Who pressed SleepButton on my laptop?

dumphblooz · 2013-10-08 07:40:51

I have Asus U33Jc laptop with 3.10.13-1-lts x86_64 and bumblebee. Sometimes I can see in logs:

окт 08 15:04:59 nuzzflonk logger[11350]: ACPI group/action undefined: hotkey / ATK0100:00
окт 08 15:04:59 nuzzflonk logger[11352]: SleepButton pressed

And my laptop going to suspend. This occurs when laptop under high cpu load. For example after ~2 minutes of

$ stress -c 2

Who pressed SleepButton? My laptop is everytime closed, has not external keyboard.

$ sudo inxi -F
System:    Host: nuzzflonk Kernel: 3.10.13-1-lts x86_64 (64 bit) Console: tty 3 Distro: Arch Linux
Machine:   Mobo: ASUSTeK model: U33Jc version: 1.0 serial: BSN12345678901234567
           Bios: American Megatrends version: U33Jc.213 date: 01/24/2011
CPU:       Dual core Intel Core i5 CPU M 460 (-HT-MCP-) cache: 3072 KB flags: (lm nx sse sse2 sse3 sse4_1 sse4_2 ssse3 vmx) 
           Clock Speeds: 1: 1199.00 MHz 2: 1199.00 MHz 3: 1199.00 MHz 4: 1199.00 MHz
Graphics:  Card-1: Intel Core Processor Integrated Graphics Controller 
           Card-2: NVIDIA GT218M [GeForce 310M] 
           X.org: 1.14.3 driver: intel tty size: 212x57 Advanced Data: N/A for root out of X
Audio:     Card: Intel 5 Series/3400 Series High Definition Audio driver: snd_hda_intel Sound: ALSA ver: k3.10.13-1-lts
Network:   Card-1: Qualcomm Atheros AR8131 Gigabit Ethernet driver: atl1c 
           IF: eth0 state: down mac: bc:ae:c5:0a:e4:48
           Card-2: Qualcomm Atheros AR9285 Wireless Network Adapter (PCI-Express) driver: ath9k 
           IF: wlan0 state: up mac: 48:5d:60:99:d6:31
Drives:    HDD Total Size: 500.1GB (64.5% used) 1: id: /dev/sda model: WDC_WD5000BEVT size: 500.1GB 
Partition: ID: / size: 35G used: 28G (86%) fs: ext4 ID: /boot size: 488M used: 59M (13%) fs: ext4 
           ID: /home size: 394G used: 273G (73%) fs: ext4 ID: swap-1 size: 8.59GB used: 0.00GB (0%) fs: swap 
Sensors:   System Temperatures: cpu: 55.0C mobo: N/A 
           Fan Speeds (in rpm): cpu: N/A 
Info:      Processes: 154 Uptime: 1:09 Memory: 483.5/7651.7MB Client: Shell (sudo) inxi: 1.9.15

Last edited by dumphblooz (2013-11-26 17:26:35)

dumphblooz · 2013-10-09 02:50:48

Same problem on systemrescuecd-x86-3.8.0. After a while stress command:

% acpi_listen
hotkey ATK0100:00 0000006d 00000000
button/sleep SBTN 00000080 00000000 K

WonderWoofy · 2013-10-09 02:56:38

Maybe it is just overheating and suspending as a safety precaution?

cfr · 2013-10-09 03:01:27

Sounds plausible but why on earth does it claim it pressed the SleepButton?

WonderWoofy · 2013-10-09 03:02:24

I dunno… I've never had a machine shut down from getting too hot… so this was entirely a guess.

cfr · 2013-10-09 03:11:44

Me neither. I had one I thought was shutting down for that reason but I'm not sure that counts. (Would be different if it thought it was shutting down for that reason.)

dumphblooz · 2013-10-09 03:38:11

WonderWoofy wrote:

Maybe it is just overheating and suspending as a safety precaution?

Maybe. But laptop works fine while running stress without suspend on systemrescuecd-x86-3.8.0. It is important to know, who press SleepButton for disable it.

WonderWoofy · 2013-10-09 04:22:04

boot with systemd.log_level=debug and see if you get more info.

dumphblooz · 2013-10-09 05:00:26

WonderWoofy wrote:

boot with systemd.log_level=debug and see if you get more info.

What can I see there? Give a tip, please.

WonderWoofy · 2013-10-09 05:04:33

man 7 kernel-command-line

WorMzy · 2013-10-09 07:42:04

dumphblooz wrote:

Who pressed SleepButton? My laptop is everytime closed, has not external keyboard.

