An assortment of very strange BSODs & crashes

Machinesbleed

Machinesbleed

Posts: 16   +1
Right, so, this is a long story-- I've been having issues with windows since December last year. It's probably best if I start from the start, since surely there's some hints in the early days of wtf'ery. So:

I moved my PC from my now-bedroom to loungeroom. It'd been disconnected for about three months, and I had exactly 0 issues with Windows beforehand. I'd bought Windows 8.1 and used it for maybe about four months before I disconnected it for that long stretch of time.

Once I moved PC to lounge, sometimes it'd completely shut off without warning while I was doing things. No BSODs or anything. In one of those instances, I noticed the heater turned off too. So I suspected faulty power. I changed location of power to bedroom by way of extension cord. Still happened. Replaced the PSU, and it solved that issue, except Windows was being disgustingly slow. I ran extensive hardware tests and found a HDD was dying, so I replaced it with a shiny new SSHD. I suspect the power issues blew them both out. Everything was okay for a short while.

Then I started getting BSODs when I was afk for ~10 minutes. If I moved my mouse before the time was up, no BSOD would happen. Suuuper weird. Then when playing WoW, in low population areas mostly, when I was just pissfarting around- occasionally my screens would turn off, a fan in my PC would be screaming (I assume my CPU, my videocard is pretty silent), and then my whole PC would shut off and restart. What.

So I did the following:
  • I reformatted windows like...3 times.
  • I checked to see if my windows install was corrupted. Nope.
  • Ran more extensive hardware tests, using: Intel Processor Diagnostic Tool, GSmartControl, Windows memory diagnostics. All came back fine.
  • Updated to windows 10.
  • Stress tested with BurnInTest, monitored everything extensively with Open Hardware Monitor, stress tested by having WoW open in a very high population area while rendering something extremely intensive in Blender. Nothing bad happened. Nothing even lagged.
  • I updated my drivers.
  • Registry repair'd.
  • Increased virtual memory size.
All those changes were informed by what sense I could make of the crash dumps + way too much time finding only vaguely helpful things via google.

None of that worked, so I went screw it: I'm duel booting Linux again. LO AND BEHOLD I have had exactly 0 issues running Arch Linux. No kernel panics, nothing. Only weird thing is sometimes when running WoW through Wine it'll crash out in high population areas but I suspect it's just a Wine and OpenGL weirdness thing since it runs like trash comparatively. It *does* come up with a WoW error saying it's using too much memory when it crashes on Linux, but running through terminal tells me it only uses 109% of 400% possible CPU and surely it's not eating all my RAM. I can AFK in Linux without issue, and it runs smoothly and wonderfully and doesn't take 500 years to load (I think my windows duel boot is being *unusually* slow, so worth noting. Not HDD failure level slow, but slower than it probably should be for an SSHD.)

Oh- I should probably state, I'm using the proprietary drivers on Linux.

I avoided Windows for a while, but I need it for more intensive WoW endeavors and game development, so I hopped over and the issues aggressively persisted. So, take two. I:
  • Found a setting that stated windows was automatically installing drivers for me. No ty. So I changed that.
  • Changed boot settings- told it not to boot a bunch of non-MS services I didn't need.
  • Re-installed drivers for the 500th time.
  • Registry repair'd

WoW behaved after. I can't say if the crashes are *fixed*, but I did go for an unusually long time without encountering any.

I managed to AFK for 22 minutes without an issue, too. New record! I went to tweet about it and then midway through my PC froze and I got a new BSOD. Looking at the crash dump it looked like an ethernet issue. So I updated the drivers for that- it was one thing I neglected in my driver pass. Sweet.

I was able to AFK for ages again today, but got a BSOD before I could come back to PC.

I'm completely stumped.

My system's specs:

Motherboard- MSI H97 gaming 3
CPU- Intel i5, 3.20 GHz
RAM- Corsair, total 16GB (2x 8GB sticks) - not sure what *type* of corsair RAM, I can't find details anywhere. It's er...very plain in appearance.
Videocard- NVIDIA GeForce GTX 760
PSU- EVGA Superova G2 Gold 750W
Duel booting Windows 10 64-bit & Arch Linux
HDDs- Windows is installed on my Seagate SSHD, WoW on a Hitachi SATA HDD.

