Windows 10 Random BSODs since December 2020

[Nomad of Norad]

I have been getting BSODs since at least December of last year. They'll just suddenly show up out of the blue when I'm in the middle of something. Sometimes it is a few days between bluescreens, occasionally its a few hours between bluescreens. There doesn't seem to be any rhyme or reason for them. I can be working on something in Daz Studio. I can be simply watching a Youtube video and/or watching a live feed in Discord. I can be sitting in SecondLife at a music-club event. I can simply be chatting in Discord or in IRC.

Often, when the BSOD happens, the sound then goes BAZZZZZZZZZZZZZ!!!!!!! the whole time.

This might be a factor, but sometimes, when I'm listening to audio, the audio will stop for a fraction of a second and go BAAZZ, and then resume where it left off.

I tried to get help over here ->> Having near-daily bluescreens in 20H2 <<- but they didn't give much, and sort of abandoned me after a bit.

Anyway, I have attached the requisite W7F zip. [A beat] Wait. "The upload file is too large for the server to process." Whut? 0o

Trying this again with it is a .7z instead.

 

[kemical]

Hi,

thanks for the upload.

you have one dump blaming the graphics driver and another blaming memory management ( the process running at the time was Discord).

Is your system overclocked? If so please return to default values until we find the culprit.

Try updating/reinstalling the latest gpu driver but use the DDU to uninstall it first as this also removes any corruption from previous drivers:
Display Driver Uninstaller (DDU) V18.0.3.6 Released. - Wagnardsoft Forum

It may also be worth updating/reinstalling Discord.

Same goes for your audio driver, if your unsure where to get it from and it's an onboard device then try your motherboard support page.
(if you post the make and model of the motherboard I'll find it for you)

Post any new dump files, the more we see the better.

 

[Nomad of Norad]

Pretty sure my machine is not overclocked... unless something ninja'd in behind my back and did it.

Motherboard is an MSI 970A-G46, (or maybe its an AMD FX-8320? Is there a way to check from inside Windows?)

Processor is an AMD FX-8320 Eight-Core Processor, 3.50 GHz
32 GB of ram
Windows 10 Pro, 64bit, version 20H2
OS build 19042.804
Experience: Windows Feature Experience Pack 120.2212.551.0

Zotac GeForce GTX 1060 card

 

[FireMarshalBill69]

I had this problem once before. Instead of screwing around, I re-imaged the machine. Then, installed questionable software one by one, each time running it heavily for a day before installing the next one. That fixed it for me after days of wasting my time. It was an app! If it blue screens without any apps, then you know where to look. I would start there rather than go looking for a hardware issue which can be more complex.

 

[kemical]

Pretty sure my machine is not overclocked... unless something ninja'd in behind my back and did it.

Motherboard is an MSI 970A-G46, (or maybe its an AMD FX-8320? Is there a way to check from inside Windows?)

Processor is an AMD FX-8320 Eight-Core Processor, 3.50 GHz
32 GB of ram
Windows 10 Pro, 64bit, version 20H2
OS build 19042.804
Experience: Windows Feature Experience Pack 120.2212.551.0

Zotac GeForce GTX 1060 card

That's great!

Your motherboard support page is here:
Link Removed

Your chipset driver page is here:
Link Removed

Remember to uninstall the original driver first. (obvious I know but many surprisingly don't)

Also as you had a bugcheck 1a it's only prudent to test the ram. The following guide will show you how to run the windows memory diagnostic:
How to Test Your Computer’s RAM for Problems.

See how you go with the above and please post any new dump files.

 

[Nomad of Norad]

I actually already had run MemTest86+ on the machine, and the first time it gave me a bunch of ram errors at a lot of memory addresses, I then rearranged the ram chips and ran it again, and it gave an error on a grand total of ONE memory address. o0

I'd give you the screengrab, but the system here keeps saying its too large, even when 7Zipped. 0o

edit: On the other hand, I just had another BSOD. Minidump attached.

