-3

I have a desktop PC I bought 10 years ago and I use it as a "server", so it's always on.

It started to suddenly shutdown few months ago, after few days or weeks being on (last time I powered it on, it lasted 46 hours to shutdown).

So, I think I have a hardware issue that is making my PC suddenly shutdown.

After that, I made some tests:

  • I searched in Windows event viewer and didn't found anything weird.

  • I stressed it with CPU-Z and watched temperature with SpeedFan, but it reached no more than 65 ºC.

Power outages weren't a problem, because I have an UPS. There are just a few power outages per year, if any. But my PC went down more than 10 times.

I also configured auto power in BIOS and manually tested it (powering on and off my electric installation) and it worked. But now isn't working auto power on. That should be a hint of what is the issue here.

I dont live at the place my computer is located, so I only can rely on logs. Also it's annoying because I have to drive there and power it on every time I need it. But I have physical access to it, so I can try anything you suggest.

I thought some ideas:

  • Looking for more sophisticated temperature tests (Prime95 or something).

  • Windows RAM test.

  • Also scanning for malware (it shouldnt have any, because the system is updated and noone can access it physically, other than me).

What would be other possible causes and what are the tests I could do?

I know I should be using Linux for this, but Windows worked just fine for years.

PC is an MSI with AMD processor and RAID SSDs (two mirrored, bought 2 years ago). It uses Windows 10 x64.

14
  • And your question is?????
    – davidgo
    Commented Feb 18, 2023 at 7:16
  • @davidgo i edited it now. I want u tell me more possible causes and tests to do. Or ur opinion on what u think is more probable Commented Feb 18, 2023 at 7:31
  • We can't realistically help you if you are not even at your machine.
    – Ramhound
    Commented Feb 18, 2023 at 7:45
  • 1
    Gamer speak is not the adequate language for a global audience where English is not the mother tongue. Why helping someone with an understandable answer who does not bother to write full words when asking?!
    – r2d3
    Commented Feb 18, 2023 at 8:25
  • 1
    In addition to running memtest, have you looked at your windows settings - eg Control Panel -> System and Security -> System -> Advanced Ssytem Setings -> Advanced -> Settings and ensure you are writing events to system log and "Automatic Restart" is disabled? Similarly, have you checked the system log? Also, have you looked at your BIOS to check if there is the option to automatically restart on power on?
    – davidgo
    Commented Feb 18, 2023 at 22:23

1 Answer 1

0

I solved it by cleaning my pc, applying thermal paste, adding a case cooler, and switching my CPU cooler for a better one. It reached 85 ºC under CPU-Z stress, but now just 70 ºC. I saw broken Windows configs (not opening), missing drivers, and Windows Update not working.

I concluded that temperature was the root cause, and Windows's broken image was its symptom.

Maybe it was updating Windows and it shutdown because of the high temperatures, and that made a broken OS image.
Believe me: that oven was 80 ºC during a simple LAN backup of 50 GB folder. Even it shut down! But I noticed all this after moving my pc to my home to diagnose and repair.

1
  • "Maybe it was updating Windows, and it shutdown because of high temperature, and that made a broken OS image." - While it's possible the system did shutdown, Windows has a substantial amount of safe guards to prevent system corruption while a Windows Update is being installed. If installing proper cooling solved the problems, the instability was simply due to heat. BLUF: You likely didn't actually have a "broken Windows image" your system was just not stable due to the heat.
    – Ramhound
    Commented Feb 28, 2023 at 10:09

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .