Question PC randomly started up by itself in the morning ?

[vdzr]

TLDR: found out my PC randomly turned on and off in the morning (shutting down screen with spinning circle) and I had to cold boot for main Nvme drive to be detected then had another random shutting down screen while using pc. Have USB Wake, Wake on lan, and fast startup disabled (all within both Windows and BIOS) smart scan found no errors, I don't see any scheduled tasks to do with shutdown .

I'm not sure what the issue is here. 2 days go I got home around 7 p.m and noticed my PC powered on (hasn't done this before and I've had it for 3 years) with the American Megatrends boot logo displaying all my drives, but realized my boot drive (SN850x Nvme) was missing from the list. The error message displayed was something about not being able to boot or find an OS which makes sense as no boot drive was in the list.

I did a cold boot which ended up working normally and went into Windows to make backups. I checked the event viewer and noticed that it powered on at 9:01 a.m (logs started appearing) and an error for unexpected shutdown at ~9:25 a.m so I assume it was sitting on the Megatrends screen until I did the cold boot at 7 p.m.

After making backup's I used my PC as normal then an hour later it went into the shutting down screen animation while I was gaming (not an instant power off).

It doesn't seem like a psu issue as that would instantly turn off with no warning but I ordered a better one anyways as I was already planning to swap it out (EVGA 700 Br to a Seasonic Prime GX-850).

At this point I'm waiting for my PSU to arrive and need some help figuring out what can be the issue here as I have Wake on lan, USB wake, and fast startup disabled (disabled in bios too).
Powercfg -lastwake results with 0 and my temps were also fine while gaming with the CPU at 51°C and GPU at 43°C.

*Specs:*

CPU=Ryzen 5 3600

GPU=RX6650 XT

Mobo=Tuf gaming B450M plus ii

Ram=48GB 3600Mhz (stable/no errors in memtest/windows memory diagnostics)

 

[Roland Of Gilead]

Hey there,

Yes, this could be a PSU issue. Replacing it will help to diagnose.

In the meantime are all your system drivers, including chipset up to date? What bios are you running?

 

[stonecarver]

Interesting I would check the front panel wires to motherboard power and reset button. Could be loose or just need a pull off than push back down kind of thing.

Also If you never use it go make sure remote desktop box is unchecked.

 

[vdzr]

Hey there,

Yes, this could be a PSU issue. Replacing it will help to diagnose.

In the meantime are all your system drivers, including chipset up to date? What bios are you running?

Most are up to date including the amd GPU drivers and I believe I do need to update the chipset drivers but avoiding for now as I don't want to risk a shutdown during the process, the bios is on Version 4402 so 2 versions behind. Will update both once I install the PSU tomorrow.

 

[vdzr]

Interesting I would check the front panel wires to motherboard power and reset button. Could be loose or just need a pull off than push back down kind of thing.

Also If you never use it go make sure remote desktop box is unchecked.

This is what I was thinking too perhaps something is shorting inside the front panel. Occasionally it takes 3 presses to power on, but the connectors look normal. I'll unplug the Front panel completely and only use the new power button I ordered by tomorrow once the PSU is installed.
I used the remote desktop feature once but made sure to disable it after.

 

[Ralston18]

I will add the suggestion to look in Reliability History/Monitor.

The timeline format may reveal some pattern of events with respect to error codes, warnings, and even informational events.

 

[vdzr]

I will add the suggestion to look in Reliability History/Monitor.

The timeline format may reveal some pattern of events with respect to error codes, warnings, and even informational events.

Oddly I only see a single event for the unexpected shutdown on 8/10 at 9:25 a.m although a shutdown also occured around 9 p.m. There's also no additional info under critical events besides that crash.

(All previous critical events are from Logitech battery monitor software)

I am seeing an event though for explorer.exe/windows explorer "stopped working" on this latest boot (8/15) but haven't seen explorer crash or freeze.

 

[vdzr]

I've installed the new PSU and power button, but on first power ON; no NVMe SSD was detected again and went straight to BIOS. Slot 1 was empty (the NVMe slot) but then it let me boot from the NVMe SSD with Windows in slot 1 after leaving the BIOS.

Will monitor if any more shutdowns/random boots occur after this.

Something else I forgot was that I updated my NVMe SSD firmware 1/2 weeks ago but was running fine since. Maybe that has something to do here with needing to cold boot for it to start?

 

[Ralston18]

If everthing is up and running then my thought is that there was some simple glitch - possibly related to a power issue.

Windows can and does fix things on its' own. However, if additional updates or installs were taking place then there may have been some sort of conflict. And the error codes themselves may or may not be helpful.

One thing that you can do is go back and compare the events on the days critical events were logged.

For example, it appears that quite a bit was going on on the 10th and again on the 15th. Try to spread out changes so there is only one change at a time. If problems occur then the problems are likely attributal to that particular change.

Unfortunately, end users may not have all that much control when it comes to updates. Overall though, you can and should be methodical while installing new applications. Wait a day or two (perhaps a couple of boots) before doing the next install. May not always be possible but that does reduce then number of possible conflicts and suspects.

Also take a look at the warnings. Warnings, especially if repeating, can be an early flag that trouble is ahead.

Informational events have value as well., Always interesting to be "informed" about something that you were not aware of or otherwise had nothing to do with.....

Generally straightforward and easy to check things out thereafter. And ask the necessary questions as warranted.

 

[vdzr]

Everything is now working like normal again. No more random shutdown or startups. I realized BIOS settings were reset, and fixed the Nvme issue by setting CSM to disabled and setting boot type from Other to UEFI Windows.

I also found the command "[I]wevtutil qe System /q:"*[System[(EventID=41) or (EventID=1074) or (EventID=6006) or (EventID=6005) or (EventID=6008)]]" /c:100 /f:text /rd:true[/I]" which came up with multiple events for "The process C:\WINDOWS\system32\winlogon.exe (DESKTOP-M7878KO) has initiated the power off of computer DESKTOP-M7878KO on behalf of user NT AUTHORITY\SYSTEM for the following reason: No title for this reason could be found". This leads me to believe it was most likely a faulty power button.

Now i just have to figure out how to get my 1.2875v CPU undervolt setting back. I only have offset mode which starts at 1.1v and setting a positive offset of 0.1875 somehow results in 1.52v!?! Perhaps a BIOS update will resolve this.