GNOME Bugzilla – Bug 360792
Crash in bacon_message_connection_new when using multiple displays or VNC
Last modified: 2008-11-29 11:27:23 UTC
Version: 2.15.9 What were you doing when the application crashed? running 2 desktop Distribution: Fedora Core release 5.92 (FC6 Test3) Gnome Release: 2.16.0 2006-09-04 (Red Hat, Inc) BugBuddy Version: 2.16.0 Memory status: size: 103157760 vsize: 0 resident: 103157760 share: 0 rss: 21458944 rss_rlim: 0 CPU usage: start_time: 1160354485 rtime: 0 utime: 60 stime: 0 cutime:58 cstime: 0 timeout: 2 it_real_value: 0 frequency: 0 Backtrace was generated from '/usr/bin/gedit' (no debugging symbols found) Using host libthread_db library "/lib/libthread_db.so.1". (no debugging symbols found) [Thread debugging using libthread_db enabled] [New Thread -1208587520 (LWP 3244)] (no debugging symbols found) 0x00cb6402 in __kernel_vsyscall ()
+ Trace 75350
Thread 1 (Thread -1208587520 (LWP 3244))
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. Thanks in advance!
*** Bug 371362 has been marked as a duplicate of this bug. ***
*** Bug 376836 has been marked as a duplicate of this bug. ***
I think this is pretty easy to reproduce: Steps to Reproduce: 1. su -c gedit on display :0 2. su -c gedit on display :1 Actual results: - gedit crashes Expected results: - no crash - gedit to use one process per display
Could you please provide a stack trace with debugging symbols? Does it crash only when using "su -c"?
Created attachment 78181 [details] gedit-root-segv.backtrace How about this one? (I think preventing gedit from running across different X display would avoid this problem FWIW.)
> I think preventing gedit from running across different X display would > avoid this problem FWIW Why? It seems a completely valid use case to me. Or you just mean for the root user? (gedit:3028): GnomeUI-WARNING **: While connecting to session manager: Authentication Rejected, reason : None of the authentication protocols specified are supported and host-based authentication failed. Xlib: connection to ":0.0" refused by server Xlib: No protocol specified I'd be more interested to know why this happen, since I think is the reason for the crash... maybe it is enforced at the system level that root cannot run gui apps remotely?
Created attachment 78197 [details] [review] minimal patch Can you check if this minimal patch at least prevents the crash? I still would like to understand if the fact that we cannot open that display is a system policy or what. About the patch, I am really not sure how to do error reporting for this condition... I put a g_warning in there for now, but this (or an eventual dialog) are displayed server-side, that is on the wrong display.
Thanks for the patch. I will try to test that soon. You may well be right that the authentication warning relates to the crash. With a root desktop session or using ssh -Y I don't see the crash. I still think it is a bit unintuitive sharing the gedit process for a user across different displays, but I guess it is considered a feature.
I am committing the above patch, because handling gracefully the case where the returned display is NULL is something we should do no matter what. I am leaving the bug open to: 1) Investigate under which conditions a display can't be opened 2) Figure out what should we do in the GUI when that happens
*** Bug 388047 has been marked as a duplicate of this bug. ***
*** Bug 390831 has been marked as a duplicate of this bug. ***
I'm not sure it is the same crash, but they have a very similar stack traces, so I'm going to file as duplicate a list of crashes happening using gedit with VNC.
*** Bug 391966 has been marked as a duplicate of this bug. ***
*** Bug 392607 has been marked as a duplicate of this bug. ***
*** Bug 394035 has been marked as a duplicate of this bug. ***
*** Bug 402525 has been marked as a duplicate of this bug. ***
*** Bug 404933 has been marked as a duplicate of this bug. ***
*** Bug 411483 has been marked as a duplicate of this bug. ***
*** Bug 412691 has been marked as a duplicate of this bug. ***
*** Bug 412762 has been marked as a duplicate of this bug. ***
*** Bug 413204 has been marked as a duplicate of this bug. ***
*** Bug 417491 has been marked as a duplicate of this bug. ***
*** Bug 419526 has been marked as a duplicate of this bug. ***
*** Bug 431038 has been marked as a duplicate of this bug. ***
*** Bug 431415 has been marked as a duplicate of this bug. ***
*** Bug 433360 has been marked as a duplicate of this bug. ***
*** Bug 446457 has been marked as a duplicate of this bug. ***
*** Bug 456291 has been marked as a duplicate of this bug. ***
*** Bug 475710 has been marked as a duplicate of this bug. ***
we got no more reports of this crash. closing.