ve SP3 keeps turning itself on and gets noisy - how to prevent?

[mmo]

My SP3 has a most annoying "feature": It keeps turning itself back on automatically on after I have sent it to sleep (using "hibernate", i.e. connected standby). This happens sometimes within seconds, sometimes after a few minutes but never later than, say, 30 minutes after switching it "off". It happens mostly, when the device is on power (i.e. charging), but occasionally also while the device is in my carrying bag.

This behavior is most annoying when having the SP3 in the same room where one is sleeping, because not only does the screen turn and illuminates the room on but the device always also starts some activity that causes the fan to go to full-speed, i.e. the device gets really noisy!

How can I find out, WHAT is causing the SP3 to switch on and how can I find out, what it is doing then? The activity always seems to stop (i.e. the fan almost immediately turns down again) as soon as I log in or start typing on it. Thus, when I e.g. bring up task manager to have a look, the device is always shown as almost 100% idle. So, this must be some background activity, that apparently automatically starts the device and then consumes LOTS of CPU.

Any idea, how to identify the culprit and stop this behavior?

 

[buers]

I had seen this as well a couple of times, last time yesterday. I have setup my surface to allow hibernation from the power menu (by some registry hack). In general both hibernation and connected standby work well.

Here is a pasted snippet of powercfg /batteryreport

<<<<
23:35:16 Active Battery 19 % 7 mWh

2015-07-17 00:05:25 Connected standby Battery 14 % 5 mWh

00:05:26 Suspended 14 % 5 mWh

00:06:23 Connected standby AC 14 % 5 mWh

00:06:23 Active AC 14 % 5 mWh

00:06:23 Connected standby AC 14 % 5 mWh

07:00:40 Connected standby Battery 101 % 41 mWh
[Due to a bug in a least the German version of powercfg, the battery capacity is displayed a factor of 1000 too low]
>>>>

At 00:05:26 I manually suspended the SP3 (hibernation). Seems it was on battery then. If I trust the report, I plugged it in then at 00:06:23, at which time it woke up again. (Type cover was over the screen)

Next morning I immediately recognized, that the 10 s delay for waking up from hibernation was missing.

Not sure, if in all cases I had the unexpected wakeup, I plugged in the SP3 after bringing it to hibernation.

I also see (only when plugged in) unexpected high CPU usage, often when I expected the SP3 to be in hibernation. I often have sysinternal's procexp running. It shows 25% System CPU usage for an elapsed time (which is an indication of exactly one hanging thread). User CPU usage is basically idle. There is never a process with hight total CPU usage besides the system process with PID 4. I have never seen significant disk usage or network at the same time. As you have described, after doing something, the high CPU usage stops immediatly.

BTW. I also had several wakeups in my back pack during connected standby and the device became hot. Not sure why. I now typically put it to hibernation, before I put it into the bag. It starts fast enough to my liking.

 

[agt499]

I think Windows needs a power management session for "When I put the PC to sleep, I want it to sleep. Really. Please.".
I had this for quite a while, and I tried a bunch of things, and it presently no longer does it for me.
In my case I would dock to charge before going to bed and would find it awake the next morning or 5 minutes later.
In the end I think it fixed itself with an update, but among the things that are worth trying:

1. Wireless/bluetooth mice and the type cover.
   This may go for your bag situation too -mouse movement can wake from sleep. There are settings in Device Manager to prevent a device from waking from sleep.
   
   
2. Scheduled maintenance (Windows)
   There are settings for this, some of which allow to wake. I don't think this is related to the wake 1-30 minutes in, as it's typically set for like 2AM. Worth trying though.
   
   
3. Scheduled tasks
   All sorts of stuff ends up in here (updaters -think Google, Adobe). There are settings for 'Allow task to wake the PC'.
   
   
4. Lastwake report
   You could try the command powercfg -lastwake :I saw some reports it would say useful thing s - never did for me though.
   
5. Read the Event Logs.
   Warning - this can drive you nuts.
   You want to jump into Event Viewer>Windows Logs>System. When all is well you should see Information Level logs with Source Kernel-power, and messages like 'The system is exiting connected standby Reason: Power Button.'.
   It might give you a clue as to what else is waking it.
   This is where I discovered that when awake the next morning it had usually woken a just few minutes after I put it to sleep.

Good luck!

 

[mmo]

I might have found the culprit! I went through the task-scheduler and inspected each and every task to see, which one had the flag "Wake up computer..." set. I found two tasks related to updating something in Windows Media Center. Apparently MS considers WMC-updates so important that it allows them to wake up the machine! ||-(
Since I use that only very occasionally (maybe once a month) I disabled those (or rather: unchecked the wake-up flag) and, satisfyingly, this night the device remained off and silent. Keeping fingers crossed that this remains so!

BTW: after wading trough all those defined tasks I am not astonished any more that the system's fan was always working like mad after the device turned itself on automatically. It is unbelievable how many tasks there are defined to run when the system is idle! I guess my system is more busy while its idle than while I am working with it. ;-)

 

[buers]

1.Wireless/bluetooth mice and the type cover.
This may go for your bag situation too -mouse movement can wake from sleep. There are settings in Device Manager to prevent a device from waking from sleep.

When talking about hibernation: can really a bluetooth device wake up the SP3?

In the bag situation I mentioned, it really was connected standby. However, I had no paired bluetooth device with me.

2.Scheduled maintenance (Windows)
3.Scheduled tasks

I will check those later

4.Lastwake report
You could try the command powercfg -lastwake :I saw some reports it would say useful thing s - never did for me though.

Same here - I have never seen anything interesting, just the number of wakes.

5.Read the Event Logs.

This was intersting.

The first log entry in Application after the unexpected wakeup was :

System 
  - Provider 
  [ Name]  Microsoft-Windows-Security-SPP 
  [ Guid]  {E23B33B0-C8C9-472C-A5F9-F2BDFEA0F156} 
  [ EventSourceName]  Software Protection Platform Service 
  - EventID 900
  [ Qualifiers]  16384 
  [...]
  caller=GWX.exe

GWX = Get Windows X=10. Due to an article in a German magazine, this is only marketing. BTW: The Timestamp is 15 seconds *before* the switch from Suspended to Connected Standby happened. However, from System event log I see 19 s time correction.

<<<
Die Systemzeit wurde von ‎2015‎-‎07‎-‎16T22:05:27.330777700Z in ‎2015‎-‎07‎-‎16T22:05:46.500000000Z geändert.

Änderungsgrund: Die Systemzeit wurde mit der Hardwareuhr synchronisiert..
>>>

According to powercfg /batteryreport, this happened, while the device was suspended. So the timestamps seem not consistent ...

Other than this, I see nothing unusal. While I am at it, I always have EventID 17 from BTHUSB when waking up from hibernation:

<<<<
Der lokale Bluetooth-Adapter ist aus einem unbekannten Grund fehlgeschlagen und wird nicht verwendet. Der Treiber wurde entladen.
>>>

My translation: the local bluetooth adapter failed for unknown reason and will not be used. The driver was unloaded.

Still, the pen and the bluetooth mouse work.

Can scheduled tasks really wake up the SP3 from hibernation? The alarm clock can't - that might be useful.