Bug 569329 – [groupwise] EDS crashed when checking contacts in Novell GroupWise Address Book

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 569329 - [groupwise] EDS crashed when checking contacts in Novell GroupWise Address Book


Summary:	[groupwise] EDS crashed when checking contacts in Novell GroupWise Address Book


Status:	RESOLVED FIXED

Product:	evolution-data-server
Classification:	Platform
Component:	Contacts
Version:	2.28.x (obsolete)
Hardware:	Other All

Importance:	High critical
Target Milestone:	---
Assigned To:	evolution-addressbook-maintainers
QA Contact:	Evolution QA team

URL:
Whiteboard:	evolution[groupwise]

Duplicates:	597255 607435 (view as bug list)
Depends on:
Blocks:

Reported:	2009-01-27 13:18 UTC by Kandepu Prasad
Modified:	2010-09-01 14:49 UTC

See Also:
GNOME target:	---
GNOME version:	2.27/2.28

Attachments
proposed eds patch (1.08 KB, patch) 2009-12-08 11:37 UTC, Milan Crha	committed	Details \| Review

Description Kandepu Prasad 2009-01-27 13:18:20 UTC

Steps to reproduce:
EDS crashed when checking contacts in Novell GroupWise Address Book


Stack trace:
(evolution-data-server-2.26:28243): libegroupwise-CRITICAL **: e_gw_item_get_item_type: assertion `E_IS_GW_ITEM (item)' failed

(evolution-data-server-2.26:28243): libegroupwise-CRITICAL **: e_gw_item_get_item_type: assertion `E_IS_GW_ITEM (item)' failed

(evolution-data-server-2.26:28243): libegroupwise-CRITICAL **: e_gw_item_get_field_value: assertion `E_IS_GW_ITEM(item)' failed

(evolution-data-server-2.26:28243): libegroupwise-CRITICAL **: e_gw_item_get_field_value: assertion `E_IS_GW_ITEM(item)' failed

Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7fffd6ff5950 (LWP 28452)]
e_gw_item_get_full_name (item=0x0) at e-gw-item.c:884
884	{
(gdb) t a a bt

+ Trace 211927

Thread 37 (Thread 0x7fffd87f8950 (LWP 28455))

#0 pread64
from /lib64/libpthread.so.0
#1 __os_io_eds
at ../os/os_rw.c line 52
#2 __memp_pgread_eds
at ../mp/mp_bh.c line 222
#3 __memp_fget_eds
at ../mp/mp_fget.c line 528
#4 __db_goff_eds
at ../db/db_overflow.c line 141
#5 __db_ret_eds
at ../db/db_ret.c line 52
#6 __db_c_get_eds
at ../db/db_cam.c line 859
#7 e_book_backend_db_cache_get_contacts
at e-book-backend-db-cache.c line 308
#8 book_view_thread
at e-book-backend-groupwise.c line 2356
#9 g_thread_create_proxy
at gthread.c line 635
#10 start_thread
at pthread_create.c line 297
#11 clone
at ../sysdeps/unix/sysv/linux/x86_64/clone.S line 112
#12 ??

Thread 36 (Thread 0x7fffd67f4950 (LWP 28454))

#0 __lll_lock_wait
at ../nptl/sysdeps/unix/sysv/linux/x86_64/lowlevellock.S line 130
#1 _L_lock_102
from /lib64/libpthread.so.0
#2 __pthread_mutex_lock
at pthread_mutex_lock.c line 86
#3 delta_thread
at e-cal-backend-groupwise.c line 358
#4 g_thread_create_proxy
at gthread.c line 635
#5 start_thread
at pthread_create.c line 297
#6 clone
at ../sysdeps/unix/sysv/linux/x86_64/clone.S line 112
#7 ??

