*** Bug 1048383 has been marked as a duplicate of this bug. ***

I'm still experiencing upower intermittently wrongly reporting battery energy as 0  and regardless that my script runs every 7 seconds to identify and restart upowerd when it starts reporting insane values, the system triggers a hibernate cycle if this happens while on battery.  It's extremely frustrating to have the system randomly hibernate while trying to work and especially so when it causes issues with the virtual machine I use for work.

Approximately once per day I"m still experiencing my laptop hibernating inappropriately when upower incorrectly reports the battery power as 0.  Is there any further information I can provide to help investigate the problem?  It's extremely frustrating as it sometimes messes up my wifi and my virtual machine guest.

Sorry to sound frustrated but I'm getting to my wits end with my system at least once a day hibernating when it's nowhere near low battery.  My script resets upower when its in this state so the battery shows the accurate amount of charge again but by that point it's too late, gnome has displayed a notification the system is below the critical point and it hibernates.  Do not have this problem in ubuntu or OSX and it's really upsetting when I'm trying to work and the system randomly hibernates.  Could someone please help?

Hello Matt,

can you check the values of /sys/bus/acpi/drivers/battery/*/power_supply/BAT*/energy_* when the issue happens? I bumped into it here too, restarting udev also helped and the values in sysfs look sane. I'd like to see if the sysfs values still look OK when the issue happens next time... Btw I'm also running on a Macbook Pro.

(In reply to Adrien Bustany from comment #5)
> Hello Matt,
> 
> can you check the values of
> /sys/bus/acpi/drivers/battery/*/power_supply/BAT*/energy_* when the issue
> happens? I bumped into it here too, restarting udev also helped and the
> values in sysfs look sane. I'd like to see if the sysfs values still look OK
> when the issue happens next time... Btw I'm also running on a Macbook Pro.

Have you tried updating the firmware on that machine (in OSX)?

Sounds like the hardware is reporting dud values to the kernel. We'll see shortly (with what Adrien is asking) whether the problem is one in UPower or in the kernel/hardware.

Hmm so I got the bug again, and checked the values in sysfs:

[abustany@localhost ~]$ cat /sys/bus/acpi/drivers/battery/PNP0C0A\:00/power_supply/BAT0/energy_now 
34840000
[abustany@localhost ~]$ cat /sys/bus/acpi/drivers/battery/PNP0C0A\:00/power_supply/BAT0/energy_full
65620000
[abustany@localhost ~]$ cat /sys/bus/acpi/drivers/battery/PNP0C0A\:00/power_supply/BAT0/energy_full_design 
63300000

That looks pretty OK to me, and restarting upower (sudo systemctl restart upower.service) makes the issue vanish. Interestingly the issue didn't appear right after resume this time, maybe 10 to 15 minutes after.

I haven't updated the firmware on this machine, but I got it fairly recently (1 month ago maybe). Not sure if they ship them with the latest firmware, I could reboot into OS X to confirm. But anyway, the values in sysfs look consistent to me (and stay the same after the upower restart of course).

Thanks for the replies.  Before I took OSX off the machine this past December it had all the updates available applied.

I've added the requested commands to my script which checks for insane values and restarts upower when required.  

The script helps when upower reports energy-full excessively high; however, when upower reports energy incorrectly as 0 the script is unable to stop my system from hibernating itself.  I will post the results of the script if it acts up tonight.

FYI this is the script:

#!/usr/bin/env perl
#
#This is a script to restart upower anytime it starts
#showing insane values for battery energy-full value
#side-note: this is my first perl script.
#borrowed some stuff from lighter.pl.  thats how i roll
#mak

use strict;
use warnings;

my $energyf;
my $energy;
my $energy_unacceptable=0;
my $energyfd_acceptable=72.6;
my $date;
my $debug=0;

sub getvalues {
    $energyf= trim(`upower -d | grep energy-full:`);
    $energy=  trim(`upower -d | grep energy:`);
    debug_msg( "upower is reporting battery energy-full as $energyf\n" );
    debug_msg( "upower is reporting battery energy as $energy\n" );
}

sub trim {
    my $string = shift;
    return if not $string;
    $string =~ s/[^.0-9]+//g;
    return $string;
}

