Windows 10 One of my previous builds is experiencing random, multiple BSODs

Captain Jack Sparrow

Anti-cryptominer
Joined
Jul 1, 2007
Messages
561
Reaction score
118
This has been a problem since upgrading to Windows 10. But it was never enough of a problem until recently, because this seems to be getting worse.

An older iteration of "The Black Pearl" which is still used as a secondary PC has a problem with random BSODs, usually during boot. It rarely happens once successfully booted, and never during gaming strangely. Firefox randomly shows the "tab crashed" page quite often, but I'm not sure if this is related.
The worst part is that there's no obvious cause, so I can't reproduce this on demand. It might boot without issues for days, then one day, it will take 2-4 attempts to boot without crashing to a BSOD.

Hardware:

Motherboard: MSI 870-C45​
CPU: AMD Phenom II X6 1100T (3.3 GHz)​
RAM: 16 GB DDR3 1333 MHz (4 x 4 GB)​
GPU: AMD Radeon RX 590 (8 GB GDDR5)​
SSD: 1 x SATA SSD (240 GB)​
HDD: 1 x SATA HDD (2 TB)​
Optical: 1 x IDE DVD RW drive​
Floppy: 1 x FDD drive​
PSU: Seasonic Industrial SSP-600ET2 (600W)​
OS: Windows 10 Education 21H2 x64​

List of BSOD codes (so far):

  • KMODE_EXCEPTION_NOT_HANDLED
  • MEMORY_MANAGEMENT
  • PAGE_FAULT_IN_NONPAGED_AREA
  • SYSTEM_THREAD_EXCEPTION_NOT_HANDLED
    If any other codes occur which are not listed, I'll add them here.

What I've tried:

  • PSU replacement
  • CPU replacement
  • GPU replacement
  • BIOS updated and settings reset
  • All drivers updated
  • RAM modules removed and re-installed
  • Windows 10 clean install
I'm running out of options here. I fear this might be motherboard related. If that's the case, it's not going to be worth replacing this motherboard. It will be very difficult to find AM3 motherboards now, even more so at a reasonable price, so I guess it may be better to ignore this?
 
Last edited:
I don't have a 'solution' for you on #1

#2. Run "windows memory diagnostic" tool


... then we'll look at the others. (or replace ram)
Hi muckshifter,

Windows Memory Diagnostic completed with no problems.
Also forgot to mention, I had previously ran the MemTest86 7.1 CD for over 48 hours, but no problems were found.

This is why I am reluctant to purchase more RAM. However, I did find something unexpected with the RAM modules.
Slots 1 and 2 are using dual rank RAM, but slots 3 and 4 are using single rank RAM, with slightly different timings.

Slots 1 and 2.png Slots 3 and 4.png

Despite having the same part numbers, these RAM modules are not the same!
Do you this this could cause issues? I'd have thought that if so, surely it would cause problems in Windows Memory Diagnostic, or MemTest86?

And it’s still really bizarre that it never crashes during gaming. Recently I’ve been playing GTA V/Online frequently. This game is quite heavy on RAM utilization, due to its open world nature. The game engine has to swap parts of the map in and out of RAM in real-time. If there were problems with RAM, I’d expect the game to crash.
 

Attachments

  • Slots 1 and 2.webp
    Slots 1 and 2.webp
    25.4 KB · Views: 36
  • Slots 3 and 4.webp
    Slots 3 and 4.webp
    25.7 KB · Views: 37
Last edited:
The reason I asked you to test your RAM is because of the error messages. I don't think that 'mismatch' would cause your errors.

I'll dig deeper.
 
This has been a problem since upgrading to Windows 10. But it was never enough of a problem until recently, because this seems to be getting worse.

An older iteration of "The Black Pearl" which is still used as a secondary PC has a problem with random BSODs, usually during boot. It rarely happens once successfully booted, and never during gaming strangely. Firefox randomly shows the "tab crashed" page quite often, but I'm not sure if this is related.
The worst part is that there's no obvious cause, so I can't reproduce this on demand. It might boot without issues for days, then one day, it will take 2-4 attempts to boot without crashing to a BSOD.

