Bug 1132905

Summary: [abrt] caja: handle_error(): caja killed by SIGTRAP
Product: [Fedora] Fedora Reporter: Patrick Monnerat <patrick>
Component: cajaAssignee: Wolfgang Ulbrich <raveit65.sun>
Status: CLOSED WORKSFORME QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: unspecified Docs Contact:
Priority: unspecified    
Version: 20CC: bcostescu, dan.mashal, gotare, ilikesci, mauropegolo, raveit65.sun
Target Milestone: ---   
Target Release: ---   
Hardware: x86_64   
OS: Unspecified   
URL: https://retrace.fedoraproject.org/faf/reports/bthash/46599e5fc2df008e2f4ea4bfad605b68dabf4c7b
Whiteboard: abrt_hash:07231181f3d41645cc1bbbbbc1532306229b893a
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2014-10-24 13:38:17 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
File: backtrace
none
File: cgroup
none
File: core_backtrace
none
File: dso_list
none
File: environ
none
File: limits
none
File: maps
none
File: open_fds
none
File: proc_pid_status
none
File: var_log_messages none

Description Patrick Monnerat 2014-08-22 08:59:22 UTC 
Description of problem:
Beacon appeared at xinetd vnc session startup ...?
No usage trouble.

Version-Release number of selected component:
caja-1.8.1-5.fc20

Additional info:
reporter:       libreport-2.2.3
backtrace_rating: 4
cmdline:        caja
crash_function: handle_error
executable:     /usr/bin/caja
kernel:         3.15.10-200.fc20.x86_64
runlevel:       N 5
type:           CCpp
uid:            1000

Truncated backtrace:
Thread no. 1 (9 frames)
 #4 handle_error at xcb_io.c:213
 #5 handle_response at xcb_io.c:325
 #8 XCloseDisplay at ClDisplay.c:61
 #9 make_root_pixmap at mate-bg.c:1305
 #10 mate_bg_create_pixmap at mate-bg.c:1194
 #11 eel_background_ensure_realized at eel-background.c:363
 #12 eel_background_set_up_widget at eel-background.c:550
 #13 background_changed_cb at eel-background.c:603
 #18 gtk_main at gtkmain.c:1257
Comment 1 Patrick Monnerat 2014-08-22 08:59:25 UTC 
Created attachment 929483 [details]
File: backtrace
Comment 2 Patrick Monnerat 2014-08-22 08:59:26 UTC 
Created attachment 929484 [details]
File: cgroup
Comment 3 Patrick Monnerat 2014-08-22 08:59:28 UTC 
Created attachment 929485 [details]
File: core_backtrace
Comment 4 Patrick Monnerat 2014-08-22 08:59:29 UTC 
Created attachment 929486 [details]
File: dso_list
Comment 5 Patrick Monnerat 2014-08-22 08:59:30 UTC 
Created attachment 929487 [details]
File: environ
Comment 6 Patrick Monnerat 2014-08-22 08:59:31 UTC 
Created attachment 929488 [details]
File: limits
Comment 7 Patrick Monnerat 2014-08-22 08:59:33 UTC 
Created attachment 929489 [details]
File: maps
Comment 8 Patrick Monnerat 2014-08-22 08:59:34 UTC 
Created attachment 929490 [details]
File: open_fds
Comment 9 Patrick Monnerat 2014-08-22 08:59:35 UTC 
Created attachment 929491 [details]
File: proc_pid_status
Comment 10 Patrick Monnerat 2014-08-22 08:59:36 UTC 
Created attachment 929492 [details]
File: var_log_messages
Comment 11 Wolfgang Ulbrich 2014-08-24 09:08:24 UTC 
Maybe i helps if you disable background transitions. (Use set instead of get).

[rave@mother ~]$ gsettings get org.mate.background background-fade
true

I don't think you need this effect for changing a background image in a vnc session. And the backtrace shows that caja chrashed at this point.
I know this because in mate-1.2/1.4 versions for my former external repo i patched out the bg transition function to get a working background for MDM (old obsolete mate fork of GDM).
You can also disable

[rave@mother ~]$ gsettings get org.mate.background draw-background
true

This key is only needed if the key 'org.mate.background show-desktop-icons' is disabled and caja don't draw the bg.
In this case mate-desktop and mate-settings-daemon draws the bg.
Which points to the real issue, imo.
The bg code is pretty distributed over 3 packages in mate. In gnome this is already changed.
Comment 12 Patrick Monnerat 2014-08-25 08:17:55 UTC 
Thanks for the hint, Wolfgang.
In fact, this transition stuff is the default (I've never changed it before).
I set the way you tell and will report here if it occurs again.
Thanks again.
Comment 13 Wolfgang Ulbrich 2014-10-23 17:38:40 UTC 
any news?
Comment 14 Patrick Monnerat 2014-10-24 09:13:06 UTC 
The gsettings trick did not make it with 1.8.1, but 1.8.2 seems OK. Thanks.
Comment 15 Wolfgang Ulbrich 2014-10-24 13:38:17 UTC 
fine, good to hear.
we can reopen the report if it comes back.