Does it happen if you don't close the laptop lid? As far as I understand it, a laptop going to sleep when you close the lid is expected behaviour..

dumphblooz · 2013-10-09 07:56:30

WorMzy wrote:

dumphblooz wrote:
Who pressed SleepButton? My laptop is everytime closed, has not external keyboard.
Does it happen if you don't close the laptop lid? As far as I understand it, a laptop going to sleep when you close the lid is expected behaviour..

Lid always opened and AC plugged.

plam · 2013-10-09 10:26:44

cfr wrote:

Sounds plausible but why on earth does it claim it pressed the SleepButton?

If it does indeed do it as a safety precaution it may be doing it because it is a really easy to implement hardware-only solution. I imagine it is really easy to monitor a thermometer and send a signal on certain treshold (the signal being sent is to the sleep button cable). I can implement this using an arduino, so I guess the engineers implemented it using some basic IC. I don't think this is the most elegant solution out there, but it is entirely possible.

WonderWoofy · 2013-10-09 15:47:32

@plam, if my shot in the dark is right at all, it wouldn't be a hardware solution, but rather a kernel based one. Otherwise, it would simply drop into some kind of a sleep state with no indication in the logs. The fact that it is notifying userspace of its move to a sleep state seems to hint towrads something within the operating system that is doing this.

I also seem to recall someone else having a similar issue some time in the past. Though at the moment I cannot find it…

cookies · 2013-10-09 16:35:52

dumphblooz wrote:

Same problem on systemrescuecd-x86-3.8.0. After a while stress command:
% acpi_listen
hotkey ATK0100:00 0000006d 00000000
button/sleep SBTN 00000080 00000000 K

Does acpi_listen return the same output if you actually press the sleep button yourself?

plam · 2013-10-09 21:16:12

WonderWoofy wrote:

@plam, if my shot in the dark is right at all, it wouldn't be a hardware solution, but rather a kernel based one. Otherwise, it would simply drop into some kind of a sleep state with no indication in the logs. The fact that it is notifying userspace of its move to a sleep state seems to hint towrads something within the operating system that is doing this.

If your shot is right, it may have been done in this way: The two cables which connect the power button to the mainboard are also connected to a logic gate, in a way in which both the logic gate ang the power button can short them to produce a button press. The logic gate is also connected to a thermometer, and shorts the power-button-circuit, if a certain treshold temperature is reached.
Again, this is just speculation, but if you don't find another explanation, this may very well be it.

/sorry I was to lazy to crop the pic/

WonderWoofy · 2013-10-09 21:27:18

That is some very reasonable speculation. Though I guess until more info is provided, nothing more than speculating can be done here...

cfr · 2013-10-09 22:40:07

Maybe I'm missing something obvious here but I'm now thoroughly confused:

dumphblooz wrote:

Same problem on systemrescuecd-x86-3.8.0. After a while stress command:
% acpi_listen
hotkey ATK0100:00 0000006d 00000000
button/sleep SBTN 00000080 00000000 K

dumphblooz wrote:

WonderWoofy wrote:
Maybe it is just overheating and suspending as a safety precaution?
Maybe. But laptop works fine while running stress without suspend on systemrescuecd-x86-3.8.0. It is important to know, who press SleepButton for disable it.

So does it or does it not happen with systemrescuecd-x86-3.8.0 running stress?

Moreover:

dumphblooz wrote:

WorMzy wrote:
dumphblooz wrote:
Who pressed SleepButton? My laptop is everytime closed, has not external keyboard.
Does it happen if you don't close the laptop lid? As far as I understand it, a laptop going to sleep when you close the lid is expected behaviour..
Lid always opened and AC plugged.

So is it closed every time? Open every time? Closed sometimes and open sometimes?

WonderWoofy · 2013-10-09 22:45:58

Yeah, I wondered about those things as well. But besides simply inconsistent, the OP has also been exceedingly sparse on their ability to both comment and provide more info, so until that happens, nothing in this thread is really relevant.

dumphblooz · 2013-10-10 14:11:38

cookies wrote:

dumphblooz wrote:
Same problem on systemrescuecd-x86-3.8.0. After a while stress command:
% acpi_listen
hotkey ATK0100:00 0000006d 00000000
button/sleep SBTN 00000080 00000000 K
Does acpi_listen return the same output if you actually press the sleep button yourself?

Almost the same. From integrated keyboard:

 $ sudo acpi_listen
button/sleep SBTN 00000080 00000000

From external keyboard:

$ sudo acpi_listen 
button/sleep SBTN 00000080 00000000 K

What is hotkey ATK0100:00 0000006d 00000000?

cookies · 2013-10-10 14:27:10

dumphblooz wrote:

What is hotkey ATK0100:00 0000006d 00000000?

