KMode_Exception_Not_Handled BSOD during Stress Test (Not Overclocked)

[Triage]

OS - Windows 10 Pro 64-bit
What OS was originally installed on the system? No
Is the OS an OEM version (came pre-installed on system) or full retail version (YOU purchased it from a retailer)? Full retail version
Age of system (hardware): <1.5 years
Age of OS installation - have you re-installed the OS? Yes, I just reinstalled the OS 4 hours or so ago.
CPU model: i7-6700K
Video Card model: 1 EVGA GeForce GTX 980 4GB Gaming ACX 2.0, 1 MSI GTX 980 4GB, and 1 MSI GTX 960 2GD5
MotherBoard: MSI Z170A Gaming M5 Motherboard
Power Supply - brand & wattage: Corsair RM1000, 1000W
Laptop or Desktop? Desktop

AIDA64: Version 5.97.4600

 

https://www.dropbox.com/s/jazeool9cuqrl32/KMode Service Exception Error.zip?dl=0

https://www.dropbox.com/s/jsr07vknkws6cqc/Kmode Service Exception Error Resource %26 Performance Monitor File.zip?dl=0

 

Whenever I run AIDA64's System Stability Test (with Stress CPU, Stress FPU, Stress cache, and Stress system memory checked), my computer will BSOD saying KMode_Exception_Not_Handled. It doesn't list a driver causing the error. None of my hardware is overclocked (CPU, RAM, etc.).  The error can occur at any time during the stress test (whether it be 9 minutes into the stress test or 39 minutes into it). So far the things I've tried to fix this issue include: Running Windows Memory Diagnostics Tool to check for errors-- found none, Reflashing the bios, Running Memtest86 to check for errors-- which it found none, Checking drivers for updates on MSI's website, and performing a clean install of Windows 10. I thought for sure that a clean install of Windows 10 would fix the issue, but that wasn't the case, unfortunately.  I don't have any antivirus installed on my clean install of Windows (except for whatever Windows comes with) so I don't believe that is causing the issue somehow. I would really appreciate some help with this BSOD error.

[Tabs]

kernel ode exceptionnot handled generally refers to one of two things.

 

1. Most commonly: Kernel ode drivers doing something they shouldn't, and then crashing in an unclean way.

 

2. Hardware errors.

 

Since 2 in this case might be easier to diagnose, I'll start there - What memory do you have installed and what speed is it running at? Do you have multiple modules and have you tried running them individually?

[Triage]

7 minutes ago, Tabs said:

kernel ode exceptionnot handled generally refers to one of two things.

 

1. Most commonly: Kernel ode drivers doing something they shouldn't, and then crashing in an unclean way.

 

2. Hardware errors.

 

Since 2 in this case might be easier to diagnose, I'll start there - What memory do you have installed and what speed is it running at? Do you have multiple modules and have you tried running them individually?

3

Thank you for the response, Tabs! I have 4 sticks of 8 GB ADATA XPG Z1 DDR4 2400MHZ. I'm running them at 2133 MHz according to my BIOS: 

 

 

I have not tried running them individually yet. MemTest86 and Windows Memory Diagnostic Tool didn't report any errors though with all 4 sticks in. Would I get a different result if I tested them individually?

[Tabs]

As long as you ran them for a decent length of time, I expect not, which rules out some of the low level hardware issues (mainly ram) that i checked for already.

 

May I ask when you first noticed the stress test causing problems? Has it always done so?

 

Have you installed/used any software since reinstalling windows, other than aida64?

 

Have you checked what temperatures your components run at during the test? Do you have airflow over your cpu area (i.e. if you have an all-in-one loop, do you have any fans nearby to keep the vrm's cool?)

[Triage]

17 minutes ago, Tabs said:

As long as you ran them for a decent length of time, I expect not, which rules out some of the low level hardware issues (mainly ram) that i checked for already.

 

May I ask when you first noticed the stress test causing problems? Has it always done so?

 

Have you installed/used any software since reinstalling windows, other than aida64?

 

Have you checked what temperatures your components run at during the test? Do you have airflow over your cpu area (i.e. if you have an all-in-one loop, do you have any fans nearby to keep the vrm's cool?)

5

 

Well, I first noticed the issue when I was trying to overclock my CPU over the last few days after I had carefully delidded my CPU with Rockit 88 Delid Tool. The results of my overclocking trials were all over the place so I decided to try running everything at stock to see if there would be any issues. My CPU temps are really solid during the test (usually under 50 C when it crashes) and none of the other components stood out as having a high temperature. My CPU cooler is an EVGA CLC 280. I use a push-pull config on it. I also have Corsair Dominator Fans on top of my RAM sticks. 

 

Edit: I haven't installed anything that should interfere with AIDA64. Just a few apps from NInite Installer and my hardware drivers is all I've installed.

[Tabs]

5 minutes ago, Triage said:

 

Well, I first noticed the issue when I was trying to overclock my CPU over the last few days after I had carefully delidded my CPU with Rockit 88 Delid Tool. The results of my overclocking trials were all over the place so I decided to try running everything at stock to see if there would be any issues. My CPU temps are really solid during the test (usually under 50 C when it crashes) and none of the other components stood out as having a high temperature. My CPU cooler is an EVGA CLC 280. I use a push-pull config on it. I also have Corsair Dominator Fans on top of my RAM sticks. 

Do you have any fans around your actual cpu area though? The fans on your memory are going to end up blowing most of their air down through the sticks and then vertically up through the channels between them, so it won't come close to the CPU VRM area.

 

The VRM is the area to the left and top of the cpu socket (with the ram at the right hand side, for positioning). You'll want at least a little airflow there to ensure they don't get too hot.

 

Also, since you mentioned you delidded, have you ensured that no capacitors or resistors on the top (near the die) or bottom (near the pads) of the cpu were damaged during the attempt? Those caps and resistors help to filter the voltage the cpu runs on, so if they get damaged the cpu may fail purely because the current it's receiving is too variable to keep it running properly.

[Triage]

14 hours ago, Tabs said:

Do you have any fans around your actual cpu area though? The fans on your memory are going to end up blowing most of their air down through the sticks and then vertically up through the channels between them, so it won't come close to the CPU VRM area.

 

The VRM is the area to the left and top of the cpu socket (with the ram at the right hand side, for positioning). You'll want at least a little airflow there to ensure they don't get too hot.

 

Also, since you mentioned you delidded, have you ensured that no capacitors or resistors on the top (near the die) or bottom (near the pads) of the cpu were damaged during the attempt? Those caps and resistors help to filter the voltage the cpu runs on, so if they get damaged the cpu may fail purely because the current it's receiving is too variable to keep it running properly.

7

My AIO cooler (with the push-pull) is directly above the VRM area. Is that enough? When I delidded, I was really careful not to damage any of the capacitors/resistors so I don't think those are causing the issue. 

 

Edit: I just ran IntelBurnTest on Very High and it passed. I'm going to run it on Maximum now and see if I get the same results then try a different stress testing application.

 

Edit #2: It passed the Maximum test a couple of times. I'm going to try one of the safer versions of Prime95 now.

[Triage]

I have a strong suspicion it might be related to my Power Supply. Is there any way for me to test if my PSU is bad? 

[Triage]

Hey LTTer's,

I'm still having random BSODs. Can someone please help me? I'm getting Kmode and Memory Management BSODs now. I'm not sure what to do. =/ This was a clean install of Windows 10 Pro April Update.

 

 

[Triage]

It seems someone else is having similar issues I think -- https://forums.aida64.com/topic/4210-aida-64-make-my-pc-bsod-kmode-execption-not-handled/