[qubes-users] Re: laptop fails to sleep, becomes very hot

2018-08-27 Thread Dave C
On Tuesday, March 13, 2018 at 3:46:03 PM UTC-7, Yuraeitha wrote:
> On Tuesday, March 13, 2018 at 10:13:45 PM UTC+1, Dave C wrote:
> > Many times in recent days, I've closed my Qubes laptop, expecting it to 
> > sleep, and put it into my backpack.  Normally it sleeps as expected.
> > 
> > Two of those times, once today and once several days ago, I reach into the 
> > backpack and find the laptop extremely hot to the touch.  I'd expect the 
> > fan to be on at these temperatures, but it isn't.
> > 
> > When I open the laptop, theres no response and similarly no response to 
> > pressing keys.  I press and hold the power button for a long time.  There's 
> > still no feedback from the computer, but it does turn off and cool down.
> > 
> > So far, I haven't noticed permanent damage, but I worry about that.  It's 
> > an unpleasant surprise to find that the battery is totally drained and I'm 
> > not sure what the computer has been trying to do.
> > 
> > Any thoughts?  Or suggestions how to troubleshoot this?
> > 
> > -Dave
> 
> I can confirm two similar cases in recent weeks, though there's been quite 
> some time between them, weeks? and I typically use suspend multiple times a 
> day, except sometimes weekends.
> 
> There are other cases, where sys-net stops responding, or the internet isn't 
> working when it's brought back from suspend, though these cases are just as 
> rare as the above no suspend issue (I'm using the wi-fi module blacklisting 
> in sys-net btw).
> 
> To me it seems like a semi-rare bug that is triggered by something yet 
> unknown, though maybe it's known on github tracking issues? 
> 
> Either way, I think this might be related to the sys-net, but I can't really 
> be sure here. Perhaps you can write a script that disables the networking, 
> shutsdown sys-net, and then starts sys-net again and re-connects networking 
> as it wakes up. 
> 
> This isn't a pretty solution, but it might work? Does your issue happen 
> frequent enough so that you can test if sys-net is shutdown, that suspend 
> then works properly? Would be ideal to know if that is indeed it before 
> spending some time on such a script.

The problem of a sleeping laptop getting *very* hot happened again today.  That 
gives you some idea how frequently: not enough to make troubleshooting easy.

I wonder if part of the cause is taking the laptop while sleeping and plugged 
in, unplugging and putting in backpack.  That is I wonder whether unplugging 
disturbs its sleep.

On reboot, my system clock is 1 hour early.  Meaning it shows 2:21 when it's 
actually 3:21.  I mention in case it's relevant.

After restarting, I used `journalctl -o short-precise -k -b -1` to get logs 
from the prior boot.  Note: this works on dom0, but on sys-net it looks like 
the "prior boot" comes from the template, so its not useful.

There are some messages in the log related to CPUs sleeping and waking.  So I 
include it here.  Maybe something will jump out to an expert here.  I'm really 
not sure what these messages mean.

Attached log shows this morning around 9am.  That's when I closed the lid.  The 
laptop was plugged in.  Around 1pm I unplugged it, put it in my backpack.  
Around 3pm, I took it out.  Noticed it was hot to the touch.  No fans blowing, 
just a really hot laptop.  Completely unresponsive to any input.  I pressed the 
power button for a while...a couple times...until it powered on (I wanted to 
get the fan running).  The battery was about 50% drained.

-- 
You received this message because you are subscribed to the Google Groups 
"qubes-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to qubes-users+unsubscr...@googlegroups.com.
To post to this group, send email to qubes-users@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/qubes-users/94db4dee-087b-4c24-adda-a1084246834b%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
Aug 27 09:06:10.441594 dom0 kernel: Freezing user space processes ... (elapsed 
0.086 seconds) done.
Aug 27 09:06:10.441802 dom0 kernel: OOM killer disabled.
Aug 27 09:06:10.441915 dom0 kernel: Freezing remaining freezable tasks ... 
(elapsed 0.102 seconds) done.
Aug 27 09:06:10.442038 dom0 kernel: Suspending console(s) (use 
no_console_suspend to debug)
Aug 27 09:06:10.442252 dom0 kernel: PM: suspend devices took 0.596 seconds
Aug 27 09:06:10.74 dom0 kernel: ACPI: EC: interrupt blocked
Aug 27 09:06:10.444624 dom0 kernel: ACPI: Preparing to enter system sleep state 
S3
Aug 27 09:06:10.444750 dom0 kernel: ACPI: EC: event blocked
Aug 27 09:06:10.444873 dom0 kernel: ACPI: EC: EC stopped
Aug 27 09:06:10.444975 dom0 kernel: PM: Saving platform NVS memory
Aug 27 09:06:10.445094 dom0 kernel: Disabling non-boot CPUs ...
Aug 27 09:06:10.445229 dom0 kernel: [ cut here ]
Aug 27 09:06:10.445380 dom0 kernel: WARNING: CPU: 1 PID: 0 at 
/home/user/rpmbuild/BUILD/kernel-4.14.57/linux-

[qubes-users] Re: laptop fails to sleep, becomes very hot

2018-03-13 Thread Yuraeitha
On Tuesday, March 13, 2018 at 10:13:45 PM UTC+1, Dave C wrote:
> Many times in recent days, I've closed my Qubes laptop, expecting it to 
> sleep, and put it into my backpack.  Normally it sleeps as expected.
> 
> Two of those times, once today and once several days ago, I reach into the 
> backpack and find the laptop extremely hot to the touch.  I'd expect the fan 
> to be on at these temperatures, but it isn't.
> 
> When I open the laptop, theres no response and similarly no response to 
> pressing keys.  I press and hold the power button for a long time.  There's 
> still no feedback from the computer, but it does turn off and cool down.
> 
> So far, I haven't noticed permanent damage, but I worry about that.  It's an 
> unpleasant surprise to find that the battery is totally drained and I'm not 
> sure what the computer has been trying to do.
> 
> Any thoughts?  Or suggestions how to troubleshoot this?
> 
> -Dave

I can confirm two similar cases in recent weeks, though there's been quite some 
time between them, weeks? and I typically use suspend multiple times a day, 
except sometimes weekends.

There are other cases, where sys-net stops responding, or the internet isn't 
working when it's brought back from suspend, though these cases are just as 
rare as the above no suspend issue (I'm using the wi-fi module blacklisting in 
sys-net btw).

To me it seems like a semi-rare bug that is triggered by something yet unknown, 
though maybe it's known on github tracking issues? 

Either way, I think this might be related to the sys-net, but I can't really be 
sure here. Perhaps you can write a script that disables the networking, 
shutsdown sys-net, and then starts sys-net again and re-connects networking as 
it wakes up. 

This isn't a pretty solution, but it might work? Does your issue happen 
frequent enough so that you can test if sys-net is shutdown, that suspend then 
works properly? Would be ideal to know if that is indeed it before spending 
some time on such a script.

-- 
You received this message because you are subscribed to the Google Groups 
"qubes-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to qubes-users+unsubscr...@googlegroups.com.
To post to this group, send email to qubes-users@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/qubes-users/694932ce-629d-4c2c-8b49-94e4936563cf%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.