Occasional BSODs with no dump file created

[arble]

My PC has had occasional blue screens for several months. Each time one happens, no dump file is created even though I have configured Windows to do so. The event log contains a volmgr error at the time of the blue screen, with text "Dump file creation failed due to error during dump creation." The EventData for this is:

\Device\HarddiskVolume8
000000000100000000000000A10004C042000400010000C000000000000000000000000000000000

Some investigation shows that HarddiskVolume8 is my C: drive (Samsung 950 PRO NVMe). I actually have the dump file configured to write to a different drive, but this doesn't seem to make a difference. Does this suggest that the C: drive is physically on its way out? The Samsung analysis tool doesn't find any errors with the drive data, but I suppose that I might just have got unlucky.

 

[Colif]

Reasons for no dump files: https://support.microsoft.com/en-au/help/130536/windows-does-not-save-memory-dump-file-after-a-crash

can you right click start
choose Disk management
take a screen shot and upload to an image sharing web site (like imgur) and share link here?

that will tell us if you have a vol 8 or not...

happen to remember BSOD code?

 

[arble]

Reasons for no dump files:

can you right click start
choose Disk management
take a screen shot and upload to an image sharing web site (like imgur) and share link here?

that will tell us if you have a vol 8 or not...

happen to remember BSOD code?

At least one of them was CRITICAL_PROCESS_DIED. My computer restarts so fast that I usually can't see the message. I've now disabled the automatic restart, but I also haven't had a BSOD since then (typical). Disk management pic is https://puu.sh/zk1KY/64908a9bc8.png. I've already ascertained that C: is volume 8 from a powershell script I found floating around: https://puu.sh/zk1Py/6f78b88edd.png

I don't think any of the cases in that linked article apply. The dump file does not already exist; the paging file is nearly 2GB and I'm requesting a small 256kB memory dump; plenty of room for the dump in the specified path; paging file is on the system root partition.

 

[Colif]

critical process died can be a windows error, not a driver error.

See if this helps

right click start button
choose powershell (admin)
type SFC /scannow and press enter
once its completed, copy/paste this command into same window:
Repair-WindowsImage -Online -RestoreHealth and press enter
SFC fixes system files, second command cleans image files, re run SFC if it failed to fix all files and restart PC

 

[arble]

critical process died can be a windows error, not a driver error.

See if this helps

right click start button
choose powershell (admin)
type SFC /scannow and press enter
once its completed, copy/paste this command into same window:
Repair-WindowsImage -Online -RestoreHealth and press enter
SFC fixes system files, second command cleans image files, re run SFC if it failed to fix all files and restart PC

OK. They reported no integrity violations and a healthy image. Trouble with this issue is that it's hard to test since the blue screens are naturally infrequent!

 

[Colif]

You have done a few of these resolutions already, so just try those you haven't: https://docs.microsoft.com/en-us/windows-hardware/drivers/debugger/bug-check-0xef--critical-process-died

I normally look at dumps to fix BSOD but thats difficult if you can't get any. Therefore I give you that link.

If we knew what process was crashing I would have some clue as to what to fix. Its possible whatever crashes is part of the process that creates the dumps, hence not getting any.

If it was caused by a driver,. there are ways to cause it to BSOD but they are pointless without dumps to record what crashed.

 

[arble]

Reasons for no dump files: https://support.microsoft.com/en-au/help/130536/windows-does-not-save-memory-dump-file-after-a-crash

can you right click start
choose Disk management
take a screen shot and upload to an image sharing web site (like imgur) and share link here?

that will tell us if you have a vol 8 or not...

happen to remember BSOD code?

Just had another with "UNEXPECTED_STORE_EXCEPTION" - is that any more or less worrying?

 

[Colif]

Do you have samsung magician installed? If so, check the SMART score

If you don't have Magician installed, download and run HDTune free version of the SSD and checked the health tab

that 2nd error could mean the ssd is the cause of crashes. SMART score may give us a hint as it is a test the ssd runs on itself. Share a screen shot here if its unclear what scores reveal.

It could also be a driver but without dumps, I can't confirm that. I wonder if moving dmp files location back onto ssd would help.

