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 522192 - gvfsd-sftp crashed with SIGSEGV in g_vfs_channel_connection_closed()
gvfsd-sftp crashed with SIGSEGV in g_vfs_channel_connection_closed()
Status: RESOLVED OBSOLETE
Product: gvfs
Classification: Core
Component: general
1.8.x
Other Linux
: Normal critical
: ---
Assigned To: gvfs-maint
gvfs-maint
: 535814 653969 (view as bug list)
Depends on:
Blocks:
 
 
Reported: 2008-03-13 09:34 UTC by Sebastien Bacher
Modified: 2015-08-06 09:04 UTC
See Also:
GNOME target: ---
GNOME version: 2.91/3.0


Attachments
Patch (1.34 KB, patch)
2008-03-24 11:48 UTC, Carlos Garcia Campos
committed Details | Review

Description Sebastien Bacher 2008-03-13 09:34:23 UTC
The bug has been opened on https://bugs.launchpad.net/ubuntu/+source/gvfs/+bug/201519

"Binary package hint: gvfs

Nautilus file manager crashes EVERYTIME I try to access an ssh server through it.

* Open nautilus.
* In the address bar, type ssh://<address of your server>
* After you are connected, try to move some files between the two systems (your local and the ssh server)
* Nautilus will crash within a few moments
* Automatic crash-report generator starts up and collects whatever data it does collect.

Package: gvfs-backends 0.1.11svn20080307-0ubuntu1

  • #0 g_vfs_channel_connection_closed
    at gvfschannel.c line 245
  • #1 send_reply_cb
    at gvfschannel.c line 547
  • #2 async_ready_callback_wrapper
    at /build/buildd/glib2.0-2.15.6/gio/goutputstream.c line 559
  • #3 IA__g_simple_async_result_complete
    at /build/buildd/glib2.0-2.15.6/gio/gsimpleasyncresult.c line 553
  • #4 write_async_cb
    at /build/buildd/glib2.0-2.15.6/gio/gunixoutputstream.c line 309
  • #5 fd_source_dispatch
    at /build/buildd/glib2.0-2.15.6/gio/gasynchelper.c line 117
  • #6 IA__g_main_context_dispatch
    at /build/buildd/glib2.0-2.15.6/glib/gmain.c line 2065
  • #7 g_main_context_iterate
    at /build/buildd/glib2.0-2.15.6/glib/gmain.c line 2698
  • #8 IA__g_main_loop_run
    at /build/buildd/glib2.0-2.15.6/glib/gmain.c line 2906
  • #9 daemon_main
    at daemon-main.c line 270
  • #10 main
    at daemon-main-generic.c line 39

Comment 1 Carlos Garcia Campos 2008-03-24 11:48:38 UTC
Created attachment 107911 [details] [review]
Patch

I can reproduce this crash by cancelling a read/write operation from the nautilus progress dialog. The problem is that, in gvfschannel, new read/write requests are processed after a cancel request, so that new read/jobs are created that try to read/write on an already cancelled stream. This patch fixes the crash.
Comment 2 Alexander Larsson 2008-03-25 15:31:39 UTC
Looks good. Please commit.
Comment 3 Carlos Garcia Campos 2008-03-25 17:35:58 UTC
Committed to svn trunk. 
Comment 4 Alexander Larsson 2009-03-06 15:14:50 UTC
Hmm. I think this patch was actually wrong and is causing bug 535814. A stream is not cancelled, just an operation on the stream. With this patch we ignore the close operation after a cancelled read operation.
Comment 5 Alexander Larsson 2009-03-06 15:15:17 UTC
*** Bug 535814 has been marked as a duplicate of this bug. ***
Comment 6 Alexander Larsson 2009-03-06 15:54:36 UTC
I reverted the patch in trunk.
I'm trying to reproduce the original crash, but is unable so far.
Comment 7 Cristian Aravena Romero 2011-07-04 19:43:07 UTC
bug #653969 duplicated(?)
Comment 8 Fabio Durán Verdugo 2011-07-05 03:05:35 UTC
*** Bug 653969 has been marked as a duplicate of this bug. ***
Comment 9 Fabio Durán Verdugo 2011-07-05 03:06:42 UTC
last dup at 1.8.x
Comment 10 Pedro Villavicencio 2011-08-15 15:49:58 UTC
we got a dup with1.9.X  https://bugs.launchpad.net/ubuntu/+source/gvfs/+bug/820930
Comment 11 Ondrej Holy 2015-08-06 09:04:54 UTC
Close as obsolete, because I don't see any similar abrt crashes on https://retrace.fedoraproject.org/faf2 . Feel free to reopen if you are still facing this issue.