Windows 10: New PC Build, Many Different BSOD Errors @ Random, Cannot Diagnose

Hello, after lurking these forums and looking into my issue I cannot find a solution. I am experiencing crashes on my new PC build.

Since my first day of building it, I have experienced BSOD errors/random crashes. I cannot seem to find something that will make it crash. In other words, I cannot replicate this problem. I cannot pinpoint it to one piece of software or action on my computer. I also do not know if this is a driver related problem, or a faulty hardware problem.

There are times where I can play a game for 3+ hours on end, and not experience a crash. There are times where I boot my pc, open up a browser, and it crashes before I can load the first webpage.

Some BSOD errors include:
IRQL NOT LESS OR EQUAL TO
KMODE EXCEPTION NOT HANDLED
DRIVER OVERRAN STACK BASED BUFFER
MEMORY MANAGEMENT
SYSTEM THREAD EXCEPTION NOT HANDLED

Attached is the zip file I created from the BSOD posting instructions
Note: Many of the errors point to LgBusEnum.sys. I believe I have corrected this error as I uninstalled the Logitech Gaming Software. This errors happened when using driver verifier. I no longer get them. Please take a look at the most recent BSOD's. I have tweaked settings/uninstalled some things during the period of these dump files.

Things I have tried/know:

• memtest86+ on a few occasions. One test included both sticks of ram with 13 consecutive passes. No errors were found.
• Tested my GPU on Unigen Heaven. It can run for hours at normal temperatures unless a random crash occurs.
• A 15 minute test on Prime95 resulted in no crashing/overheating etc
• I have tried my best to make sure that I have all up-to-date drivers, but I cannot say that I am 100% up to date.
• Driver verifier will run right now without finding any immediate errors (I fixed the LGBusEnum/ScpVBus errors). I can even play games on it without it finding an error/creating a BSOD. Something is odd though: After I try to shut down my PC with verifier on, a BSOD will appear as soon as I click the shut down button.
• One BSOD error was caused by a "single bit error" Doesnt this usually indicate faulty ram? I got no errors on MemTest though....
• I did a clean install of windows after getting fed up with it crashing. It does not seem like it helped at all.

I understand that a lot of these BSOD errors hint that memory might be defective. Although I have tested the memory, is it possible that it can still be faulty?

Thanks for looking at this. This community is awesome, and the support it provides is outstanding. I apologize if my BSOD errors are lacking. I reinstalled windows again and lost some of my earlier ones. I can post the latest dump files as my continues to crash.

:)

 

Getting Lots of different BSOD Errors at Random, Cannot diagnose the problem.

Hi,

Since it is a new build and you are getting a variety of different Bug_Check codes it sure looks like hardware. Remove ALL power then reseat the cards, memory, and cables (on both ends when possible) - actually remove and replace them - do not just snug.

Heat is also a possibility - add a fan blowing into the intake or case. Heat related does not always mean excessive levels as a single component could be too sensitive to even the normal levels of heat.

Memtest needs to run at least 4+ hours. Also memory often needs to be swapped in and out in different combinations and patterns - faster memory needs to be in front of slower and even tiny difference (even within the specs) can make a difference. Try using
the minimum memory required to boot the system.

Be sure BIOS, chipset drivers, and all major drivers are updated.

Checking the minidumps now.

Hope this helps.

--------------------------------------------------------------------------------------------

Rob Brown -
Microsoft MVP <- profile - Windows and Devices for IT : Bicycle - Mark Twain said it right.

 

Getting Lots of different BSOD Errors at Random, Cannot diagnose the problem.

My pc has these random BSOD crashes. This was a new build I did about 3 weeks ago, and it has been crashing ever since. I have tried to include as many details as possible. I am using windows 10 pro edition. Please read the following:

Components:

ASUS gtx 1080 strix OC edition

ASUS z170 pro gaming aura

1tb Samsung 850 evo SSD

i7 6700k cpu

evga g2 650 watt psu

g.skill ripjaws v 16gb (8 x 2) 3200 ram

Symptoms:

Computer randomly blue screens with several different BSOD errors. Does not have to be under load/gaming/etc. Sometimes BSOD appears right after booting and signing in. Sometimes I can game for 2 hours without a BSOD.

These blue screens include and are not limited to:

-kernel security check error

-driver overran stack-based buffer

-irql not less or equal

-memory management

Tests:

ran memtest86 for 3 consecutive passes. No errors.

ran memtest86 for 5 consecutive passes. No errors.

Stress-tested GPU on Unigen Heaven and except for the random crashes, it ran for about 3 hours straight under normal temps.

Stress-tested CPU on Prime95 for 15 minutes. Observed normal everything.

With verifier on, about 10-15 secs after logging in, I get BSOD of SYSTEM THREAD EXCEPTION NOT HANDLED. Is this related to verifie

Did a clean install of windows 10. Downloaded latest nvidia driver, tested in unigen heaven. Crashed again in 15-20mins.

Other notes:

-No antivirus installed.

-This is a clean copy of windows 10 with newest drivers on GPU and motherboard.

-Was doing on first installation of windows 10. Still doing it.

-Computer is brand new. BSOD's first appeared on second gaming session.

-I saw that hidclass.sys was in one of the dumps, so since then I made sure that I had the most recent drivers for my peripherals. I am only using a keyboard and mouse as of now.

Questions:

Can this be the ram, despite it fairing well on memtest86?

Could simply reseating the ram/gpu fix this?

What else can be faulty? Hardware/software?

Did I miss a driver?

Here is a link to all of my dump files. I have shared this before, so this is easy for me, but if I must zip them up and attach them to this posting then I can:

dumps

 

Hi HighMule,

Welcome to the 10forums and apologies for the delay.

