_valist
    at gsignal.c line 2977
  • #13 IA__g_signal_emit
    at gsignal.c line 3034
  • #14 dbus_g_proxy_emit_remote_signal
    at dbus-gproxy.c line 1718
  • #15 dbus_g_proxy_manager_filter
    at dbus-gproxy.c line 1286
  • #16 dbus_connection_dispatch
    at dbus-connection.c line 4406
  • #17 message_queue_dispatch
    at dbus-gmain.c line 101
  • #18 g_main_dispatch
    at gmain.c line 2144
  • #19 IA__g_main_context_dispatch
    at gmain.c line 2697
  • #20 g_main_context_iterate
    at gmain.c line 2778
  • #21 IA__g_main_loop_run
    at gmain.c line 2986
  • #22 IA__gtk_main
    at gtkmain.c line 1200
  • #23 main
    at gui/main.cpp line 4821 

Other information:
Comment 1 Peter Robinson 2008-10-30 20:21:49 UTC 
Created attachment 121688 [details]
ekiga -d5 -u4

ekiga debug output from the crash
Comment 2 Peter Robinson 2008-10-30 23:09:48 UTC 
The crash info in bugz 556764 and 558591 and possibly 556498 seem to share a lot in common with this bug as well.
Comment 3 Peter Robinson 2008-11-13 15:04:26 UTC 
Created attachment 122582 [details]
Ekiga crash backtrace

Ekiga to ekiga call on Fedora 10 (rawhide).
Comment 4 Peter Robinson 2008-11-13 15:43:15 UTC 
Created attachment 122584 [details]
ekiga crash using avahi instead of SIP

Same crash but using avahi neighbours instead of SIP. This BT has a few missing debuginfo packages installed.
Comment 5 Damien Sandras 2008-11-17 21:36:00 UTC 
After some experimentation, we noticed it was due to an API breakage in Fedora, hence not in Ekiga. Marking as NOTGNOME, a patch has been provided.

Thanks Peter for the efficient debugging.