 

[Nomad of Norad]

Bah, it crashed again. The earlier one today, it was while I was navigating around in Windows Explorer. This time it was while I was scrolling through a chat in Telegram. In the latter case, I then had to wait through a few minutes of it frozen up and going BAZZZZZZZZZ!!!! before it finally went to the BSOD error. This is getting infuriating.

edit: I have since run the Windows Memory Diagnostic, and it came back with no errors.

 

[kemical]

I actually already had run MemTest86+ on the machine, and the first time it gave me a bunch of ram errors at a lot of memory addresses, I then rearranged the ram chips and ran it again, and it gave an error on a grand total of ONE memory address. o0

Hmm... If your getting errors during memtest86 then your RAM is almost certainly suspect.

I know you say Windows memory diag didn't find anything but it often doesn't, thats why we use memtest.

Have you tried determining which stick is the culprit?

 

[Nomad of Norad]

Well, the weird thing is, it was giving a LOT more errors, spread across a LOT of different parts of the memory, when I tested it with the sticks in the positions I originally had them in. I pulled them all out, and put them all back in with the older pair now where the newer pair had been and vice versa, ran the test again, and got WAY less error addresses in it. In fact, in that last test, ONLY one memory address gave an error. I tried to post a picture of the last test from the previous instance, and the last picture from the newer test.... but the forum software said the file was too big, even when I tried sticking it in a zip file.

Anyway, the machine just crashed again, so posting a new minidump.

On a different note, for some reason, sometimes at startup, one of my chat programs, Pidgin, shows up with all the interface showing complete gibberish, as in the type-font looks like it got replaced with complete junk, and some other file-explore windows then show blank white instead of the file names. 0o Exiting out of Pidgin and relaunching it usually sorts it, but this is another apparent symptom of whatever malfunction is going on. On the other hand, this time, on relaunching Pidgin, everything showed up with [][][][][] in it. 0O

Edit: Annnd, BlueScreenView is showing me blank white space instead of readout. And Paint.net is missing parts of its interface, too. Bah!

 

[kemical]

Hi,
your last three dumps have all been the same BUGCHECK_CODE: 1ca:

Link Removed

The above relates to VM (virtual machines and hyper V). Try running these scans to see if Windows has become corrupted:

File scans
Open an admin command prompt. Type:
sfc /scannow
press enter and await results

In the same command prompt and after the above scan has finished type:
dism /online /cleanup-image /restorehealth
Press enter and await results (longer this time).

If the first scan found files it could not repair but the second scan is successful, run the first scan again using the same command prompt box and this time it should repair the files found.


Regarding the RAM tests, I mean really there shouldn't be any errors at all, if there is then as I wrote, the ram may be bad?

Lastly apologies about the forum errors in uploading items. Hopefully it will be resolved soon.

 

[Nomad of Norad]

Bah, I'm trying to launch Windows PowerShell (Admin), and it shows up briefly as a slender, tall, mostly white window that then vanishes away after 3 or 4 seconds.

As for VM and hyper V stuff, could any of this have to do with Docker Demon (Docker Desktop)? I have a sub-task running a small virtual Linux distro inside Windows to mine Pi Coin, but it seems to have crashed earlier or something, and so wasn't running today.

edit: Okay, rebooted, and GUI interface stuff behaved more normal. Did the scannow stuff, and attached is the screengrab of it all.

 

[kemical]

As for VM and hyper V stuff, could any of this have to do with Docker Demon (Docker Desktop)? I have a sub-task running a small virtual Linux distro inside Windows to mine Pi Coin, but it seems to have crashed earlier or something, and so wasn't running today.

Hmm... I'd say that sounds like a vm alright. A virtual machine running Linux..

Did you start mining in December when the bsod's started?

Anyhoo it looks like the sfc scanner repaired some stuff so may help but if not I'd seriously checkout the VM as a possible cause. It would be simple enough, simply stop running the vm and see if the bsod's stop.

