Bug 566735 – Evolution crashed while startup

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 566735 - Evolution crashed while startup


Summary:	Evolution crashed while startup


Status:	RESOLVED FIXED

Product:	evolution-mapi
Classification:	Applications
Component:	Mail
Version:	0.25.x
Hardware:	Other All

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

URL:
Whiteboard:

Duplicates:	568539 (view as bug list)
Depends on:
Blocks:

Reported:	2009-01-06 06:57 UTC by Kandepu Prasad
Modified:	2010-06-10 04:00 UTC

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

Attachments
No need for summary during deletion, (1.06 KB, patch) 2009-01-30 05:07 UTC, Johnny Jacob	committed	Details \| Review

Description Kandepu Prasad 2009-01-06 06:57:02 UTC

Steps to reproduce:
Just started evolution from a terminal


Stack trace:
Distribution: SUSE Linux Enterprise Desktop 11 (i586)
Gnome Release: 2.24.1 2008-12-08 (SUSE)
BugBuddy Version: 2.24.1

System: Linux 2.6.27.8-1-pae #1 SMP 2008-12-08 03:55:28 +0100 i686
X Vendor: The X.Org Foundation
X Vendor Release: 10502000
Selinux: No
Accessibility: Disabled
GTK+ Theme: Gilouche
Icon Theme: Gilouche

Memory status: size: 162590720 vsize: 162590720 resident: 24195072 share: 18382848 rss: 24195072 rss_rlim: 18446744073709551615
CPU usage: start_time: 1231223820 rtime: 84 utime: 76 stime: 8 cutime:0 cstime: 0 timeout: 0 it_real_value: 0 frequency: 100

Backtrace was generated from '/usr/bin/evolution'

[?1034h[Thread debugging using libthread_db enabled]
[New Thread 0xaf0d5b90 (LWP 19109)]
[New Thread 0xafadbb90 (LWP 19105)]
0xffffe430 in __kernel_vsyscall ()

+ Trace 211308

Thread 2 (Thread 0xaf0d5b90 (LWP 19109))

#0 __kernel_vsyscall
#1 __lll_lock_wait
at ../nptl/sysdeps/unix/sysv/linux/i386/i486/lowlevellock.S line 136
#2 _L_lock_89
from /lib/libpthread.so.0
#3 __pthread_mutex_lock
at pthread_mutex_lock.c line 86
#4 <signal handler called>
#5 camel_folder_summary_load_from_db
at camel-folder-summary.c line 945
#6 camel_mapi_summary_new
at camel-mapi-summary.c line 155
#7 camel_mapi_folder_new
at camel-mapi-folder.c line 1578
#8 mapi_get_folder
at camel-mapi-store.c line 425
#9 camel_store_get_folder
at camel-store.c line 345
#10 mail_tool_uri_to_folder
at mail-tools.c line 331
#11 get_folder_exec
at mail-ops.c line 1214
#12 mail_msg_proxy
at mail-mt.c line 520
#13 g_thread_pool_thread_proxy
at gthreadpool.c line 265
#14 g_thread_create_proxy
at gthread.c line 635
#15 start_thread
at pthread_create.c line 297
#16 clone
at ../sysdeps/unix/sysv/linux/i386/clone.S line 130

Comment 1 Kandepu Prasad 2009-01-06 08:36:04 UTC

MAPI rpm version

evolution-mapi-provider-0.25.3-58.3
evolution-mapi-provider-debuginfo-0.25.3-58.3

Comment 2 Kandepu Prasad 2009-01-07 08:50:49 UTC

Guess there was some problem with the build rpms given for testing. I updated back to the RC1 rpms and the crash does not seem to happen again. I also tested with the latest updated trunk mapi build and the crash does not occur. So closing this as invalid.

Comment 3 Kandepu Prasad 2009-01-29 11:43:38 UTC

Hmm I can reproduce this bug in trunk consistently. So reopening the bug with new traces.

Evolution-mapi revision: 86

Steps to reproduce:
1. Create a folder with character '&' in the name (like.....'xyz&abc'). When seen in outlook the name contains the '&' character but it seems evolution-mapi does not recongnize this really. (xyzabc)
2. So if you try to rename this folder, it says folder does not exist.
3. Now try to delete that folder then Evolution is crashed with below trace. 
Note: The folder is deleted when seen from outlook

libexchangemapi-Message: exchange-mapi-connection.c(1559): exchange_mapi_remove_folder: lock(connect_lock)

Folder with id EE788DE208010001 was emptied 
Folder with id EE788DE208010001 was deleted libexchangemapi-Message: exchange-mapi-connection.c(1614): exchange_mapi_remove_folder: unlock(connect_lock)

exchange-mapi-connection.c(1616): Leaving exchange_mapi_remove_folder 
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7fffdddf7950 (LWP 13172)]
camel_folder_summary_load_from_db (s=0x7fffd80a9870, ex=0x7fffdddf6e70) at camel-folder-summary.c:960
960		ret = camel_folder_summary_header_load_from_db (s, s->folder->parent_store, s->folder->full_name, ex);
(gdb) t a a bt

+ Trace 211982

Thread 23 (Thread 0x7fffdddf7950 (LWP 13172))

#0 camel_folder_summary_load_from_db
at camel-folder-summary.c line 960
#1 camel_mapi_summary_new
at camel-mapi-summary.c line 155
#2 mapi_delete_folder
at camel-mapi-store.c line 505
#3 camel_store_delete_folder
at camel-store.c line 495
#4 remove_folder_rec
at mail-ops.c line 1453
#5 remove_folder_exec
at mail-ops.c line 1475
#6 mail_msg_proxy
at mail-mt.c line 520
#7 g_thread_pool_thread_proxy
at gthreadpool.c line 265
#8 g_thread_create_proxy
at gthread.c line 635
#9 start_thread
at pthread_create.c line 297
#10 clone
at ../sysdeps/unix/sysv/linux/x86_64/clone.S line 112
#11 ??

(gdb)

Comment 4 Johnny Jacob 2009-01-29 11:55:21 UTC

(In reply to comment #3)
> Hmm I can reproduce this bug in trunk consistently. So reopening the bug with
> new traces.
> 
The traces are different. #1 is because of binary incompat. And this one is a bug in the code.

Comment 5 Johnny Jacob 2009-01-30 05:07:57 UTC

Created attachment 127514 [details] [review]
No need for summary during deletion,

Comment 6 Johnny Jacob 2009-02-01 11:22:39 UTC

committed. r89

Comment 7 Bharath Acharya 2010-06-10 04:00:34 UTC

*** Bug 568539 has been marked as a duplicate of this bug. ***