nih-dbus-tool: dbus_connection_send_with_reply() returns NULL pending_call even when returning TRUE if connection is not connected

[James Michael DuPont]

Binary package hint: mountall

Running mountall produces this output, it never stops.

This is normally a bug in some application using the D-Bus library.
process 12925: arguments to dbus_pending_call_set_notify() were incorrect, assertion "pending != NULL" failed in file dbus-pending-call.c line 596.

Description: Ubuntu 9.10
Release: 9.10
mountall:
  Installed: 1.0
  Candidate: 1.0
  Version table:
 *** 1.0 0
        500 http://de.archive.ubuntu.com karmic/main Packages
        100 /var/lib/dpkg/status

ProblemType: Bug
Architecture: i386
Date: Sat Nov 7 03:34:23 2009
DistroRelease: Ubuntu 9.10
InstallationMedia: Ubuntu-Netbook-Remix 9.10 "Karmic Koala" - Release i386 (20091028.4)
NonfreeKernelModules: nvidia
Package: mountall 1.0
ProcEnviron:
 LANG=en_US.UTF-8
 SHELL=/bin/bash
ProcVersionSignature: Ubuntu 2.6.31-14.48-generic
SourcePackage: mountall
Tags: ubuntu-unr
Uname: Linux 2.6.31-14-generic i686

[redguy (mateusz-kijowski)]

Any workarounds available? I currently cannot boot Ubuntu due to this bug...

[Scott James Remnant (Canonical) (canonical-scott)]

There's no workaround, because the cause of this bug is unknown.

It implies that dbus_connection_send_with_reply returns TRUE, but doesn't set the pending call

[Bjørn Johansen (bjornlljohansen)]

Me too, i just rebooted and cannot get into ubuntu again

[Scott James Remnant (Canonical) (canonical-scott)]

Oh, I see, it's a weirdism of dbus; it'll return TRUE but sending pending_call to NULL if the connection isn't actually connected

[Scott James Remnant (Canonical) (canonical-scott)]

Fixed in libnih 1.0.0, note that this fix will still need to be merged into software using libnih or the software updated to dynamically link to libnih instead.

[Scott James Remnant (Canonical) (canonical-scott)]

mountall 1.1 used the fixed libnih

mountall (1.1) lucid; urgency=low

  * Update to use external libnih.
  * Updated autoconf details at same time to match libnih.
  * rcS is a job, not an event. LP: #456806.

 -- Scott James Remnant <email address hidden> Sun, 29 Nov 2009 20:16:10 +0000

[Nicolas Delvaux (malizor)]

This bug may be fix on Lucid, but it is not on Karmic.

A friend of mine is encountering this issue.
What can I suggest to him? To reinstall?

We tried to update mountall to the Lucid version but it turned out that there is too many missing dependencies.

[Scott James Remnant (Canonical) (canonical-scott)]

On Thu, 2010-03-11 at 13:58 +0000, Nicolas Delvaux wrote:

> This bug may be fix on Lucid, but it is not on Karmic.
>
> A friend of mine is encountering this issue.
> What can I suggest to him? To reinstall?
>
He could always update to the Lucid Beta 1 when it's out next week; if
it's a desktop machine, that's probably as stable as Karmic

Scott
--
Have you ever, ever felt like this?
Had strange things happen? Are you going round the twist?

[Nicolas Delvaux (malizor)]

This is not exactly the fix that I expected... I'm also running Lucid on a laptop and, at least for now, it breaks regularly on upgrades.

The best would be to backport the fix to Karmic or to find another workaround.

Karmic is still maintained for more than a year.
I mean, even if this bug happen rarely, the fact that there is no fix but a reinstall is unacceptable.

[Scott James Remnant (Canonical) (canonical-scott)]

On Thu, 2010-03-11 at 20:05 +0000, Nicolas Delvaux wrote:

> Karmic is still maintained for more than a year.
> I mean, even if this bug happen rarely, the fact that there is no fix but a reinstall is unacceptable.
>
You're quite welcome to apply for an SRU via the usual procedure.

Scott
--
Have you ever, ever felt like this?
Had strange things happen? Are you going round the twist?

[Nicolas Delvaux (malizor)]

The fact is that we finally succeeded in installing Lucid packages (mountall 1.1 and its dependencies: libnih1 and libnih-dbus1 1.01-1).

Indeed, the "infinite error loop" is gone but we now run into an "unable to mount root file system". Ok, so we now have basic command line access without "live-cd + chroot" but nothing more...

This new error makes me think that a simple backport may not be sufficient (perhaps some architecture changes between Karmic and Lucid?)

I must admit that this is beyond my knowledge. Anyway the computer will be reinstalled soon (it is needed to work).
I just hope that this (random) bug is as rare as it seems...

[Scott James Remnant (Canonical) (canonical-scott)]

On Thu, 2010-03-11 at 21:36 +0000, Nicolas Delvaux wrote:

> This new error makes me think that a simple backport may not be
> sufficient (perhaps some architecture changes between Karmic and Lucid?)
>
A near-rewrite (again) of this part of the boot system; it's why it's
hard to backport this kind of change.

Scott
--
Have you ever, ever felt like this?
Had strange things happen? Are you going round the twist?

[Allan Anderson (rubel)]

Ironically, I get booted out of the "sudo do-release-upgrade --devel-release" process when trying to upgrade a Server install. 'mountall' produces that lovely error.