Debian Bug report logs - #569058
helper not setuid; breaks gnome-power-manager

Toggle useless messages

Message #5 received at submit@bugs.debian.org (full text, mbox, reply):

From: Joey Hess <joeyh@debian.org>

To: Debian Bug Tracking System <submit@bugs.debian.org>

Subject: helper not setuid; breaks gnome-power-manager

Date: Tue, 9 Feb 2010 13:43:44 -0500

[Message part 1 (text/plain, inline)]

Package: dbus
Version: 1.2.20-2
Severity: normal

gnome-power-manager is not working, devkit-power has the same
problem:

(devkit-power:10045): devkit-power-gobject-WARNING **: Couldn't enumerate devices: The permission of the setuid helper is not correct

Is this supposed to be setuid?

-rwxr-xr-x 1 root root 39K Feb  3 22:19 /usr/lib/dbus-1.0/dbus-daemon-launch-helper*

Downgrading dbus to testing fixed it for me:

-rwsr-xr-- 1 root messagebus 39K Jul 15  2009 /usr/lib/dbus-1.0/dbus-daemon-launch-helper*

Then I re-upgraded, and the permissions were preserved.

Two possibilities:

1. Btrfs, which I use for my root filesystem, is more experimental than
   expected and somehow responsible for the lost suid bits and group
   ownership. Possible, but seems unlikely.
2. Somehow dbus.postinst's chown failed. Maybe I had a dpkg-statoverride
   entry for the helper from before (if so, I had not added it manually,
   and I don't have it now).

-- System Information:
Debian Release: squeeze/sid
  APT prefers unstable
  APT policy: (500, 'unstable'), (500, 'testing'), (500, 'stable')
Architecture: i386 (i686)

Kernel: Linux 2.6.31-1-686 (SMP w/2 CPU cores)
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/bash

Versions of packages dbus depends on:
ii  adduser                       3.112      add and remove users and groups
ii  libc6                         2.10.2-5   Embedded GNU C Library: Shared lib
ii  libdbus-1-3                   1.2.20-2   simple interprocess messaging syst
ii  libexpat1                     2.0.1-7    XML parsing C library - runtime li
ii  libselinux1                   2.0.89-4   SELinux runtime shared libraries
ii  lsb-base                      3.2-23     Linux Standard Base 3.2 init scrip

dbus recommends no packages.

Versions of packages dbus suggests:
ii  dbus-x11                      1.2.20-2   simple interprocess messaging syst

-- debconf-show failed

-- 
see shy jo

[signature.asc (application/pgp-signature, inline)]

Message #10 received at 569058@bugs.debian.org (full text, mbox, reply):

From: Michael Biebl <biebl@debian.org>

To: Joey Hess <joeyh@debian.org>, 569058@bugs.debian.org

Subject: Re: [Pkg-utopia-maintainers] Bug#569058: helper not setuid; breaks gnome-power-manager

Date: Tue, 09 Feb 2010 19:54:28 +0100

[Message part 1 (text/plain, inline)]

Hi Joey!

Joey Hess wrote:
> Package: dbus
> Version: 1.2.20-2
> Severity: normal
> 
> gnome-power-manager is not working, devkit-power has the same
> problem:
> 
> (devkit-power:10045): devkit-power-gobject-WARNING **: Couldn't enumerate devices: The permission of the setuid helper is not correct
> 
> Is this supposed to be setuid?
> 
> -rwxr-xr-x 1 root root 39K Feb  3 22:19 /usr/lib/dbus-1.0/dbus-daemon-launch-helper*
> 
> Downgrading dbus to testing fixed it for me:
> 
> -rwsr-xr-- 1 root messagebus 39K Jul 15  2009 /usr/lib/dbus-1.0/dbus-daemon-launch-helper*

Yes, the helper should be setuid.

> Then I re-upgraded, and the permissions were preserved.
> 
> Two possibilities:
> 
> 1. Btrfs, which I use for my root filesystem, is more experimental than
>    expected and somehow responsible for the lost suid bits and group
>    ownership. Possible, but seems unlikely.

Possible.

> 2. Somehow dbus.postinst's chown failed. Maybe I had a dpkg-statoverride
>    entry for the helper from before (if so, I had not added it manually,
>    and I don't have it now).

From which version did you upgrade?
You can find the current postinst at [1] along with its history. Afaics the last
time this particular code has been touched is over 2 years ago. I highly doubt
that such a bug would have been unnoticed for so long.