sub debug_msg {
    print shift if $debug==1;
}

for( ; ; )
{
   getvalues();

   if ($energy == $energy_unacceptable) {
      debug_msg( "Energy level insane at $energy, restarting upower\n\n" );
      my $date = localtime;
      system "echo $date restarting upowerd because >> /var/log/upower_restart";
      system "echo energy was reporting unacceptable value at $energy >> /var/log/upower_restart";
      system "upower -d >> /var/log/upower_restart";
      system "grep -r . /sys/class/power_supply/* >> /var/log/upower_restart";
      system "cat /sys/bus/acpi/drivers/battery/PNP0C0A\:00/power_supply/BAT0/energy_now >> /var/log/upower_restart";
      system "cat /sys/bus/acpi/drivers/battery/PNP0C0A\:00/power_supply/BAT0/energy_full >> /var/log/upower_restart";
      system "cat /sys/bus/acpi/drivers/battery/PNP0C0A\:00/power_supply/BAT0/energy_full_design >> /var/log/upower_restart";
      system "systemctl restart upower";
      sleep 7;
   }

   else {
      debug_msg( "Energy level sane at $energy, checking energy_full value \n\n" );

      if ($energyf < $energyfd_acceptable) {
         debug_msg( "Energy Full level sane as $energyf < $energyfd_acceptable , doing nothing \n\n" );
         sleep 7;
      }

      else {
         debug_msg( "Energy Full level insane at $energyf > $energyfd_acceptable , restarting upower\n\n" );
         my $date = localtime;
         system "echo $date restarting upowerd because >> /var/log/upower_restart";
         system "echo energy-full was $energyf greater than acceptable $energyfd_acceptable >> /var/log/upower_restart";
         system "upower -d >> /var/log/upower_restart";
         system "grep -r . /sys/class/power_supply/* >> /var/log/upower_restart";
         system "cat /sys/bus/acpi/drivers/battery/PNP0C0A\:00/power_supply/BAT0/energy_now >> /var/log/upower_restart";
         system "cat /sys/bus/acpi/drivers/battery/PNP0C0A\:00/power_supply/BAT0/energy_full >> /var/log/upower_restart";
         system "cat /sys/bus/acpi/drivers/battery/PNP0C0A\:00/power_supply/BAT0/energy_full_design >> /var/log/upower_restart";
         system "systemctl restart upower";
         sleep 7;
      }
   }      

}

exit 0;


Matt

Well that didnt take long.  I believe this shows the error is with upower, here's the output of the script taken before the system hibernated.  Too bad there was no way to abort the hibernation cycle, even though upower is restarted by the script and the battery indicator goes back to normal the system still has to hibernate.

[katmeef@katmeefs ~]$ cat /var/log/upower_restart 
Wed Mar 5 09:16:59 2014 restarting upowerd because
energy was reporting unacceptable value at 0
Device: /org/freedesktop/UPower/devices/battery_BAT0
  native-path:          BAT0
  vendor:               DPONz45165UF957A7123456789ABCDE
  model:                bq20z45165UF957A7123456789ABCDE
  power supply:         yes
  updated:              Wed 05 Mar 2014 09:16:53 AM EST (6 seconds ago)
  has history:          yes
  has statistics:       yes
  battery
    present:             yes
    rechargeable:        yes
    state:               discharging
    energy:              0 Wh
    energy-empty:        0 Wh
    energy-full:         65.6 Wh
    energy-full-design:  66 Wh
    energy-rate:         0 W
    voltage:             12.216 V
    percentage:          0%
    capacity:            98.5%
  History (charge):
    1394029013	0.000	discharging
    1394028983	93.000	discharging
    1394028923	94.000	discharging
  History (rate):
    1394029013	0.000	discharging
    1394028983	14.178	discharging
    1394028953	9.842	discharging
    1394028923	9.072	discharging

Device: /org/freedesktop/UPower/devices/line_power_ADP1
  native-path:          ADP1
  power supply:         yes
  updated:              Wed 05 Mar 2014 08:59:47 AM EST (1032 seconds ago)
  has history:          no
  has statistics:       no
  line-power
    online:             no

