Question Random BSODs caused by ntoskrnl.exe or Ntfs.sys ?

[cocosoy29]

Hi guys, I've been experiencing random BSODs recently (but I haven't changed any hardware in a while). I have 5 mini dump files, 3 of them saying it's caused by ntoskrnl.exe and the other two by Ntfs.sys.

My PC had a total breakdown last month and I had to do a clean installation of windows. Now the BSOD is back again 🙁 What bothers me more is when I restart or turn on the PC, it will randomly decide to do disk scan and repair (indicating there is an issue with my hard drive?)

Also, I feel it's more prone to BSOD when idling. I haven't encountered any BSOD during gaming sessions. (This seems like a common occurrence in other posts I've read so far)

I tried to do sfc scans, and also Dism restore (However, the DISM restore always stuck at 62.3% even reading from a local iso).

I have attached a screenshot of my PC spec, and all 5 mini dump files and a BlueScreenView summary screenshot. (also added an energy report)

Should I be starting with doing memtest to check if RAM is at fault?

Thanks in advance!

Minidump Files: https://drive.google.com/drive/folders/1gf3et43slc7f5b_fMTnEN5Pz1r245bSH?usp=sharing

p.s. Some additional info that might be helpful, I'm new to this so let me know if any other information should be provided.

I have a MSI 1000w PSU (new with this build). The PC has been heavily used for about 1.5 years since it's built. (mostly heavy gaming and also a lot of AI training/stable diffusion/LLM ect.).

Harddrives:

- 4 internal SSD (all different brands: Samsung 980Pro for main drive, Sk hynix P41, Crucial P5 plus and Kingston NV2)
- 1 external SSD plugged in 24/7
- 1 external HHD plugged in 24/7 (this one has its own power supply, a 16TB backup device)

RAM:
2 sets of (64GB total):
G.Skill Trident Z5 RGB Series (Intel XMP) 32GB (2 x 16GB) 288-Pin SDRAM DDR5 6400 CL32-39-39-102 1.40V Dual Channel Desktop Memory F5-6400J3239G16GA2

I had trouble running those RAM stable at 6400, I'm running them at 6000 now. They did pass the memtest when I just build this machine.

 

[ubuysa]

A full set of specs would be useful please.

From what you're saying the system was having BSODs, then you had some sort of 'total breakdown' (can you elaborate on what that was?), then you clean installed Windows and now you're still getting BSODs?

You also very clearly have an issue with a storage drive...

What bothers me more is when I restart or turn on the PC, it will randomly decide to do disk scan and repair (indicating there is an issue with my hard drive?)

You need to determine exactly which drive this is, because if it's your system drive then that may well be the root cause of your problems. Do you know which drive it's trying to check on each reboot?

We need to clear up this storage drive issue first, before we move on to look at anything else.

 

[cocosoy29]

A full set of specs would be useful please.

From what you're saying the system was having BSODs, then you had some sort of 'total breakdown' (can you elaborate on what that was?), then you clean installed Windows and now you're still getting BSODs?

You also very clearly have an issue with a storage drive...

You need to determine exactly which drive this is, because if it's your system drive then that may well be the root cause of your problems. Do you know which drive it's trying to check on each reboot?

We need to clear up this storage drive issue first, before we move on to look at anything else.

So the total breakdown was I couldn't boot into Windows at all, the PC just kept going into the blue screen where it give you the option to restore/troubleshoot/etc. I could get into BIOS, but never Windows. I tried to boot into safe mode and that didn't work too. So I was like screw it, I formatted my system drive and did a clean install. (The clean install wasn't smooth either, I had to do it like 3 times before it finally succeeded. You know when you do a fresh install, the PC would automatically restart at some point. But it wouldn't resume the installation process after restarting. Only worked on my 3rd try). Also, when the first two times the clean install failed. I opened up the PC and cleaned all the dust, re-inserted RAMs and GPU (not sure if that made a difference). It has been running fine for about 2-3 months since that breakdown.

For the drive scan/repair, it always checks the C: drive and maybe 1 other drives (Can't remember exactly which one, but C: gets checked every time). But the scan/repair doesn't happen every single time, usually only happens after i got a BSOD.

I've tried chkdsk commands on the drives, but that didn't find any issue. I've also tried CrstalDiskinfo, but that didn't provide anything useful too (just says Good 100%). What's the best way to check whether my SSD is at fault? Let me know if there is any report I can run and provide to you guys. Thank you!!