Attached are some of the crashdumps, anything before march is prior to my wave two of attempted fixes.

I've looked at my motherboard and there's exactly 0 signs of visual damage- looked over it for blown capacitors and fractures. I'm thinking it could possibly be the cause still though. Or some weird really obscure software compatibility issues in Windows only. I'm guessing the latter, hence the prefix choice.

Thoughts? And thank you very much! I swear my computer has become sentient. Perhaps this is the beginning of the machine uprising.
 

Attachments

  • 16.03.19.rosescrashdumpfiles.zip
    910.8 KB · Views: 1
First, I would make a semi-permanent restore point as it sounds like you have cleared up a lot of the issues.

AFK BSOD... have you changed your power options yet?

Since I tinker (the sign says ' minidumps are a mystery'), this is all I can offer.

As you can AFK in Linux without crash, it is almost certain to be Windows OS software or something which only runs along with that.

Anything in Event Viewer 'critical error or error'?
 
Cycloid Torus said:
First, I would make a semi-permanent restore point as it sounds like you have cleared up a lot of the issues.

AFK BSOD... have you changed your power options yet?
Click to expand...

Yeah, I made a restore point preeeetty damn fast after realizing I could afk for longer, aha. Which power options do you suggest I change? I'm new to tinkering with that side of things.

Cycloid Torus said:
Since I tinker (the sign says ' minidumps are a mystery'), this is all I can offer.

As you can AFK in Linux without crash, it is almost certain to be Windows OS software or something which only runs along with that.

Anything in Event Viewer 'critical error or error'?
Click to expand...

The sign says they're a mystery; you mean you can't find much usefulness out of the crash dumps either? I'll check now re: event viewer. I *did* go through things and couldn't find much I could make use of, but lets see what I can find now...

All critical errors are just "The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly." Real helpful, windows.

There is a whole lot of errors, here's a bunch around the time of the critical power loss ones:
  • Failed with 0x490 modifying AppModel Runtime status for package Microsoft.Windows.Photos_16.302.8200.0_x64__8wekyb3d8bbwe for user REX\Rose (current status = 0x0, desired status = 0x20).
  • Failed with 0x490 modifying AppModel Runtime status for package Microsoft.People_10.0.10500.0_x64__8wekyb3d8bbwe for user REX\Rose (current status = 0x0, desired status = 0x20).
  • Failed with 0x490 modifying AppModel Runtime status for package Microsoft.ZuneVideo_3.6.17801.0_x64__8wekyb3d8bbwe for user REX\Rose (current status = 0x0, desired status = 0x20).
  • Failed with 0x490 modifying AppModel Runtime status for package Microsoft.MicrosoftSolitaireCollection_3.8.3092.0_x64__8wekyb3d8bbwe for user REX\Rose (current status = 0x0, desired status = 0x20).
  • Failed with 0x490 modifying AppModel Runtime status for package Microsoft.CommsPhone_2.15.9015.0_x64__8wekyb3d8bbwe for user REX\Rose (current status = 0x0, desired status = 0x20).
  • Failed with 0x490 modifying AppModel Runtime status for package BallardAppCraftery.CraftyEMLViewer_1.1.0.27_neutral__epyrqhfctk40t for user REX\Rose (current status = 0x0, desired status = 0x20).
Appears harmless enough, since I don't even use the apps listed there? Except for that last one. But I'm admittedly not very good at the event log side of things, either. Also worth noting some of those happened *after* power losses, so I don't think it's consistent enough to really be considered a problem. When I used CrafyEMLViewer it didn't error out or crash or anything.
 
