shutdown after plugging in AC with "critical battery"

[towsonu2003]

Binary package hint: gnome-power-manager

This is similar to bug # 52345, but not the same...

It started to happen in the last few weeks. I do not know why.

Steps:
1. Unplug AC
2. Wait for a couple of minutes
3. Plug AC

Actual result:
shuts down automatically with a message "battery is critically low"... (happened about 20 minutes ago)

user.log has these two entries:
Nov 4 16:03:46 localhost gnome-power-manager: Shutting down computer because we are critically low for the primary battery
Nov 4 16:03:46 localhost gnome-power-manager: Shutting down computer because we are critically low for the primary battery

Expected Result:
should not shut down... battery was full when it was unplugged. It did not drain. The battery can hold charge for about 1 hour without AC.

Reproduceable:
Occured twice in the last few weeks (not always / seldom)

I'm using an up-to-date Dapper.

[towsonu2003 (towsonu2003)]

Subscribed ubuntu-kernel-acpi thinking that this might be a weird interplay between acpi and the gnome power manager.

Note: today, when I plugged in the AC power (as per above steps), a notification popped up but it disappeared so quickly I couldn't read it. It looked like the "battery is critical" notice (I saw the word "data" in there) but it was too quick... And the laptop didn't shut down. So, this still is occurring seldom.

[towsonu2003 (towsonu2003)]

I saw a few lines in /var/log/acpid that made me think this might as well be an acpid bug. I'm clueless, so it's up to you :)

[Sat Nov 4 16:02:41 2006] received event "battery BAT1 00000080 00000001"
#probably when I plugged *out* the power?
[Sat Nov 4 16:02:41 2006] notifying client 3945[108:108]
[Sat Nov 4 16:02:41 2006] executing action "/etc/acpi/power.sh"
[Sat Nov 4 16:02:41 2006] BEGIN HANDLER MESSAGES
[Sat Nov 4 16:02:41 2006] END HANDLER MESSAGES
[Sat Nov 4 16:02:41 2006] action exited with status 0
[Sat Nov 4 16:02:41 2006] completed event "battery BAT1 00000080 00000001"

[Sat Nov 4 16:03:35 2006] received event "battery BAT1 00000080 00000001"
#and this should be when I plugged *in* the power.
[Sat Nov 4 16:03:35 2006] notifying client 3945[108:108]
[Sat Nov 4 16:03:35 2006] executing action "/etc/acpi/power.sh"
[Sat Nov 4 16:03:35 2006] BEGIN HANDLER MESSAGES
[Sat Nov 4 16:03:36 2006] END HANDLER MESSAGES
[Sat Nov 4 16:03:36 2006] action exited with status 0
[Sat Nov 4 16:03:36 2006] completed event "battery BAT1 00000080 00000001"
[Sat Nov 4 16:03:58 2006] exiting

[Sat Nov 4 16:05:00 2006] starting up
#now the computer restarted...

[Sebastien Bacher (seb128)]

Thank you for your bug report. Do you still get the issue? Could you stop gnome-power-manager and restart it using the --verbose and --no-daemon options and attach the log to the bug?

[towsonu2003 (towsonu2003)]

Just to make sure you know, I'm using Dapper :)

Attached are my log files, thanks a lot for your interest.

[Daniel T Chen (crimsun)]

Does this symptom still occur using the 8.10 alpha 6 desktop image?

[Id2ndR (id2ndr)]

Same with 8.10 beta

[Loïc Minier (lool)]

The acpi-support scripts might deal with battery events, but acpid doesn't so I'm reassigning; I'd recommend adding debug to these scripts (e.g. with 'set -x; exec 2>"$(mktemp -t foobar.XXXXXXXX)"' at the top of a shell script).

[Gaetan Nadon (memsize)]

Thanks for reporting this bug and any supporting documentation. Since this bug has enough information provided for a developer to begin work, I'm going to mark it as confirmed and let them handle it from here.
The problem can also be reproduced in Intrepid
Thanks for taking the time to make Ubuntu better!
BugSquad

[Steve Langasek (vorlon)]

marking invalid for acpi-support; this is a gnome-power-manager issue only, acpi-support is not triggering this shutdown.

[Karim Ammouche (karim-ammouche)]

100% occurence with a Dell Vostro 1710 laptop using Ubuntu 9.04 (Linux 2.6.28-14-generic).

Battery indicates a 0.7% charge.

Does not occur when AC is plug/unplugged under a terminal.

This bug is very annoying because my AC connector is broken and disconnect from time to time.

[Scott Howard (showard314)]

Thank you for taking the time to report this bug and helping to make Ubuntu better. The issue that you reported is one that should be reproducible with the live environment of the Desktop CD of the development release - Karmic Koala. It would help us greatly if you could test with it so we can work on getting it fixed in the next release of Ubuntu. You can find out more about the development release at http://www.ubuntu.com/testing/. Thanks again and we appreciate your help.

(set back to confirmed if it exists in Karmic. Karmic's GPM is a significantly different version than in Jaunty, and we should make sure that this bug doesn't propagate any further)

[David Coles (dcoles)]

This is still an issue for Karmic Alpha 5.

I suspended my Dell XPS M1530 at low battery charge and about one hour later connected the AC power and then resumed. I warning that the battery was now at critical level and shutdown would occur appeared, but I ignored it since the charging icon was present. After a few seconds the the laptop shut down.

Unfortunately gnome-power-daemon doesn't appear to have logged anything in /var/log/messages so I can't give the output of gnome-power-manager right at the moment.

[Scott Howard (showard314)]

Thanks for confirming in Karmic! Do get GPM's debugging messages go to:
https://wiki.ubuntu.com/DebuggingGNOMEPowerManager

Can you follow the steps for:
Getting DBUS info
Getting info from GPM

And post the output here for recreating the bug?