This month I have a few exams, I don't know exactly when but please don't hesitate to leave a PM if I haven't responded in 2 days *Smile


<div>Please disable fast startup</div><div>Turn On or Off Fast Startup in Windows 10</div>

Please make sure your pc is configured for small dump files How to configure minidump | Windows 10 tutorial







Diagnostics Test

HDD TEST
*Arrow Check Disk (chkdsk) - Read Event Viewer Log - Windows 7 Help Forums

 

Thanks a lot for the reply, Axe0.

I am away from my desktop until this evening. I forgot to mention that I am using a single SSD for all of the storage on this rig. Will it be okay to run all of the tests for the SSD?

I am obviously going to skip the SeaTools test.

 

The mentioned storage tests can be run on all kinds of storage devices, the exceptions are mentioned *Smile
To be straight, it is OK to run the tests on your SSD *Smile

 

Thanks for the speedy reply. I can't to find out what my problem is!

 

Axe0,
I have conducted all of the tests you recommended, aside from the SeaTools test.
I think my SSD looks good.

Also attached is my most recent crash -> I had one more since starting this thread. Sorry that I did not configure the minidump before this happened. I looked at the debugging info, and I saw "memory corruption" inside of it several times.

 

I just experienced another crash while gaming. Attached is the .dmp.

Debug read this:

*** WARNING: Unable to verify timestamp for nvhda64v.sys*** ERROR: Module load completed but symbols could not be loaded for nvhda64v.sysProbably caused by : memory_corruptionFollowup: memory_corruption

Right clicking on the driver file in file manager and device manager reports that the driver is version 1.3.34.17. I did not install NVIDIA's graphics driver release from this week, but I have the next-latest one. In the documentation, my driver matches the second-latest NVIDIA released audio driver.

In my device manager, I do have 3 items under the sound, video, and game controllers:
-NVIDIA High Definition Audio
-NVIDIA Virtual Audio Device (Wave Extensible) (WDM)
-Realtek High Definition Audio

Could these be interfering and causing crashes?

What is going on here? Does this indicate that RAM might be faulty? Please let me know if I am wrong to suspect RAM even if memtest86+ did 13 consecutive passes. I am considering just using a single memory stick in a combination of slots to see if there is a faulty stick/slot.

There is a different error every​ time! It just happened to the the audio driver that failed this time.

Edit: Computer crashed a couple more times. Both are labelled with "memory corruption" stuff.

 

There is a new BIOS version 2103 Improve system stability , you are still on 0311 first release.

One of the dumpfiles names Memory management
You are using 3200 mhz RAM did you enable xmp profile.
At what voltage are your modules running.
Check the timings.

How long ago did you run your memtest86+?

 

The model is 3200, but the RAM is running it SPD 2133.
@HighMule, it more likely indicates that the NVIDIA audio driver caused memory corruption. The nvhda64v.sys is not a graphics driver, but an audio driver.
You could uninstall the audio drivers of NVIDIA, they are not necessary.

 

@axeO I know thats why i said using 3200 mhz modules, because on some forums there are people mentioning problems if they have enabled xmp profiles on this mobo.

 

The memtest was done recently (week to 10 days ago). It was during the midst of all of this crashing. The system is now near brand-new. I believe I have done it before and after reinstalling windows on the machine. For diagnostic purposes, will it be worth it to run the test again?
@axe0 just to clear up some confusion, I did understand initially that this was an audio driver, not a graphics one. I referenced the most recent installments of Nvidia's drivers, and saw that my driver version matched the second latest release. My bad if I worded that incorrectly. I will be uninstalling this tonight to see if it improves my system. A good question here is, was the driver actually the cause, or did something else cause the driver to be flagged? In other words, is the dump very accurate at blaming drivers? I might try enabling verifier again....

And both of you two are correct about the memory settings. I did try to enable XMP settings to see if the stability of my system would change, and from what I recall, it didn't improve anything, so I decided to revert it back to the "auto" settings.

A good question here is,what memory settings should I be using during this diagnosis? I have read on numerous occasions to turn off all overclocks, so should I NOT use XMP settings? If so, how should I set the ram timings/voltage/speed? Shall I leave them on "auto" (I recall them being ever so slightly faster than my ram's specs), or should I set them per my ram's specifications? Shall I leave it at 2133Mhz/1.2V?

Lastly, I will update my BIOS version tonight in hopes of improving the stability.

This just seems so odd to me. Why cant this stuff just work right out of the box?

Sorry for all of the questions, but I do have one more good one. What are the chances that I might have a a faulty piece of hardware? If the tests indicate that my memory and SSD are "good," could they still be defective? What about my other components?

Thanks in advance! I am looking forward to another reply.

 

@lifetec, my bad. Didn't read properly *Redface

We can never with 100% accuracy tell if a driver is truly the cause, reason being that hardware problems can be difficult to find out from time to time. IME from the past year, in the cases where hardware was the root cause lots of drivers were blamed as the cause.

Stock settings. Meaning the memory uses its default settings that is compatible with the CPU and mobo.

Difficult to tell, I've seen plenty users who have lots of outdated drivers as the cause but also quite some who have hardware problems.
If the tests indicate that the RAM and SSD are fine, there is still a (small) chance that they are defective. This is because no test is 100%.

 

The fact that results are not 100% makes this quite a difficult diagnosis.

Like i mentioned previously, I am going to do this tonight:

• Update Bios on motherboard
• Use stock settings for ram (includes tweaking auto timings of mobo to match ram's)
• uninstall the NVIDIA audio driver

If I continue to get crashes, then I will play around with the memory sticks. If the memory is faulty, I will make the assumption that it is only one stick, and use one stick at a time in different slots in an attempt to find the bad stick or slot.

If none of this works, then I will be completely dead in the water.

I will reply back with my results.

Thanks for the help and advice!