 

[Nomad of Norad]

No, the mining app (that uses Docker Desktop) was installed less than a month ago, as an optional extra coming off of and linked up with the Pi Coin desktop app, which I didn't even join up with until mid January or so. For some reason the Docker thingy keeps crashing at startup lately, though, and I have to go manually restart Docker Desktop to get it going again. 0o

I've been having these BSODs off and on since late December, usually once every few days, though occasionally two or more on a given day. Yesterday, I was having more trouble than usual, and I suppose its possible this time it was because I simply ignored Docker having given the error and simply exited out of the An error occurred box the first occasion and didn't bother trying to launch Docker again. I guess it then left the thing in a not-running-fully state of being on subsequent machine restarts or something.

 

[kemical]

Since you've ran the sfc anymore crashes?

 

[Nomad of Norad]

Not so far. I think I ONCE heard the sound go BAAAZ for a half-second while I was listening to something... but no other hints of trouble other than that. Normally it can be 4 or 5 days between bluescreens.

 

[kemical]

Just for information have you monitored your cpu temp?

Try using HWiNFO app as this gives a kind of history. Keep it running over the day and just use the system as you would normally. I'd be interested to see what temps are like.

 

[Nomad of Norad]

I don't get the impression its gone outside of the 60 to 68 F range, looking at the CPU-temperature thingy showing on my task bar, but I've not been watching that like a hawk either. If I remember to launch HWINFO next time I fire the machine up, I'll try looking at that.

 

[Nomad of Norad]

I've been wondering for days if the reason I was having at least the more recent bluescreen errors was because the Docker mining daemon wasn't successfully launching at startup, and some other background task was poking and prodding at it saying "Why aren't you replying to me?!?" ...but since around the time I did the scannow thing, Docker was instead stuck forever at trying to launch.... but I left it more or less alone all week and haven't had BSOD problems since.

Well, today I was trying to get the Docker daemon program going again.... and someone had suggested I try uninstalling Docker Desktop and reinstalling it, but that seemed to get stuck at trying to remove the related content first, without showing any disk activity, so I killed out of it in Task Manager, and ran the installer for Docker Desktop again.... which simply told me I already had the current version installed.... and didn't seem to fix the daemon. Then a couple hours later, I got another bluescreen while in SL.... after NO such problems for a WEEK!

So now I'm wondering if trying to FIX the thing in Docker in fact UNdid the fixxes scannow did earlier. 0o

On the other hand, after the reboot caused by the BSOD, I find the pi-consensus container/app in Docker Desktop is running again.

Anyway, mini-dump attached.

 

[kemical]

Hi,

the dump was again a bugcheck 1CA which we had kinda linked to your VM. So it looks like this is your culprit especially as you've had no bsod until you tried to get it going.

Also if it is the culprit why is it making the system fall over? I wonder if it has something to do with cpu temps?

My reasons for this are because I've heard the sound you have mentioned that the audio makes and with me it was an overheating cpu. You stated your cpu is at around 60c-68c?

AMD state that the Tmax for this chip is 62c, so you might be close to the margin which is why i asked you to monitor the temp.

If the system hasn't been cleaned out recently then you may have some dust build up?

 

[Nomad of Norad]

Well, the VM thingy was running fine for a few weeks, the day I was having multiple BSODs was one where the daemon hadn't launched correctly, and wasn't running at all, which is what made me wonder if it was the LACK of it that was causing the hangs and crashes, due to some other program refusing to accept that it wasn't there.

Come to that, when this last BSOD happened, the thing in Docker wasn't running yet either. After I tried getting the Docker thing back in place, I checked Docker Desktop, it still wasn't showing the needed dock running, at least until I gave up on it and went on to do other things on the machine. It was only running again properly AFTER the machine came back up from the restart that was triggered by the BSOD.

Is there an app that will record the CPU temperature ups and down into a logfile for later perusal? Something that retains, say, the last 5 minutes of the readings, from session to session, storing 5 or 10 consecutive logfiles on a rotational basis?