Also, could you type the following into the terminal:
apport-collect 70293

This bug has already been reported upstream, but is lacking useful information:
https://bugzilla.gnome.org/show_bug.cgi?id=581598

We can forward your info upstream.

On Tue, Sep 8, 2009 at 9:41 PM, David Coles<email address hidden> wrote:
> This is still an issue for Karmic Alpha 5.
>
> I suspended my Dell XPS M1530 at low battery charge and about one hour
> later connected the AC power and then resumed. I warning that the
> battery was now at critical level and shutdown would occur appeared, but
> I ignored it since the charging icon was present. After a few seconds
> the the laptop shut down.
>
> Unfortunately gnome-power-daemon doesn't appear to have logged anything
> in /var/log/messages so I can't give the output of gnome-power-manager
> right at the moment.
>
> --
> shutdown after plugging in AC  with "critical battery"
> https://bugs.launchpad.net/bugs/70293
> You received this bug notification because you are a direct subscriber
> of the bug.
>
> Status in “acpi-support” package in Ubuntu: Invalid
> Status in “gnome-power-manager” package in Ubuntu: Confirmed
>
> Bug description:
> Binary package hint: gnome-power-manager
>
> This is similar to bug # 52345, but not the same...
>
> It started to happen in the last few weeks. I do not know why.
>
> Steps:
> 1. Unplug AC
> 2. Wait for a couple of minutes
> 3. Plug AC
>
> Actual result:
> shuts down automatically with a message "battery is critically low"... (happened about 20 minutes ago)
>
> user.log has these two entries:
> Nov  4 16:03:46 localhost gnome-power-manager: Shutting down computer because we are critically low for the primary battery
> Nov  4 16:03:46 localhost gnome-power-manager: Shutting down computer because we are critically low for the primary battery
>
> Expected Result:
> should not shut down... battery was full when it was unplugged. It did not drain. The battery can hold charge for about 1 hour without AC.
>
> Reproduceable:
> Occured twice in the last few weeks (not always / seldom)
>
> I'm using an up-to-date Dapper.
>
>

[David Coles (dcoles)]

Architecture: amd64
DistroRelease: Ubuntu 9.10
NonfreeKernelModules: nvidia
Package: gnome-power-manager 2.27.92-0ubuntu1
PackageArchitecture: amd64
ProcEnviron:
 SHELL=/bin/bash
 PATH=(custom, user)
 LANG=en_AU.UTF-8
 LANGUAGE=en_AU.UTF-8
ProcVersionSignature: Ubuntu 2.6.31-9.29-generic
Uname: Linux 2.6.31-9-generic x86_64
UserGroups: adm admin cdrom dialout lpadmin plugdev sambashare

[David Coles (dcoles)]

There's the apport log.
I'll have a look at writing up some steps for reproducing and the DBUS and GPM output in the morning once I run the battery down again. :)

[David Coles (dcoles)]

Here's the output from a test this morning. Karmic doesn't appear to have a org.freedesktop.PowerManagement interface running on the session bus, but I've included a log from the org.freedesktop.DeviceKit.Power interface on the system bus as well as the GPM logs.

Steps used to reproduce:
1. Unplug AC power and let battery discharge
2. Wait for power to reach critical level. The system should notify you that the battery is at a critical level and the computer is about to be shut down.
3. Immediately reconnect the power supply. The power tray icon should show that the system is charging.
4. After a few seconds a message in any open console will inform you that the system is going down for halt NOW!.
(5. Optional: Curse if you had anything unsaved open)

Note: This isn't over a suspend, but certainly easier to reproduce than trying to catch it right at the critical moment.

[Scott Howard (showard314)]

I submitted a new bug report at:
https://bugzilla.gnome.org/show_bug.cgi?id=594739. You can add yourself to the cc: list there and add any additional information if it is missing it. Thanks for your help!

[matthewc (matthewc)]

FYI, same problem here, added my own logs to the GNOME Bugzilla entry.

[Konstantin Lavrov (lacosta)]

Still can confirm with lubuntu beta

g-p-m Version 2.29.91

[Calmarius (david15b)]

This bug always happen to me. Even if my battery is fully charged if I pull out the power cable and plug in again immediately it shuts down almost always (eg when somebody plugs my cable out to plug a power distributor in a plugs my cable in again).

[Konstantin Lavrov (lacosta)]

Hi Calmarius,

Try workaround from 531190 to avoid critical low action on AC-cord unplugged:

gconftool-2 --type bool --set /apps/gnome-power-manager/general/use_time_for_policy false

[Jeremy Foshee (jeremyfoshee)]

removed deprecated team subscription

~Jfo

[Venkatesan Srinivasan (venkatesan-sriniva)]

I get a similar situation.
Using a Dell Vostro 1310 laptop with a 6 cell battery.
Lucid Lynx(10.04) and Maverick(10.10) 32 bit and 64 bit updated.
When the battery is not fully charged and you remove the AC cord, a critical battery warning is raised. But the battery has enough charge so as not to raise a critical battery warning under normal circumstances.
Even if the power cord is attached quickly, the laptop still goes to hibernate mode.

Have been experiencing this from the time 10.04 was launched. Was not seen before that in the previous versions. Haven't checked up on the previous versions now.

[tuna (tunaturkmen)]

Hi I got a similar situation but it occurs both when running on AC and unplugged as well as when running on battery and plugged. I began having the problem after upgrading the karmic koala and have it since then in every version.

release 11.04
kernel linux 2.6.38-8-generic
gnome 2.32.1

Acer One
intel atom CPU N270 @1.70GHz