Cheers,
Michael


[1]
http://svn.debian.org/wsvn/pkg-utopia/packages/unstable/dbus/debian/dbus.postinst
-- 
Why is it that all of the instruments seeking intelligent life in the
universe are pointed away from Earth?

[signature.asc (application/pgp-signature, attachment)]

Message #15 received at 569058@bugs.debian.org (full text, mbox, reply):

From: Joey Hess <joeyh@debian.org>

To: Michael Biebl <biebl@debian.org>

Cc: 569058@bugs.debian.org

Subject: Re: [Pkg-utopia-maintainers] Bug#569058: helper not setuid; breaks gnome-power-manager

Date: Tue, 9 Feb 2010 14:10:34 -0500

[Message part 1 (text/plain, inline)]

Michael Biebl wrote:
> From which version did you upgrade?

2010-02-07 19:00:44 upgrade dbus 1.2.16-2 1.2.20-2

> You can find the current postinst at [1] along with its history. Afaics the last
> time this particular code has been touched is over 2 years ago. I highly doubt
> that such a bug would have been unnoticed for so long.

I can't see how what happened could happen unless dpkg-statoverride == true.
And if the helper was statoverridden before, why would it not be now.

-- 
see shy jo

[signature.asc (application/pgp-signature, inline)]

Message #20 received at 569058@bugs.debian.org (full text, mbox, reply):

From: Baggett Jonas <Jonas.Baggett@edu.hefr.ch>

To: "569058@bugs.debian.org" <569058@bugs.debian.org>

Date: Sat, 20 Feb 2010 18:50:42 +0100

Hi

I recently converted my root partition from ext4 to btrfs and I experimented the same problem.
I cannot tell if the reason was the upgrade of dbus (1.2.16-2 to 1.2.20-2)
or the conversion of my root partition because they happened almost the same time.
At first glance this bug seems to exist only with btrfs.

Bye
Jonas

Message #25 received at 569058@bugs.debian.org (full text, mbox, reply):

From: Michael Biebl <biebl@debian.org>

To: Baggett Jonas <Jonas.Baggett@edu.hefr.ch>, 569058@bugs.debian.org, Joey Hess <joeyh@debian.org>

Subject: Re: [Pkg-utopia-maintainers] Bug#569058:

Date: Sat, 20 Feb 2010 23:27:49 +0100

[Message part 1 (text/plain, inline)]

Baggett Jonas wrote:
> Hi
> 
> I recently converted my root partition from ext4 to btrfs and I experimented the same problem.
> I cannot tell if the reason was the upgrade of dbus (1.2.16-2 to 1.2.20-2)
> or the conversion of my root partition because they happened almost the same time.
> At first glance this bug seems to exist only with btrfs.

Joey,

given this additional information, I very much assume that this indeed is a
btrfs issue and I can't see any bug in dbus' handling of the setuid helper.

Are you ok with closing this bug?

Michael


-- 
Why is it that all of the instruments seeking intelligent life in the
universe are pointed away from Earth?

[signature.asc (application/pgp-signature, attachment)]

Message #30 received at 569058@bugs.debian.org (full text, mbox, reply):

From: Joey Hess <joeyh@debian.org>

To: Michael Biebl <biebl@debian.org>

Cc: Baggett Jonas <Jonas.Baggett@edu.hefr.ch>, 569058@bugs.debian.org

Subject: Re: [Pkg-utopia-maintainers] Bug#569058:

Date: Thu, 25 Feb 2010 01:08:58 -0500

[Message part 1 (text/plain, inline)]

Michael Biebl wrote:
> given this additional information, I very much assume that this indeed is a
> btrfs issue and I can't see any bug in dbus' handling of the setuid helper.
> 
> Are you ok with closing this bug?

If this bug is really a bug in btrfs that causes it to lose track of
mode settings done by dpkg, it really needs to be brought to the
attention of the btfs developers. Ideally before btrfs becomes
non-experimental..

I'm unsure about drawing that conclusion yet since the problem is not
easily reproducible and both of us who saw it did so after the same
1.2.16-2 -> 1.2.20-2 dbus upgrade.

