#PC restarting itself for no apparent reason
1 messages · Page 1 of 1 (latest)
Event Viewer → Windows Logs → System. Sort by Level and scroll to the top where you have the 'Critical' events. Screenshot.
They are all Kernel power
sorted by level
sorted by date and time
@umbral ore Selec the top Kernel Power event and then click the Details tab
Look for BugCheckCode. If it's not 0x0, tell me what it is
Just arrow key down to the next one and quickly check all of them
the 2 criticals error that ive had today their bugcheckcode is 26 and 59
Ok, so you are getting BSODs then
Go to C:\Windows\Minidump (Sometimes Minidumps for some reason) and check if you have any minidump files. If you do, go back to the Windows folder and copy the Minidump folder itself to the Downloads folder (You can use the desktop if you don't have OneDrive syncing files). Zip the copied folder and attach it to a post.
Memory errors.
Are you also getting shutdowns that don't have a bugcheckcode?
i think so
Check in the list
I just checked and no, they all have bugcheckcodes
@umbral ore You have a lot more crashes than dump files (It only saves the last 5, but two of the dump files are from August, so it just recorded three since then)
So the question is why it's not able to create dump files.
It's usually a sign of a storage issue
Storage issue?
Yes
Ok, we can check if you have any WHEA events then
Go back to Event Viewer → Windows Logs → System
Then on the right hand side, select Filter Current Log...
In the menu that opens, find the Event Sources dropdown menu and select "WHEA-logger" in the list
Then click Ok to apply the filter.
If the list is empty, you don't have any events
Its empty
I dont think its related to my ssd, i have already switched it and i still get the same unexpected crashes
What did you switch?
I used to have an old ssd nvme and i had the same issue so i replaced it for a newer one
@umbral ore Let's look at the crashes where we don't have a dump file.
Do these instructions again, but instead of selecting WHEA-logger, let's select Kernel-Power as the event source.
That way we can sort them by date so they are easier to look at
These are the dates we have dump files from
Then give me the BugCheckCode (If they have one) of the kernel-power events that don't have a dump file associated with them.
Did you clean install or clone the previous install btw?
Clean
@umbral ore Get those other crash errors if you can. If they aren't storage related we can go through just general memory error things.
This just feels so much like a storage issue, but that you already replaced it is a big point against it. The M.2 slot can go bad, but it's so uncommon.
It sure does feel like it
I have tried EVERYTHING and i really mean it
The only thing i cant try is replacing parts with different ones
But i have used memtest86, tried both rams alone and together, updated bios, played a bit with the ram voltage, looked into the ssd settings and even uninstalling and reinstalling the gpu and ssd drivers
Reseated the cpu, tried different ram slots
My latest update on it is that afther four hours of heavy use it has crashed with only 1 ram stick
Rigth now im running occt test and afther that im trying a different nvme slot
I have realized that my ssd(gen4) is on the gen5 slot, i know its retro compatible but its one of the few things i havent tried yet
It passed the occt test and it diditn matter when i changed the ssd to a different slot
@wary ibex what can i do now?
@umbral ore I asked for those other codes that didn't create dump files. If you want though, I am a part of a different tech support community and in that community we made a tool that gathers system info and a bunch of logs from Windows. I completely understand not wanting to run random software though so this is 100% voluntary. It just automates many of things we can check.
The community is the reddit.com/r/techsupport community, but it's more used for the Discord live chat we have for it. The tool is on github so anyone can check it out.
With that tool, I can just check all this myself if you found it confusing
Specify: a lightweight and fast tool to gather information about a Windows computer, designed for support purposes - Spec-ify/specify
You want Specify_noring0.exe
Click on Settings, then remove usernames
Before starting the collection.
No directly storage related BSODs. And it looks like you have shutdowns as well so this is a mixed issue.
It shouldn't be CPU/PCIe because if those are causing the shutdowns, it would usually be the CPU ordering the shutdown. And if it does that it would generate a WHEA report on the next startup (Windows Hardware Error Architecture). And you don't have any WHEA reports.
No freezes happening?
And I'll check the two new dump files since the last time you provided dump files
it freezes then crashes
One of the new crashes is this
This is again a memory error, but it's very often caused by errors to the page file and not RAM.
if im running a game it will close it, wont let me open it again then afther a minute or two to where my pc freezes and resets
And when looking at the log of commands sent for the crash (What the PC was doing right when it crashed) it shows this
ST_STORE is storage commands.