Last edited:
Oh- I understand event viewer, just not like...the error messages, and in my googling I also found that thread you'd linked and didn't seem to find anything useful in it. :/

I doubt it's the cause, in either case. I'm getting constant error messages of same source in event viewer, those were just around the time of the sudden dying of my comp. And like I'd said-- some were after it. I mean, if anyone who knows more on it thinks otherwise I'm obviously going to listen, but at this point I don't think event viewer is any use in trying to solve wtf is happening. The BSODs will tell us a lot more. Well- maybe.

I mean, I couldn't find much in them other than "drivers maybe" and I tried fixing that and nope.

I'll try flashing my BIOS, doubtful it'll help, but whilst I wait for any other advice it can't hurt I suppose. And thank you for the replies!
 
Oh, thank you!
That would explain the monitors turning off before PC crashes in WoW. I wonder why this is an issue, though? Should I perhaps downgrade my drivers to an older version?
 
Well; I have the updated driver already. So I ran *another* windows install scan (sfc /scannow) - probably my third - and it found corrupt files this time?? I've attached the log. Hopefully that's fixed it, combined with my driver updates. This has been bizarre.

If there's anything worth noting in the repair log please let me know! I want to learn more about wtf even happened. I'm leery and not sure this will fix the underlying issue, but I'll test it by AFKing as frequently as possible over the next few days. If it still BSODs I'll come back here to scream about it. And if it's fixed now, I'll also come back to let you know.

Thank you for the help thus far! <3
 

Attachments

  • 16.03.21.roseswindowsrepairlogs.zip
    880.4 KB · Views: 0
Last edited:
Aaaaand it bluescreened on first attempt to AFK. ~25 minutes. Seems to be fairly consistent with time, which is sus to me. It used to be more like 10 mins which is also weird. Attached is the crashdump. Ran another system file checker tool scan- and it found nothing corrupt this time around.

It doesn't seem to matter what's running (in the foreground, at least) when it happens. Really weird.
 

Attachments

  • 16.03.21.rosescrashdumpfile.zip
    89.4 KB · Views: 2
Yeah; I can see the BSODs, and I can vaguely understand them. I just don't know how to fix it or what the root cause is.
 
I'm gonna try backtracking my GPU drivers and installing a version from before this all started, if anyone has any other ideas for fixes It'd be much appreciated. Thanks!
 
Right, so, update:

I hadn't got around to backtracking drivers, and WoW did it's crash thing, except this time it was just the screens turning off- and the fan screaming that happened during *is* my GPU's after all. Opened up case to monitor it. This time, no restart. I had to manually do it. So I think it's safe to say my windows install corruption was what was causing the restarts in this instance. No idea why it didn't pick up on the corruption earlier though.

Check event log, it had spat out a wall of new actually useful errors, all of which were as follows:
The description for Event ID 13 from source nvlddmkm cannot be found. Either the component that raises this event is not installed on your local computer or the installation is corrupted. You can install or repair the component on the local computer.

So yeah, you were right- videocard driver issues. This is enough information for me now to find anything useful, since before I hadn't narrowed it down which made it difficult. It's apparently a common issue with NVIDIA drivers. I'm fairly certain I'll be able to fix it; if I do, I'll reply with what I did for future ref, if not...I'll come back and yell more.

Thank you for the guidance!
 
You are welcome, that video card may be an issue itself... I have never had an issue with Nvidia drivers or cards. AMD is a very different story
 
Yeah idk hey. I used to have an AMD comp and it was d I s g u s t I n g. I've had this videocard for maybe two years? And it's been fine this whole time. It passes all health and stress tests. And drivers for Windows 10 are around, and recent, so they definitely support it. Apparently nvidia are just trash at drivers thesedays because people have been complaining of issues like mine since august last year. Also all using windows 10. I just don't get how you mess that up so badly and for so long.

Backtracked to drivers from October 2015 because my comp was off for the months between then and when I started getting issues, so not sure whether I should trust the ones inbetween. Time to AFK test...!
 
Yeah, my GPU is physically fine and well dusted.

