Bug 669226 – Crash when removing calendar component

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 669226 - Crash when removing calendar component


Summary:	Crash when removing calendar component


Status:	RESOLVED FIXED

Product:	evolution-mapi
Classification:	Applications
Component:	Calendar
Version:	3.2.x
Hardware:	Other Linux

Importance:	Normal critical
Target Milestone:	---
Assigned To:	evolution-mapi-maint
QA Contact:	evolution-mapi-maint

URL:
Whiteboard:

Depends on:
Blocks:

Reported:	2012-02-02 09:23 UTC by Milan Crha
Modified:	2012-02-03 13:31 UTC

See Also:
GNOME target:	---
GNOME version:	---

Attachments
ema patch (726 bytes, patch) 2012-02-03 13:28 UTC, Milan Crha	committed	Details \| Review

Description Milan Crha 2012-02-02 09:23:42 UTC

Moving this from a downstream bug report:
https://bugzilla.redhat.com/show_bug.cgi?id=786547

[abrt] evolution-data-server-3.2.3-1.fc16: strcmp: Process /usr/libexec/e-calendar-factory was killed by signal 11 (SIGSEGV)

libreport version: 2.0.8
abrt_version:   2.0.7
backtrace_rating: 4
cmdline:        /usr/libexec/e-calendar-factory
comment:        Connected to MS Exchange server, trying to delete many
recurring instances of calendar appointments. Stopped working and ABRT popped
up and said it crashed.
crash_function: strcmp
executable:     /usr/libexec/e-calendar-factory
kernel:         3.2.2-1.fc16.x86_64
reason:         Process /usr/libexec/e-calendar-factory was killed by signal 11
(SIGSEGV)
time:           Wed 01 Feb 2012 01:08:51 PM EST

+ Trace 229581

Thread 1 (Thread 0x7fe8237fe700 (LWP 9480))

#0 __strcmp_ssse3
at ../sysdeps/x86_64/strcmp.S line 214
#1 g_str_equal
at gstring.c line 116
#2 ecbm_remove_object
at e-cal-backend-mapi.c line 1988
#3 ecbm_operation_cb
at e-cal-backend-mapi.c line 2768
#4 thread_func_cb
at em-operation-queue.c line 144
#5 g_thread_pool_thread_proxy
at gthreadpool.c line 319
#6 g_thread_create_proxy
at gthread.c line 1962
#7 start_thread
at pthread_create.c line 309
#8 clone
at ../sysdeps/unix/sysv/linux/x86_64/clone.S line 115

Comment 1 Milan Crha 2012-02-03 13:28:56 UTC

Created attachment 206683 [details] [review]
ema patch

for evolution-mapi;

I cannot reproduce this myself, but looking into the place of the crash and a circumstance (it was called with 'rid' being NULL), and it succeeded with component removal on the server, and then, when removing it also from the local cache, it crashed on 'rid' comparison. The test didn't make much sense to me anyway, because it prevented notifications of removed components for certain cases, where this is not correct, because e_cal_backend_notify_component_removed() makes sure that only affected views are updated with the change.

Comment 2 Milan Crha 2012-02-03 13:30:06 UTC

Created commit e321b33 in evo master (3.3.5+)

Comment 3 Milan Crha 2012-02-03 13:31:18 UTC

err, *ema* master, of course.