Daemon:
  daemon-version:  0.9.23
  on-battery:      yes
  on-low-battery:  yes
  lid-is-closed:   no
  lid-is-present:  yes
  is-docked:       no
/sys/class/power_supply/ADP1/type:Mains
/sys/class/power_supply/ADP1/power/control:auto
/sys/class/power_supply/ADP1/power/async:disabled
/sys/class/power_supply/ADP1/power/wakeup_abort_count:0
/sys/class/power_supply/ADP1/power/wakeup_active:0
/sys/class/power_supply/ADP1/power/wakeup_total_time_ms:0
/sys/class/power_supply/ADP1/power/wakeup_active_count:1
/sys/class/power_supply/ADP1/power/runtime_enabled:disabled
/sys/class/power_supply/ADP1/power/runtime_active_kids:0
/sys/class/power_supply/ADP1/power/runtime_active_time:0
/sys/class/power_supply/ADP1/power/wakeup_max_time_ms:0
/sys/class/power_supply/ADP1/power/wakeup_count:1
/sys/class/power_supply/ADP1/power/wakeup_last_time_ms:2336
/sys/class/power_supply/ADP1/power/wakeup:enabled
/sys/class/power_supply/ADP1/power/runtime_status:unsupported
/sys/class/power_supply/ADP1/power/runtime_usage:0
/sys/class/power_supply/ADP1/power/wakeup_expire_count:0
/sys/class/power_supply/ADP1/power/runtime_suspended_time:0
/sys/class/power_supply/ADP1/online:0
/sys/class/power_supply/ADP1/uevent:POWER_SUPPLY_NAME=ADP1
/sys/class/power_supply/ADP1/uevent:POWER_SUPPLY_ONLINE=0
/sys/class/power_supply/BAT0/type:Battery
/sys/class/power_supply/BAT0/power_now:9593000
/sys/class/power_supply/BAT0/alarm:0
/sys/class/power_supply/BAT0/power/control:auto
/sys/class/power_supply/BAT0/power/async:disabled
/sys/class/power_supply/BAT0/power/wakeup_abort_count:0
/sys/class/power_supply/BAT0/power/wakeup_active:0
/sys/class/power_supply/BAT0/power/wakeup_total_time_ms:0
/sys/class/power_supply/BAT0/power/wakeup_active_count:16
/sys/class/power_supply/BAT0/power/runtime_enabled:disabled
/sys/class/power_supply/BAT0/power/runtime_active_kids:0
/sys/class/power_supply/BAT0/power/runtime_active_time:0
/sys/class/power_supply/BAT0/power/wakeup_max_time_ms:0
/sys/class/power_supply/BAT0/power/wakeup_count:16
/sys/class/power_supply/BAT0/power/wakeup_last_time_ms:33484306
/sys/class/power_supply/BAT0/power/wakeup:enabled
/sys/class/power_supply/BAT0/power/runtime_status:unsupported
/sys/class/power_supply/BAT0/power/runtime_usage:0
/sys/class/power_supply/BAT0/power/wakeup_expire_count:0
/sys/class/power_supply/BAT0/power/runtime_suspended_time:0
/sys/class/power_supply/BAT0/capacity:93
/sys/class/power_supply/BAT0/status:Discharging
/sys/class/power_supply/BAT0/voltage_now:12237000
/sys/class/power_supply/BAT0/uevent:POWER_SUPPLY_NAME=BAT0
/sys/class/power_supply/BAT0/uevent:POWER_SUPPLY_STATUS=Discharging
/sys/class/power_supply/BAT0/uevent:POWER_SUPPLY_PRESENT=1
/sys/class/power_supply/BAT0/uevent:POWER_SUPPLY_TECHNOLOGY=Unknown
/sys/class/power_supply/BAT0/uevent:POWER_SUPPLY_CYCLE_COUNT=0
/sys/class/power_supply/BAT0/uevent:POWER_SUPPLY_VOLTAGE_MIN_DESIGN=11210000
/sys/class/power_supply/BAT0/uevent:POWER_SUPPLY_VOLTAGE_NOW=12237000
/sys/class/power_supply/BAT0/uevent:POWER_SUPPLY_POWER_NOW=9593000
/sys/class/power_supply/BAT0/uevent:POWER_SUPPLY_ENERGY_FULL_DESIGN=66000000
/sys/class/power_supply/BAT0/uevent:POWER_SUPPLY_ENERGY_FULL=65010000
/sys/class/power_supply/BAT0/uevent:POWER_SUPPLY_ENERGY_NOW=60990000
/sys/class/power_supply/BAT0/uevent:POWER_SUPPLY_CAPACITY=93
/sys/class/power_supply/BAT0/uevent:POWER_SUPPLY_MODEL_NAME=bq20z45165UF957A7123456789ABCDE
/sys/class/power_supply/BAT0/uevent:POWER_SUPPLY_MANUFACTURER=DPONz45165UF957A7123456789ABCDE
/sys/class/power_supply/BAT0/uevent:POWER_SUPPLY_SERIAL_NUMBER=
/sys/class/power_supply/BAT0/energy_now:60990000
/sys/class/power_supply/BAT0/model_name:bq20z45165UF957A7123456789ABCDE
/sys/class/power_supply/BAT0/manufacturer:DPONz45165UF957A7123456789ABCDE
/sys/class/power_supply/BAT0/technology:Unknown
/sys/class/power_supply/BAT0/cycle_count:0
/sys/class/power_supply/BAT0/energy_full_design:66000000
/sys/class/power_supply/BAT0/voltage_min_design:11210000
/sys/class/power_supply/BAT0/energy_full:65010000
/sys/class/power_supply/BAT0/present:1
60990000
65010000
66000000
[katmeef@katmeefs ~]$