Thread 35 (Thread 0x7fffd6ff5950 (LWP 28452))

#0 e_gw_item_get_full_name
at e-gw-item.c line 884
#1 populate_full_name
at e-book-backend-groupwise.c line 663
#2 fill_contact_from_gw_item
at e-book-backend-groupwise.c line 1238
#3 update_address_book_deltas
at e-book-backend-groupwise.c line 3122
#4 g_thread_create_proxy
at gthread.c line 635
#5 start_thread
at pthread_create.c line 297
#6 clone
at ../sysdeps/unix/sysv/linux/x86_64/clone.S line 112
#7 ??

Thread 34 (Thread 0x7fffd77f6950 (LWP 28451))

#0 fsync
from /lib64/libpthread.so.0
#1 e_xml_save_file
at e-xml-utils.c line 103
#2 e_xmlhash_write
at e-xml-hash-utils.c line 349
#3 e_file_cache_remove_object
at e-file-cache.c line 436
#4 e_file_cache_replace_object
at e-file-cache.c line 410
#5 delta_thread
at e-cal-backend-groupwise.c line 470
#6 g_thread_create_proxy
at gthread.c line 635
#7 start_thread
at pthread_create.c line 297
#8 clone
at ../sysdeps/unix/sysv/linux/x86_64/clone.S line 112
#9 ??

Thread 33 (Thread 0x7fffd7ff7950 (LWP 28450))

#0 __lll_lock_wait
at ../nptl/sysdeps/unix/sysv/linux/x86_64/lowlevellock.S line 130
#1 _L_lock_102
from /lib64/libpthread.so.0
#2 __pthread_mutex_lock
at pthread_mutex_lock.c line 86
#3 delta_thread
at e-cal-backend-groupwise.c line 358
#4 g_thread_create_proxy
at gthread.c line 635
#5 start_thread
at pthread_create.c line 297
#6 clone
at ../sysdeps/unix/sysv/linux/x86_64/clone.S line 112
#7 ??

Thread 9 (Thread 0x7fffe37fe950 (LWP 28322))

#0 pthread_join
at pthread_join.c line 89
#1 g_thread_join_posix_impl
at gthread-posix.c line 385
#2 IA__g_thread_join
at gthread.c line 703
#3 e_book_backend_groupwise_authenticate_user
at e-book-backend-groupwise.c line 3403
#4 ORBit_small_invoke_adaptor
at orbit-small.c line 846
#5 ORBit_POAObject_handle_request
at poa.c line 1357
#6 ORBit_POAObject_invoke_incoming_request
at poa.c line 1427
#7 giop_thread_queue_process
at giop.c line 792
#8 giop_request_handler_thread
at giop.c line 502
#9 g_thread_pool_thread_proxy
at gthreadpool.c line 265
#10 g_thread_create_proxy
at gthread.c line 635
#11 start_thread
at pthread_create.c line 297
#12 clone
at ../sysdeps/unix/sysv/linux/x86_64/clone.S line 112
#13 ??

(gdb) 



Other information:
Found this crash in trunk build
Evolution revision: 37139
EDS revision: 9971

Comment 1 Kandepu Prasad 2009-06-08 08:58:11 UTC

I got this crash again in 2.27.3 build.

Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0xadafcb90 (LWP 15732)]
0xb6632eba in e_gw_item_get_full_name (item=0x81dee68) at e-gw-item.c:891
(gdb) t a a bt

+ Trace 215914

Thread 31 (Thread 0xadafcb90 (LWP 15732))

#0 e_gw_item_get_full_name
at e-gw-item.c line 891
#1 fill_contact_from_gw_item
at e-book-backend-groupwise.c line 1180
#2 update_address_book_deltas
at e-book-backend-groupwise.c line 3064
#3 g_thread_create_proxy
at gthread.c line 635
#4 start_thread
from /lib/libpthread.so.0
#5 clone
from /lib/libc.so.6

Comment 2 Akhil Laddha 2009-10-05 11:53:28 UTC

*** Bug 597255 has been marked as a duplicate of this bug. ***

Comment 3 Milan Crha 2009-12-08 11:37:47 UTC

Created attachment 149325 [details] [review]
proposed eds patch

for evolution-data-server;

I do not have groupwise, thus this is just a blind shot into the darkness...
It seems that the fetching of contact-list details failed for some reason and was used previously freed EGWItem pointer in reading. This is trying to check for failure conditions. Such failed contact-lists will not have members, I guess, but I'm not sure, maybe only not all, hard to tell.