-- 
see shy jo

[signature.asc (application/pgp-signature, inline)]

Message #35 received at 569058@bugs.debian.org (full text, mbox, reply):

From: Timo Juhani Lindfors <timo.lindfors@iki.fi>

To: 569058@bugs.debian.org

Subject: Re: [Pkg-utopia-maintainers] Bug#569058:

Date: Tue, 09 Mar 2010 21:58:12 +0200

Hi,

I think I have hit this bug at least twice last year. I did not report
it since I thought that the helper was not supposed to be suid and
having it suid was only my own local hack. (I'm running
freesmartphone.org software on my openmoko that everyone else runs as
root but I run as a normal user...)

All this time I have been using ext3 on a 2G microSD card so I don't
think you can blame btrfs alone.

From backups I can recover the following info:

$ sudo grep dbus-daemon-launch /home/share/neo/20*/var/log/auth.log
/home/share/neo/2009-05-29/var/log/auth.log:May 29 01:40:49 ginger sudo:    lindi : TTY=pts/5 ; PWD=/usr/share/doc/dbus ; USER=root ; COMMAND=/bin/chmod u-x /usr/lib/dbus-1.0/dbus-daemon-launch-helper

$ sudo grep dbus-daemon-launch /home/share/neo/20*/home/lindi/.bash_history
/home/share/neo/2009-05-29/home/lindi/.bash_history:ls -l  /usr/lib/dbus-1.0/dbus-daemon-launch-helper
/home/share/neo/2009-05-29/home/lindi/.bash_history:chown root.dbus /usr/lib/dbus-1.0/dbus-daemon-launch-helper
/home/share/neo/2009-05-29/home/lindi/.bash_history:chown root:dbus /usr/lib/dbus-1.0/dbus-daemon-launch-helper
/home/share/neo/2009-05-29/home/lindi/.bash_history:chown root:messagebus /usr/lib/dbus-1.0/dbus-daemon-launch-helper
/home/share/neo/2009-05-29/home/lindi/.bash_history:chmod u+s /usr/lib/dbus-1.0/dbus-daemon-launch-helper
/home/share/neo/2009-05-29/home/lindi/.bash_history:ls -l  /usr/lib/dbus-1.0/dbus-daemon-launch-helper
/home/share/neo/2009-05-29/home/lindi/.bash_history:chmod o-rx  /usr/lib/dbus-1.0/dbus-daemon-launch-helper
/home/share/neo/2009-05-29/home/lindi/.bash_history:chmod o+r  /usr/lib/dbus-1.0/dbus-daemon-launch-helper
/home/share/neo/2009-05-29/home/lindi/.bash_history:ls -l  /usr/lib/dbus-1.0/dbus-daemon-launch-helper
/home/share/neo/2009-05-29/home/lindi/.bash_history:dpkg -L /usr/lib/dbus-1.0/dbus-daemon-launch-helper