Here's the script output from the other failure scenario where energy_full reports a nonsensically high value.  This failure scenario only results in the battery showing lower than it should in gnome but it does not cause the system to hibernate.

Thu Mar 6 01:19:56 2014 restarting upowerd because
energy-full was 948.306 greater than acceptable 72.6
Device: /org/freedesktop/UPower/devices/battery_BAT0
  native-path:          BAT0
  vendor:               DPONz45165UF957A7123456789ABCDE
  model:                bq20z45165UF957A7123456789ABCDE
  power supply:         yes
  updated:              Thu 06 Mar 2014 01:19:55 AM EST (1 seconds ago)
  has history:          yes
  has statistics:       yes
  battery
    present:             yes
    rechargeable:        yes
    state:               discharging
    energy:              948.306 Wh
    energy-empty:        0 Wh
    energy-full:         948.306 Wh
    energy-full-design:  66 Wh
    energy-rate:         18.563 W
    voltage:             11.23 V
    time to empty:       51.1 hours
    percentage:          100%
    capacity:            96.9848%
  History (charge):
    1394086795	100.000	discharging
    1394086735	51.000	discharging
  History (rate):
    1394086795	18.563	discharging
    1394086765	16.821	discharging
    1394086735	22.588	discharging
    1394086705	30.230	discharging

Device: /org/freedesktop/UPower/devices/line_power_ADP1
  native-path:          ADP1
  power supply:         yes
  updated:              Thu 06 Mar 2014 01:03:20 AM EST (996 seconds ago)
  has history:          no
  has statistics:       no
  line-power
    online:             no

