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 45698 - Clicking option to enable, disable, enable desktop crashes Nautilus
Clicking option to enable, disable, enable desktop crashes Nautilus
Status: RESOLVED DUPLICATE of bug 76342
Product: nautilus
Classification: Core
Component: Desktop
0.x.x [obsolete]
Other Linux
: High minor
: 1.1.x
Assigned To: Nautilus Maintainers
Nautilus Maintainers
Depends on:
Blocks:
 
 
Reported: 2001-01-16 21:23 UTC by bfrank
Modified: 2004-12-22 21:47 UTC
See Also:
GNOME target: ---
GNOME version: 2.0


Attachments
i recreated the crash in 1.1.x (2.00 KB, text/plain)
2002-04-17 04:53 UTC, Dave Bordoley [Not Reading Bug Mail]
Details

Description bfrank 2001-09-10 00:51:56 UTC
I've reliably been able to crash Nautilus by:

1) Turning on the "have Nautilus manage the desktop" option in the user-level
preference menu
2) Turn the option off
3) Turn the option back on again

I've done these operations in fairly quick succession...but it crashes almost
every time.



------- Additional Comments From darin@bentspoon.com 2001-01-17 10:59:10 ----

If true, this sounds like it might even be a PR3 blocker.



------- Additional Comments From engber@eazel.com 2001-01-17 11:15:10 ----

I'm having trouble reproducing this on my machine. Brian was able to show it to
me on his machine - but it wasn't that easy to do.



------- Additional Comments From don@eazel.com 2001-01-17 13:59:59 ----

On the table for PR3 while Mike investigates this ...




------- Additional Comments From don@eazel.com 2001-01-18 10:49:51 ----

Looks like the X server fix made MOST of this go away.  Mike will continue to
investigate this for 1.0.




------- Additional Comments From engber@eazel.com 2001-01-25 10:30:45 ----

I don't think this is a desktop specific problem, I think it could happen to any
nautilus window. It's just easy to repro by turning on/off the desktop.



------- Additional Comments From sullivan@eazel.com 2001-01-29 13:47:59 ----

This could be a "state machine" bug that Darin has already fixed or knows how to
fix. If this still happens, can we get a backtrace in the bug report?



------- Additional Comments From sullivan@eazel.com 2001-01-29 13:53:37 ----

  • #0 __kill
    from /lib/libc.so.6
  • #1 raise
    at signals.c line 64
  • #2 nautilus_stop_in_debugger
    at nautilus-debug.c line 43
  • #3 nautilus_stop_after_default_log_handler
  • #4 g_logv
    at gmessages.c line 386
  • #5 g_log
  • #6 fm_directory_view_begin_loading
    at fm-directory-view.c line 2102
  • #7 finish_loading
    at fm-directory-view.c line 3839
  • #8 finish_loading_if_all_metadata_loaded
    at fm-directory-view.c line 3896
  • #9 metadata_for_files_in_directory_ready_callback
    at fm-directory-view.c line 3932
  • #10 ready_callback_call
    at nautilus-directory-async.c line 1512
  • #11 call_ready_callbacks
    at nautilus-directory-async.c line 1990
  • #12 nautilus_directory_async_state_changed
    at nautilus-directory-async.c line 3209
  • #13 metafile_read_mark_done
    at nautilus-directory-async.c line 463
  • #14 metafile_read_done
    at nautilus-directory-async.c line 470
  • #15 metafile_read_done_callback
    at nautilus-directory-async.c line 611
  • #16 read_file_succeeded
    at nautilus-file-utilities.c line 750
  • #17 read_file_read_callback
    at nautilus-file-utilities.c line 823
  • #18 dispatch_read_callback
    at gnome-vfs-job.c line 231
  • #19 dispatch_job_callback
    at gnome-vfs-job.c line 542
  • #20 g_idle_dispatch
    at gmain.c line 1367
  • #21 g_main_dispatch
    at gmain.c line 656
  • #22 g_main_iterate
    at gmain.c line 877
  • #23 g_main_run
    at gmain.c line 935
  • #24 gtk_main
    at gtkmain.c line 476
  • #25 bonobo_main
    at bonobo-main.c line 275
  • #26 main
    at nautilus-main.c line 265

Here's a stack trace. It looks like the finish_loading callback doesn't take
into account the possibility that the FMDirectoryView might get destroyed in the
meantime.



------- Additional Comments From eli@eazel.com 2001-02-09 11:28:19 ----

Duane is now the proud owner for Desktop QA.



------- Additional Comments From darin@bentspoon.com 2001-02-22 13:46:45 ----

The stack trace in this bug report doesn't happen any more. But I am running
into other assertions when I do this quickly.



------- Additional Comments From darin@bentspoon.com 2001-02-23 12:15:58 ----

The new problem seems to be some kind of trouble with widget destruction.
Sometimes we die inside Bonobo's sync. machinery and other times when destroying
the actual desktop. The bad object seems to be either the menu bar, or a menu or
toolbar item.

I'm having trouble figuring out the cause.

Note that this doesn't usually result in a crash, but sometimes it does. Since
it's some kind of double-destroy it's hard to figure out what the worst symptom
could be. Perhaps we can defer this to 1.2 -- it's terrible, but probably OK to
say "don't bash on that button in 1.0".



------- Additional Comments From don@eazel.com 2001-02-23 13:49:28 ----

Yeah, although this is a crash, I think it can be safely deferred to 1.2 since I
believe it unlikely that people will actually be caffienated enough to do this.




------- Additional Comments From eli@eazel.com 2001-03-26 11:19:20 ----

SPAAAAAAAAAM! 

(Jon Allen has taken these components; QA Assigning bugs to him.)



------- Bug moved to this database by unknown@bugzilla.gnome.org 2001-09-09 20:51 -------

The original reporter (bfrank@eazel.com) of this bug does not have an account here.
Reassigning to the exporter, unknown@bugzilla.gnome.org.

Comment 1 aaron 2001-10-08 19:14:14 UTC
Confirming and prioritizing.
Comment 2 Dave Bordoley [Not Reading Bug Mail] 2002-04-17 04:53:40 UTC
Created attachment 7747 [details]
i recreated the crash in 1.1.x
Comment 3 Dave Bordoley [Not Reading Bug Mail] 2002-04-17 04:54:36 UTC
ccing naut-qa
Comment 4 Dave Bordoley [Not Reading Bug Mail] 2002-04-17 04:59:02 UTC

*** This bug has been marked as a duplicate of 76342 ***