[Bug 858122] Re: incomplete migration to /run

[Steve Langasek]

On Sat, Sep 24, 2011 at 09:05:17PM -0000, Tom Chiverton wrote:

> fstab:
<snip>

> //bookcase.house/falken /home/falken/smb4k/bookcase-falken cifs noexec,credentials=/home/falken/.cifsPassword,uid=falken,noauto 0 0
> //bookcase.house/scratch /home/falken/smb4k/bookcase-scratch cifs noexec,credentials=/home/falken/.cifsPassword,uid=falken,noauto 0 0
> //bookcase.house/mp3 /media/bookcase-mp3 cifs noexec,credentials=/home/falken/.cifsPassword,uid=falken,noauto 0 0
> //bookcase.house/video /home/falken/smb4k/bookcase-video cifs noexec,credentials=/home/falken/.cifsPassword,uid=falken,noauto 0 0

This makes me suspect an effect of bug #221631.  Although the dbus package
in oneiric includes a fix for this bug, since the dbus service can't be
restarted, the old stop condition is in effect for the job at the time of
the post-upgrade reboot, with the result that dbus shuts down as soon as
'runlevel 6' triggers, and if you use Network Manager, your cifs filesystems
will not cleanly unmount.

However, the behavior seen here is a timeout of the mount after 300 seconds
and a forced unmount, so I don't know why these would have blocked a clean
unmount in your case.

Do you often see errors on this machine at reboot about filesystems not
being cleanly unmounted?

Apologies if I've already asked this question, but is /etc/init.d/umountroot
an unmodified conffile?  (i.e., does 'debsums -s -e initscripts' return
warnings, and what does 'ls -l /etc/init.d/umountroot*' show?)

What is the output of ls -l /etc/rc6.d/ on this system?

> If you've tried to do it safely at shutdown, and it didn't work, I think
> it's OK to try less safely early in the boot process, then reboot once
> it's done.  Certainly I did this by hand, and no ill effects occurred.

It may come to this, but that's easier said than done.  The risks of doing
it here include hanging the system on boot (potentially very bad if it's
remotely managed), and filesystem corruption due to an unclean shutdown.  We
do want to resolve this bug, but we shouldn't be hasty to change things that
might make it worse.

-- 
Steve Langasek                   Give me a lever long enough and a Free OS
Debian Developer                   to set it on, and I can move the world.
Ubuntu Developer                                    http://www.debian.org/
slangasek at ubuntu.com                                     vorlon at debian.org

-- 
You received this bug notification because you are a member of Ubuntu
Foundations Bugs, which is subscribed to sysvinit in Ubuntu.
https://bugs.launchpad.net/bugs/858122

Title:
  incomplete migration to /run

Status in “sysvinit” package in Ubuntu:
  Incomplete

Bug description:
  I had the exact same issue documented as
  https://bugs.launchpad.net/ubuntu/+source/initramfs-tools/+bug/807306
  on upgrade from 11.04 to 11.10. Opening new bug as req.

  Removing the existing /var/run and /var/run/lock and replacing with
  sym links as per comment 76 there fixed it.

  The impact of this was a total lack of system (wireless) networking
  coming up - this gives a >! minute pause during boot, as well as a
  subsequent total failure to be able to log into X (KDE or GNOME
  sessions give a cryptic error related to DBUS).

  I've just run an apt-get update/upgrade after *finally* getting back
  into my new 11.10 beta 2 system, and no updates to the mountall show
  up... therefore I assume this bug is still present.

  ProblemType: Bug
  DistroRelease: Ubuntu 11.10
  Package: initscripts 2.88dsf-13.10ubuntu4
  ProcVersionSignature: Ubuntu 3.0.0-11.18-generic 3.0.4
  Uname: Linux 3.0.0-11-generic i686
  ApportVersion: 1.23-0ubuntu1
  Architecture: i386
  Date: Sat Sep 24 10:49:58 2011
  ProcEnviron:
   LANGUAGE=
   PATH=(custom, user)
   LANG=en_US.UTF-8
   SHELL=/bin/bash
  SourcePackage: sysvinit
  UpgradeStatus: Upgraded to oneiric on 2011-09-23 (0 days ago)

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/sysvinit/+bug/858122/+subscriptions