 

[arble]

I do have the Samsung tool. I have 13.4TB written, which is well below the rated endurance of 200 for this model, so I'll feel pretty unlucky if the drive is on its way out already. It reports the drive condition as good, with the following data: https://i.imgur.com/l9sz8wX.png. Nothing in those scores looks particularly worrying, but I'm no storage expert so I can't tell for sure. Another thread I posted elsewhere also suggested moving the dump location back to the SSD and disabling the automatic restart upon BSOD, so I've done that for now and I'm just waiting for another one.

 

[Colif]

i wish Magician scaled, its soo tiny on a 4k screen

My usage is only 15.82 so i doubt its a hardware problem. Was the image a smart score or did you run a performance test?

Guess we wait and hope having it on ssd lets it record it. I guess if crash is too quick, its best to have dump close and not send it around system at end.

 

[arble]

i wish Magician scaled, its soo tiny on a 4k screen

My usage is only 15.82 so i doubt its a hardware problem. Was the image a smart score or did you run a performance test?

Guess we wait and hope having it on ssd lets it record it. I guess if crash is too quick, its best to have dump close and not send it around system at end.

The image was just the info brought up by default when you run the Samsung magician software. I didn't run any sort of performance test.

I had a repeat occurrence, this time with UNEXPECTED_STORE_EXCEPTION. No dump file was created even after moving the dump file location back to %SystemRoot%. Are there any other pieces of hardware that are implicated in this kind of BSOD? My PC is coming up to the time where I'd think about refreshing some parts anyway

 

[Colif]

that error always seems to point at storage so i would suggest to check sata cables right here... except its an NVME.

is that latest version of Magician? I just updated mine from an ancient version and if you click on SMART under the size of the drive it will show a screen with scores on the NVME, I assume you ran a performance benchmark? Even if it says everything is ok, run HDTune as a 2nd opinion

The lack of dump file could be because it is the boot drive crashing and PC can't record anything.

Try running PC in safe mode for a while and see if it crashes there, that will at least tell us if it is hardware or not. No drivers in safe mode so crash should be hardware if it occurs.

 

[samyakjain222]

There are only two solution for this problem mate.
1.Install/Reinstall/Update bios.
2.Create windows 10 media and reinstall windows.
Note:-you can download bios flashing setup from your motherboard manufacturer's website.
PS:I also suffered from this problem and i fixed 2 times by each of these steps.

 

[arble]

that error always seems to point at storage so i would suggest to check sata cables right here... except its an NVME.

is that latest version of Magician? I just updated mine from an ancient version and if you click on SMART under the size of the drive it will show a screen with scores on the NVME, I assume you ran a performance benchmark? Even if it says everything is ok, run HDTune as a 2nd opinion

The lack of dump file could be because it is the boot drive crashing and PC can't record anything.

Try running PC in safe mode for a while and see if it crashes there, that will at least tell us if it is hardware or not. No drivers in safe mode so crash should be hardware if it occurs.

I was able to trigger a BSoD with WhoCrashed that successfully wrote out a dump file, so they *can* be written, just apparently not for the exceptions I was getting. Anyway, new twist in the tale: I updated the NVMe driver and since then, I haven't had any BSoDs. What I have had instead for about a week now is that when I power the computer on for the first time each day, within the first 10-15min it will totally freeze up except for the mouse pointer (i.e. the cursor moves but cannot interact with anything on screen, all windows freeze etc). When I power cycle the machine I then have no trouble for the rest of the day. I had never seen this behaviour before the NVMe update and although it's difficult to be certain since it was intermittent anyway, it looks like the former BSoDs have been replaced with this behaviour. Does that suggest any particular cause? I'm willing to replace bits of hardware, but it would be good to have some idea of which pieces to start with.

 

[Colif]

All of the errors - as I said before - point at storage. Fact they were fixed with the NVMe driver rings true to it being a driver. Was it a motherboard driver or something from Samsung?

Curious if you get freeze if running in safe mode. It seems odd it happens at a specific time after start up.

Did you ever look at SMART score of SSD?

What are rest of specs of the PC?