Mount of samba share always failes IF password are saved

[PM]

Binary package hint: nautilus

Mounting samba shares fails consistenly if, and only if, password are saved. I have reported this bug previosly in https://bugs.launchpad.net/bugs/530605. However, this has apparently been fixed with libgnome-keyring as the root cause. On my test machine, a completely updated system Lucid-beta2 (9th april 2010), this still fails. Everything works fine in Ubuntu Karmic (updated) and from Window clients.

Details:
Mounting a NAS (Synology) share in Nautilus works fine, but only as long as I choose NOT to save passwords. I can mount multiple shares on this server without problem. However, if I choose to remember the password (either until logout or permanently), every attempt to mount fails with this pop-up: "Could not open location 'smb://xxxx@192.168.2.10/xxxx/' DBus error org.freedesktop.DBus.Error.NoReply: Message did not receive a reply (timeout by message bus)". Spesificaly, mounting every shares on the server fails. Rebooting does not solve this, but if I delete the stored passwords in Seahorse everything works. That is... until another attempt to save passwords.

I have dumped the Nautilus output (attachments). Furthermore I have run "/usr/lib/gvfs/./gvfsd -r", which output "ERROR:gkr-operation.c:169:gkr_operation_set_result: assertion failed: ((int) res != INCOMPLETE)" whenever a mount fails as described.

This bug may be in Seahorse. It is a nasty bug!

ProblemType: Bug
DistroRelease: Ubuntu 10.04
Package: nautilus 1:2.30.0-0ubuntu4
ProcVersionSignature: Ubuntu 2.6.32-19.28-generic 2.6.32.10+drm33.1
Uname: Linux 2.6.32-19-generic i686
Architecture: i386
CheckboxSubmission: fb9b92a485987b5d5d3a348263ec9590
CheckboxSystem: 13d06691acf655d5e27c3bafb5fa8769
Date: Fri Apr 9 21:04:00 2010
ExecutablePath: /usr/bin/nautilus
InstallationMedia: Ubuntu 10.04 "Lucid Lynx" - Beta i386 (20100318)
ProcEnviron:
 LANGUAGE=nb_NO:nb:no_NO:no:nn_NO:nn:en
 LANG=nb_NO.utf8
 SHELL=/bin/bash
SourcePackage: nautilus
XsessionErrors:
 (polkit-gnome-authentication-agent-1:1546): GLib-CRITICAL **: g_once_init_leave: assertion `initialization_value != 0' failed
 (gnome-terminal:2052): Gtk-CRITICAL **: gtk_accel_map_unlock_path: assertion `entry != NULL && entry->lock_count > 0' failed

[Peng (pengwg)]

I have this problem on a laptop but not on a 64bit desktop. Could this problem be associate with wifi?

[malbo (pique-sel)]

Il have similar problem when i chose to store the password. I receive the message "DBus error org.freedesktop.DBus.Error.NoReply: Message did not receive a reply (timeout by message bus)"
Apport doesn't works and apport.log contain these 3 lines :
apport (pid 1869) Wed Apr 14 12:20:12 2010: called for pid 1867, signal 6
apport (pid 1869) Wed Apr 14 12:20:12 2010: executable: /usr/lib/gvfs/gvfsd-smb (command line "/usr/lib/gvfs/gvfsd-smb --spawner :1.6 /org/gtk/gvfs/exec_spaw/5")
apport (pid 1869) Wed Apr 14 12:20:16 2010: wrote report /var/crash/_usr_lib_gvfs_gvfsd-smb.1000.crash

I assume that the file _usr_lib_gvfs_gvfsd-smb.1000.crash is'nt correct : i have delete some bad lines with too many characters and i join an extract of this file in attachment without these bad lines.

[malbo (pique-sel)]

Post #3 is obtained on PC HP Vectra
But I have the same on my PC Medion. So I join attachment with this last PC ( I have also delete some bad lines in the middle of the original file)

[malbo (pique-sel)]

All ready on my Medion, some time i have trouble with gvfsd-smb-browse process instead of gvfsd-smb process. In this case, apport.log contain these 3 lines :
apport (pid 2100) Wed Apr 14 17:44:46 2010: called for pid 2098, signal 6
apport (pid 2100) Wed Apr 14 17:44:46 2010: executable: /usr/lib/gvfs/gvfsd-smb-browse (command line "/usr/lib/gvfs/gvfsd-smb-browse --spawner :1.6 /org/gtk/gvfs/exec_spaw/6")
apport (pid 2100) Wed Apr 14 17:44:47 2010: wrote report /var/crash/_usr_lib_gvfs_gvfsd-smb-browse.1000.crash

And the corresponding file (_usr_lib_gvfs_gvfsd-smb-browse.1000.crash) contain wrong lines at the end. So I join in attachment a .txt file that copy only the clean lines.

[PM (pm77)]

I can confirm that all is working now (as of 2010-04-15), This includes mounting/unmounting without delay. No more "DBus error org.freedesktop.DBus.Error". Good job!

[malbo (pique-sel)]

after update today, it's works really better. Now, it's possible to browse the samba shares with password stored. But i still have the bug sometime on both PC's (see comments #3 to #5). In this case, if I repeat the request for mount, I obtain finally the share. It's a very good improvement. A little bit effort, and the bug will be solved.

[malbo (pique-sel)]

When th bug happens on my HP Vectra :
Apport.log :
apport (pid 2202) Fri Apr 16 14:34:41 2010: called for pid 2200, signal 6
apport (pid 2202) Fri Apr 16 14:34:41 2010: executable: /usr/lib/gvfs/gvfsd-smb (command line "/usr/lib/gvfs/gvfsd-smb --spawner :1.6 /org/gtk/gvfs/exec_spaw/25")
apport (pid 2202) Fri Apr 16 14:34:46 2010: wrote report /var/crash/_usr_lib_gvfs_gvfsd-smb.1000.crash
An extract of the corresponding file (_usr_lib_gvfs_gvfsd-smb.1000.crash) is attached

[Nikolaus Filus (nfilus)]

With the updates until 2010-04-15 it's not all-or-nothing for me, but it still fails randomly for different shares. apport.log shows crashes of_usr_lib_gvfs_gvfsd-smb.1000.crash like with the other reporters.

Therefore I will continue with bug #554103.

[PM (pm77)]

Hmmm...I may have been to quick. Lucid still crashes sometimes with ""DBus error org.freedesktop.DBus.Error". It's a completely updated system as of 17th April 2010.

[Sebastien Bacher (seb128)]

Thanks for the bug report. This particular bug has already been reported, but feel free to report any other bugs you find.

[Sebastien Bacher (seb128)]

see bug #560588 for example

[Nikolaus Filus (nfilus)]

Sebastian, sorry to bother you but just marking as invalid and abandoning this report is not helping anyone. As the last change to libgnome-keyring did affect this report, I'm changing the component affected. Comments are welcome...

[Sebastien Bacher (seb128)]

the bug has been closed because it's a duplicate and we don't need several bugs registered about the same issue