GNOME Bugzilla – Bug 300157
LDAP search crashes e-d-s
Last modified: 2013-09-14 16:49:05 UTC
Distribution: Debian 3.1 Package: Evolution-Data-Server Severity: major Version: GNOME2.8.3 unspecified Gnome-Distributor: Debian Synopsis: e-d-s crashed when opening new mail window Bugzilla-Product: Evolution-Data-Server Bugzilla-Component: Contacts Bugzilla-Version: unspecified BugBuddy-GnomeVersion: 2.0 (2.8.1) Description: Description of the crash: e-d-s crashed when opening new mail window. I think it could be caused by LDAP and its autocompletion. I also have LDAP offline synchronization enabled. I hope the back-trace will help. How often does this happen? It doesn't happen everytime. Debugging Information: Backtrace was generated from '/usr/libexec/evolution-data-server-1.2' Using host libthread_db library "/lib/tls/i686/cmov/libthread_db.so.1". (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) [Thread debugging using libthread_db enabled] [New Thread -1218776416 (LWP 11417)] [New Thread -1550021712 (LWP 27063)] [New Thread -1281455184 (LWP 27062)] [New Thread -1272927312 (LWP 26887)] [New Thread -1239077968 (LWP 11459)] [New Thread -1222132816 (LWP 11418)] (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) 0xffffe410 in ?? ()
+ Trace 58012
------- Bug moved to this database by unknown@bugzilla.gnome.org 2005-04-11 04:45 ------- Unknown platform unknown. Setting to default platform "Other". Unknown milestone "unknown" in product "Evolution-Data-Server". Setting to default milestone for this product, '---' The original reporter of this bug does not have an account here. Reassigning to the person who moved it here, unknown@bugzilla.gnome.org. Previous reporter was jan.mynarik@systinet.com. Setting to default status "UNCONFIRMED". Setting qa contact to the default for this product. This bug either had no qa contact or an invalid one.
Thanks for the bug report. Looks like the stack trace in bug 273286 to me, can someone else confirm?
Actually, I (the reporter) have these 100% CPU problems too. It takes several minutes (IBM R40, 1.8GHz). But when this 100% thing happens, my evo doesn't hang. I suspect LDAP addressbook replication for offline usage is the problem.
I switched off LDAP addressbook replication and I still experience this bug at least TWICE A DAY. The problem is still present in evolution 2.2.2 (and appropriate version of e-d-s etc.). Now the evo itself crashes right after the e-d-s crash and some alert windows.
This is related to LDAP access, it doesn't matter if it's from 'compose mail' window or from 'schedule meeting'. When I want to be at least a bit sure that my evo won't crash next time I'll use LDAP, I need to go to addressbook, click on LDAP account and then push 'clear' button. It causes to load all LDAP items (100% CPU for 40 seconds), it's about 150 items in our company's LDAP. But even this magic doesn't work always, sooner or later evo (e-d-s, to be specific) crashes. Actually, evolution crashed 2-6 times a day. Many times it 'disables' my tasks and calendars during this crash (it's reported via dialogs) and I need to enable them (using checkboxes) after evolution restart.
This bug is really annoying. Today my evo even froze (without crash and without using 100% CPU either). Is there any chance to help you to resolve this bug? I desperately need it fixed.
Corrected summary as per request by Jan.
Still present in e-d-s 1.3.6. Is there any way I can help? I wouldn't use evolution (and experiance crash almost everytime I try to write an email) If I wasn't a huge GNOME fan. New back-trace: Backtrace was generated from '/usr/libexec/evolution-data-server-1.4' Using host libthread_db library "/lib/tls/i686/cmov/libthread_db.so.1". (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) [Thread debugging using libthread_db enabled] [New Thread -1219446560 (LWP 16955)] [New Thread -1292506192 (LWP 17154)] [New Thread -1267254352 (LWP 17153)] [New Thread -1231262800 (LWP 17151)] [New Thread -1239766096 (LWP 16992)] [New Thread -1222866000 (LWP 16956)] (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) 0xffffe410 in ?? ()
+ Trace 62022
Thread 4 (Thread -1231262800 (LWP 17151))
I'll bump up the priority/severity since it's a crasher and update the version numbers. If you could get a stack trace with debugging symbols that would probably help the developers, but otherwise, you'll have to wait to hear back from one of them...