$ grep dbus /home/share/neo/2009-05-29/var/log/dpkg.log
2009-05-27 14:15:02 upgrade libdbus-1-dev 1.2.12-1 1.2.14-2
2009-05-27 14:15:02 status half-configured libdbus-1-dev 1.2.12-1
2009-05-27 14:15:03 status unpacked libdbus-1-dev 1.2.12-1
2009-05-27 14:15:03 status half-installed libdbus-1-dev 1.2.12-1
2009-05-27 14:15:05 status half-installed libdbus-1-dev 1.2.12-1
2009-05-27 14:15:08 status unpacked libdbus-1-dev 1.2.14-2
2009-05-27 14:15:08 status unpacked libdbus-1-dev 1.2.14-2
2009-05-27 14:15:09 upgrade libdbus-1-3 1.2.12-1 1.2.14-2
2009-05-27 14:15:09 status half-configured libdbus-1-3 1.2.12-1
2009-05-27 14:15:09 status unpacked libdbus-1-3 1.2.12-1
2009-05-27 14:15:10 status half-installed libdbus-1-3 1.2.12-1
2009-05-27 14:15:10 status half-installed libdbus-1-3 1.2.12-1
2009-05-27 14:15:12 status unpacked libdbus-1-3 1.2.14-2
2009-05-27 14:15:12 status unpacked libdbus-1-3 1.2.14-2
2009-05-27 14:15:50 upgrade dbus 1.2.12-1 1.2.14-2
2009-05-27 14:15:50 status half-configured dbus 1.2.12-1
2009-05-27 14:15:51 status unpacked dbus 1.2.12-1
2009-05-27 14:15:52 status half-installed dbus 1.2.12-1
2009-05-27 14:15:53 status half-installed dbus 1.2.12-1
2009-05-27 14:15:54 status half-installed dbus 1.2.12-1
2009-05-27 14:15:55 status unpacked dbus 1.2.14-2
2009-05-27 14:15:56 status unpacked dbus 1.2.14-2
2009-05-27 14:17:06 upgrade dbus-x11 1.2.12-1 1.2.14-2
2009-05-27 14:17:06 status half-configured dbus-x11 1.2.12-1
2009-05-27 14:17:07 status unpacked dbus-x11 1.2.12-1
2009-05-27 14:17:07 status half-installed dbus-x11 1.2.12-1
2009-05-27 14:17:07 status half-installed dbus-x11 1.2.12-1
2009-05-27 14:17:08 status half-installed dbus-x11 1.2.12-1
2009-05-27 14:17:09 status unpacked dbus-x11 1.2.14-2
2009-05-27 14:17:10 status unpacked dbus-x11 1.2.14-2
2009-05-27 14:18:16 upgrade libdbus-glib-1-dev 0.80-3 0.80-4
2009-05-27 14:18:16 status half-configured libdbus-glib-1-dev 0.80-3
2009-05-27 14:18:17 status unpacked libdbus-glib-1-dev 0.80-3
2009-05-27 14:18:17 status half-installed libdbus-glib-1-dev 0.80-3
2009-05-27 14:18:18 status half-installed libdbus-glib-1-dev 0.80-3
2009-05-27 14:18:20 status unpacked libdbus-glib-1-dev 0.80-4
2009-05-27 14:18:21 status unpacked libdbus-glib-1-dev 0.80-4
2009-05-27 14:18:21 upgrade libdbus-glib-1-2 0.80-3 0.80-4
2009-05-27 14:18:21 status half-configured libdbus-glib-1-2 0.80-3
2009-05-27 14:18:22 status unpacked libdbus-glib-1-2 0.80-3
2009-05-27 14:18:22 status half-installed libdbus-glib-1-2 0.80-3
2009-05-27 14:18:23 status half-installed libdbus-glib-1-2 0.80-3
2009-05-27 14:18:26 status unpacked libdbus-glib-1-2 0.80-4
2009-05-27 14:18:26 status unpacked libdbus-glib-1-2 0.80-4
2009-05-27 14:35:11 install libqt4-dbus 4.4.3-2 4.5.1-2
2009-05-27 14:35:11 status half-installed libqt4-dbus 4.4.3-2
2009-05-27 14:35:14 status unpacked libqt4-dbus 4.5.1-2
2009-05-27 14:35:14 status unpacked libqt4-dbus 4.5.1-2
2009-05-29 01:56:29 configure libdbus-1-3 1.2.14-2 1.2.14-2
2009-05-29 01:56:29 status unpacked libdbus-1-3 1.2.14-2
2009-05-29 01:56:29 status half-configured libdbus-1-3 1.2.14-2
2009-05-29 01:56:31 status installed libdbus-1-3 1.2.14-2
2009-05-29 01:56:36 configure dbus 1.2.14-2 1.2.14-2
2009-05-29 01:56:36 status unpacked dbus 1.2.14-2
2009-05-29 01:56:37 status unpacked dbus 1.2.14-2
2009-05-29 01:56:37 status unpacked dbus 1.2.14-2
2009-05-29 01:56:38 status unpacked dbus 1.2.14-2
2009-05-29 01:56:38 status unpacked dbus 1.2.14-2
2009-05-29 01:56:39 status half-configured dbus 1.2.14-2
2009-05-29 01:57:09 status installed dbus 1.2.14-2
2009-05-29 01:58:47 configure dbus-x11 1.2.14-2 1.2.14-2
2009-05-29 01:58:47 status unpacked dbus-x11 1.2.14-2
2009-05-29 01:58:48 status unpacked dbus-x11 1.2.14-2
2009-05-29 01:58:48 status half-configured dbus-x11 1.2.14-2
2009-05-29 01:58:48 status installed dbus-x11 1.2.14-2
2009-05-29 01:59:41 configure libdbus-glib-1-2 0.80-4 0.80-4
2009-05-29 01:59:41 status unpacked libdbus-glib-1-2 0.80-4
2009-05-29 01:59:41 status half-configured libdbus-glib-1-2 0.80-4
2009-05-29 01:59:43 status installed libdbus-glib-1-2 0.80-4
2009-05-29 02:05:46 configure libdbus-1-dev 1.2.14-2 1.2.14-2
2009-05-29 02:05:46 status unpacked libdbus-1-dev 1.2.14-2
2009-05-29 02:05:47 status half-configured libdbus-1-dev 1.2.14-2
2009-05-29 02:05:47 status installed libdbus-1-dev 1.2.14-2
2009-05-29 02:05:48 configure libdbus-glib-1-dev 0.80-4 0.80-4
2009-05-29 02:05:48 status unpacked libdbus-glib-1-dev 0.80-4
2009-05-29 02:05:48 status half-configured libdbus-glib-1-dev 0.80-4
2009-05-29 02:05:48 status installed libdbus-glib-1-dev 0.80-4
2009-05-29 02:08:22 configure libqt4-dbus 4.5.1-2 4.5.1-2
2009-05-29 02:08:22 status unpacked libqt4-dbus 4.5.1-2
2009-05-29 02:08:23 status half-configured libqt4-dbus 4.5.1-2
2009-05-29 02:08:25 status installed libqt4-dbus 4.5.1-2

