5

Sometimes my computer shuts down unexpectedly, usually when some process is taxing the CPU. This answer mentions that "when a system shuts down in a normal way, it is either a user with root privileges or an acpi script", and "an acpi shutdown can be caused by power button press, overheating or low battery (laptop)." So my guess is that it was overheating, but I'd like to be sure.

A comment in the question linked above says that /var/log/acpid would contain some information, but that file doesn't exist on my machine. Does that log get placed somewhere else in Ubuntu? Do I need to configure something to make it show up?

  • Note: the methods suggested here didn't return any relevant information (dmesg | grep -i acpi didn't return timestamps, and /var/log/kern.log didn't have any entries for the period when the computer shut down). I found these relevant entries in /var/log/syslog, but would like something cleaner/more informative, which I assume acpid would provide — if I could find it...
Improve this question
2

While I'm not sure what more info you are looking for, your pastebin seem to provide your answer. There are several logs that show high CPU temps. The BIOS will shut the computer down if the CPU's get to hot to protect them from burning out. The overheating could be caused by several things:

  • The CPU fan could be dirty or not running at speed
  • The case fan could be dirty or not running at speed
  • If it is a laptop, the vents tend to get clogged with dirt and hair
  • The heat-sink on the CPU could be dirty and clogged
  • The thermal compound between your CPU and the heat-sink could be old and dried out so it is not transferring the heat like it should

In short I would appear from your logs that the system is shutting down due to overheating. The fact that this only happens when you are doing CPU intensive programs also strongly suggests that (in fact that was my first thought before I read your pastebin)

You can likely clean it yourself if you're careful: clean the outside vents with a dry toothbrush. If you open the case make sure that you don't have any static buildup on your body (touch something metal to discharge it), as any static electric spark can fry electronic chips. The best way to clean the inside of the case is with air, most stores sell canned air just for this purpose in the electronic section. If you use an air compressor, make sure the pressure is turned way down. Blow air into the fans and the heat-sink until the dust and hair is gone, make sure to blow off the motherboard afterwards, to clean off any dust you blew onto it. Just make sure that the power is OFF before you open the case — I recommend unplugging it!!

If the computer is old enough that the thermal compound is bad, you can buy one of those at electronics stores, but if you are unsure that you can replace it, I would recommend letting a shop do it for you. If it is done wrong, at best the problem will continue, at worst it could kill the CPU.

I realize that this doesn't answer your question, but it may solve your problem. I'm not sure there is one: if the system shut down due to overheating, I'm not sure you would get much more info that a hardware error, and that you are already showing in you posted log...

In fact the post you linked to says just that in the accepted answer:

Recently I had a system that started repeatedly to power off ungracefully, turned out that it was overheating and the mobo was configured to just power off early. The system didn't have a chance to save logs, but fortunately monitoring the system's temperature showed it was starting to increase just before powering off.

Improve this answer
5
  • Yes, I can indeed deduct the information that overheating was what caused the shutdown, but I wondered if there was a way to get this information in a more direct way. The guy who asked the question whose answer you quoted, commented saying that he found relevant info in /var/log/acpid, so I was hoping it would contain something more informative (perhaps something like "[timestamp] CPU temperature too high (xxxº), shutting down"). In any case, I am curious why that file doesn't seem to exist in my machine. – waldyrious Jan 18 '13 at 15:28
  • Ok, taking into account the "The system didn't have a chance to save logs" line, maybe something more realistic would be "[timestamp] CPU temperature dangerously high (xxxº), computer might shut down unexpectedly". But I don't even know if I'd find such info in the /var/log/acpid — regardless, I'd like to have a chance to check it out. – waldyrious Jan 18 '13 at 15:38
  • @waldir Looking around in my log files did not find any logs that might help you so I still can't give you a location. Keep in mind that if the shutdown is initiated by the motherboard or bios, the system may not get these kind of notifications so there would be nothing to log. It is likely that it would have the same result as pushing the powerbutton, just a system power off signal, if there was a graceful shut down or like pulling the plug if it was not graceful. Sorry I was not able to be of more help.... – TrailRider Jan 19 '13 at 0:31
  • thanks for taking the time to look for that. I realize that the ACPI log might not be of much use for the specific overheating problem, but regardless of that, I would like to know why I can't find such file in my machine. I'll keep looking, but I can't accept your answer. I hope you understand. I'n any case, I'll upvote it :) – waldyrious Jan 21 '13 at 14:51
  • I understand, thanks for the upvote. Hope you can get an answer that solves your problem – TrailRider Jan 21 '13 at 21:30

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.