Can anyone test this, please?

Comment 4 Magnus Boman 2009-12-08 21:07:40 UTC

This patch fixes the issue for me. Thanks!!

Comment 5 Magnus Boman 2009-12-17 20:06:50 UTC

After upgrading to e-d-s 2.28.2, this issue is back:


Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7f2866fda910 (LWP 6869)]
e_gw_item_get_full_name (item=0x0) at e-gw-item.c:909
909	{
(gdb) thread apply all bt

+ Trace 219661

Thread 10 (Thread 0x7f28657d7910 (LWP 6870))

#0 g_unichar_isalnum
from /usr/lib64/libglib-2.0.so.0
#1 read_attribute
at e-vcard.c line 535
#2 parse
at e-vcard.c line 654
#3 e_vcard_construct
at e-vcard.c line 774
#4 e_contact_new_from_vcard
at e-contact.c line 1245
#5 e_book_backend_db_cache_get_contacts
at e-book-backend-db-cache.c line 301
#6 book_view_thread
at e-book-backend-groupwise.c line 2296
#7 ??
from /usr/lib64/libglib-2.0.so.0
#8 start_thread
from /lib64/libpthread.so.0
#9 clone
from /lib64/libc.so.6
#10 ??

Thread 9 (Thread 0x7f2866fda910 (LWP 6869))

#0 e_gw_item_get_full_name
at e-gw-item.c line 909
#1 populate_full_name
at e-book-backend-groupwise.c line 662
#2 fill_contact_from_gw_item
#3 update_address_book_deltas
at e-book-backend-groupwise.c line 3014
#4 ??
from /usr/lib64/libglib-2.0.so.0
#5 start_thread
from /lib64/libpthread.so.0
#6 clone
from /lib64/libc.so.6
#7 ??

Thread 7 (Thread 0x7f2868b74910 (LWP 6861))

#0 pthread_join
from /lib64/libpthread.so.0
#1 pthread_attr_setdetachstate
from /usr/lib64/libgthread-2.0.so.0
#2 g_thread_join
from /usr/lib64/libglib-2.0.so.0
#3 e_book_backend_groupwise_authenticate_user
at e-book-backend-groupwise.c line 3341
#4 ORBit_small_invoke_adaptor
from /usr/lib64/libORBit-2.so.0
#5 ??
from /usr/lib64/libORBit-2.so.0
#6 ??
from /usr/lib64/libORBit-2.so.0
#7 giop_thread_queue_process
from /usr/lib64/libORBit-2.so.0
#8 ??
from /usr/lib64/libORBit-2.so.0
#9 ??
from /usr/lib64/libglib-2.0.so.0
#10 ??
from /usr/lib64/libglib-2.0.so.0
#11 start_thread
from /lib64/libpthread.so.0
#12 clone
from /lib64/libc.so.6
#13 ??

Comment 6 Akhil Laddha 2010-03-19 03:49:13 UTC

*** Bug 607435 has been marked as a duplicate of this bug. ***

Comment 7 Akhil Laddha 2010-03-19 03:50:05 UTC

There is a patch attached in bug 607435 as well.

https://bugzilla.gnome.org/show_bug.cgi?id=607435#c7

Comment 8 Milan Crha 2010-03-26 15:28:25 UTC

(In reply to comment #5)
> After upgrading to e-d-s 2.28.2, this issue is back

Did you apply the patch too? Also, the above patch also prevents using of uninitialized memory, whereas the patch from the bug #607435 seems to check only for NULL/not-NULL values, but the uninitialized 'item' variable can be pretty anything, not only NULL. Maybe both can be used.

Comment 9 Milan Crha 2010-09-01 14:49:17 UTC

As spoken on IRC:

Created commit 2130b68 in eds master (2.31.92+)