On Friday, 25 November 2016 07:11:15 UTC+11, Loren Rogers  wrote:
> On 11/24/2016 01:39 PM, Loren Rogers wrote:
> >
> >
> > On 11/24/2016 03:57 AM, Pawel Debski wrote:
> >> W dniu środa, 23 listopada 2016 00:34:09 UTC+1 użytkownik Drew White 
> >> napisał:
> >>> On Tuesday, 22 November 2016 13:41:30 UTC+11, Loren Rogers  wrote:
> >>>> On 11/21/2016 06:24 PM, Drew White wrote:
> >>>>> On Tuesday, 22 November 2016 06:04:43 UTC+11, Loren Rogers  wrote:
> >>>>>> On 11/21/2016 11:04 AM, Loren Rogers wrote:
> >>>>>>> On 11/21/2016 12:42 AM, Drew White wrote:
> >>>>>>>> On Sunday, 20 November 2016 04:56:03 UTC+11, Loren Rogers  wrote:
> >>>>>>>>> Another correlation I've noticed is that my machine randomly 
> >>>>>>>>> shuts
> >>>>>>>>> itself down without warning when I'm browsing in the 
> >>>>>>>>> Anon-Whonix VM. It
> >>>>>>>>> seems that simply having the Whonix browser open causes the 
> >>>>>>>>> problem.
> >>>>>>>>> I've not been able to pin down an exact cause, but it seems to 
> >>>>>>>>> happen
> >>>>>>>>> after about 5-20min. When this happens, the machine sometimes 
> >>>>>>>>> ends
> >>>>>>>>> up in
> >>>>>>>>> a hung state (black screen) at the end of the shutdown process.
> >>>>>>>>>
> >>>>>>>>> I've also noticed that the fan speeds up right at it starts to
> >>>>>>>>> shutdown.
> >>>>>>>>> (The screen turns to the Qubes logo with the progress bar, 
> >>>>>>>>> then the fan
> >>>>>>>>> cranks up.) Sometimes the bar makes it all the way to the end, 
> >>>>>>>>> other
> >>>>>>>>> times it seems to simply crash to a hault. As I mentioned 
> >>>>>>>>> elsewhere,
> >>>>>>>>> the
> >>>>>>>>> Thinkpad X201t is known to have overheating issues, but I'm 
> >>>>>>>>> not sure if
> >>>>>>>>> this is related. I'm not working the machine particularly hard 
> >>>>>>>>> (just
> >>>>>>>>> browsing articles on the web), and the hardware is not 
> >>>>>>>>> particularly hot
> >>>>>>>>> to the touch.
> >>>>>>>> When it gets to the qubes logo screen, press ESC to see what it's
> >>>>>>>> actually doing.
> >>>>>>>>
> >>>>>>>> If you wish to always know what it's doing, turn off rhgb and 
> >>>>>>>> quiet
> >>>>>>>> in the boot config.
> >>>>>>>>
> >>>>>>>> Then you will see where the issue is.
> >>>>>>> Thanks, I'll give that a shot next time it happens. I feel like 
> >>>>>>> it'll
> >>>>>>> go by too quickly for me to see what's happening; does it also 
> >>>>>>> write
> >>>>>>> its activity to a log somewhere?
> >>>>>> I can now confirm that it's an over heating issue. When it went 
> >>>>>> into the
> >>>>>> automatic shutdown sequence, I pressed escape and managed to take 
> >>>>>> note
> >>>>>> of a few of the messages. One of the very first ones was 
> >>>>>> something about
> >>>>>> "thermal_zone_0 critical temperature reached: 128C", which I 
> >>>>>> assume is
> >>>>>> the cause. (This isn't an exact quote, since I noted it from 
> >>>>>> memory.)
> >>>>>>
> >>>>>> This raises some questions:
> >>>>>> - What could be causing this overheating issue in Whonix?
> >>>>>> - Is 128C a normal temperature for the safety shutdown to kick in?
> >>>>>> - Does Qubes have a warning / alert system for potential 
> >>>>>> overheat? (Like
> >>>>>> low battery)
> >>>>> It is a high temperature, but does it ONLY happen in Whonix?
> >>>>> Or if you push the PC does it happen also?
> >>>>> Have you tried limiting the threads Whonix can use?
> >>>>>
> >>>>> Sometimes CPUs have shutdown at 99 degrees.
> >>>>> So 128 degrees is a bit high in my own opinion.
> >>>>>
> >>>>> I recommend you check the CPU Fan and heatsinks (if it has them).
> >>>> Thanks for the input - I just dusted out the fan, and we'll see if it
> >>>> helps. It wasn't too bad, but we'll see if there's an improvement.
> >>>>
> >>>> No, it also randomly goes into auto-shutdown when backing up VMs.
> >>>> However, that happens about 20% of the time. Whonix seems to do it 
> >>>> about
> >>>> 80% of the time, the other 20% I figure I shut it down before it 
> >>>> does so
> >>>> on its own. I figure there may be something in the Whonix VM that's
> >>>> causing my processor to over work itself. The auto-shutdowns may be
> >>>> ultimately linked to dust in the fan or something like that, but if
> >>>> there's something processor intensive in Whonix, it may be worth 
> >>>> looking
> >>>> into.
> >>>>
> >>>> Also, a heat warning message would be nice. I assume the thresholds 
> >>>> are
> >>>> set via the bios - is there a standard way of monitoring this? (I'm 
> >>>> not
> >>>> particularly well versed in this sort of thing.)
> >>>
> >>>
> >>> I recommend you get your HDD checked, and your RAM.
> >>>
> >>> Test both thoroughly.
> >>> Could be some bad sectors.
> >>> Also run a smartd check.
> >>>
> >>> Some PCs have system diagnostics built in for RAM in the startup 
> >>> sequence.
> >> Thinkpads have known problem that after long time of usage GPU 
> >> radiator glued to the chip goes loose and X201 is rather older model. 
> >> Not sure how to measure GPU temp but if this is the case you can 
> >> improve it yourself with superglue & silver smear or any repair shop 
> >> around shall do it for a few coppers.
> > Thanks everyone -- this is great info.
> >
> > I think the dust removal was really helpful. I've noticed that the fan 
> > is working a lot less, and it's not crashing during backups like it 
> > did before. I'll check the HDD and RAM soon, and I'll keep the GPU 
> > radiator issue in mind. (I had no idea this was an issue - thanks for 
> > the pointer!)
> >
> > Any thoughts on a temperature warning message?
> 
> Also, returning to the original issue, I've attached a photo of the 
> output logs when the machine hangs during shutdown. (I pressed esc while 
> the Qubes logo was showing, before it locked up.) The last few messages 
> confirm that I was having an overheating issue, but the device-mapper 
> messages are new to me. Does anyone have a lead on what this may be?

I would recommend taking it to a real computer shop and getting it checked out.

There could be several things wrong.. It may need a reseat of the heatsinks, 
and also a check of the sensors. They can sometimes be old and go bad and read 
wrong.

If you have a temperature sensor that you can use, check the temperature of the 
pc case near the CPU and GPU just to see how hot it is on the case. That may 
give you a good indication of something being very wrong.

But it's cold to the touch, then either it'd doing well in moving the heat, or 
the sensor is knackered.

Either way, it may be a good idea to get someone to give it a quick checkup.

-- 
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/93adc025-7daa-4138-b30d-01558a75fbf5%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to