Bug 323239 – crash if two identical appointments in database

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 323239 - crash if two identical appointments in database


Summary:	crash if two identical appointments in database


Status:	RESOLVED INCOMPLETE

Product:	Evolution Exchange
Classification:	Deprecated
Component:	Connector
Version:	2.4.x
Hardware:	Other All

Importance:	High major
Target Milestone:	2.5
Assigned To:	Chenthill P
QA Contact:	Ximian Connector QA

URL:
Whiteboard:

Depends on:
Blocks:	327514

Reported:	2005-12-05 01:25 UTC by Malcolm Caldwell
Modified:	2006-03-08 13:21 UTC

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

Description Malcolm Caldwell 2005-12-05 01:25:32 UTC

Steps to reproduce:
1. Start up evolution - exchange connector crashes... SOMETIMES

OR

2. Go to an email with a meeting invite - connector crashes ALWAYS



Stack trace:
This is from the 'crash at startup' case.

(gdb) thread apply all bt

+ Trace 64422

Thread 5 (Thread -1219581008 (LWP 4726))

#0 g_str_hash
from /usr/lib/libglib-2.0.so.0
#1 g_hash_table_lookup
from /usr/lib/libglib-2.0.so.0
#2 e_cal_backend_exchange_add_object
#3 todos_backend_exchange_factory_get_type
#4 todos_backend_exchange_factory_get_type
#5 check_for_send_options
#6 e_cal_backend_sync_open
from /usr/lib/libedata-cal-1.2.so.1
#7 e_cal_backend_sync_open
from /usr/lib/libedata-cal-1.2.so.1
#8 e_cal_backend_open
from /usr/lib/libedata-cal-1.2.so.1
#9 e_data_cal_get_type
from /usr/lib/libedata-cal-1.2.so.1
#10 _ORBIT_skel_small_GNOME_Evolution_Calendar_Cal_open
from /usr/lib/libedata-cal-1.2.so.1
#11 ORBit_POA_setup_root
from /usr/lib/libORBit-2.so.0
#12 ORBit_OAObject_invoke
from /usr/lib/libORBit-2.so.0
#13 ORBit_small_invoke_adaptor
from /usr/lib/libORBit-2.so.0
#14 ORBit_POAObject_post_invoke
from /usr/lib/libORBit-2.so.0
#15 ORBit_POAObject_post_invoke
from /usr/lib/libORBit-2.so.0
#16 giop_thread_queue_process
from /usr/lib/libORBit-2.so.0
#17 giop_thread_queue_process
from /usr/lib/libORBit-2.so.0
#18 g_thread_pool_free
from /usr/lib/libglib-2.0.so.0
#19 g_static_private_free
from /usr/lib/libglib-2.0.so.0
#20 start_thread
from /lib/libpthread.so.0
#21 clone
from /lib/libc.so.6



Other information:
This software is almost unusable for me - maybe its my setup?

This morning I started up about 10 times before I could get this to run.

Comment 1 Malcolm Caldwell 2005-12-05 01:48:44 UTC

Here is the bt for 'viewing a meeting invite'
Looks the same.
(gdb) thread apply all bt

+ Trace 64423

Thread 4 (Thread -1211065424 (LWP 8240))

#0 g_str_hash
from /usr/lib/libglib-2.0.so.0
#1 g_hash_table_lookup
from /usr/lib/libglib-2.0.so.0
#2 e_cal_backend_exchange_add_object
#3 todos_backend_exchange_factory_get_type
#4 todos_backend_exchange_factory_get_type
#5 check_for_send_options
#6 e_cal_backend_sync_open
from /usr/lib/libedata-cal-1.2.so.1
#7 e_cal_backend_sync_open
from /usr/lib/libedata-cal-1.2.so.1
#8 e_cal_backend_open
from /usr/lib/libedata-cal-1.2.so.1
#9 e_data_cal_get_type
from /usr/lib/libedata-cal-1.2.so.1
#10 _ORBIT_skel_small_GNOME_Evolution_Calendar_Cal_open
from /usr/lib/libedata-cal-1.2.so.1
#11 ORBit_POA_setup_root
from /usr/lib/libORBit-2.so.0
#12 ORBit_OAObject_invoke
from /usr/lib/libORBit-2.so.0
#13 ORBit_small_invoke_adaptor
from /usr/lib/libORBit-2.so.0
#14 ORBit_POAObject_post_invoke
from /usr/lib/libORBit-2.so.0
#15 ORBit_POAObject_post_invoke
from /usr/lib/libORBit-2.so.0
#16 giop_thread_queue_process
from /usr/lib/libORBit-2.so.0
#17 giop_thread_queue_process
from /usr/lib/libORBit-2.so.0
#18 g_thread_pool_free
from /usr/lib/libglib-2.0.so.0
#19 g_static_private_free
from /usr/lib/libglib-2.0.so.0
#20 start_thread
from /lib/libpthread.so.0
#21 clone
from /lib/libc.so.6

