Hard crashes and lock-ups under under Linux and possibly Win10 - HW failure?

[pogofcz]

Hello, for the past year or so I'm having these mystery crashes in random intervals. Sometimes it can be 5-8 a day, sometimes it takes a week between them. So far I have been unable to pinpoint what is the cause. I am dual-booting Pop_OS and Windows 10, doing the same thing on my Lenovo Legion, using the same programs on both + the crashes were present even shortly after I wiped the drive with fresh install, my laptop is however 100% fine.

 

It doesn't matter what Im doing at the time. I can be editing a document in writer, watch youtube video or play runescape. The crashes and lockups are not exactly the same every time but very similar.

Few examples:

• PC would just reboot on its own
• PC would freeze and last 0.5-1s audio would repeat then it would reboot on its own in 5-10s
• Same thing but after 30s I would just reboot it myself
• My desktop would freeze but I can still move my mouse for 1-3s then it would lock up completely and either reboot or I would have to reboot it.

 

As I dont use Windows 10 as much and for as long I dont have many instances of BSOD but two that I can remember are

• KMODE_EXCEPTION_NOT_HANDLED
• HYPERVISOR_ERROR

In both instances would my other two screens turn corrupted (deep-fry meme style) and constant sound/deep buzz(?) would sound from my speakers.


What I need now is some kind of solution/direction. Im starting to think its HW related but I dont want to just blindly buy and swap parts I would rather know with some level of certainty what is the cause. I am unable to reproduce the error by using any sort of stress tests.

 

MemTest64 - clean after several rounds of testing

stressapptest - clean after two times running it
SSD swapped places - no change

Parts list:

Ryzen 3700X + included cooler
EVGA GeForce 2060SUPER

MSI x570 gaming edge wifi - latest BIOS (updated by me few days ago)

Hyperx 16GB 3200MHz
Crucial P2 M.2 1000GB - Linux drive
Intel 665p 1000GB - Windows drive
WD 2TB HDD

WD 8TB HDD
Corsair HX750 750W PSU

Akasa USB 3.0 PCIe card

 

 

What was present in the PC in the past:

Corsair 32GB 3200MHz - Faulty; Memtest64 broken sectors

AverMedia Live Gamer HD - Was not using it
EVGA 750BQ 750W PSU - Swapped as I thought power issues were the cause
Aliexpress RGB strip thing - Unplugged during troubleshooting; had no issues with it
DVD drive TS-H653- Unplugged during troubleshooting
Fractal design 120mm RGB fans - RGB unplugged, but still blowing  

 
Peripherals:

Corsair K55 keyboard

Logitech M650 mouse
Two Asus vx279h-w - 1 HDMI, 1 DVI
behringer xenyx 302usb USB soundboard
Modecom speakers - 3.5mm jack
Valve index

HP laserjet 1020
Samsung S27E591C - DP [Added after crashes started]
Elgato CamLink [Added After crashes started]
Misc. USB thumb drives, yubikey, SD card reader, controlles...

 

journalctl-1.txt journalctl-2.txt journalctl-3.txt journalctl-5.txt journalctl-6.txt SysnativeFileCollectionApp.zip

[Whatisthis]

You definitely have a hardware fault that’s causing this. I’m not great at reading logs to pinpoint, so hope someone more knowledgeable comes in.

[pogofcz]

On 5/29/2023 at 3:24 PM, Whatisthis said:

You definitely have a hardware fault that’s causing this. I’m not great at reading logs to pinpoint, so hope someone more knowledgeable comes in.

Hopefully someone can point me the right direction

[pogofcz]

Update: I have just bit the bullet and bought Ryzen 5 4600G as a trial replacement. I have 30 days to return it, i used my original cooler so I would just have to clean it well, however I will keep it anyway for my GF's new PC if it doesn't work out. I'll update you in 30 days or as soon as the problems continue.

Long term update: It was definitely the CPU. After almost a month of using the swapped CPU, I have not seen any crashes, lockups or other weirdness, apart from what I would consider to be standard issues (had some issues with Gnome, but that's 100% unrelated).

I would still love to know what exactly was the problem. It seems like it was getting sort of worse over time and im sure it was not present when I bought the parts. All benchmarks, tests and stress tests I have done were not able to reproduce the issue. Maybe one day I'll figure it out.