Daemon:
  daemon-version:  0.9.23
  on-battery:      yes
  on-low-battery:  no
  lid-is-closed:   no
  lid-is-present:  yes
  is-docked:       no
/sys/class/power_supply/ADP1/type:Mains
/sys/class/power_supply/ADP1/power/control:auto
/sys/class/power_supply/ADP1/power/async:disabled
/sys/class/power_supply/ADP1/power/wakeup_abort_count:0
/sys/class/power_supply/ADP1/power/wakeup_active:0
/sys/class/power_supply/ADP1/power/wakeup_total_time_ms:0
/sys/class/power_supply/ADP1/power/wakeup_active_count:1
/sys/class/power_supply/ADP1/power/runtime_enabled:disabled
/sys/class/power_supply/ADP1/power/runtime_active_kids:0
/sys/class/power_supply/ADP1/power/runtime_active_time:0
/sys/class/power_supply/ADP1/power/wakeup_max_time_ms:0
/sys/class/power_supply/ADP1/power/wakeup_count:1
/sys/class/power_supply/ADP1/power/wakeup_last_time_ms:2336
/sys/class/power_supply/ADP1/power/wakeup:enabled
/sys/class/power_supply/ADP1/power/runtime_status:unsupported
/sys/class/power_supply/ADP1/power/runtime_usage:0
/sys/class/power_supply/ADP1/power/wakeup_expire_count:0
/sys/class/power_supply/ADP1/power/runtime_suspended_time:0
/sys/class/power_supply/ADP1/online:0
/sys/class/power_supply/ADP1/uevent:POWER_SUPPLY_NAME=ADP1
/sys/class/power_supply/ADP1/uevent:POWER_SUPPLY_ONLINE=0
/sys/class/power_supply/BAT0/type:Battery
/sys/class/power_supply/BAT0/power_now:24420000
/sys/class/power_supply/BAT0/alarm:0
/sys/class/power_supply/BAT0/power/control:auto
/sys/class/power_supply/BAT0/power/async:disabled
/sys/class/power_supply/BAT0/power/wakeup_abort_count:0
/sys/class/power_supply/BAT0/power/wakeup_active:0
/sys/class/power_supply/BAT0/power/wakeup_total_time_ms:0
/sys/class/power_supply/BAT0/power/wakeup_active_count:2
/sys/class/power_supply/BAT0/power/runtime_enabled:disabled
/sys/class/power_supply/BAT0/power/runtime_active_kids:0
/sys/class/power_supply/BAT0/power/runtime_active_time:0
/sys/class/power_supply/BAT0/power/wakeup_max_time_ms:0
/sys/class/power_supply/BAT0/power/wakeup_count:2
/sys/class/power_supply/BAT0/power/wakeup_last_time_ms:63534591
/sys/class/power_supply/BAT0/power/wakeup:enabled
/sys/class/power_supply/BAT0/power/runtime_status:unsupported
/sys/class/power_supply/BAT0/power/runtime_usage:0
/sys/class/power_supply/BAT0/power/wakeup_expire_count:0
/sys/class/power_supply/BAT0/power/runtime_suspended_time:0
/sys/class/power_supply/BAT0/capacity:51
/sys/class/power_supply/BAT0/status:Discharging
/sys/class/power_supply/BAT0/voltage_now:11156000
/sys/class/power_supply/BAT0/uevent:POWER_SUPPLY_NAME=BAT0
/sys/class/power_supply/BAT0/uevent:POWER_SUPPLY_STATUS=Discharging
/sys/class/power_supply/BAT0/uevent:POWER_SUPPLY_PRESENT=1
/sys/class/power_supply/BAT0/uevent:POWER_SUPPLY_TECHNOLOGY=Unknown
/sys/class/power_supply/BAT0/uevent:POWER_SUPPLY_CYCLE_COUNT=0
/sys/class/power_supply/BAT0/uevent:POWER_SUPPLY_VOLTAGE_MIN_DESIGN=11210000
/sys/class/power_supply/BAT0/uevent:POWER_SUPPLY_VOLTAGE_NOW=11156000
/sys/class/power_supply/BAT0/uevent:POWER_SUPPLY_POWER_NOW=24420000
/sys/class/power_supply/BAT0/uevent:POWER_SUPPLY_ENERGY_FULL_DESIGN=66000000
/sys/class/power_supply/BAT0/uevent:POWER_SUPPLY_ENERGY_FULL=64010000
/sys/class/power_supply/BAT0/uevent:POWER_SUPPLY_ENERGY_NOW=32650000
/sys/class/power_supply/BAT0/uevent:POWER_SUPPLY_CAPACITY=51
/sys/class/power_supply/BAT0/uevent:POWER_SUPPLY_MODEL_NAME=bq20z45165UF957A7123456789ABCDE
/sys/class/power_supply/BAT0/uevent:POWER_SUPPLY_MANUFACTURER=DPONz45165UF957A7123456789ABCDE
/sys/class/power_supply/BAT0/uevent:POWER_SUPPLY_SERIAL_NUMBER=
/sys/class/power_supply/BAT0/energy_now:32650000
/sys/class/power_supply/BAT0/model_name:bq20z45165UF957A7123456789ABCDE
/sys/class/power_supply/BAT0/manufacturer:DPONz45165UF957A7123456789ABCDE
/sys/class/power_supply/BAT0/technology:Unknown
/sys/class/power_supply/BAT0/cycle_count:0
/sys/class/power_supply/BAT0/energy_full_design:66000000
/sys/class/power_supply/BAT0/voltage_min_design:11210000
/sys/class/power_supply/BAT0/energy_full:64010000
/sys/class/power_supply/BAT0/present:1
32650000
64010000
66000000
[katmeef@katmeefs ~]

when running matt's script on f19/rMBP 2012 i see the following lines on stderr:

grep: /sys/class/power_supply/ADP1/power/autosuspend_delay_ms: Input/output error
grep: /sys/class/power_supply/BAT0/power/autosuspend_delay_ms: Input/output error

don't know if that's relevant, though.

