[SOLVED] Random restarts since September -- I've tried a lot of fixes, details inside

balisong

Distinguished
Aug 26, 2012
21
0
18,520
Hi, all. Been a long time since I stopped by these forums, but I've always known you good folks to be a collective godsend when the going gets tough (in the realm of PCs, anyway).

So, I'm sure you're all busy, and I do not wish to waste your time, so enough prelude.

Update: My issue seems similar to this person's, except I do get the occasional freeze: https://forums.tomshardware.com/threads/getting-a-lot-of-random-restarts-with-my-new-build.1381506/

MY BUILD:

GPU:
EVGA GTX 1080 Ti Black Edition
CPU: i7 8700K
Memory: 16GB Corsair Vengeance LPX DDR4 (2666MHz, I believe)
MOBO: ASUS Z370 Plus TUF Gaming
Storage: 2X SSD, one 970 EVO Pro PCIe NVMe, the other an older SATA SSD. Also an EVO, I believe. Both 500GB.

PROBLEM: My system has been randomly restarting and/or freezing since the 25th of September. I know it was this particular day because that's when the first Kernel-Power (event ID 41) error shows up (in EventVwr).

These freezes/crashes do NOT generate crash dumps, and they are NOT the result of a BSOD. I've tried running WhoCrashed to no avail as well.

These crashes primarily seem to happen while gaming, though on one occasion it occurred for no discernible reason shortly after booting and launching my web browser, and just today, it happened the second I booted a game (but it usually takes a couple hours or much more).

So, here's what I've ruled out:

-Overheating. Neither my GPU or CPU are overheating, even when I've run a very intense game for hours. The GPU rarely goes above 70 C, and CPU never goes above 60. I've tried stress testing my CPU using Prime95, and my GPU usine Unigine Heaven and Furmark.

-Memory. I've run both Win. Memory Diag. and Memtest86 (3 passes -- I know I could do more, but I consider this sufficient for now. In my experience, if a RAM stick has significant stability problems, you are going to get flooded with errors).

-Power Supply(*). Based on my initial research, a failing or insufficient power supply can be the cause of these random restarts; particularly if they occur during gaming. So, I replaced my old EVGA 650W (Silver-rated) with a new EVGA 750W (Bronze-rated) one. The issue remains. It IS possible I got two faulty PSUs in a row, but given that I've had this system put together for about a year and a half with no issues prior to my new PSU purchase, it's unlikely.

-Storage. I've run every SSD/HDD health checking utility I can find, and nothing has turned up any errors. I'm also not seeing any other signs of a failing drive, such as errors when transferring files, unresponsiveness, etc. My system is just as speedy and efficient as ever.

-Drivers. I've clean uninstalled my GPU drivers using DDU (run properly -- in safe mode, as administrator, etc) and reinstalled them three times.

-Viruses. I've run rootkit scans and malware scans galore. Nothing found.

-Overclock. The issue occurs even without my system being overclocked, and I only recently started overclocking my CPU. I've never overclocked my GPU.

-Improper seating. I've re-seated all the hardware I can, except my CPU, and that's only because I do not have any thermal paste or rubbing alcohol handy.

-Bad wall socket. I've tried plugging my system into a surge protector, directly into the wall, and into different power sockets in my house. No dice.

There have been no major or minor hardware changes before this error occurred.

Any help would be appreciated.
 
Last edited:
Solution
OK, so, I think I've solved my problem.

Around the time my random restarts began, I'd moved my PC to another part of my room, and plugged it into different wall sockets. When the problem started, I didn't pay much mind to this possibility at first. Eventually I tried plugging it into different sockets on the same outlet to no avail, so I gave up on that avenue.

Now, I've temporarily put my PC (not my desk, yet -- just sitting on top of a dresser) back where it was before, plugging it into different sockets.

NO RANDOM CRASHES. Been somewhat uncomfortably gaming for around an hour now and no issues.

I'm starting to think that, with the addition of my new 2080 Ti, that wall socket (the problematic one) simply couldn't handle the...

PC Tailor

Illustrious
Ambassador
-Memory. I've run both Win. Memory Diag. and Memtest86 (3 passes -- I know I could do more, but I consider this sufficient for now.
I would run for at least 4 passes, I have had multiple instances where instability in RAM has only showed single digit errors that can appear right at the very end of the last pass.

-Power Supply(*). Based on my initial research, a failing or insufficient power supply can be the cause of these random restarts; particularly if they occur during gaming. So, I replaced my old EVGA 650W (Silver-rated) with a new EVGA 750W (Bronze-rated) one.
What exact models are they? The bronze is only an efficiency rating, not necessarily representative of quality.

-Storage. I've run every SSD/HDD health checking utility I can find, and nothing has turned up any errors.
Have you run SeaTools as per these instructions: https://www.sysnative.com/forums/threads/hard-drive-hdd-diagnostics-sea-tools-for-dos-ssd-test.4072/

-Drivers. I've clean uninstalled my GPU drivers using DDU
Have you updated all other drivers? (Not using a driver updater app)

-Overclock. The issue occurs even without my system being overclocked,
Does this include not having XMP/DOCP enabled?

-Improper seating. I've re-seated all the hardware I can, except my CPU
Have you checked the CPU/Socket for bent pins or contamination at all?
 
  • Like
Reactions: balisong

balisong

Distinguished
Aug 26, 2012
21
0
18,520
I would run for at least 4 passes, I have had multiple instances where instability in RAM has only showed single digit errors that can appear right at the very end of the last pass.


What exact models are they? The bronze is only an efficiency rating, not necessarily representative of quality.


Have you run SeaTools as per these instructions: https://www.sysnative.com/forums/threads/hard-drive-hdd-diagnostics-sea-tools-for-dos-ssd-test.4072/


Have you updated all other drivers? (Not using a driver updater app)


Does this include not having XMP/DOCP enabled?


Have you checked the CPU/Socket for bent pins or contamination at all?

Thanks for the suggestions.

I have indeed run SeaTools, but not using those instructions. I believe I ran the .exe version, so I will try that method next.

I don't believe I have XMP enabled, or at least I did not manually enable it, but I will check to be certain.

I have not checked my CPU socket for bent pins or contamination. I don't have any thermal paste handy, so I'm trying to rule out the other possibilities before popping it out and giving it a look.

And I will try running MemTest overnight this time, perhaps for 6-8 passes, to see if any errors are detected.

Forgive me for the lack of clarity regarding my PSU models. Frankly, it's the one part of my system I'm the least familiar with (everything is plugged in properly, though).

The first PSU I purchased was the EVGA 650 N1, the second (and current) is the EVGA 750 BQ.

I'll give your suggestions (barring the CPU thing, until I can get more thermal paste) a shot and report back.

Regarding drivers, what other drivers might I want to update, and how would I go about doing it?
 

PC Tailor

Illustrious
Ambassador
I have not checked my CPU socket for bent pins or contamination. I don't have any thermal paste handy, so I'm trying to rule out the other possibilities before popping it out and giving it a look.
Understood, a single slightly bent pin can cause all manner of mayhem so it's worth a look when you can.

Forgive me for the lack of clarity regarding my PSU models. Frankly, it's the one part of my system I'm the least familiar with (everything is plugged in properly, though).

The first PSU I purchased was the EVGA 650 N1, the second (and current) is the EVGA 750 BQ.
Nothing to apologise for my friend :)
So overall the BQ line aren't fantastic, they're a budget line and have very mixed outcomes with voltage regulation for example. So it can't be ruled out. Typically with random restarts the list of culprits tend to go (once you've eliminated software/overheating):
PSU > GPU > RAM > MB > CPU.

Can you also confirm if the random restarts occur even if you're in the BIOS or in Safe mode?
Do youalso have latest BIOS installed? If not i would do so, however if there is a possiblity of a shutdown occuring during the update, then I would avoid it.

Regarding drivers, what other drivers might I want to update, and how would I go about doing it?
Mostly motherboard/chipset drivers, which you will find for yours here: https://www.asus.com/Motherboards/TUF-Z370-PLUS-GAMING/HelpDesk_Download/
Just don't use driver updater apps :)
 

balisong

Distinguished
Aug 26, 2012
21
0
18,520
Understood, a single slightly bent pin can cause all manner of mayhem so it's worth a look when you can.


Nothing to apologise for my friend :)
So overall the BQ line aren't fantastic, they're a budget line and have very mixed outcomes with voltage regulation for example. So it can't be ruled out. Typically with random restarts the list of culprits tend to go (once you've eliminated software/overheating):
PSU > GPU > RAM > MB > CPU.

Can you also confirm if the random restarts occur even if you're in the BIOS or in Safe mode?
Do youalso have latest BIOS installed? If not i would do so, however if there is a possiblity of a shutdown occuring during the update, then I would avoid it.


Mostly motherboard/chipset drivers, which you will find for yours here: https://www.asus.com/Motherboards/TUF-Z370-PLUS-GAMING/HelpDesk_Download/
Just don't use driver updater apps :)

Hi there,

Wanted to thank you for all your help so far.

I currently have thermal paste on order (should arrive before the end of the weekend), and I'm planning to purchase a Uninterruptible Power Supply (this one: https://www.amazon.com/CyberPower-CP1500PFCLCD-Sinewave-Outlets-Mini-Tower/dp/B00429N19M/?th=1), which should also arrive before the end of the weekend.

I recently upgraded to a 2080 Ti, largely because I wanted to, but also partially because I was very curious to see whether or not a new GPU would solve or otherwise affect my problem.

Indeed, it seems to have made it worse. I now random restart even MORE than before, after just ~10 minutes of gaming.

This leads me to believe it is indeed something related to my power supply, or wall power in general.

I'm now starting to wonder if perhaps my problem is related to the wall outlet I'm using. I have tried other outlets, but I mostly noticed this problem after I moved my desk to the other side of the room. I have not tried my old outlets again for a while.

Do you have any thoughts? The fact that purchasing a more power hungry card instantly leads to quicker restarts has me wondering.
 

balisong

Distinguished
Aug 26, 2012
21
0
18,520
OK, so, I think I've solved my problem.

Around the time my random restarts began, I'd moved my PC to another part of my room, and plugged it into different wall sockets. When the problem started, I didn't pay much mind to this possibility at first. Eventually I tried plugging it into different sockets on the same outlet to no avail, so I gave up on that avenue.

Now, I've temporarily put my PC (not my desk, yet -- just sitting on top of a dresser) back where it was before, plugging it into different sockets.

NO RANDOM CRASHES. Been somewhat uncomfortably gaming for around an hour now and no issues.

I'm starting to think that, with the addition of my new 2080 Ti, that wall socket (the problematic one) simply couldn't handle the demands of my system for whatever reason. I have no idea why, but the fact that the frequency of the crashes escalated as soon as I installed the new GPU effectively pushed me to test this theory.

So, for now, I'm tentatively considering this resolved.

To any random bystanders who, like me, have been scouring the internet for fixes to similar issues for nearly 2 months with no results, ask yourself whether a bad wall socket could be the culprit. If so, try moving to somewhere else in your house.

New question for anyone willing to answer: would a appropriate-wattage UPS help me continue using the bad socket, or should I just buy a particularly long surge protector and drape my cords across the room.
 
Last edited:
Solution