Bug 213431 – Crash: evolution-mail deadlock/hang

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 213431 - Crash: evolution-mail deadlock/hang


Summary:	Crash: evolution-mail deadlock/hang


Status:	RESOLVED NOTXIMIAN

Product:	GAL
Classification:	Deprecated
Component:	ETable
Version:	unspecified
Hardware:	Other All

Importance:	Normal blocker
Target Milestone:	1.0.x
Assigned To:	Chris Lahey
QA Contact:	Evolution QA team

URL:
Whiteboard:

Duplicates:	213620 (view as bug list)
Depends on:
Blocks:

Reported:	2001-10-24 20:10 UTC by Michael Thome
Modified:	2001-11-01 15:06 UTC

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

Attachments
current set of rpms (13.68 KB, patch) 2001-10-25 12:40 UTC, Michael Thome	none	Details \| Review
ps, gdb bt/threads of various procs (14.01 KB, text/plain) 2001-10-30 15:00 UTC, Michael Thome		Details

Description Michael Thome 2001-10-24 20:10:29 UTC

Package: Evolution
Priority: Major
Version: 0.15
Synopsis: evolution-mail deadlock/hang
Bugzilla-Product: Evolution
Bugzilla-Component: Mailer

Description:
Several times a day, evolution hangs solid (no refresh, no activity). 
This crash is the result of doing "killall -SEGV evolution-mail" in the
hopes that it will give you some clue as to what is going on.  Please
let me know if there is anything more useful I can pass along.




Debugging Information:

[New Thread 1024 (LWP 2808)]
[New Thread 2049 (LWP 2835)]
0x40b685c9 in __wait4 () from /lib/i686/libc.so.6

+ Trace 12290

Thread 1 (Thread 1024 (LWP 2808))

#0 __wait4
from /lib/i686/libc.so.6
#1 __DTOR_END__
from /lib/i686/libc.so.6
#2 waitpid
at wrapsyscall.c line 172
#3 gnome_segv_handle
at gnome-init.c line 659
#4 segv_redirect
at main.c line 67
#5 pthread_sighandler
at signals.c line 97
#6 <signal handler called>
#7 __libc_read
from /lib/i686/libc.so.6
#8 __DTOR_END__
from /lib/i686/libpthread.so.0
#9 _IO_file_read
at fileops.c line 838
#10 _IO_new_file_underflow
at fileops.c line 542
#11 __underflow
at genops.c line 345
#12 _IO_file_xsgetn
at fileops.c line 1045
#13 _IO_sgetn
at genops.c line 480
#14 _IO_fread
at iofread.c line 44
#15 gzread
from /usr/lib/libz.so.1
#16 xmlParserInputBufferGrow
at xmlIO.c line 372
#17 xmlOldParserInputGrow
at parserold.c line 205
#18 xmlOldParseStartTag
at parserold.c line 5289
#19 xmlOldParseElement
at parserold.c line 5626
#20 xmlOldParseContent
at parserold.c line 5546
#21 xmlOldParseElement
at parserold.c line 5696
#22 xmlOldParseContent
at parserold.c line 5546
#23 xmlOldParseElement
at parserold.c line 5696
#24 xmlOldParseDocument
at parserold.c line 6313
#25 xmlParseDocument
at parser.c line 8782
#26 xmlSAXParseFile
at parser.c line 10725
#27 xmlParseFile
at parser.c line 10752
#28 bonobo_ui_node_from_file
at bonobo-ui-node.c line 477
#29 bonobo_ui_util_new_ui
at bonobo-ui-util.c line 1212
#30 bonobo_ui_util_set_ui
at bonobo-ui-util.c line 1258
#31 ui_add
at folder-browser-ui.c line 180
#32 folder_browser_ui_add_list
at folder-browser-ui.c line 328
#33 control_activate
at folder-browser-factory.c line 92
#34 gtk_marshal_NONE__BOOL
at gtkmarshal.c line 181
#35 gtk_handlers_run
at gtksignal.c line 1917
#36 gtk_signal_real_emit
at gtksignal.c line 1477
#37 gtk_signal_emit
at gtksignal.c line 552
#38 impl_Bonobo_Control_activate
at bonobo-control.c line 211
#39 _ORBIT_skel_Bonobo_Control_activate
at Bonobo-skels.c line 6262
#40 ORBit_POA_handle_request
at orbit_poa.c line 507
#41 ORBit_handle_incoming_request
at server.c line 90
#42 ORBit_handle_incoming_message
at server.c line 160
#43 giop_main_handle_connection
at connection.c line 1211
#44 orb_handle_connection
at oaf-mainloop.c line 69
#45 g_io_unix_dispatch
at giounix.c line 137
#46 g_main_dispatch
at gmain.c line 656
#47 g_main_iterate
at gmain.c line 877
#48 g_main_run
at gmain.c line 935
#49 gtk_main
at gtkmain.c line 524
#50 bonobo_main
at bonobo-main.c line 283
#51 main
at main.c line 141
#52 __libc_start_main
at ../sysdeps/generic/libc-start.c line 129
#0 __wait4
from /lib/i686/libc.so.6
#0 __wait4
from /lib/i686/libc.so.6
#1 __DTOR_END__
from /lib/i686/libc.so.6
#2 waitpid
at wrapsyscall.c line 172
#3 gnome_segv_handle
at gnome-init.c line 659
#4 segv_redirect
at main.c line 67
#5 pthread_sighandler
at signals.c line 97
#6 <signal handler called>
#7 __libc_read
from /lib/i686/libc.so.6



