Question BSOD: DRIVER_IRQL_NOT_LESS_OR_EQUAL (Problematic driver: ntoskrnl.exe) ?

[Contraquest]

Hello,

I have been trying to fix an issue that I keep getting resulting in BSOD every time I export a video after rendering it or sometimes when I am playing games... I have located the bad driver through Bluescreen Viewer.

Steps tried( Could be missing some):

- I have checked memtest86 for my RAM and it appeared fine.
- I have tried updating drivers both in Windows and GPU.
- I have tried updating drivers through my manufacturer of ASUS.
- I have located the problematic driver of ntoskrnl.exe.
- etc.

Please can someone help me with this and locating the problem or how to update the driver... I have been at this for so long just running around in circles.

Minidump file: https://1drv.ms/f/s!Aj3HgdzJg5ALiYkXItzVbs3l8BCoEg?e=yW95yr

Thank you!!!!

 

[ubuysa]

Just so's you know for the future , ntoskrnl.exe is the Windows kernel. Basic tools like BlueScreeenView report this because that's the component that raised the bugcheck. However, when a third-party driver fouls up, or when there is a hardware issue that fouls up the driver, the problem often isn't discovered until control returns to the kernel and it validates what the driver is asking for. Because the kernel cannot correct driver code it has no choice but to BSOD - with the kernel (ntoskrnl.exe) indicated as the failing component. This is why we always need to see the full dump for all BSODs to be able to back-track to locate the problem driver (or hardware).

In your case, and taking all the dumps as a whole (because together they paint a fuller picture) this would seem to be a RAM issue. The clues that lead me in that direction are these..

• None of the dumps reference any third-party drivers on the call stack leading to the bugcheck - that's often an indication of a hardware issue.
• One dump fails with a 0xC000005 exception code - that's a memory access violation (referencing a page that is either not allocated, paged out, or is bad).
• One dump fails because of a bad instruction pointer.
• One dump fails because of a bad stack-pointer.

It's very difficult to find any other common cause for these particular dumps other than bad RAM. Memtest86 is good, but you need to run it at least twice in succession, and more than twice to find some RAM issues.

You appear to have only a single 32GB RAM stick installed, which isn't really the most sensible config (because you can't take advantage of dual-channel memory) but it also means you can't test your RAM by removing one stick.

I suggest you run Memtest86 (free) up to four times in quick succession; don't leave a gap between ending one set of tests and restarting Memtest86, you need to keep the RAM stressed continuously to find the really niggly problems.

If it passes all four runs of Memtes86 then we'll start looking elsewhere.

 

[Contraquest]

Thank you for getting back to me.

I initially thought that it was RAM and have not fully ruled it out. It just feels weird as I was doing tests with my RAM (I have 2X32GB DDR5 sticks). I have tested each individually and together. I have tested each stick in slots B1, A2, and B2, and the BSOD was still persistent. When I would bluescreen I would swap out the next stick and try again and so on and so forth. That would mean that both sticks are bad no?

I did end up ordering new RAM with a different manufacturer. The current RAM that I do have is from Corsair and my new RAM coming is from Kingston. I have cross referenced the correct RAM that is specified for my motherboard and it shows that they should both be working with it. Do you think XMP could be playing a role with this?

I disabled XMP on my current RAM setup as it was causing more BSOD.

 

[ubuysa]

When you get the new Kingston RAM, disable XMP, and any overclock/undervolt you may have on the CPU, so that everything is running at stock frequencies and voltages. If it BSODs after that then please download the SysnativeBSODCollectionApp, run that and upload the zip file it outputs to the cloud with a link to it here.

This app collects all the troubleshooting info we're likely to need, so it saves coming back again and again asking for something else. It does not collect any personally identifying information - you can look though everything it's collected before uploading if you like, but please don't change anything in there.

 

[Contraquest]

Okay, I will do that. The RAM should be here around the 8th so I will get back to you once I know more.

Thank you again for your help on this and guiding me through these issues!

 

[Contraquest]

So after extensive testing of the new RAM and old RAM. I have found the issue.

As I was going through BIOS options and through the Intel Extreme Tuning Utility I found that my motherboard was overclocking (unstably I presume) to 5.5GHz. When I left it where it was but just with new RAM it did bluescreen again so that also led me to the fact that my RAM was not faulty and that the overclock was the culprit. When I lowered it to 5GHz with better voltages and such I was unable to bluescreen my PC. Even after clearing CMOS, it still overclocked it to 5.5GHz... So I had to manually fix that.

Even though the error codes were pointing at a hardware issue, above that, it was stating it could also be overclocking. I didn't realize that it was overclocked as I never did it myself... So great to know they do that automatically but not stably...

Also, thank you @ubuysa for pointing out your last message of "When you get the new Kingston RAM, disable XMP, and any overclock/undervolt you may have on the CPU, so that everything is running at stock frequencies and voltages." That helped me also realize that I should check overclocking.

TLDR: My system automatically overclocked my CPU improperly so the BSOD's that I was getting was not related to RAM but to the overclock itself. Once removed or lowered everything is okay.

Thank you for your help @ubuysa!!

 

[ubuysa]

Very well spotted! Overclocking is a common source of BSODs but that's the first time I've come across an 'automatic' overclock. Strange.