Here is a screenshot of chkdsk result: https://drive.google.com/file/d/1f98rjSj3LPs5lpgE5kk0lOKoOcfqEzEs/view?usp=drive_link

Here is my full set of specs:

CPU: 13900K
MOBO: ROG STRIX Z790F LGA1700
GPU: MSI 4090
RAM: GSkillz Trident Z5 (16GB x 4) 6400MT/s CL32-39-39-102 1.40V
Cooler: Lian Li Galahad 240
PSU: MSI MPG A1000G PCIE 5.0 Gold (1000W)

Internal Storage:
1) SK Hynix Platinum P41 2TB M2 (system drive)
2) Kingston NV2 1TB M2
3) Crucial P5 P{lus 2TB Gen 4 M2

External Storage (plugged in 24/7 for backups):
1) Samsung SSD t7 Protable 1TB
2) WD160EMFZ (Western Digital Element) 16TB backup drive (this one has its own power supply)

Let me know if I missed anything.

Thank you!

 

[ubuysa]

The struggles to which you refer when trying to install Windows are almost certainly being caused by a hardware issue. It's most likely that it's the same hardware issue that is causing you problems now.

I still think that a storage drive may well be the root cause here. I know from experience that M.2 drives can be temperamental if they're not seated properly. As a first off then, remove all your M.2 drives, ensure that the M.2 sockets are clean of dust with a blower (don't blow on them with your mouth!) and that the M.2 drive pins are also clean. Then reinsert each drive firmly, ensuring that they are fully home.

If that doesn't stop the BSODs, and since you think they may happen more when idling, then in the BIOS disable C-States if you can (some BIOSes don't support this. This will stop the CPU processors entering low power states. I've seen some issues caused by a CPU that is a bit tardy coming out of low power stets - though mostly these are AMD CPUs.

If you can't disable C-States in the BIOS then go into your power profile in Windows and set the Processor Power State values so that the minimum processor state is 99% and the maximum processor state is 99%. This also stops the processors entering low power states.

If none of that helps then disconnect the two external drives and see whether they were causing the BSODs.

If you're still getting BSODs then I would suggest removing all drives except the system drive (the Hynix) and clean install Windows onto there (deleting existing partitions in the install). This will ensure that the installer doesn't put the boot data on a different drive - as it is wont to do. Run Windows Update repeatedly (and across reboots) until no more updates are found. Then check in Device Manager for any devices with a yellow triangle containing a black exclamation mark. If you find any then check first in the 'view optional updates' link of Windows Updates, expand the Driver Updates in there are see whether there is a driver for these devices in there. If not then visit the website for your motherboard and look for drivers in there.

Now test the PC as much as you can and for as long as you can in this stripped-down and pristine state. If it still BSODs then remove all but one RAM card and test on that one card. Swap that one card for one of the others is you still get BSODs.

If at this point you still get BSODs then physically remove everything you can do without; the GPU, any other PCIe cards, anything else that is USB connected so that you really are running the absolute bare minimum of hardware.

 

[cocosoy29]

The struggles to which you refer when trying to install Windows are almost certainly being caused by a hardware issue. It's most likely that it's the same hardware issue that is causing you problems now.

I still think that a storage drive may well be the root cause here. I know from experience that M.2 drives can be temperamental if they're not seated properly. As a first off then, remove all your M.2 drives, ensure that the M.2 sockets are clean of dust with a blower (don't blow on them with your mouth!) and that the M.2 drive pins are also clean. Then reinsert each drive firmly, ensuring that they are fully home.

If that doesn't stop the BSODs, and since you think they may happen more when idling, then in the BIOS disable C-States if you can (some BIOSes don't support this. This will stop the CPU processors entering low power states. I've seen some issues caused by a CPU that is a bit tardy coming out of low power stets - though mostly these are AMD CPUs.

If you can't disable C-States in the BIOS then go into your power profile in Windows and set the Processor Power State values so that the minimum processor state is 99% and the maximum processor state is 99%. This also stops the processors entering low power states.

If none of that helps then disconnect the two external drives and see whether they were causing the BSODs.

If you're still getting BSODs then I would suggest removing all drives except the system drive (the Hynix) and clean install Windows onto there (deleting existing partitions in the install). This will ensure that the installer doesn't put the boot data on a different drive - as it is wont to do. Run Windows Update repeatedly (and across reboots) until no more updates are found. Then check in Device Manager for any devices with a yellow triangle containing a black exclamation mark. If you find any then check first in the 'view optional updates' link of Windows Updates, expand the Driver Updates in there are see whether there is a driver for these devices in there. If not then visit the website for your motherboard and look for drivers in there.

Now test the PC as much as you can and for as long as you can in this stripped-down and pristine state. If it still BSODs then remove all but one RAM card and test on that one card. Swap that one card for one of the others is you still get BSODs.

If at this point you still get BSODs then physically remove everything you can do without; the GPU, any other PCIe cards, anything else that is USB connected so that you really are running the absolute bare minimum of hardware.

Thank you for the detailed action plan! Hopefully I can find some free time this weekend and start with reconnecting all the M.2 drives,and proceed with all your recommendations!

My M.2 drives are a bit harder to remove because I used the heat sink that came with the MOBO and sticked the drive to the heat sink. So I need to carefully peel off the sticker before I can remove them. Well, it is what it is. Just need to find the time to do that.

Thank you for the help again!!