Unknown reporter: mthome@bbn.com, changed to bugbuddy-import@ximian.com.

Comment 1 Luis Villa 2001-10-24 21:07:37 UTC

Can you guys take a look at this?

Comment 2 Not Zed 2001-10-24 21:09:25 UTC

First, i'd suggest updating.

Second, this looks like bonobo or zlib.

Comment 3 Dan Winship 2001-10-25 00:46:24 UTC

what version of gnome-xml do you have installed?

Comment 4 Michael Thome 2001-10-25 12:39:11 UTC

Currently, I'm running with:
bonobo-1.0.7-ximian.3
libbonobo-conf0-0.12-ximian.1
bonobo-devel-1.0.7-ximian.3
bonobo-conf-0.12-ximian.1
zlib-1.1.3-12
zlib-devel-1.1.3-12
libxml-devel-1.8.15-ximian.2
libxml-1.8.15-ximian.2
FYI, I'll also attach my *current* set of rpms.
I'll also plan on doing a red-carpet update later today

Comment 5 Michael Thome 2001-10-25 12:40:06 UTC

Created attachment 40430 [details] [review]
current set of rpms

Comment 6 Michael Thome 2001-10-26 12:32:36 UTC

I've done a red-carpet update.  Here are the current versions of the
above packages:
bonobo-1.0.9-ximian.1
libbonobo-conf0-0.12-ximian.1
bonobo-devel-1.0.9-ximian.1
bonobo-conf-0.12-ximian.1
zlib-1.1.3-12
zlib-devel-1.1.3-12
libxml-devel-1.8.15-ximian.2
libxml-1.8.15-ximian.2
evolution-0.16-ximian.2

see bug 213620 for a post-update crash done the same way.

Additional info - it seems to most often occur after reading a message
in one folder and then selecting a different folder.

Comment 7 Jeffrey Stedfast 2001-10-27 23:38:54 UTC

*** bug 213620 has been marked as a duplicate of this bug. ***

Comment 8 Not Zed 2001-10-29 21:35:06 UTC

Can you attach with gdb instead of usihng SEGV?

Need to find out if its looping, or waiting for non-existant
resources.

But unless you know how to drive gdb, i can't see how we can do this.

Comment 9 Michael Thome 2001-10-30 15:00:15 UTC

Created attachment 40463 [details]
ps, gdb bt/threads of various procs

Comment 10 Jeffrey Stedfast 2001-10-30 21:32:19 UTC

looks like etable?

Comment 11 Chris Lahey 2001-10-30 22:32:14 UTC

gnome-xml bug.

Comment 12 Dan Winship 2001-10-31 17:19:21 UTC

more specific, Chris? Is this a *fixed* gnome-xml bug? And if
not, is it workaroundable? And has it been reported?

Comment 13 Michael Thome 2001-11-01 15:06:33 UTC

Could this be due to libxml as non-thread-safe?  libxml2.4.7 is the
first version which claims to be thread safe.  If evolution is parsing
in parallel, it could easily explain the visible behavior.

If this is the problem, it could certainly be viewed as a
ximian bug (use of a lib in an unsupported way).  In any case, if
ximian cannot resolve the bug, you'll at least need to document
evolution as being unstable in SMP environments.