Intermittent BSOD - Possible root cause?

albigularis · 15 Dec 2016

So I've been having quite a time with this computer. I upgraded a bunch of the hardware inside, nothing major, a 750ti/i3 setup just to game lightly on.

It started out fine and over the course of a few weeks, I got more and more frequent BSODs, usually with MEMORY_MANAGEMENT, IRQL_DRIVER_NOT_LESS_OR_EQUAL, BAD_POOL_HEADER, FAULTY_HARDWARE_CORRUPTED_PAGE or KERNEL_SECURITY_CHECK_FAILURE.

They seemed to get more frequent and running System File Checker returned that there was corrupted files which could not be repaired. I then tried to do the built-in system reset, which actually failed. At this point I fully re-installed Windows 10 using the ISO from the Microsoft website. System ran perfectly for maybe two days, then built up to maybe 1 BSOD per day, now I'm on between 4-6.

I've ran memtest on 10 parses overnight, nothing found.

Drivers are all fresh installs and are most recent. The issues I had last time still occured on the previous versions of the same drivers which are installed.

Dumps are included below, but before examining them, is it likely that with a fresh install and the issue getting worse, it's actually a hardware issue?

I ran the SFC checker again and it gave the error- "Windows Resource Protection found corrupt files but was unable to fix some of them". I have included the logs it directed me to in the upload below, inside the "CBS" folder.

albigularis · 15 Dec 2016

Slight update, although unsure how relevant.

Was sitting listening to music, so had iTunes open whilst opening the Battle.net client. The computer done its usual bluescreen ritual where the sound stops and repeats the last few milliseconds of audio. Except this time it didn't blue screen and the audio reset to my music after about 5 seconds. The screen froze for about 25s and it became extremely unresponsive, but it settled down after about 2-min. Whilst unresponsive, the cursor was taking like 15s to register movement etc. I checked task manager and the only thing using high amounts of anything was "Service Host: Local System" which was using about 50% of Disk capacity.

Had about 4 or 5 Bluescreens today, all have been "MEMORY_MAMANGEMENT" I think. Happened when playing games, and whilst just browsing in Chrome.

lifetec · 16 Dec 2016

Your on BIOS version F5 of your mobo H81M-H.
You said you installed a new cpu, this is a 4th Generation Intel® Core™ i3 Processors.
According to your supportpage BIOS version F7 has Support New 4th Generation Intel Core Processors and
Enhance DDR capability

Update your BIOS to the latest version.

albigularis · 16 Dec 2016

lifetec said:

Your on BIOS version F5 of your mobo H81M-H.
You said you installed a new cpu, this is a 4th Generation Intel® Core™ i3 Processors.
According to your supportpage BIOS version F7 has Support New 4th Generation Intel Core Processors and
Enhance DDR capability

Update your BIOS to the latest version.

Thank you for taking the time to help me out :). I've updated the BIOS to F7 now, I will report back if the issues persist with more dumps!

albigularis · 16 Dec 2016

Bad news already, had another MEMORY_MANAGEMENT bluescreen a few minutes ago.

lifetec · 16 Dec 2016

The last dumpfile is again memory management as you said.

Code:

BugCheck 1A, {41793, fffff700d83b2ff8, 200, 1ff}

Probably caused by : memory_corruption

Followup: memory_corruption

Usual causes:* Device driver, memory, kernel

The only older driver your system has is the Logitech gaming software driver it is from 2015, try to update it.

Memory: you said you already run Memtest with no errors.
Acccording to the QVL list your memory needs to be placed in the 3_1 socket/channel A., see supportpage.
And you have your memory placed in 3_2 socket/channel B

Code:

 Memory Device (Type 17) - Length 34 - Handle 0043h]
  Physical Memory Array Handle  0007h
  Memory Error Info Handle      [Not Provided]
  Total Width                   64 bits
  Data Width                    64 bits
  Size                          8192MB
  Form Factor                   09h - DIMM
  Device Set                    [None]
  Device Locator                ChannelB-DIMM0
  Bank Locator                  BANK 2
  Memory Type                   18h - Specification Reserved
  Type Detail                   0080h - Synchronous
  Speed                         1600MHz
  Manufacturer                  0215
  Serial Number                         
  Asset Tag Number                        
  Part Number                   CMV8GX3M1A1600C11

As a test replace your memory in Socket/channel A.

Kernel:. rerun the SFC command again, use this tutorial.
SFC
Use SFC to check and repair windows systemfiles.

See this tutorial SFC /SCANNOW Command - Run in Windows 8
SFC Command - Run in Windows 10 - Windows 10 Tutorials

Note
If SFC could not fix something, then run the command again to see if it may be able to the next time. Sometimes it may take running the sfc /scannow command 3 times restarting the PC after each time to completely fix everything that it's able to.

If not, then run the Dism /Online /Cleanup-Image /RestoreHealth command to repair any component store corruption, restart the PC afterwards, and try the sfc /scannow command again.

albigularis · 16 Dec 2016

Amazingly, that Logitech driver got updated only yesterday! That was the last version they made. Only a year between updates... Updated it and switched the RAM stick to the other slot. I'll go through the SFC stuff now and continue using the computer afterwards, hoping for the best haha.

albigularis · 16 Dec 2016

So SFC returned clean this time, first time it's ever done so. Updated those drivers and changed the RAM to the other slot.

Just got another one, this time "FAULTY_HARDWARE_CORRUPTED_PAGE". Log attached.

lifetec · 17 Dec 2016

The last dumpfile points again to your memory, this time even with a hardware memory fault

Code:

   BugCheck 12B, {ffffffffc00002c4, 7a9, 2cb47900, ffffa700946c6000}

*** WARNING: Unable to verify timestamp for win32k.sys
*** ERROR: Module load completed but symbols could not be loaded for win32k.sys
Probably caused by : memory_corruption

Followup: memory_corruption
---------

2: kd> !analyze -v
*******************************************************************************
*                                                                             *
*                        Bugcheck Analysis                                    *
*                                                                             *
*******************************************************************************

FAULTY_HARDWARE_CORRUPTED_PAGE (12b)
This bugcheck indicates that a single bit error was found in this page.  This is a hardware memory error.
Arguments:
Arg1: ffffffffc00002c4, virtual address mapping the corrupted page
Arg2: 00000000000007a9, physical page number
Arg3: 000000002cb47900, zero
Arg4: ffffa700946c6000, zero

Reseat it to the second slot of channel A.
Check the settings of your memory in the BIOS.

Memory Size
8GB Kit (1 x 8GB)
SPD Latency
9-9-9-24
SPD Speed
1333MHz
SPD Voltage
1.5V
Speed Rating
PC3-12800 (1600MHz)
Tested Latency
11-11-11-30
Tested Speed
1600MHz
Tested Voltage
1.5V

Test it again with Memtest 86+.
Memory Test.
Please start with diagnosing the system RAM first. You can use the free tool "Memtest86+" to stress test the RAM.

How to run "Memtest" - RAM - Test with Memtest86+ - Windows 7 Help Forums
MemTest86+ - Test RAM - Windows 10 Tutorials
Make sure you are using the latest version 5.01 - Memtest86+ - Advanced Memory Diagnostic Tool

You can stop the test even if a single error is found during the test. Otherwise continue the test for 8-10 passes and post a screenshot with next reply. 8 passes is the recommended bare minimum but more passes will give better result.

albigularis · 17 Dec 2016

Each channel only has one slot on this board. So it's in the right channel now, I'll try another 8 parses and see if I get anything else showing up this time.