SMKM_STORE as well
So this was likely the page file
The other of the two new crashes was a memory error, this time with NTFS.sys involved which is a storage driver
So it just continues looking like a storage issue
what can i do now then?
The unknown for me is the times it doesn't BSOD.
If that's part of the same issue or if it's a separate issue
If it's part of the same issue (Unless it's freezes, those can be from storage), storage is probably not the root cause.
Storage doesn't cause shutdown unless you see WHEA events.
But can cause freezes so it's hard to tell just from the data I have here
And because you already replaced the storage, the second question is if you have two faulty drives in a row or if it's the motherboard.
You already tried different slots, but it could be a larger part that affects both M.2 slots having an issue
There isn't really any way to test this without just replacing one of them with a known good part and seeing if it still crashes.
@umbral ore TL;DR: Storage or motherboard.
im screwed then
That's my analysis
i mean
Do you have an old SATA SSD you could throw in there? Or a drive you could steal form an old laptop
it used to be stable and not crash with only 1 ram stick before i updated the bios recently
nope
but i dont think its gonna fix it since changing into a brand new ssd didint solve it
It could be RAM because the page file is so intertwined with memory/RAM, but RAM rarely causes freezes and shutdowns.
And these dump files are often pointing to storage drivers/page file. It just looks more like storage than RAM.
btw im checking my ssd temps and im seeing that one of the temps is 72 celsius in idle, is it normal?
No
Screenshot
70°C is the max temp for SSDs
And NVMe gets a lot hotter during load so 70 already at idle is not good
@umbral ore Is the CPU water cooled? Water cooling builds usually have less fans blowing over the board so you can see M.2 SSDs overheat on those builds.
nope its got a generic heatsink
I don't know what each sensor refers to. Try CrystalDiskInfo if that shows 70°C. That tool should pick a relevant sensor
hold on my pc is dying as we speal
It being slow and stuttery and freezing before a crash is also such a classic storage sign.
It just did that
This is why it's really annoying if it's not storage, because everything looks like storage
If it turns out to be storage what could i do?
Do i get a new ssd?
Note sure
I would return the current one using the warranty.
But first maybe see if you can move a fan to blow more air over the SSD
Or if you have a heat shield you can install on it
thats the nand controller temp... the 71C
and its normal...for the controller to be that hot...
the drives temp appears to be normal...
It has his own small heatsink that came with the board
What could it be then?
the drives can fail in other ways as well...it doesnt have to be temps..
I assume you reseated it and used different slots...
so what you can do...is check for SSD firmware updates..
download kingston SSD manager app...and check for ssd firmware update...
Yeah i did change slots
Alr
@robust cove no firmware updates available
then I guess...rma it..if its under warranty...
you dont have a spare drive to test?...
Nope no spare drives to test but the issue was present in an older ssd too
This one i got new because of this issue
same issue?... was the older drive...a different model or the same one?..
It was a different model and brand
A samsung 250gb ssd nvme
sata or nvme?..
how often are the crashes?....
Kinda random
But in between 2-4 hours of use
crashes during gaming...or randomly?..
Randomly
It will crash either gaming or idle
It starts acting strange and then freezes and crashes
did you clean install windows...when you switched SSDs?..or cloned it from the old SSD?...
Clean install
I havent checked the dumps... but according to bjoolz..its storage ...if NTFS module is all over the stack....
not sure how two different SSD models end up having the same issue..
this could mean...
-
either you repeated the same activity you did on the previous windows install...hence its the same error...uninstall wallpaper engine...
-
its either coming from the CPU or the motherboard... in which case a firmware update..can be done...if available..
try using a SATA ssd if above doesnt work...
I get the cpu and motherboard but what does wallpaper engine have anything to do with this?
@robust cove i did just update the bios to the latest version the other day
you have GPU driver crash ...reported on one of your dumps...
while it looks like...storage...but your case with the old SSD having the same issue...changed the course...
I already uninstalled my graphics drivers with ddu and reinstalled them but its still the same
uninstall wallpaper engine..
also...since you did bios updates and such... either you can clean install windows...and dont install apps like wallpaper engine or use any debloating...and continue testing...
or take it to a repair shop...
or if its an issue with...nvme protocol...then you can also try SATA ssds..
What a bummer
I guess i got no more options then
I just gotta find my way to get my pc into a repair shop without a car
Great it now crashes every hour
To be clear, it doesn't directly point to storage, but many of the crashes have storage drivers involved and one points directly to the page file.