BSOD pointing to SSD but no drive scan errors

Pointing to SSD (Samsung 970 EVO Plus) related error:

• BSODs are usually random CRITICAL PROCESS DIED but sometimes WHEA UNCORRECTABLE ERROR usually from idle/web browsing. I also got REFERENCE BY POINTER a few times when manually restarting.
• I cannot get a dump (even minidump to write) for CRITICAL/WHEA errors, it gets stuck at 0% on BSOD screen.
• A couple of times before the PC would actually BSOD, the SSD drive in Windows would become inaccessible for a few seconds in This PC (freezes when opening drive/files)
• A few times I would see windows error messages of - application error "The instruction at <CODE> referenced memory at <CODE>. The required data was not placed into memory because of an I/O error status of <CODE>." right before it BSODed.
• A few times (not consistently) I would get "no disk drive found" DOS error on boot (pointing to SSD), on restarting again it would boot fine.
• BSODs are random, generally at idle or in browser. Sometimes multiple times per day, sometimes a couple of times a week.
• Demanding games don't seem trigger it (e.g. Elden Ring), CPU, RAM or GFX card stress tests don't trigger it or report any errors.
• Similar to this thread, after reinstalling Windows it seemed to have almost disappeared but has begun to get worse and worse (BSOD more frequently) - BSOD Critical Process Died constantly -&gt; Reinstall Win10 -&gt; BSOD

However benchmarks / stress test of the SSD turn up no errors either ...

• Hard Drive Sentinel reports PERFECT SMART score and doesn't fail on random seek (it BSODed once but I couldn't replicate it).
• Samsung Magician reports no errors on full scan.

I haven't yet left SSD benchmark run overnight but I could also try this as well.

As I don't have access to spare parts, I'm trying to get some degree of certainty it's the SSD before I replace it. Is there any reason to think it could be the motherboard or another component potentially?

Have a read of this article.

The collected log files reported problems related to drives.

Please perform the following steps in sequence:



1) The CPU supports RAM speeds up to Up to 3200MHz
At the time the V2 log collector ran the RAM speed was clocked at 3600MHz.

Please make sure that there is no overclocking during the troubleshooting.
Turn off XMP.



2) Run:

a) https://www.tenforums.com/attachmen...-files-use-bsod-debugging-tuneup_plus_log.bat

b) https://www.tenforums.com/attachmen...atch-files-use-bsod-debugging-diskparinfo.bat

Post share links into this thread using one drive, drop box, or google drive.




3) Open administrative command prompt and copy and paste:

msdt.exe -id WindowsUpdateDiagnostic
msdt.exe -id PowerDiagnostic
msdt.exe -id DeviceDiagnostic

For each Windows troubleshooter click: view detailed information
Post images or share links into this thread using one drive, drop box, or google drive.



4) Run HD Tune: (free or trial version)(test all drives)
Post images for results on these tabs:
a) Health
b) Benchmark
c) Full error scan



5) Run Sea Tools for Windows (test all drives)
Long generic test



6) Open administrative command prompt and type or copy and paste:

chkdsk /b /v

Use the syntax chkdsk /b /v C: or chkdsk /b /v D: or chkdsk /b /v E: etc.


C:\WINDOWS\system32>chkdsk /b /v
The type of the file system is NTFS.
Cannot lock current drive.

Chkdsk cannot run because the volume is in use by another
process. Would you like to schedule this volume to be
checked the next time the system restarts? (Y/N)

Type: Y
Reboot as needed.

Test all drives.

It can take substantial time to run the chkdsk switches on the windows drive so plan to run overnight while sleeping.



7) These steps / tests can run overnight:
a) HD Tune full error scan
b) Sea Tools for Windows long generic test
c) Chkdsk switches



8) HD Sentinel (free or trial version) (make sure that steps 1 - 5 above are performed before performing these tests)
Post images for results on these tabs
a) Overview
b) Temperature
c) SMART

Okay so HD Tune - benchmark of the NVME drive seems to reliably trigger a BSOD. Interestingly, only at the end of the test though. After having run the benchmark and triggered BSOD, the PC gets more unresponsive (sometimes it BSODs directly at login screen).

Rather than causing BSOD, I once caught it giving the below application error and the NVME becoming inaccessible. This is heavily pointing me to the NVME but I'm not sure if it could also be the motherboard instead or another related component.


However HD Tune - error scan and Seatools and Samsung Magician scan complete successfully with no detected errors and do not cause BSOD (will post later, my PC is currently unusable).

