GNOME Bugzilla – Bug 333738
LDAP backend regularly crashes
Last modified: 2013-09-14 16:49:42 UTC
Steps to reproduce: 1. Have a slow remode LDAP addressbook active. 2. Fill an address. 3. Before LDAP completes reply, press Backspace or other key to abort search. The LDAP backend crashes. This bug is not easy to reproduce, but it crashes several times in a week. The bug is reproducible in all previous version of evolution. Stack trace: Backtrace from version 1.5.92 (SuSE Linux 10.1 beta6, x86_64): Backtrace was generated from '/opt/gnome/libexec/evolution-data-server-1.6' Using host libthread_db library "/lib64/libthread_db.so.1". [Thread debugging using libthread_db enabled] [New Thread 47147893310160 (LWP 4588)] [New Thread 1191504224 (LWP 22174)] [New Thread 1233467744 (LWP 22172)] [New Thread 1250253152 (LWP 22170)] [New Thread 1275431264 (LWP 22168)] [New Thread 1183111520 (LWP 22161)] [New Thread 1090791776 (LWP 4635)] [New Thread 1074006368 (LWP 4596)] 0x00002ae1792ef72f in __libc_waitpid (pid=22165, stat_loc=0x7fff3547d84c, options=0) at waitpid.c:41 41 int result = INLINE_SYSCALL (wait4, 4, pid, stat_loc, options, NULL);
+ Trace 66751
Thread 1 (Thread 47147893310160 (LWP 4588))
0x00002ae1792ef72f 41 int result = INLINE_SYSCALL (wait4, 4, pid, stat_loc, options, NULL); Other information: More backtraces and comments can be found in: https://bugzilla.novell.com/show_bug.cgi?id=117813
Not happening for me in Suse 10.0. Also traces are not showing any details of ldap backend. Also, your older traces in Novell bugzilla look different than this.
Oops. This crash occurred during the exactly the same operation (abort of auto-complete operation by pressing Enter on one of found entries), but in different place of code. But https://bugzilla.novell.com/show_bug.cgi?id=117813#c13 is not so old (1.5.90) and crashed in the LDAP backend. Not easy to reproduce, it happens only few times in a week. I am using just now 7 auto-completion address books: Default address book (local) Private address book (local) SuSE address book (local, contains lists and some users) Automatic contacts (local) SuSE.DE (LDAP) SuSE.CZ (LDAP, server sends a query to other LDAP servers) Novell (GroupWise Address Book) I will continue in observing of this bug.
And there is a LDAP backend crash. evolution-2.5.92-6, evolution-data-server-1.5.92-5, SuSE Linux 10.1 beta7 AMD64 Backtrace was generated from '/opt/gnome/libexec/evolution-data-server-1.6' Using host libthread_db library "/lib64/libthread_db.so.1". [Thread debugging using libthread_db enabled] [New Thread 47724718035664 (LWP 4119)] [New Thread 1250253136 (LWP 20576)] [New Thread 1208289616 (LWP 20569)] [New Thread 1099184464 (LWP 4148)] [New Thread 1074006352 (LWP 4124)] 0x00002b67c69dc78f in __libc_waitpid (pid=20575, stat_loc=0x7fffe7d90cec, options=0) at waitpid.c:41 41 int result = INLINE_SYSCALL (wait4, 4, pid, stat_loc, options, NULL);
+ Trace 66821
Thread 1 (Thread 47724718035664 (LWP 4119))
And another one. It is a little different, but again in the LDAP backend: Backtrace was generated from '/opt/gnome/libexec/evolution-data-server-1.6' Using host libthread_db library "/lib64/libthread_db.so.1". [Thread debugging using libthread_db enabled] [New Thread 47406787478224 (LWP 20620)] [New Thread 1225075024 (LWP 20815)] [New Thread 1241860432 (LWP 20804)] [New Thread 1082399056 (LWP 20779)] [New Thread 1074006352 (LWP 20621)] 0x00002b1dc07a778f in __libc_waitpid (pid=20793, stat_loc=0x7fffedfc63bc, options=0) at waitpid.c:41 41 int result = INLINE_SYSCALL (wait4, 4, pid, stat_loc, options, NULL);
+ Trace 66822
Thread 1 (Thread 47406787478224 (LWP 20620))
New backtrace for SuSE Linux 10.1 pre-RC1 packages (weekend build). evolution-2.6.0-10 evolution-data-server-1.6.0-9 Backtrace was generated from '/opt/gnome/libexec/evolution-data-server-1.6' Using host libthread_db library "/lib64/libthread_db.so.1". [Thread debugging using libthread_db enabled] [New Thread 47236555182800 (LWP 11358)] [New Thread 1208289616 (LWP 4101)] [New Thread 1099184464 (LWP 4092)] [New Thread 1090791760 (LWP 11568)] [New Thread 1074006352 (LWP 11359)] 0x00002af61dd7c80f in __libc_waitpid (pid=4094, stat_loc=0x7fff909ee44c, options=0) at waitpid.c:41 41 int result = INLINE_SYSCALL (wait4, 4, pid, stat_loc, options, NULL);
+ Trace 67275
Thread 1 (Thread 47236555182800 (LWP 11358))
*** Bug 339955 has been marked as a duplicate of this bug. ***
There is an another backtrace. This time, evolution-data-server did not crash, but evolution experienced freeze while trying to complete e-mail: evolution backtrace: Backtrace was generated from '/opt/gnome/bin/evolution-2.6' Using host libthread_db library "/lib64/libthread_db.so.1". [Thread debugging using libthread_db enabled] [New Thread 47554769297200 (LWP 1258)] [New Thread 1124096320 (LWP 1368)] [New Thread 1141147968 (LWP 1275)] [New Thread 1140881728 (LWP 1273)] [New Thread 1132489024 (LWP 1272)] [New Thread 1115703616 (LWP 1269)] [New Thread 1107310912 (LWP 1268)] [New Thread 1098918208 (LWP 1266)] [New Thread 1090525504 (LWP 1264)] [New Thread 1082132800 (LWP 1263)] 0x00002b403175ed3f in waitpid () from /lib64/libpthread.so.0
+ Trace 68725
Thread 1 (Thread 47554769297200 (LWP 1258))
evolution-data-server backtrace: (note: both backtraces were generated by kill -11 after several minutes of no response of evolution) Backtrace was generated from '/opt/gnome/libexec/evolution-data-server-1.6' Using host libthread_db library "/lib64/libthread_db.so.1". [Thread debugging using libthread_db enabled] [New Thread 47189023707232 (LWP 28859)] [New Thread 1199896896 (LWP 3450)] [New Thread 1191504192 (LWP 3449)] [New Thread 1183111488 (LWP 3448)] [New Thread 1174718784 (LWP 3447)] [New Thread 1149540672 (LWP 3446)] [New Thread 1225075008 (LWP 3445)] [New Thread 1233467712 (LWP 3444)] [New Thread 1208289600 (LWP 3442)] [New Thread 1275431232 (LWP 3435)] [New Thread 1250253120 (LWP 3432)] [New Thread 1099184448 (LWP 28883)] [New Thread 1074006336 (LWP 28860)] 0x00002aeb0c9d473f in waitpid () from /lib64/libc.so.6
+ Trace 68726
Thread 1 (Thread 47189023707232 (LWP 28859))
*** Bug 345689 has been marked as a duplicate of this bug. ***
Created attachment 71522 [details] [review] Patch This patch should fix most of the issues mentioned in the bug. Committing this patch to head, please check with head and see if you still observe any issues.