Red Hat Bugzilla – Bug 991074
[abrt] Unnecessary crash due to g_assert() call
Last modified: 2014-01-02 05:50:24 EST
Description of problem:
happens most probably on unclean exit when e-c-f gets sigterm from some external process.
Version-Release number of selected component:
libreport version: 2.0.9
:Thread no. 1 (3 frames)
: #4 cal_backend_finalize at e-cal-backend.c
: #6 g_hash_table_foreach at /lib64/libglib-2.0.so.0
: #7 e_data_cal_factory_finalize at e-data-cal-factory.c
Created attachment 781626 [details]
(In reply to David Jaša from comment #0)
> happens most probably on unclean exit when e-c-f gets sigterm from some
> external process.
True, true, I was just about to fix this (remove the assert), while I found your bug. I do not know why I wasn't auto-CC'ed in the first place. I definitely should be.
a) kill all evolution processes (e-calendar-factory, e-addressbook-factory,
evolution-alarm-notify, and evolution itself)
b) run evolution in calendar view:
$ evolution -c calendar
c) let it show events, the best if there are any "slow" calendars, like
the Zimbra calendar accessed through CalDAV
d) run --force-shutdown to close the processes
$ evolution --force-shutdown
Evolution is closed, the same as other processes, but the calendar factory may eventually crash with this message (it's not shown in the David's backtrace):
> Server is up and running...
> Received quit signal...
> libedata-cal:ERROR:e-cal-backend.c:269:cal_backend_finalize: assertion
> failed: (priv->clients == NULL)
> Aborted (core dumped)
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.
For information on the advisory, and where to find the updated
files, follow the link below.
If the solution does not work for you, open a new bug report.