dpkg fails to install acpi-support 0.137

Bug #671214 reported by DH
12
This bug affects 1 person
Affects Status Importance Assigned to Milestone
acpid (Ubuntu)
Invalid
Undecided
Unassigned
upstart (Ubuntu)
Invalid
Undecided
Unassigned

Bug Description

Binary package hint: acpi-support

Ubuntu Maverick version of acpi-support was replaced with Debian sid by accident on my machine. Successfully manually removed apci-support acpi-support-base and acpi-fakekey using dpkg -P --force-all <name> . Now, trying to revert to Maverick version in apt cache or manually downloaded from us.archive.ubuntu.com pool results in dpkg hanging:

$sudo dpkg -i --force-all acpi-support_0.137_amd64.deb
(Reading database ... 192568 files and directories currently installed.)
Preparing to replace acpi-support 0.137 (using acpi-support_0.137_amd64.deb) ...
Unpacking replacement acpi-support ...
Setting up acpi-support (0.137) ...

It will stay hanging on that line all night, dpkg process shows no cpu / cputime activity in top > 1GB free ram and <.5 loadavg:

9303 root 20 0 77516 57m 1120 S 0 1.6 0:00.48 dpkg

Until I ctrl-c out of it.

Setting dpkg --debug=72700 it appears to hang on one of two lines with equal probability:

D000200: conffderef in='/etc/acpi/events/sony-eject' current working='/etc/acpi/events/sony-eject'

or

D000200: conffderef in='/etc/acpi/sonybright.sh' current working='/etc/acpi/sonybright.sh'

so I tried to remove acpi-support again:

---------------------------
 $ sudo dpkg -P --force-all acpi-support
(Reading database ... 192556 files and directories currently installed.)
Removing acpi-support ...
Purging configuration files for acpi-support ...
Processing triggers for man-db ...
Processing triggers for ureadahead ...
---------------------------------

Which changes the failure step lines to 100% of the time:

---------------------------------------
Configuration file `/etc/init.d/acpi-support', does not exist on system.
Installing new config file as you requested.

------------------------------
/etc/init.d/acpi-support is correctly removed / deleted by dpkg each time -P is used...

$ aptitude show dpkg
Package: dpkg
Essential: yes
State: installed
Automatically installed: no
Version: 1.15.8.4ubuntu3
Priority: required
Section: admin
Maintainer: Ubuntu Developers <email address hidden>
Uncompressed Size: 6,263k
PreDepends: libbz2-1.0, libc6 (>= 2.11), libselinux1 (>= 1.32), zlib1g (>= 1:1.1.4), coreutils (>= 5.93-1), xz-utils
Suggests: apt
Breaks: apt (< 0.7.7), aptitude (< 0.4.7-1), dpkg-dev (< 1.14.16), emacs21 (< 21.4a+1-5.7), emacs21-nox (< 21.4a+1-5.7), emacs22 (<= 22.2-0ubuntu2),
        emacs22-gtk (<= 22.2-0ubuntu2), emacs22-nox (<= 22.2-0ubuntu2), jed (< 1:0.99.18+dfsg.1-13), jed-extra (<= 2.5.3-2), konqueror (<=
        4:4.2.96-1), libdpkg-perl (< 1.15.8), pinfo (< 0.6.9-3.1), tkinfo (< 2.8-3.1), xemacs21-support (< 21.4.22-2), xjed (< 1:0.99.18+dfsg.1-13)
Description: Debian package management system
 This package provides the low-level infrastructure for handling the installation and removal of Debian software packages.

 For Debian package development tools, install dpkg-dev.
Homepage: http://wiki.debian.org/Teams/Dpkg

I've googled like crazy on this, found other anecdotal reports of this package failing on upgrade, but no real answers . I have no idea what else to try / test / debug... any ideas why this .deb is failing.

Revision history for this message
DH (dave-higherform) wrote :

Tried the Natty version .deb with the same results...

A little more info on my system, as I saw a kernel config issue in https://bugs.launchpad.net/ubuntu/+source/acpi-support/+bug/585100/comments/2 :

uname -a
Linux saturn 2.6.36-1-generic #7~lucid1-Ubuntu SMP Fri Oct 22 00:21:17 UTC 2010 x86_64 GNU/Linux

Acer 3820T with Core i3-350M is the machine...

Revision history for this message
Steve Langasek (vorlon) wrote :

Please show the relevant section of 'pstree' output on this system when the hang occurs. The problem is almost certainly not in dpkg itself, but somewhere in a subprocess of the maintainer script; since the maintainer script is fairly simple, and no one else is reporting this problem, we'll need to see precisely where it gets stuck to be able to debug this.

Changed in acpi-support (Ubuntu):
status: New → Incomplete
Revision history for this message
DH (dave-higherform) wrote :

acpi related lines from pstree during sudo dpkg -i :

--─acpid
─hald─┬─hald-runner─┬─hald-addon-acpi
─konsole─┬─bash───dpkg───acpi-support.po───invoke-rc.d───acpid───restart

so I tried "sudo restart acpid", which also hangs, but with this output.

update-rc.d: warning: /etc/init.d/acpi-support missing LSB information
update-rc.d: see <http://wiki.debian.org/LSBInitScripts>

Attaching acpid, acpid.dpkg-new and acpi-support scripts from /etc/init.d/ ...

Revision history for this message
DH (dave-higherform) wrote :
Revision history for this message
DH (dave-higherform) wrote :
Revision history for this message
Steve Langasek (vorlon) wrote :

Ok, this indicates some sort of a problem communicating with upstart and getting the acpid process to restart. What does 'status acpid' return on this system?

Revision history for this message
DH (dave-higherform) wrote :

'status acpid' returns 'acpid start/spawned, process 1245' during the dpkg -i hang. The output of status changes to 'stop/spawned' if 'sudo stop -v acpid' is issued in another shell, even though the stop command never exits and gives no output. Ctrl-c cancelling the stop command and issuing a sudo start -v acpid will also change the output of status back to start/spawned. Start also hangs/no exit with no output. Also, issuing two sudo start -v acpid in a row gives 'start: Job is already running: acpid' Two stop commands in a row gives 'stop: Job has already been stopped: acpid'

Let me know what else you need to t/s this...

Revision history for this message
Steve Langasek (vorlon) wrote :

Ok, there's at least enough information here to indicate that this is not a bug in acpi-support, but in acpid and/or upstart. Reassigning to those packages.

affects: acpi-support (Ubuntu) → acpid (Ubuntu)
Changed in acpid (Ubuntu):
status: Incomplete → New
Revision history for this message
Bryan Quigley (bryanquigley) wrote :

The Ubuntu release this was reported with has reached End of Life since you reported the bug. Is this still an issue for you on a supported release? (Try Ubuntu 12.04 or 13.04)

https://wiki.ubuntu.com/Releases

Changed in upstart (Ubuntu):
status: New → Incomplete
Changed in acpid (Ubuntu):
status: New → Incomplete
Revision history for this message
DH (dave-higherform) wrote :

I am currently running Bodhi 2.x, based on 12.04. Acpi-support appears to be installed and functioning correctly. Unable to duplicate the above documented behavior on my current system. Recommend closure of this bug.

Revision history for this message
Bryan Quigley (bryanquigley) wrote :

I'm glad this was resolved, sorry it wasn't tracked in the bug.. Closing.

Changed in acpid (Ubuntu):
status: Incomplete → Invalid
Changed in upstart (Ubuntu):
status: Incomplete → Invalid
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.