Endless BSODs + Tasks Being Killed

bjgrenke · 25 Apr 2019

Just completed 8 passes of memtest86+ with no errors.
Attachment 231705

I also ran the Intel Processor Diagnostic Tool, which came back clean.
Attachment 231706

As for the BIOS update, I updated it yesterday, but the problems persisted exactly the same before and after I updated the BIOS.

Here is the PoW32kWatchdog-20190424-1934.dmp

After collecting this info, I got a long freeze-up followed by a DPC_WATCHDOG_VIOLATION BSOD. Here is the memory.dmp containing that dump.

Nothing is under warranty, I built this system more than 6 years ago. I don't have any other hardware available for swap testing. If I can be certain the CPU is the only issue, I can get another 2500k for cheap. Otherwise I will wait a bit and do a full upgrade of CPU/RAM/Motherboard.

Just started running Prime95, stable so far with nothing over 50C. I'll let it run for a few hours.

Edit: After around 40 minutes of Prime95, I got BSOD. Minidump here Attachment 231708

zbook · 25 Apr 2019

1) Make a backup image using Macrium:
Macrium Software | Your Image is Everything

2) Save the backup image to another disk drive (not another partition on the same drive) or to the cloud.

3) Make a new restore point:
Create System Restore Point in Windows 10 | Tutorials

4) Windows driver verifier is a tool that is used to find misbehaving drivers:
Enable and Disable Driver Verifier in Windows 10 | Tutorials
After each BSOD you will use the computer's windows advanced troubleshooting menu to turn off or recover from using the tool.
After each BSOD when back on the desktop post a new BETA zip and memory.dmp share link into the thread.

While using the tool the computer performance may become sluggish.
The boot may also become slow.

All non-Microsoft drivers are tested simultaneously.
Start testing with the 3 customized tests in this link:
Enable and Disable Driver Verifier in Windows 10 | Tutorials

The mini dump was bugcheck 50.
The memory dump was bugcheck 133 parameter 1.
The kernel dump and the above dumps did not display misbehaving drivers.

Please use the BETA log collector to collect files after each BSOD and manually zip the memory.dmp

bjgrenke · 25 Apr 2019

After starting Driver Verifier and rebooting, I immediately got a chain of BSODs. IRQL_NOT_LESS_OR_EQUAL, PAGE_FAULT_IN_NONPAGED_AREA, and multiple SYSTEM_SERVICE_EXCEPTIONs.

Attachment 231712

Memory.dmp

zbook · 25 Apr 2019

Please restart windows driver verifier to create more dump files for debugging.

Create multiple memory dump files.

See post #3: uncheck overwrite existing file and reboot to maintain settings.

Also check the box to prevent automatic deletion.

If there are no BSOD using Windows driver verifier with the current 3 tests, there are additonal tests to produce BSOD.

bjgrenke · 25 Apr 2019

I unchecked the overwrite existing file box and for some reason the Memory.dmp file is not being updated. I've had another string of BSODs with Driver Verifier running, but the memory.dmp file hasn't been modified since I last posted it.

It seems with Driver Verifier running I get BSODs in bunches, I'll get one then another one or two immediately on reboot, then it will be stable with driver verifier running for 20-30 minutes. This has happened twice, but the memory dump hasn't changed.

I'll check the box to prevent automatic deletion and try to get more BSODs with Driver Verifier running to see if the memory.dmp gets updated. Which other tests can I enable to produce more BSOD?

I've been running different games to add more stress, but the games will just crash every 5 minutes and no BSOD.

Edit: I actually checked the overwrite existing file box again and got another 2 BSODs with Driver Verifier running. This time the memory.dmp updated, along with 2 new minidumps have been added.

Zip file here.

- - - Updated - - -

After running stable playing games with Driver Verifier off for 2+ hours, I got another wave of BSODs in a row.

Memory.dmp and 2 more minidumps

zbook · 25 Apr 2019

There are currently 3 simultaneous tests with the Ten Forums link.

This Microsoft link has approximately 19 simultaneous tests:

Driver Verifier-- tracking down a mis-behaving driver. - Microsoft Community

Increase the number of simultaneous tests by 3 - 4 / hour.
If there are performance or boot problems you can remove some and keep adding so that you nearly reach or reach the 19 simultaneous tests.

So far there have been no drivers identified in the dump files(less the ones that were just posted).

If there are no drivers then testing will be planned in safe mode with networking to evaluate stability / instability.

bjgrenke · 26 Apr 2019

Can't seem to get any bsod's from Verifier anymore. With all the tests enabled, things are stable yet sluggish for 5-10 minutes, then the PC will freeze up, no BSOD.

I've booted into safemode with networking, running Prime95 again. Any specific testing I should perform in safemode?

zbook · 26 Apr 2019

There is no special testing in safe mode.
It is by duration of time.
Is the computer stable or unstable (unexpected shutdown or restart with or without BSOD.

