Created attachment 319833 [details] GDB session during crash Somebody on #fedora-qa reported evolution instability with a Google calendar conencted. I was quickly able to reproduce. Stack trace attached. Seems to come from ea-cal-view.c, line 280: g_signal_emit_by_name (ea_cal_view, "children_changed", NULL); I've reproduced the crash several times pretty easily, typically by just "playing around" with the calendar stuff, changing the view, enabling/disabling, etc... Crash always comes from line above. I would recommend to mark this a F10 blocker.
Thanks for the backtrace. Looks like this is accessibility-related, and I see some atk symbols missing from the trace. Would you mind installing atk-debuginfo and attempting to reproduce the crash again?
Created attachment 319864 [details] GDB session, all debuginfo installed. I also recompiled at-spi with '-g3 -O0' for extra info. This bug is easily reproducible.
Great! Thanks for the extra info.
"valgrind -q evolution" shows invalid reads in the same at-spi code section.
Changing subject, as it doesn't appear to be strictly related to the Google feature.
Possible connection upstream?
This [1] is more accurate upstream bug, than the old one (357608). Close this as upstream? There is some patch already, done by Matt, awaiting the review. [1] http://bugzilla.gnome.org/show_bug.cgi?id=530776
This bug appears to have been reported against 'rawhide' during the Fedora 10 development cycle. Changing version to '10'. More information and reason for this action is here: http://fedoraproject.org/wiki/BugZappers/HouseKeeping
According to [1] this bug should be fixed now in Fedora 11. Please reopen if this is not the case. [1] http://bugzilla.gnome.org/show_bug.cgi?id=560329