nautilus inhibits logout

Bug #419184 reported by Luca Ferretti
96
This bug affects 14 people
Affects Status Importance Assigned to Milestone
nautilus (Ubuntu)
Fix Released
Medium
Ubuntu Desktop Bugs

Bug Description

Binary package hint: nautilus

Steps to reproduce
 * Select a file or directory
 * Trash it (or delete)
 * Try to logout

Current result
  gnome-session reports Nautilus action (or actions) are still running, inhibiting logout. But relevant files was yet deleted or trashed or moved or copied other files while session was running

Notes
 This seems retaled to recent addition of "inhibit sleep while performing action".

ProblemType: Bug
Architecture: i386
Date: Wed Aug 26 14:17:50 2009
DistroRelease: Ubuntu 9.10
Package: nautilus 1:2.27.91-0ubuntu1
ProcEnviron:
 LANG=it_IT.UTF-8
 SHELL=/bin/bash
ProcVersionSignature: Ubuntu 2.6.31-7.27-generic
SourcePackage: nautilus
Uname: Linux 2.6.31-7-generic i686

Related branches

Revision history for this message
Luca Ferretti (elle.uca) wrote :
Revision history for this message
A. Walton (awalton) wrote :

We definitely added uninhibit calls at the job's finish, so either we're leaking jobs or for whatever reason the Uninhibit call isn't working.

Does valgrind complain we're leaking *Job structs? Can you add a break in nautilus-file-operations.c:finalize_common and make sure we're actually calling it at the end of jobs (and thusly nautilus_uninhibit_power_manager?) If you run nautilus in a terminal does it complain that it can't uninhibit the power manager? Does dbus-monitor --session show the outgoing "Uninhibit" call?

Changed in nautilus (Ubuntu):
assignee: nobody → Ubuntu Desktop Bugs (desktop-bugs)
importance: Undecided → Medium
status: New → Incomplete
Revision history for this message
Luca Ferretti (elle.uca) wrote :

Andrew, by now I was able to perform only the dbus-monitor test, there is the log:

method call sender=:1.79 -> dest=org.freedesktop.DBus serial=18 path=/org/freedesktop/DBus; interface=org.freedesktop.DBus; member=AddMatch
   string "type='signal',sender='org.gnome.SessionManager',path='/org/gnome/SessionManager',interface='org.gnome.SessionManager'"
method call sender=:1.79 -> dest=org.freedesktop.DBus serial=19 path=/org/freedesktop/DBus; interface=org.freedesktop.DBus; member=GetNameOwner
   string "org.gnome.SessionManager"
method call sender=:1.79 -> dest=org.gnome.SessionManager serial=20 path=/org/gnome/SessionManager; interface=org.gnome.SessionManager; member=Inhibit
   string "Nautilus"
   uint32 0
   string "Trashing Files"
   uint32 5
signal sender=:1.3 -> dest=(null destination) serial=86 path=/org/gnome/SessionManager; interface=org.gnome.SessionManager; member=InhibitorAdded
   object path "/org/gnome/SessionManager/Inhibitor1"
method return sender=:1.3 -> dest=:1.79 reply_serial=20
   uint32 614526181
method call sender=:1.37 -> dest=:1.3 serial=34 path=/org/gnome/SessionManager; interface=org.gnome.SessionManager; member=IsInhibited
   uint32 8
method return sender=:1.3 -> dest=:1.37 reply_serial=34
   boolean false
method call sender=:1.48 -> dest=org.freedesktop.DBus serial=416 path=/org/freedesktop/DBus; interface=org.freedesktop.DBus; member=RemoveMatch
   string "type='signal',sender=':1.83',path='/',interface='org.freedesktop.DBus.Introspectable'"

No "Uninhibit" call :(

PS it seems that "nautilus -q" don't prevent nautilus to be respawned. Is this related to /desktop/gnome/session/required_components_list ?

Revision history for this message
Luca Ferretti (elle.uca) wrote :

One more test launching nautilus from terminal. Here is the "log"

luca@luca-virtual:~$ nautilus
Initializing nautilus-gdu extension

** (nautilus:3602): WARNING **: No marshaller for signature of signal 'UploadFinished'

** (nautilus:3602): WARNING **: No marshaller for signature of signal 'DownloadFinished'

** (nautilus:3602): CRITICAL **: nautilus_uninhibit_power_manager: assertion `cookie < 0' failed

Revision history for this message
Jon Arnold (jonarnoldsemail) wrote :

Just an addition- this bug keeps me from doing a full soft-shutdown. I have to power off.

Revision history for this message
A. Walton (awalton) wrote :

** (nautilus:3602): CRITICAL **: nautilus_uninhibit_power_manager: assertion `cookie < 0' failed

This was all I needed to see. Another g_return_if_fail() with the test in the wrong direction. Fixed in trunk.

Changed in nautilus (Ubuntu):
status: Incomplete → Fix Committed
Revision history for this message
kylea (kylea) wrote :

After latest updates on 27/8 the Nautilus "wait trashing" message is gone, however the PC will not fully shutdown, have to press and hold power button.

Does anyone else have this behaviour?

Revision history for this message
Ryan (ubuntu-draziw) wrote :

Kylea - yes. Same thing here.

Revision history for this message
Ryan (ubuntu-draziw) wrote :

Spoke too soon - back to the inhibiting logout message on this shutdown. The hang usually shows something about sending programs terminate signals, then just hangs, but will echo keyboard input. Don't think it's related to the nautilus shutdown/logout issue.

Revision history for this message
A. Walton (awalton) wrote :

The shutdown thing should have nothing to do with Nautilus. The issue with the screenshot is fixed in HEAD now Ryan (and is actually the bug reported here). The machine not shutting down should not be a Nautilus issue, as you can always press "log out anyways" and that will kill Nautilus from the session manager.

Revision history for this message
Chris Coulson (chrisccoulson) wrote :

People experiencing issues with shutting down are experiencing a known bug with the 2.6.31-7 kernel, which is already reported somewhere on Launchpad, and it is unrelated to the Nautilus issue here.

Revision history for this message
kylea (kylea) wrote :

"Nautilus "wait trashing" message is gone," - has returned.

Copy the point on shutdown

I think is this

https://bugs.launchpad.net/ubuntu/+source/linux/+bug/418509

Revision history for this message
Luca Ferretti (elle.uca) wrote :

Kylea, the nautilus bug was fixed on git.gnome.org, not yet available in karmic. In fact the status of this bug is Fix Commited, not Fix Released :)

The shutdown issues, intead, should be now fixed in karmic after kernel update. At least in my karmic virtual machine.

Revision history for this message
kylea (kylea) wrote :

Luca Ok thanks - all good will wait

Revision history for this message
Launchpad Janitor (janitor) wrote :

This bug was fixed in the package nautilus - 1:2.27.91-0ubuntu2

---------------
nautilus (1:2.27.91-0ubuntu2) karmic; urgency=low

  * debian/patches/90_git_change_fix_inhibit_issue.patch:
    - git change to fix an inhibit issue (lp: #419184)

 -- Sebastien Bacher <email address hidden> Mon, 31 Aug 2009 11:24:38 +0200

Changed in nautilus (Ubuntu):
status: Fix Committed → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.