BSOD (WHEA_UNCORRECTABLE_ERROR & ntoskrnl.exe)

Hi Guys,


I'm here hoping for some help in diagnosing a Windows 10 BSOD crash I keep getting, I have read the pre-post instructions so will download the Log collector tool once I get home and update thread.
System:
Windows 10 Ultimate x64
200 GB SSD for OS.
32GB DDR3 1600 Hyper Savage RAM
i7 4760k processor with Corsair cooler
ASUS GeForce GTX 750 Ti Graphics Card.

Crash details:
The crashes are quite random Windows
randomly BSOD'ing with an error related to WHEA_UNCORRECTABLE_ERROR.
Initial BSOD viewer dump file [log] suggests the problem is related to the following .exe file ntoskrnl.exe
There doesn't appear to be any pattern to the crash in terms of software used/spefific times etc.Sometimes I get multiple BSOD crashes one after the next and sometimes the OS will run for several hours/days before then suddenly crashing.


Details of Minidump attached.



What I have tried so far:
Ran Memtest 64 for circa 9hrs with no issues found.
Ran CPU stress test for circa 4 hrs with no issues.
Ran GPU stress test for circa 1hr with no issues.
Replaced thermal paste on processor and re-seated cooler.
Replaced ATI graphics card with ASUS Geforce.
Updated BIOS to latest firmware and turned off any OC.
Applied all Windows updates (I keep these updated each week).

I have tried lots of forum threads but unfortunately still get the crash after all of my troubleshooting, would really appreciate some help.

Thanks in advance
Tylor

Hello and welcome to TenForums :)

The minidump text you have provided does not help. Please post a full set of files as requested at the top of the page.

philc43 said:

Hello and welcome to TenForums :)

The minidump text you have provided does not help. Please post a full set of files as requested at the top of the page.

Apologies, I managed to get the server to stay on long enough to run the tool (results attached).

Kind Regards
Tylor
Attachment 110939

These hardware errors are quite difficult to track down and you have to be very thorough with your stress tests. I would repeat the memory test because 9h is nowhere near long enough to complete the eight or more passes required for 32GB of memory. You would need 1-2 days of continuous testing.

Use the tutorials on this site for further information on all the following tests and report back the results.

Memtest86+

Prime95 - Stress Test Your CPU

FurMark - GPU Stress Test

SeaTools for DOS - Hard Drive Diagnostic

Thanks for your reply philc43, much appreciated.

I was experiencing major issues last night which hindered my testing, the machine was shutting down after only a few minutes. Before this happened I did try taking out all 4 ram sticks and replacing one at a time. Unfortunately the pc still crashed whichever stick I used. Not sure I can rule this out in terms of good or bad memory!

I re-seated the CPU albeit i'm not 100% sure what is a good solid fit as first time I have done this. I think I am going to struggle running prolonged stress tests with the time the system is staying up :-(

Could this be related to the SSD at all, I'm reluctant to keep throwing money at parts if they could be fine. Not sure if any of the errors in the log files could point towards SSD?

Thanks again
Tylor

Try to get the system running with as few components attached as possible, also remove the graphics card and use on board graphics if you can. Then you can try a clean boot process to eliminate as much of the software as possible.

If you can get a stable system, gradually add hardware and software back until you find the culprit.

I will do just that, I guess there's also a minimum amount of hardware you physically need to boot as well. I have taken out the graphics card so currently running onboard. Hopefully not but if there is an issue with the processor how will I know as it's not like you can remove for testing (don't have a spare to try)

See if you can get it stable. If not it would suggest motherboard or CPU is the problem, but post a new log file for any new BSOD events and we can see if there are any new clues. There are usually diagnostic programs provided for the SSD which you can use to test that component.

Ok latest update, After more frustration I tried doing a reset of windows from the advanced startup which advised it would wipe software before windows was installed but keep my files. I then managed to keep the pc online for all of yesterday and slowly added some of the essential software back, firewall/email etc. Things were going well with no BSOD until I woke up this morning and found it crashed overnight at 2am.

Luckily I could log straight back in unlike the last attempts and I have ran another log to attach, I can manage to keep the machine up for most of today I will try and run some memory tests.
Attachment 111119

Having said that I think the last software i used was memtest 4.6, is memtest86 the best to use?

Yes, you should use Memtest86+ but I've looked at your error report and can extract the following from the WHEA error record, I've highlighted in red the important bits:

Code:

kd> !errrec ffffb60edb8dc028
===============================================================================
Common Platform Error Record @ ffffb60edb8dc028
-------------------------------------------------------------------------------
Record Id     : 01d244e5c2fc2eb6
Severity      : Fatal (1)
Length        : 928
Creator       : Microsoft
Notify Type   : Machine Check Exception
Timestamp     : 11/23/2016 2:49:11 (UTC)
Flags         : 0x00000000

===============================================================================
Section 0     : Processor Generic
-------------------------------------------------------------------------------
Descriptor    @ ffffb60edb8dc0a8
Section       @ ffffb60edb8dc180
Offset        : 344
Length        : 192
Flags         : 0x00000001 Primary
Severity      : Fatal

Proc. Type    : x86/x64
Instr. Set    : x64
Error Type    : Cache error
Operation     : Generic
Flags         : 0x00
Level         : 0
CPU Version   : 0x00000000000306c3
Processor ID  : 0x0000000000000005

===============================================================================
Section 1     : x86/x64 Processor Specific
-------------------------------------------------------------------------------
Descriptor    @ ffffb60edb8dc0f0
Section       @ ffffb60edb8dc240
Offset        : 536
Length        : 128
Flags         : 0x00000000
Severity      : Fatal

Local APIC Id : 0x0000000000000005
CPU Id        : c3 06 03 00 00 08 10 05 - bf fb fa 7f ff fb eb bf
                00 00 00 00 00 00 00 00 - 00 00 00 00 00 00 00 00
                00 00 00 00 00 00 00 00 - 00 00 00 00 00 00 00 00

Proc. Info 0  @ ffffb60edb8dc240

===============================================================================
Section 2     : x86/x64 MCA
-------------------------------------------------------------------------------
Descriptor    @ ffffb60edb8dc138
Section       @ ffffb60edb8dc2c0
Offset        : 664
Length        : 264
Flags         : 0x00000000
Severity      : Fatal

Error         : DCACHEL0_EVICT_ERR (Proc 5 Bank 1)
  Status      : 0xbb80000000000174

This error points to the CPU cache, it would suggest that the problem may be a faulty CPU but be aware that other things can trigger this such as overheating or overclocking.