Did a kernel upgrade in Fedora yesterday. Rebooted today. Wouldn’t boot, had errors. Rather than simple reboot, I decided to shut the whole machine down, then turn it on again. Now it’s dead.
Guess I’ll try swapping the power supply first. If that doesn’t work, sounds like a motherboard or CPU problem.
[Update a few minutes later]
Aaaaaaand, I can’t find any spare supplies. Have to run over to Fry’s to buy one, that probably isn’t the problem…
[Update after returning from Fry’s]
Welp, before I opened the new PS, I tried firing it up again. It booted without complaint. I guess I scared it with the new PS. #HappyHalloween
Seriously, though, it’s probably still a symptom of an incipient problem, probably from overheating.
Ouch! I guess it never hurts to have a spare PSU around, if that turns out not to be the problem. I picked up a cheap power supply tester a while ago; that’s been a helpful tool. Might not be a bad thing to have in the tool kit for a few bucks.
D’oh!
I actually have one somewhere, and had forgotten about it. I haven’t opened the new PS yet, I’ll go look for it.
Yuck. Could be a failing motherboard, too.
You can try running “cpuburn” with the temperature meter (BIOS or other) to see how it withstands the temperature.
It’s probably a good idea to vacuum the fans and things like that. Probably filled with gunk.
It’s an open case.
You have a CPU and PSU fan right?
Yes.
I once knew a guy whose computer kept black screening. He said he had an open case as well. Turns out he had the CPU fan disconnected from the start. Another guy I knew had the heatsink improperly attached.
That’s not my problem. It’s been running fine for years.
How old is this laptop?
It’s not a laptop, it’s a desktop. Mobo is a couple years old.
Ah. Homebuilt or prebuilt? Supposedly the TIM can dry out after a couple of years and lead to overheating.
I assume you’ve checked for dust & for other things that can block vents, and that the CPU fan hasn’t failed.
Completely homebuilt. There are no obvious reasons for cooling problems.
Rand still running with the unplugged case fan? Or did you finally replace it? You know, the one you said wasn’t there? 😛
I know this might sound a like bit of pernicious advice, but have you ever thought of investing in two systems? The production one that has to stay up and the one you can install new software on and kick the tires? Otherwise both systems are running identical hardware and software. Just a thought… Aerospace had this concept long long ago in a galaxy far far away….
Dried-out TIM (oh, I forgot to define the acronym for those who don’t know, which may not include you: Thermal Interface Material, the goop you put between the CPU and heat sink) isn’t obvious, it just results in decreased heat removal ability of your heat sink/fan combo over time, which can eventually cause the CPU to throttle. It’d have to be real bad to do worse, but the next worse thing is “the CPU heats up so fast it shuts itself off a couple seconds after you turn the power on”, which, going back to the post, isn’t what you described.
Did you use your PSU tester yet (or did you mean you have a spare PSU)?
I didn’t find the PS tester, and I bought a new PS, but I’m probably going to return it unopened. I haven’t had any problems since that morning (including a reboot). Weird.
I’ve had issues with the computer booting once, that happened to be solved by pulling the DRAM modules out and putting them back in again, over and over, to remove oxidation in the contacts.