Here is a new back-trace with debugging symbols. Say if it's still not enough :-) Backtrace was generated from '/usr/libexec/evolution-data-server-1.4' Using host libthread_db library "/usr/lib/debug/libthread_db.so.1". [Thread debugging using libthread_db enabled] [New Thread 16384 (LWP 31495)] [New Thread 32769 (LWP 31496)] [New Thread 16386 (LWP 31497)] [New Thread 49156 (LWP 31528)] [New Thread 311307 (LWP 31584)] 0xb75f884b in waitpid () from /usr/lib/debug/libc.so.6
+ Trace 62029
Thread 5 (Thread 311307 (LWP 31584))
Approximately how many user objects your server is having?
There are currently 147 contacts. 95% of them have a photo assigned in LDAP.
*** Bug 306872 has been marked as a duplicate of this bug. ***
*** Bug 308889 has been marked as a duplicate of this bug. ***
Not reproducible in our set up. Moving to 2.4.1
I can provide you with more information. Just say what you need to know. Was the last backtrace enough? Maybe I can compile more things with debug symbols. This bug is really annoying here. And still reproducible with newest packages (e-d-s 1.3.8, evo 2.3.8). I had it on my old laptop with Debian Sid and I have it with new laptop and Ubuntu Breezy too.
Created attachment 51660 [details] [review] patch - with the mutex for ldap connection. can someone try this patch? I am suspecting this is because of ldap connection time out.
Just tried it. It still crashes but it seems to happen a bit later. I can still crash e-d-s using one new-message-window and trying to add more people to (or delete from) 'To' of 'CC' fields. New backtrace from bug-buddy attached. The backtrace mentions e-book-backend-ldap.c:3441, it is this part of source file /* XXX should the call to op->handler be protected by the lock? */ g_static_rec_mutex_unlock (&bl->priv->op_hash_mutex); Backtrace was generated from '/usr/libexec/evolution-data-server-1.4' Using host libthread_db library "/lib/tls/i686/cmov/libthread_db.so.1". [Thread debugging using libthread_db enabled] [New Thread -1219893568 (LWP 2972)] [New Thread -1269109840 (LWP 3071)] [New Thread -1241637968 (LWP 2984)] [New Thread -1224455248 (LWP 2973)] 0xffffe410 in ?? ()
+ Trace 62705
Thread 4 (Thread -1224455248 (LWP 2973))
*** Bug 268360 has been marked as a duplicate of this bug. ***
*** Bug 320076 has been marked as a duplicate of this bug. ***
Jan Mynarik, can you please check if you are finding the same issue in 2.4.2?
Still the same: evolution: 2.5.4-0ubuntu4, e-d-s: 1.5.4-0ubuntu3. I've had the problem with all versions (in ubuntu Breeze or Dapper) since my last reported version (e-d-s 1.3.8, evo 2.3.8). Backtrace was generated from '/usr/libexec/evolution-data-server-1.6' (no debugging symbols found) Using host libthread_db library "/lib/tls/i686/cmov/libthread_db.so.1". (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) [Thread debugging using libthread_db enabled] [New Thread -1220794688 (LWP 14933)] [New Thread -1233056848 (LWP 14047)] [New Thread -1249842256 (LWP 13981)] [New Thread -1259340880 (LWP 13975)] [New Thread -1284518992 (LWP 10126)] [New Thread -1224397904 (LWP 14934)] 0xffffe410 in __kernel_vsyscall ()
+ Trace 65156
Thread 4 (Thread -1259340880 (LWP 13975))
*** This bug has been marked as a duplicate of 274240 ***