System hibernates despite "hibernate after" set to "never"

steve108 said:

Post your output of "powercfg /requestsoverride" please.

C:\>powercfg /requestsDISPLAY:None.SYSTEM:[SERVICE] \Device\HarddiskVolume3\Program Files\Macrium\Common\MacriumService.exe (MacriumService)[PROCESS] \Device\HarddiskVolume3\pctools\disk-utilities\disk-imagers-cloners\Macrium\v7.3.5555\ReflectBin.exeAWAYMODE:[SERVICE] \Device\HarddiskVolume3\Program Files\Macrium\Common\MacriumService.exe (MacriumService)[PROCESS] \Device\HarddiskVolume3\pctools\disk-utilities\disk-imagers-cloners\Macrium\v7.3.5555\ReflectBin.exeEXECUTION:None.PERFBOOST:[DRIVER] Legacy Kernel CallerPower ManagerACTIVELOCKSCREEN:None.


stand by... this is a mess; let me fix the formatting

- - - Updated - - -

[C:\>powercfg /requests
DISPLAY:
None.

SYSTEM:
[SERVICE] \Device\HarddiskVolume3\Program Files\Macrium\Common\MacriumService.exe (MacriumService)
[PROCESS] \Device\HarddiskVolume3\pctools\disk-utilities\disk-imagers-cloners\Macrium\v7.3.5555\ReflectBin.exe

AWAYMODE:
[SERVICE] \Device\HarddiskVolume3\Program Files\Macrium\Common\MacriumService.exe (MacriumService)
[PROCESS] \Device\HarddiskVolume3\pctools\disk-utilities\disk-imagers-cloners\Macrium\v7.3.5555\ReflectBin.exe

EXECUTION:
None.

PERFBOOST:
[DRIVER] Legacy Kernel Caller
Power Manager

ACTIVELOCKSCREEN:
None.

That's not the command/option I requested - just copy/paste it out of my post

steve108 said:

While the backup is running (without your .bat file), what is the output of "powercfg /requests"?

I would expect it to show that the backup app is running and will prevent sleep/hibernation. Are you saying the backup never completes if you don't run your .bat file? Or is it possible the backup does complete, then the PC sleeps/hibernates afterwards?

BTW, the bat file includes a call to the backup prog between the powercfg /h on|off calls. It's not independent, in case there was some confusion. But that should not matter. At this point I am doing tests without the bat file altogether.

steve108 said:

That's not the command/option I requested - just copy/paste it out of my post

You can leave the above results there for the option you did run.

steve108 said:

That's not the command/option I requested - just copy/paste it out of my post

Sorry. Too many things going on at once.

C:\>powercfg /requestsoverride
[SERVICE]

[PROCESS]

[DRIVER]

- - - Updated - - -

steve108 said:

That's not the command/option I requested - just copy/paste it out of my post

I had been still responding to "You want to run "powercfg /requests" while backup is running".

- - - Updated - - -

I again tested that it still fails as I claimed -- sometimes you go back and the problem is no longer there. Now I will reboot and see if it persists.

Everything looks good - I'm perplexed. If you laptop is usually connected to AC, you could consider permanently disabling hibernation and just rely on sleep as I do. Hibernation seems to often be a general nuisance, so I don't use it. I use sleep, and shutdown if rarely necessary.

Have you confirmed that the BIOS and drivers are up to date from PC manufacturer website or their update software app?

steve108 said:

Everything looks good - I'm perplexed. If you laptop is usually connected to AC, you could consider permanently disabling hibernation and just rely on sleep as I do. Hibernation seems to often be a general nuisance, so I don't use it. I use sleep, and shutdown if rarely necessary.

Have you confirmed that the BIOS and drivers are up to date from PC manufacturer website or their update software app?

Yes, latest bios was updated just a few days ago. Same with Windows. In fact, if I didn't know that the problem has been occurring for at least a month (based on backup file created, modified dates), I would blame the updates.

Also, I just finished rebooting and confirm that the problem persists.

Could you address my question about disabling all power requests (as opposed to individual overrides)? I vaguely remember that maybe Brink posted how to do that, unless I saw it in some random other place. I should check that for whatever reason that is not the current state. I keep a pretty detailed log so I don't think I did that, but maybe I did in some moment of desperation.

My other option is to disable system unattended sleep timeout (assuming setting it to zero does that) and allow the system idle timeout (sleep|hibernate after) to rule what happens (assuming that would actually happen). That still has the problem that if the backup takes longer than the timeout value, I still lose.

I don't generally like leaving the laptop in "sleep" state because it stays warm and frequently I want it in or under something that won't allow good heat dissipation. I use sleep on my desktops, but the laptop doesn't really sleep in 'sleep.'

Thanks.

apb said:

..........
Could you address my question about disabling all power requests (as opposed to individual overrides)? I vaguely remember that maybe Brink posted how to do that, unless I saw it in some random other place. .......

I'm not aware of a way to do that. If you find it, I'd like to see how.

steve108 said:

I'm not aware of a way to do that. If you find it, I'd like to see how.

All I was able to find was a group policy thing, which I tested and it doesn't help:

Allow applications to prevent automatic sleep (plugged in) (from Allow applications to prevent automatic sleep (plugged in) )

They claim that "If you disable or do not configure this policy setting, applications, services, or drivers do not prevent Windows from automatically transitioning to sleep. Only user input is used to determine if Windows should automatically sleep."

However, I think that other than the equivalent registry setting to that policy, there was some other thing in the registry that allegedly overrode all power requests, but for me it's lost to the mists of time.

Anyway, thanks for all the help trying to figure out why this is happening. I'm starting to think that maybe the power requests no longer apply to hibernation. Unfortunately, I don't have another win 10 machine to see if the behavior is the same.

--peter