System has hibernated twice on me this evening when there was plenty of battery life...  When upower falsely triggers hibernation I have scripts running to rmmod the wl driver and hibernate my QEMU VM to prevent issues upon resuming the system, too bad there's not also a command I could run to abort the hibernation...

Jan I see the same error if I run 'grep -r . /sys/class/power_supply/* >> /var/log/upower_restart', not sure if it's relevant

Sorry if this sounds unappreciative but I've been having this problem since January and the dev this bug is assigned to has not replied once.  What are the chances someone will look into this or should I just be looking for a new distro?  It makes using my system for my job quite unpleasant when it spontaneously hibernates with a half-full or more battery.

(In reply to Matt Keith from comment #14)
> Sorry if this sounds unappreciative but I've been having this problem since
> January and the dev this bug is assigned to has not replied once.  What are
> the chances someone will look into this or should I just be looking for a
> new distro?  It makes using my system for my job quite unpleasant when it
> spontaneously hibernates with a half-full or more battery.

Given that I'm not able to reproduce the bugs, that the problem only occurs on 2 machines that I know of, and that other distros use the exact same UPower, I doubt changing distros would fix the problem.


(In reply to Adrien Bustany from comment #7)
> Hmm so I got the bug again, and checked the values in sysfs:
> 
> [abustany@localhost ~]$ cat
> /sys/bus/acpi/drivers/battery/PNP0C0A\:00/power_supply/BAT0/energy_now 
> 34840000
> [abustany@localhost ~]$ cat
> /sys/bus/acpi/drivers/battery/PNP0C0A\:00/power_supply/BAT0/energy_full
> 65620000
> [abustany@localhost ~]$ cat
> /sys/bus/acpi/drivers/battery/PNP0C0A\:00/power_supply/BAT0/
> energy_full_design 
> 63300000
> 
> That looks pretty OK to me, and restarting upower (sudo systemctl restart
> upower.service) makes the issue vanish. Interestingly the issue didn't
> appear right after resume this time, maybe 10 to 15 minutes after.
> 
> I haven't updated the firmware on this machine, but I got it fairly recently
> (1 month ago maybe). Not sure if they ship them with the latest firmware, I
> could reboot into OS X to confirm. But anyway, the values in sysfs look
> consistent to me (and stay the same after the upower restart of course).

And the value in capacity is? That's where the percentage is gathered.

In any case, if one of you that can reproduce the problem isn't going to root cause it, we'll need the sections of "/usr/libexec/upowerd -v" output when reproducing the problem. Or everything if it's small enough to be attached to this bug.

I'm now running the 3.12 COPR repo (for a couple of months now), which brings an updated upower. I don't have any issue anymore with that one, so I'd say the issue is fixed for me.

i'm now on f20 (upower-0.9.23-2.fc20.x86_64) and use matt's script for restarting upowerd. i don't see intermittent hibernation anymore (w/ the script), but the battery levels still fluctuate. can we move the bug onto f20?

Hasn't happened for a while but just did again this evening.  I've modified my script to grap /usr/libexec/upowerd -v next time it happens, I'll wait for it to happen once more then I'll look into to the 3.12 COPR repo as Adrien has done

with upower-0.9.23-2.fc20.x86_64 and current updates things are running smooth for the last months. please close.

I gave up on Fedora many months ago over this issue, can't confirm.

(In reply to jan p. springer from comment #19)
> with upower-0.9.23-2.fc20.x86_64 and current updates things are running
> smooth for the last months. please close.

Great, closing then.

(In reply to Matt Keith from comment #20)
> I gave up on Fedora many months ago over this issue, can't confirm.

Seeing as UPower in other distributions will eventually get updated to the same version...