Bug 420462 – evolution crashed when trying to view folder subscriptions

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 420462 - evolution crashed when trying to view folder subscriptions


Summary:	evolution crashed when trying to view folder subscriptions


Status:	RESOLVED FIXED

Product:	evolution
Classification:	Applications
Component:	Mailer
Version:	2.10.x (obsolete)
Hardware:	Other Linux

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

URL:
Whiteboard:

Depends on:
Blocks:

Reported:	2007-03-20 09:42 UTC by Sebastien Bacher
Modified:	2007-08-13 05:15 UTC

See Also:
GNOME target:	---
GNOME version:	2.17/2.18

Attachments
Proposed patch (960 bytes, patch) 2007-08-10 08:38 UTC, Srinivasa Ragavan	committed	Details \| Review

Description Sebastien Bacher 2007-03-20 09:42:24 UTC

The bug has been opened on https://launchpad.net/bugs/93987

"Binary package hint: evolution

When attempting to view folder subscriptions on an IMAP server, Evolution was locking up, when I clicked 'Refresh' it crashed.

ProblemType: Crash
Architecture: i386
Date: Tue Mar 20 01:51:51 2007
DistroRelease: Ubuntu 7.04
ExecutablePath: /usr/bin/evolution
InterpreterPath: /usr/bin/evolution-2.10
Package: evolution 2.10.0-0ubuntu1
PackageArchitecture: i386
ProcCmdline: evolution --component=mail
...
.

+ Trace 120302

Thread 1 (process 29464)

#0 IA__gtk_tree_path_get_indices
at gtktreemodel.c line 619
#1 _gtk_tree_view_find_node
at gtktreeview.c line 8903
#2 IA__gtk_tree_view_expand_row
at gtktreeview.c line 11825
#3 sub_folderinfo_free
at em-subscribe-editor.c line 413
#4 mail_msg_free
at mail-mt.c line 193
#5 mail_msgport_replied
at mail-mt.c line 475
#6 g_io_unix_dispatch
at giounix.c line 162
#7 IA__g_main_context_dispatch
at gmain.c line 2045
#8 g_main_context_iterate
at gmain.c line 2677
#9 IA__g_main_loop_run
at gmain.c line 2881
#10 bonobo_main
at bonobo-main.c line 311
#11 main
at main.c line 611

Comment 1 Srinivasa Ragavan 2007-08-03 06:18:17 UTC

Thanks for taking the time to report this bug.
Unfortunately, that stack trace is missing some elements that will help a lot to solve the problem, so it will be hard for the developers to fix that crash. Can you get us a stack trace with debugging symbols? Please see http://live.gnome.org/GettingTraces for more information on how to do so and reopen this bug or report a new one. Thanks in advance!

Comment 2 Sebastien Bacher 2007-08-03 10:23:18 UTC

the thread of the crash has debug informations, what other informations do you need?

Comment 3 Srinivasa Ragavan 2007-08-03 10:42:16 UTC

Seb, looks like the other threads has corrupted information and the thread that has debug info has no relevance of a crash.

Comment 4 Srinivasa Ragavan 2007-08-03 11:53:43 UTC

Here is the full trace :-)


Memory status: size: 230731776 vsize: 0 resident: 230731776 share: 0 rss: 29495296 rss_rlim: 0
CPU usage: start_time: 1186136519 rtime: 0 utime: 4799 stime: 0 cutime:4419 cstime: 0 timeout: 380 it_real_value: 0 frequency: 14

Backtrace was generated from '/home/sragavan/opt/gnome218/bin/evolution'

Using host libthread_db library "/lib/libthread_db.so.1".
[Thread debugging using libthread_db enabled]
[New Thread -1233271104 (LWP 4356)]
[New Thread -1362949232 (LWP 29730)]
[New Thread -1390273648 (LWP 28548)]
[New Thread -1381233776 (LWP 4370)]
[New Thread -1371341936 (LWP 4367)]
[New Thread -1354556528 (LWP 4364)]
[New Thread -1346163824 (LWP 4363)]
[New Thread -1336874096 (LWP 4362)]
0xb7f678b2 in _dl_sysinfo_int80 () from /lib/ld-linux.so.2

+ Trace 152535

Thread 1 (Thread -1233271104 (LWP 4356))

#0 _dl_sysinfo_int80
from /lib/ld-linux.so.2
#1 __waitpid_nocancel
from /lib/libpthread.so.0
#2 libgnomeui_segv_handle
at gnome-ui-init.c line 872
#3 <signal handler called>
#4 _dl_sysinfo_int80
from /lib/ld-linux.so.2
#5 raise
from /lib/libc.so.6
#6 abort
from /lib/libc.so.6
#7 g_logv
at gmessages.c line 497
#8 g_log
at gmessages.c line 517
#9 g_return_if_fail_warning
#10 gtk_tree_view_expand_row
from /home/stable/opt/gnome2/lib/libgtk-x11-2.0.so.0
#11 sub_folderinfo_free
at em-subscribe-editor.c line 413
#12 mail_msg_free
at mail-mt.c line 193
#13 mail_msgport_replied
at mail-mt.c line 475
#14 g_io_unix_dispatch
at giounix.c line 162
#15 g_main_context_dispatch
at gmain.c line 2045
#16 g_main_context_iterate
at gmain.c line 2677
#17 g_main_loop_run
at gmain.c line 2881
#18 bonobo_main
at bonobo-main.c line 311
#19 main
at main.c line 594
#0 _dl_sysinfo_int80

Comment 5 Srinivasa Ragavan 2007-08-10 08:38:26 UTC

Created attachment 93416 [details] [review]
Proposed patch

Comment 6 Srinivasa Ragavan 2007-08-10 08:48:16 UTC

This bug has occured post public-folder optimization. It has changed a few things in the subscribe editor. Subscription editor depends a lot on CAMEL_FOLDER_NOCHILDREN flag now and it is not set well by the imap provider. 

The EMFolderTree uses the folderinfo->child to determine this. So I patched the imap provider, when ever a FI is created it has NOCHILDREN and when the FI tree is built, if the child is present, I unset that flag in the FI that way the flag represents the state in the node.

I may be wrong, I dont know how to set this in the provider directly, any better solutions are welcome.

This crash is just a minor issue, there are more. I saw a infinite loop and hangs due to this. 

Btw, It I haven't tested NNTP. There could be some brokenness in subscription.

Comment 7 Srinivasa Ragavan 2007-08-13 05:15:32 UTC

Committed to head for now.