I have no idea, but it looks like this one gets send whenever your sleep button gets pressed by magic. So, as a workaround you could check for this hotkey and ignore the sleep button if it's present. Also, if you don't use an external keyboard (or you don't use the sleep button on the external one), you could check for that "K" at the end of the sleep button line and use that to decide whether you actually want your laptop to go to sleep or just ignore the request.

Edit: Since this strange going to sleep mode might be a very nice protection against overheating (my laptop just cuts the power, that's not very nice): Does your laptop get hot before it goes to sleep all by itself? You might also want to install lm_sensors for closer examination.

Last edited by cookies (2013-10-10 14:32:57)

dumphblooz · 2013-11-15 16:23:31

Hmm. I think

acpi_osi=linux

solves problem

WonderWoofy · 2013-11-15 17:59:36

How funky. It seems strange that not telling acpi that you are Linux would randomly push the sleep button.

In any case, I have to use acpi_osi="!Windows 2012" to tell my machine that it is not windows 8. Apparently there has been some kind of change in how the backlight is handled in windows, which doesn't adhere to the acpi spec. So of course, the vendors go ahead and change their machines to fit the windows way, correct or not. But your machine should be just old enough to not have to deal with this particular backlight issue.

Glad you seem to have gotten with worked out. If you are satisfied with your own solution there, please mark the thread as [Solved], which can be done by editing the first post. That way others who might be seeking a similar answer will be able to know that this thread has a potential fix.

dumphblooz · 2013-11-18 03:14:27

WonderWoofy wrote:

How funky. It seems strange that not telling acpi that you are Linux would randomly push the sleep button.

Not randomly. Button pushed under heavy CPU load.

WonderWoofy wrote:

Glad you seem to have gotten with worked out. If you are satisfied with your own solution there, please mark the thread as [Solved], which can be done by editing the first post. That way others who might be seeking a similar answer will be able to know that this thread has a potential fix.

I'm not entirely sure that solved. I need more time for testing.

dumphblooz · 2013-11-26 17:27:21

Nothing solved Latop still going sleep under high cpu load.

Arch Linux

#1 2013-10-08 07:40:51

Who pressed SleepButton on my laptop?

#2 2013-10-09 02:50:48

Re: Who pressed SleepButton on my laptop?

#3 2013-10-09 02:56:38

Re: Who pressed SleepButton on my laptop?

#4 2013-10-09 03:01:27

Re: Who pressed SleepButton on my laptop?

#5 2013-10-09 03:02:24

Re: Who pressed SleepButton on my laptop?

#6 2013-10-09 03:11:44

Re: Who pressed SleepButton on my laptop?

#7 2013-10-09 03:38:11

Re: Who pressed SleepButton on my laptop?

#8 2013-10-09 04:22:04

Re: Who pressed SleepButton on my laptop?

#9 2013-10-09 05:00:26

Re: Who pressed SleepButton on my laptop?

#10 2013-10-09 05:04:33

Re: Who pressed SleepButton on my laptop?

#11 2013-10-09 07:42:04

Re: Who pressed SleepButton on my laptop?

#12 2013-10-09 07:56:30

Re: Who pressed SleepButton on my laptop?

#13 2013-10-09 10:26:44

Re: Who pressed SleepButton on my laptop?

#14 2013-10-09 15:47:32

Re: Who pressed SleepButton on my laptop?

#15 2013-10-09 16:35:52

Re: Who pressed SleepButton on my laptop?

#16 2013-10-09 21:16:12

Re: Who pressed SleepButton on my laptop?

#17 2013-10-09 21:27:18

Re: Who pressed SleepButton on my laptop?

#18 2013-10-09 22:40:07

Re: Who pressed SleepButton on my laptop?

#19 2013-10-09 22:45:58

Re: Who pressed SleepButton on my laptop?

#20 2013-10-10 14:11:38

Re: Who pressed SleepButton on my laptop?

#21 2013-10-10 14:27:10

Re: Who pressed SleepButton on my laptop?

#22 2013-11-15 16:23:31

Re: Who pressed SleepButton on my laptop?

#23 2013-11-15 17:59:36

Re: Who pressed SleepButton on my laptop?

#24 2013-11-18 03:14:27

Re: Who pressed SleepButton on my laptop?

#25 2013-11-26 17:27:21

Re: Who pressed SleepButton on my laptop?

Board footer