Windows driver verifier testing can be discontinued.
Make sure the tool is off.

After prime95 testing has completed:

Run FurMark - GPU Stress Test Windows 10 Performance Maintenance Tutorials

FurMark - GPU Stress Test Windows 10 Performance Maintenance Tutorials:
FurMark - GPU Stress Test | Tutorials

Please make note of the importance of temperature monitoring and aborting the test in the GPU link.
a) Record the maximum temperature and post the maximum temperature into the thread.
b) Record the test duration and post the uninterrupted test duration into the thread.
c) Aim for testing 1 hour.
d) Use the Microsoft snipping tool to post images into the thread.

Take Screenshot in Windows 10 Windows 10 General Tips Tutorials
Take Screenshot in Windows 10 | Tutorials

Open administrative powershell and copy and paste:

Code:

[Cmdletbinding()] 
Param( 
    [string]$Computername = "localhost" 
) 
cls 
$PysicalMemory = Get-WmiObject -class "win32_physicalmemory" -namespace "root\CIMV2" -ComputerName $Computername 
 
Write-Host "Memore Modules:" -ForegroundColor Green 
$PysicalMemory | Format-Table Tag,BankLabel,@{n="Capacity(GB)";e={$_.Capacity/1GB}},Manufacturer,PartNumber,Speed -AutoSize 
 
Write-Host "Total Memory:" -ForegroundColor Green 
Write-Host "$((($PysicalMemory).Capacity | Measure-Object -Sum).Sum/1GB)GB" 
 
$TotalSlots = ((Get-WmiObject -Class "win32_PhysicalMemoryArray" -namespace "root\CIMV2" -ComputerName $Computername).MemoryDevices | Measure-Object -Sum).Sum 
Write-Host "`nTotal Memory Slots:" -ForegroundColor Green 
Write-Host $TotalSlots 
 
$UsedSlots = (($PysicalMemory) | Measure-Object).Count  
Write-Host "`nUsed Memory Slots:" -ForegroundColor Green 
Write-Host $UsedSlots 
 
If($UsedSlots -eq $TotalSlots) 
{ 
    Write-Host "All memory slots are filled up, none is empty!" -ForegroundColor Yellow 
}

When these have completed > right click on the top bar or title bar of the administrative command prompt box > left click on edit then select all > right click on the top bar again > left click on edit then copy > paste into the thread

bjgrenke · 26 Apr 2019

Just trying different things here, I went into a clean boot according to this guide (disabled all non-microsoft services) and still managed to recreate the BSOD. Here is the memory.dmp and minidump for that BSOD, if it helps at all.

Here is the PowerShell output

Code:

Memore Modules:

Tag               BankLabel Capacity(GB) Manufacturer PartNumber         Speed
---               --------- ------------ ------------ ----------         -----
Physical Memory 1 BANK 1               4 029E         CML8GX3M2A1600C9    1333
Physical Memory 3 BANK 3               4 029E         CML8GX3M2A1600C9    1333


Total Memory:
8GB

Total Memory Slots:
4

Used Memory Slots:
2
PS C:\Windows\system32>

Prime95 again showed no signs of instability whether it be in a normal boot, clean boot, or safemode. I will run FurMark in each of these scenarios to test for any instabilities.

I'm finding that the BSODs are very easy to recreate just by opening multiple applications such as Chrome, Discord, and a game. It will be stable until Chrome extensions and/or pages crash, then applications may close or game will crash, then eventually a BSOD.

Edit: Managed to recreate the BSOD while in safemode just by opening multiple applications and waiting. Memory.dmp and minidump zip

zbook · 26 Apr 2019

After posting images of the FurMark results move the RAM modules to untested DIMM.
Repeat RAM testing so that all DIMM are tested.
Post new images into the thread.

There have been many tests to find misbehaving drivers using windows driver verifier and debugging both mini and memory dumps.
There have been no misbehaving drivers seen in the debugging.

Complete FurMark and DIMM testing then plan to run the computer in safe mode for several days.

Run HD Tune: (free version) (all drives)
Post images into the thread for results on these tabs:
a) Health (SMART)
b) Benchmark
c) Full error scan

Run Sea Tools long generic testing on any HD:
http://www.seagate.com/support/downl...ls-win-master/
How to use SeaTools for Windows | Seagate Support US

If SMART results are not displayed using HD on any drive then run any of these additional tests:

HD Sentinel (trial or free version)
Hard Disk Sentinel - HDD health and temperature monitoring

Crystal Disk
Download Crystal Dew World
CrystalDiskInfo Crystal Dew World
Download Crystal Dew World
Software Crystal Dew World

Speecy
Speccy - Free Download

Samsung Magician
Samsung Magician Software | Samsung V-NAND SSD | Samsung Semiconductor Global Website