Thread 3 (Thread -1211065424 (LWP 8219))

#0 g_str_hash
from /usr/lib/libglib-2.0.so.0
#1 g_hash_table_lookup
from /usr/lib/libglib-2.0.so.0
#2 e_cal_backend_exchange_add_object
#3 todos_backend_exchange_factory_get_type
#4 todos_backend_exchange_factory_get_type
#5 check_for_send_options
#6 e_cal_backend_sync_open
from /usr/lib/libedata-cal-1.2.so.1
#7 e_cal_backend_sync_open
from /usr/lib/libedata-cal-1.2.so.1
#8 e_cal_backend_open
from /usr/lib/libedata-cal-1.2.so.1
#9 e_data_cal_get_type
from /usr/lib/libedata-cal-1.2.so.1
#10 _ORBIT_skel_small_GNOME_Evolution_Calendar_Cal_open
from /usr/lib/libedata-cal-1.2.so.1
#11 ORBit_POA_setup_root
from /usr/lib/libORBit-2.so.0
#12 ORBit_OAObject_invoke
from /usr/lib/libORBit-2.so.0
#13 ORBit_small_invoke_adaptor
from /usr/lib/libORBit-2.so.0
#14 ORBit_POAObject_post_invoke
from /usr/lib/libORBit-2.so.0
#15 ORBit_POAObject_post_invoke
from /usr/lib/libORBit-2.so.0
#16 giop_thread_queue_process
from /usr/lib/libORBit-2.so.0
#17 giop_thread_queue_process
from /usr/lib/libORBit-2.so.0
#18 g_thread_pool_free
from /usr/lib/libglib-2.0.so.0
#19 g_static_private_free
from /usr/lib/libglib-2.0.so.0
#20 start_thread
from /lib/libpthread.so.0
#21 clone
from /lib/libc.so.6


Also noticed that if I ever view the calendar at all I get a crash as well.

Comment 2 Christian Kirbach 2005-12-06 10:33:53 UTC

This seems to be a unique stack trace. 

Thanks for the bug report. Unfortunately, that stack trace is not very useful in 
determining the cause of the crash. Can you get us one with debugging symbols? 
Please see http://live.gnome.org/GettingTraces for more information on how to do 
so.

Comment 3 Malcolm Caldwell 2005-12-07 00:51:43 UTC

This is a bt with evolution-connector with debug symbols:

+ Trace 64459

#0 g_str_hash
from /usr/lib/libglib-2.0.so.0
#1 g_hash_table_lookup
from /usr/lib/libglib-2.0.so.0
#2 e_cal_backend_exchange_add_object
at e-cal-backend-exchange.c line 501
#3 add_vevent
at e-cal-backend-exchange-calendar.c line 170
#4 add_ical
at e-cal-backend-exchange-calendar.c line 256
#5 open_calendar
at e-cal-backend-exchange-calendar.c line 473
#6 e_cal_backend_sync_open
from /usr/lib/libedata-cal-1.2.so.1
#7 e_cal_backend_sync_open
from /usr/lib/libedata-cal-1.2.so.1
#8 e_cal_backend_open
from /usr/lib/libedata-cal-1.2.so.1
#9 e_data_cal_get_type
from /usr/lib/libedata-cal-1.2.so.1
#10 _ORBIT_skel_small_GNOME_Evolution_Calendar_Cal_open
from /usr/lib/libedata-cal-1.2.so.1
#11 ORBit_POA_setup_root
from /usr/lib/libORBit-2.so.0
#12 ORBit_OAObject_invoke
from /usr/lib/libORBit-2.so.0
#13 ORBit_small_invoke_adaptor
from /usr/lib/libORBit-2.so.0
#14 ORBit_POAObject_post_invoke
from /usr/lib/libORBit-2.so.0
#15 ORBit_POAObject_post_invoke
from /usr/lib/libORBit-2.so.0
#16 giop_thread_queue_process
from /usr/lib/libORBit-2.so.0
#17 giop_thread_queue_process
from /usr/lib/libORBit-2.so.0
#18 g_thread_pool_free
from /usr/lib/libglib-2.0.so.0
#19 g_static_private_free
from /usr/lib/libglib-2.0.so.0
#20 start_thread
from /lib/libpthread.so.0
#21 clone
from /lib/libc.so.6



