Bug 634084 – ECal reports loaded state too early

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 634084 - ECal reports loaded state too early


Summary:	ECal reports loaded state too early


Status:	RESOLVED FIXED

Product:	evolution-data-server
Classification:	Platform
Component:	Calendar
Version:	2.32.x (obsolete)
Hardware:	Other Linux

Importance:	Normal critical
Target Milestone:	---
Assigned To:	evolution-calendar-maintainers
QA Contact:	Evolution QA team

URL:
Whiteboard:

Depends on:
Blocks:

Reported:	2010-11-05 14:49 UTC by Milan Crha
Modified:	2010-11-09 09:20 UTC

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

Attachments
eds patch (5.34 KB, patch) 2010-11-08 12:46 UTC, Milan Crha	committed	Details \| Review

Description Milan Crha 2010-11-05 14:49:02 UTC

Moving this from a downstream bug report:
https://bugzilla.redhat.com/show_bug.cgi?id=650160

abrt version: 1.1.13
architecture: x86_64
Attached file: backtrace
cmdline: /usr/libexec/e-calendar-factory
component: evolution-data-server
crash_function: e_intervaltree_search
executable: /usr/libexec/e-calendar-factory
kernel: 2.6.35.6-48.fc14.x86_64
package: evolution-data-server-2.32.0-3.fc14
rating: 4
reason: Process /usr/libexec/e-calendar-factory was killed by signal 11
(SIGSEGV)
release: Fedora release 14 (Laughlin)
time: 1288960793
uid: 500

Core was generated by `/usr/libexec/e-calendar-factory'.
Program terminated with signal 11, Segmentation fault.

+ Trace 224510

Thread 1 (Thread 10176)

#0 e_intervaltree_search
at e-cal-backend-intervaltree.c line 526
#1 e_cal_backend_file_start_query
at e-cal-backend-file.c line 1891
#2 impl_DataCalView_start
at e-data-cal-view.c line 309
#3 _e_gdbus_gdbus_cclosure_marshaller_BOOLEAN__OBJECT
at e-gdbus-marshallers.c line 202
#4 g_closure_invoke
at gclosure.c line 766
#5 signal_emit_unlocked_R
at gsignal.c line 3252
#6 g_signal_emit_valist
at gsignal.c line 2993
#7 g_signal_emit
at gsignal.c line 3040
#8 handle_method_call
at e-gdbus-egdbuscalview.c line 1078
#9 call_in_idle_cb
at gdbusconnection.c line 4397
#10 g_main_dispatch
at gmain.c line 2149
#11 g_main_context_dispatch
at gmain.c line 2702
#12 g_main_context_iterate
at gmain.c line 2780
#13 g_main_loop_run
at gmain.c line 2988
#14 main
at e-data-cal-factory.c line 922

Comment 1 Milan Crha 2010-11-08 12:46:14 UTC

Created attachment 174053 [details] [review]
eds patch

for evolution-data-server;

I cannot reproduce this myself, but I see what's going on:
a) when opening calendar asynchronously, the ECal reported the load state
   as 'loaded', though it wasn't yet, which caused b)
b) the EIntervalTree has some checking on non-NULL 'tree', but the first thing
   in the public functions was to access its 'priv' member, which lead to
   the crash

This patch is fixing both things.

Comment 2 Milan Crha 2010-11-08 12:48:55 UTC

Created commit 3e433f5 in eds master (2.91.3+)
Created commit c6d844f in eds gnome-2-32 (2.32.1+)