I'm happy to provide more information including complete root
filesystem tarballs for days of interest.

Message #40 received at 569058@bugs.debian.org (full text, mbox, reply):

From: Christian Weeks <cpw@weeksfamily.ca>

To: 569058@bugs.debian.org

Subject: Clearly NOT just btrfs

Date: Tue, 09 Mar 2010 14:52:18 -0500

Because I am running on ext3 and I experienced the problem on two
different systems.

/dev/mapper/alliecore-root on / type ext3 (rw,errors=remount-ro)

2010-02-18 10:26:49 upgrade dbus 1.2.16-2 1.2.20-2


I have LVM under my filesystem. Dunno if that's a factor.

My workaround, make it x for all:

-rwsr-xr-x 1 root messagebus 45792 Feb  3
22:45 /usr/lib/dbus-1.0/dbus-daemon-launch-helper

Other systems have worked fine however. It's a peculiar issue for sure.

Thanks Joey for linking this bug, it's been annoying me for days!

Christian

Message #45 received at 569058-done@bugs.debian.org (full text, mbox, reply):

From: Joey Hess <joey@kitenet.net>

To: cwillu <cwillu@cwillu.com>

Cc: 575891@bugs.debian.org, 569058-done@bugs.debian.org

Subject: Re: btrfs issues

Date: Tue, 6 Apr 2010 14:33:19 -0400

[Message part 1 (text/plain, inline)]

cwillu wrote:
> New information on http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=575891
> 
> Turns out to have been an unsafe assumption on dpkg's part with
> apparently astronomic odds of being triggered on most filesystems.
> 
> Putting /var/lib/dpkg on an ext3 mount (I used a bind mount from my
> /boot) works around it until the problem can be fixed in dpkg.

Cwillu, thanks for following up.

That does seem to explain #569058, closing that bug.

At the moment, my other two problems, #567135 and #568908, do not
seem likely to be caused by the dpkg bug.

-- 
see shy jo

[signature.asc (application/pgp-signature, inline)]

Message #50 received at 569058@bugs.debian.org (full text, mbox, reply):

From: Alexander Kurtz <kurtz.alex@googlemail.com>

To: 569058@bugs.debian.org

Subject: Re: btrfs issues

Date: Wed, 14 Apr 2010 16:54:21 +0200

[Message part 1 (text/plain, inline)]

Hi,

I also experienced this issue with dbus and btrfs and fixed it by simply
reinstalling the dbus package.

I also think it's not the first time this happens: I've had a few
experiences where different error messages lead me to reinstalling a
package, which fixed that particular problem.

From my experience this happens particularly in situations where the
btrfs file system is under heavy load (this doesn't mean much data is
read/written, but that there are are *lots* of open files which are
read/written to simultaneously) like upgrading a lot of packages with
dpkg.

I _guess_ that something roughly like that happens:
 * dpkg does a lot of small file modifications
   (creating, changing perms, etc.)
 * btrfs caches these transactions
 * other file modifications happen
 * the original dpkg transactions are delayed over and over again
 * the machine is rebooted and the cache is lost

However just guessing without having looked at the code.

So for provoking this bug I would:
 * create/remove lots of small files
 * change their permissions and possibly content
 * maybe do a snapshot
 * repeat the whole procedure

Best Regards

Alexander Kurtz

[signature.asc (application/pgp-signature, inline)]

Send a report that this bug log contains spam.