0

Recently, my laptop has begun to randomly switch off. I assumed it was overheating, but monitoring the cpu temperatures doesn't seem to indicate this. It stays around or below 50*C in normal usage, which is pretty good I would think.

What else could cause it to switch off like this? It makes a sort of a "beep" noise when it turns off. Dell tech support plainly said "overheating" and are sending someone down since I still have warranty on this, but I'd like to still be prepared for other issues. What other hardware failure could cause this?

$ sudo sensors
acpitz-virtual-0
Adapter: Virtual device
temp1:        +45.5°C  (crit = +100.0°C)

nouveau-pci-0100
Adapter: PCI adapter
temp1:        +50.0°C  (high = +95.0°C, crit = +105.0°C)

coretemp-isa-0000
Adapter: ISA adapter
Core 0:       +48.0°C  (high = +95.0°C, crit = +105.0°C)
Core 2:       +48.0°C  (high = +95.0°C, crit = +105.0°C)
FranciscoD
  • 103
  • 3
  • What about other components? Chipset temperatures, battery temperatures, etc. – vcsjones Apr 03 '13 at 00:39
  • I use the "sensors" tool on my Fedora/Linux system. Whatever it showed seemed normal: cpu + gpu I think. The battery isn't warm at all. I don't quite know how to monitor temperatures from the other parts. – FranciscoD Apr 03 '13 at 00:42
  • If I switch it back on immediately after it goes off, it'll switch off again in about a minute. It isn't an OS issue: I let it stay at the BIOS screen and it still went off. On this "restart", the fan does run pretty hard. – FranciscoD Apr 03 '13 at 00:44
  • Try to take out on of the rams if you have two. – Dilshod Apr 03 '13 at 19:35

1 Answers1

1

What other hardware failure could cause this?

Virtually any other component, if malfunctioning, could cause this behavior. A short across the Mini-PCI bus (Wireless card), a short across the drive controller (bad hard drive or optical drive)... I suppose it is even possible for a bad stick of Ram to cause the laptop to shut down. A faulty power adapter, DC jack or even a bad battery could cause it. A component on the motherboard that isn't removable... I suppose even a faulty lid switch could give the impression of it turning off.

As far as being able to monitor your temperature, what are you using to do this? Not all laptops include chipsets that allow for temperature monitoring (although almost all newer ones do). Not all software can access those sensors. I've even seen where Speedfan caused a machine to BSOD every time it tried.

Bon Gart
  • 12,950
  • 1
  • 25
  • 36
  • I'm using "sensors" (http://www.lm-sensors.org/). I've added the output to my question description. – FranciscoD Apr 03 '13 at 00:55
  • Thats one of those cases where you miss the Windows "blue screen of death". The infamous screen is the system's last attempt of briefing the user with what may have caused this irrecuperable hardware accident. Linux don't do that, it just drop dead. – Havenard Apr 03 '13 at 01:26
  • LM-sensors appears to need extensive configuration, and their motherboard support (chip support actually) is still based on feedback data. The configuration modifications based on specific motherboards seem to be more about labeling the outputs and upper/lower limits though. Supposedly, the auto chip detection doesn't work for Dell, and you are supposed to use the i8k drivers... which could mean if you used the wrong sensor drivers, you might be getting incorrect readings. – Bon Gart Apr 03 '13 at 01:31
  • I did run `sensors-detect` as [this post](http://superuser.com/questions/25176/how-can-i-monitor-the-cpu-temperature-under-linux) details. I'm running the system without battery to see if the battery could be causing it. – FranciscoD Apr 03 '13 at 01:44
  • Then you can see where the data available at lm-sensors.org isn't all that complete and/or is still a work in progress. I agree, if the information provided by the sensors is accurate, and the sensors were properly detected, then it may not be overheating. – Bon Gart Apr 03 '13 at 01:58
  • Well, the Dell guy came and changed the fan and cleaned some vents ("overheating, of course" is what he said). 30 minutes after he left, the system crashed. It won't turn on at all now. So much for their diagnostics. I guess what we do know is that it wasn't the processor overheating. Thanks for all your help folks. He'll come in again tomorrow and take another look at it. – FranciscoD Apr 04 '13 at 01:01