Hardware:

Motherboard: MSI 870-C45​
CPU: AMD Phenom II X6 1100T (3.3 GHz)​
RAM: 16 GB DDR3 1333 MHz (4 x 4 GB)​
GPU: AMD Radeon RX 590 (8 GB GDDR5)​
SSD: 1 x SATA SSD (240 GB)​
HDD: 1 x SATA HDD (2 TB)​
Optical: 1 x IDE DVD RW drive​
Floppy: 1 x FDD drive​
PSU: Seasonic Industrial SSP-600ET2 (600W)​
OS: Windows 10 Education 21H2 x64​

List of BSOD codes (so far):

  • KMODE_EXCEPTION_NOT_HANDLED
  • MEMORY_MANAGEMENT
  • PAGE_FAULT_IN_NONPAGED_AREA
  • SYSTEM_THREAD_EXCEPTION_NOT_HANDLED
    If any other codes occur which are not listed, I'll add them here.

What I've tried:

  • PSU replacement
  • CPU replacement
  • GPU replacement
  • BIOS updated and settings reset
  • All drivers updated
  • RAM modules removed and re-installed
  • Windows 10 clean install
I'm running out of options here. I fear this might be motherboard related. If that's the case, it's not going to be worth replacing this motherboard. It will be very difficult to find AM3 motherboards now, even more so at a reasonable price, so I guess it may be better to ignore this?
I would start checking everything step by step, starting with the motherboard and continuing to check the RAM! You can take it from a friend for a test and change it and you will find the cause of the malfunction!
 
I would start checking everything step by step, starting with the motherboard and continuing to check the RAM! You can take it from a friend for a test and change it and you will find the cause of the malfunction!
Thanks, but it would help if I actually had friends who use PCs! Even if they did, they're much richer than me, with PS5 and Xbox Series X consoles, so I don't think they would still have such an old platform in 2023.

What we know so far is that the CPU, GPU and PSU are confirmed to be good. So that leaves the motherboard, RAM and storage.
I checked the BSOD crash dumps in more detail, but unfortunately they all referenced ntoskrnl.exe. I did somewhat expect this, but it's not really helpful for troubleshooting.

In Event Viewer, this is where things get interesting. There were a large amount of errors listed with Event ID 1005.
Windows cannot access the file for one of the following reasons: there is a problem with the network connection, the disk that the file is stored on, or the storage drivers installed on this computer; or the disk is missing. Windows closed the program [program] because of this error.
This is even more weird, because all of these programs are installed on the OS SSD. There isn't really any pattern I can see, but these are the affected programs:
  • Adobe Genuine Software Integrity Service
  • AMD Software: Host Application
  • Firefox
  • Malwarebytes Anti-Ransomware
  • Microsoft Software Protection Platform Service
  • OriginWebHelperService
  • Trellix Endpoint Security
So I think that there might be a problem with the SSD intermittantly becoming inaccessible. If it's inaccessible while Windows tries to access the page file, it might be plausible that this could cause BSOD crashes?

I'm going to run a full CHKDSK /R on the SSD to see what happens.
 
Last edited:
So I ran CHKDSK which didn't find any problems. However, I did find that the SATA connector on the SSD enclosure was damaged, so I replaced the drive enclosure. But it didn't solve the problem, got a BSOD immediately after.
... then we'll look at the others. (or replace ram)
As a last ditch attempt, I paid £28 for 4 x 4GB Hynix HMT351U6BFR8C-H9 DDR3 1333 MHz modules from eBay. This time, all modules are exactly the same, and the part number is on the motherboard's compatibility list.

It's beginning to look like @muckshifter was right about replacing the RAM. Already the PC seems more stable, and those weird Application Error messages in Event Viewer are gone. Still, it's eBay though, so for peace of mind, I'll run MemTest86 for 48 hour followed by Windows Memory Diagnostic.
Despite having the same part numbers, these RAM modules are not the same!
Do you this this could cause issues? I'd have thought that if so, surely it would cause problems in Windows Memory Diagnostic, or MemTest86?
Feel like I might have been right about this. I really hope replacing the RAM with identical modules solves this problem, which has plagued this PC for over 3 years. Over the next couple of weeks, I'll check the overall stability. If all goes well, hopefully I can finally call this solved.
 
Last edited:
Well, it has been 2 weeks with no more BSODs or application crashes. I think we can confirm that RAM was the problem!

Sad to say, but I'll be adding Corsair products to my "do-not-buy" list. Purely because selling completely different and incompatible RAM modules with the same part number is a d*** move! They shouldn't be doing that, and there's no excuse.
 
What are the common troubleshooting steps to identify and address random BSODs, and how can one effectively test and replace faulty RAM modules in a computer system?
 
What are the common troubleshooting steps to identify and address random BSODs
If there is an error code or message, start researching it.
Disconnect all peripherals and reconnect them one at a time.
Update Windows.
Update drivers and software.

how can one effectively test and replace faulty RAM modules in a computer system?

 
Back
Top