My reading of this is I have corupt data on the exchange server.  Perhaps I can
work arround this by deleting the data.  Even so, corrupt data should not crash
an application.

Comment 4 Malcolm Caldwell 2005-12-07 02:11:44 UTC

I went into owa and confirmed that there were two copies of that appointment,
and owa itself had troubles opening either copy.  Deleted them both and now
connector does not crash.

Comment 5 Christian Kirbach 2005-12-07 12:07:01 UTC

Thanks a lot for tracking down the cause.

I'll confirm the report because of the good trace and the known cause.
Let the maintainers decide what to do.

Comment 6 Sushma Rai 2005-12-12 06:41:19 UTC

Looks like problem with the hash table when 2 items exist.
Chen, can we look into it?

Comment 7 Poornima 2005-12-12 09:15:03 UTC

Malcolm: How did two copies of one appointment got created ?. Is it possible for
you to mail that appointment to  me/sushma/chentil. ? The same appointment would
help maintainers to fix this bug fast.

Comment 8 Sushma Rai 2006-01-10 12:31:41 UTC

Malcolm: can you please provide these details?

Comment 9 Malcolm Caldwell 2006-01-11 00:17:23 UTC

The two copies probably got created by the pilot sync code.

I deleted the appointment - sorry, but I needed to be able to use my calendar...

Comment 10 Sushma Rai 2006-01-11 05:38:52 UTC

varadhan, can we replicate this scenario?

Comment 11 Veerapuram Varadhan 2006-01-11 08:34:44 UTC

from comment 9: Can you tell us about what exactly did you sync?  Was it a fresh-calendar-in-pda during sync?  Did you do a copy-to-pilot/copy-from-pilot/synchronize? Did you only sync with evolution?

Comment 12 Malcolm Caldwell 2006-01-11 11:27:48 UTC

Sorry, this was from a palm pilot that had been corrupted by sucessive bad evolution syncs.  I do not think I could possibly every replicate this.
I am not sure what fres-calendar-in pad means.

It would have been synchronize rather than copy to or from.

Comment 13 Veerapuram Varadhan 2006-01-11 13:07:31 UTC

By fresh-calendar-in-pda, I meant a "empty" Datebook/Calendar database in palm.

Also, evolution-2.4.x does not support exchange calendar syncing. So, I am not sure how you got a corrupted record in your palm pilot.

Did you upgrade your evolution from 1.4?

I am just trying to figure out a scenario to reproduce the bug.

Comment 14 Malcolm Caldwell 2006-01-11 16:52:21 UTC

OK, I guess I was wrong.  Perhaps the corrupt data did not come form a palm sync but rather from a migration of our old lotus notes data into our exchange calendars.

I wish I could help you find a way to reproduce this.

Comment 15 Veerapuram Varadhan 2006-01-12 17:07:10 UTC

Malcolm: Do you have any "backed-up" test lotus notes data that I can import in exchange and check?  If so, can you mail it to vvaradhan@novell.com?

Comment 16 Sushma Rai 2006-01-14 06:41:11 UTC

Lowering the severity.

Comment 17 Sushma Rai 2006-03-08 13:21:36 UTC

Closing the bug as we could not reproduce this. Please re-open the bug
with the data, if it happens to you again.