Alright, so:
  • Reset VM settings (turned off, restarted, turned on again, restarted)
  • Downgraded and upgraded Nvidia drivers again, since first try didn't improve anything- none of the drivers seem to be helping
  • Set HDD to never turn off, in power settings (*immediately* got a BSOD after saving those options, and it was set to 20 minutes- about as long as afk BSODs take, interestingly)
  • Set the service User-mode Driver Framework to automatic instead of manual (it was suggested in a forum topic I'd found.)
  • Disabled my second monitor, since some people reported it being an issue for recent Nvidia drivers.
.....aaaand I'm getting more aggressive crashes & BSODs. Attached are recent crash dumps. But, er, progress? More broken means more likely to find cause, I hope.

I ran a memtest + a HDD test as a precaution and they turned up clean as per usual.

There's some new event log errors that I've never seen. Two in particular. Here they are:

[LEFT]The description for Event ID 263 from source Win32k cannot be found. Either the component that raises this event is not installed on your local computer or the installation is corrupted. You can install or repair the component on the local computer.
If the event originated on another computer, the display information had to be saved with the event.
The following information was included with the event:
The specified resource type cannot be found in the image file

&

The driver \Driver\WudfRd failed to load for the device ROOT\WPD\0000[/LEFT]​
 

Attachments

  • 16.03.23.rosescrashdumpfiles.zip
    347.9 KB · Views: 2
I'll focus on trying to fix that first, then, yeah- the gpu weirdness is annoying but maybe the driver is borked bc of the boot drive issues. I'll do some research and report back. And ty!
 
I tried fixing things via bootrec- found windows ran a loooot smoother, I think gnu grub was messing with my windows boot drive (apparently known to happen.) I still got BSODs though, so I reformatted, and individually installed each driver/thing and AFK'd between. I've had one BSOD, and it was right after I installed chrome-- it took an hour, and I only caught it at the motherboard boot screen. Crash dump initialization failed (found that info out via the event log), and some other weird things were in the event log-

It'd somehow messed the time up, my PC was on only from 11am-8pm at that point and the time before the BSOD the clock in event log was saying it was 2am...? The time was correct the instant I reformatted, so like...weird. It had multiple instances of loss of internet in there.

My first suspect was chrome/pandora, waaaaay back, since I always have both running, and I was half right. Apparently the killer networking suite that comes with the drivers is dodgy and known to cause issues, especially with chrome, for some reason. I *did* notice my internet would drop out a lot more than usual during these issues. I still got BSODs with chrome closed, though, so I figured I was wrong- but I guess it has background processes running at all times?

I wasn't able to get it to BSOD even with the killer suite installed since that one time I juuuuust missed it, AFKing for like an hour at a time. It took a few days after fresh installs the last few times for it to start playing up, but the fact that it happened *right* after chrome was installed was enough for me to uninstall the killer suite and get the standalone drivers. I hope that fixes it.

Seems it was/is a really complicated mix of things. I'm logging what I install & when just in case it's not fixed yet, so I can pinpoint exactly when things start acting weird. I'm no longer duel booting windows & linux, and am staying with windows 8.1 for now. I'll report back with findings!
 
I'm going to call this fixed; I've not had a single BSOD since removing the killer network suite. Chrome is acting slightly buggier, but only by freezing momentarily sometimes, which confirms it being a weird chrome-induced issue I guess. I've AFK'd for like...8-10 hours without issue repeatedly over last few days and have kept PC on without a single restart. Yay!

Thanks for the guidance. That really was a combination of very odd problems. I'm uh...not going to upgrade to windows 10 again for a while, just in case.
 
  • Like
Reactions: Cycloid Torus
You must log in or register to reply here.

Similar threads

N
Very long delay in receiving incoming emails in outlook 2021
Replies
0
Views
281
Neitanod
N
midian182
Windows and Microsoft Office are being replaced with Linux and LibreOffice in this German state
Replies
20
Views
222
amghwk
amghwk
E
Linux could have been brought down by backdoor found in widely used utility
2
Replies
31
Views
475
ZedRM
ZedRM

Latest posts

Back