I'm heavily leaning towards buying a new NVME right now and RMAing the existing one.

===============

Some other thoughts:

• NVME temperature looks quite hot - daily maximum is between 60 - 80 C in HD Tune. The reason is I have it in an SFFPC where it's sandwiched against the graphics card. I'm buying a cooling pad and will install it to see if there is any improvement.
• Maybe the BSOD is triggered by something to do with temperature dropping (after HD Tune benchmark)? Or maybe frequently high temperature has the made the NVME generally unstable?
• All drives except are external except for the NVME so I've removed them. Still getting BSODs.

==============

Results to items below - Note, I have a custom Windows 11 install (Phoenix LiteOS 11 Pro+) which might be preventing me from running some diagnostics. The same Windows image is running on a Surface Pro 5 I have without issues so I don't think it's related to this.

1. I've turned off XMP and set RAM down to 3200, still got a BSOD subsequently.
2. Output of DiskParInfo.bat attached. Tuneup_plus_log.bat isn't working for me since sfc /scannow does not seem to generate a CBS.log file (it does say corrupt files are found that cannot be repaired). I've tried multiple online solutions (repointing it to output to different locations) but none of these help.
3. I unfortunately can't get these to work with my Windows as Update is disabled.
4. Rest of the results:

   1. https://imgur.com/a/gf1dgDE

Please post share links for each image.

For tuneup plus please post images of any error messages or codes.

Post an image or share link of Disk management:
(make sure both rows and columns are widened so that characters are not cutoff)

For each of these disk drives:
a) Post share links for HD Tune
b) Post share links for Sea Tools for Windows

Disk 0 : SSD. It has 3 partition(s). Model: Samsung SSD 970 EVO Plus
Disk 1 : USB. It has 2 partition(s). Model: Seagate BUP BL
Disk 2 : USB. It has 2 partition(s). Model: Seagate Expansion
Disk 3 : HDD. It has 1 partition(s). Model: WD Elements 2620
Disk 4 : USB. It has 1 partition(s). Model: Lexar USB Flash Drive


When the chkdsk switches have completed for all disk drives please run this script and post a share link into this thread:

When available post share links for the view detailed information windows troubleshooters.
(What was used to disable Windows updates?)



Run the V2 log collector and post a new share link into the newest post.

Thanks for your help.

Disk 2 and 3 are TrueCrypt volumes so they don't appear in Disk Management. As mentioned, I've confirmed that they are not related as I get the same BSOD with them disconnected (all except C: are external hard drives).



Running chkdsk offline (I can't seem to trigger it to run on boot):

Was the recovery partition deliberately or accidentally deleted?



Please run these chkdsk switches on all drives: chkdsk /b /v

Please use the syntax chkdsk /b /v C: or chkdsk /b /v D: or chkdsk /b /v E:


C:\WINDOWS\system32>chkdsk /b /v
The type of the file system is NTFS.
Cannot lock current drive.

Chkdsk cannot run because the volume is in use by another
process. Would you like to schedule this volume to be
checked the next time the system restarts? (Y/N)


Type: Y

Reboot as needed.




The images were difficult to view in the prior post.
If possible please post for each drive a share link for HD Tune results

a) Health
b) Benchmark
c) Full error scan


Another share link for Sea Tools Long generic test.

The script in the earlier post should capture all of the chkdsk results.


Once all of the chkdsk results have completed please post a new V2 share link into the newest post.

For share links please use one drive, drop box, or google drive.


What method(s) have been used to disable Windows updates?

Were any of the Windows troubleshooters able to run?
Please post images of the error messages and codes.

Thanks for your help.

I've run chkdsk on the NVME which cleaned up some errors and it's been quite stable for a few days now.

I'm fairly sure it's not related to the other external drives (besides NVME) as it still BSODs with them disconnected.

I remembered I have a spare SSD and SATA to USB cable so if the errors persist I will try disconnecting NVME and running Windows to Go off the spare to see if the BSODs continue. If I still have issues I'll come back with all the diagnostic results you mentioned.

The Windows image I installed (Phoenix LiteOS 11 Pro+ ? Windows 11 22H2 LiTE (22621.1)) doesn't come with a recovery partition and I think Windows Update is disabled as certain components (e.g. telemetry) have been stripped out intentionally to prevent them being restored.

The drive tests on the non windows drives can be performed anytime.
The drive tests on the Windows drives can be performed overnight while sleeping.

It is important that you test all drives to address the findings seen in the log files.