Question Computer fails to boot

[AO23]

Hi,

I bought this custom PC off of eBay. The first thing I see when I open the computer are these white and green lines, and then it redirects to the BSOD screen and restarts.

I clean installed the GPU drivers with DDU in safe mode, and it got rid of the problem for 2 days, but now it's back. The drivers I'm using are 19.9.2. I've done a chkdsk /f already and it found no problems. I've already stopped Windows from automatically updating drivers. CrystalDiskInfo says my SSD is good. I've already planned a memory diagnostic once I restart, but I don't want to do it yet since the computer might not boot up again. I tried downloading DirectX from Microsoft's website, but it said I already had the updated version.


Here are my specs:
http://speccy.piriform.com/results/kKX3Ugb47wHPkAa0g60nDdD

Here are my minidump details from WhoCrashed. I bought the PC in October. I had to do System Restore last week so I actually lost some of the minidumps.

On Sat 11/16/2019 6:38:12 PM your computer crashed or a problem was reported
crash dump file: C:\Windows\MEMORY.DMP
This was probably caused by the following module: atikmdag.sys (atikmdag+0x722BA)
Bugcheck code: 0xEA (0xFFFF8E8C7AB3A080, 0x0, 0x0, 0x0)
Error: THREAD_STUCK_IN_DEVICE_DRIVER
file path: C:\Windows\System32\DriverStore\FileRepository\u0346940.inf_amd64_1ea00c8019a8594e\B346681\atikmdag.sys
product: ATI Radeon Family
company: Advanced Micro Devices, Inc.
description: ATI Radeon Kernel Mode Driver
Bug check description: This indicates that a thread in a device driver is endlessly spinning.
This appears to be a typical software driver bug and is not likely to be caused by a hardware problem.
A third party driver was identified as the probable root cause of this system error. It is suggested you look for an update for the following driver: atikmdag.sys (ATI Radeon Kernel Mode Driver, Advanced Micro Devices, Inc.).
Google query: atikmdag.sys Advanced Micro Devices, Inc. THREAD_STUCK_IN_DEVICE_DRIVER



On Sat 11/16/2019 6:38:12 PM your computer crashed or a problem was reported
crash dump file: C:\Windows\Minidump\111619-7171-01.dmp
This was probably caused by the following module: dxgkrnl.sys (dxgkrnl+0x40DB5)
Bugcheck code: 0x100000EA (0xFFFF8E8C7AB3A080, 0x0, 0x0, 0x0)
Error: THREAD_STUCK_IN_DEVICE_DRIVER_M
file path: C:\Windows\system32\drivers\dxgkrnl.sys
product: Microsoft® Windows® Operating System
company: Microsoft Corporation
description: DirectX Graphics Kernel
Bug check description: This indicates that a thread in a device driver is endlessly spinning.
This appears to be a typical software driver bug and is not likely to be caused by a hardware problem.
The crash took place in a Microsoft module. Your system configuration may be incorrect. Possibly this problem is caused by another driver on your system that cannot be identified at this time.



On Tue 11/12/2019 6:42:39 PM your computer crashed or a problem was reported
crash dump file: C:\Windows\Minidump\111219-7718-01.dmp
This was probably caused by the following module: dxgkrnl.sys (dxgkrnl+0x40DB5)
Bugcheck code: 0x100000EA (0xFFFF9586E4EB1040, 0x0, 0x0, 0x0)
Error: THREAD_STUCK_IN_DEVICE_DRIVER_M
file path: C:\Windows\system32\drivers\dxgkrnl.sys
product: Microsoft® Windows® Operating System
company: Microsoft Corporation
description: DirectX Graphics Kernel
Bug check description: This indicates that a thread in a device driver is endlessly spinning.
This appears to be a typical software driver bug and is not likely to be caused by a hardware problem.
The crash took place in a Microsoft module. Your system configuration may be incorrect. Possibly this problem is caused by another driver on your system that cannot be identified at this time.



On Tue 7/30/2019 5:44:34 PM your computer crashed or a problem was reported
crash dump file: C:\Windows\Minidump\073019-7156-01.dmp
This was probably caused by the following module: dxgkrnl.sys (dxgkrnl+0x40D35)
Bugcheck code: 0x100000EA (0xFFFFB985CDF92180, 0x0, 0x0, 0x0)
Error: THREAD_STUCK_IN_DEVICE_DRIVER_M
file path: C:\Windows\system32\drivers\dxgkrnl.sys
product: Microsoft® Windows® Operating System
company: Microsoft Corporation
description: DirectX Graphics Kernel
Bug check description: This indicates that a thread in a device driver is endlessly spinning.
This appears to be a typical software driver bug and is not likely to be caused by a hardware problem.
The crash took place in a Microsoft module. Your system configuration may be incorrect. Possibly this problem is caused by another driver on your system that cannot be identified at this time.

 

[Ralston18]

What PSU is installed: make, model, wattage, age, condition? Reason: the PSU may simply be unable to provide the necessary power needed by the PC.

Reliability History and Event Viewer may be capturing error codes and warnings related to the boot failures.

Reliability History is much more user friendly so look there first.

Go to Task Manager > Startup. Disable any apps that are not immediately needed or necessary. Simplify boot up.

The computer may have an updated version of DirectX. However; that does not rule out that the current installation could be buggy or corrupted.

I would reinstall anyway. Maybe with the previous version as a starting point.

Try running the built in Windows' Troubleshooters - the troubleshooters may find and fix something.

Likewise try running "sfc /scannow" via the command prompt. Doing so may find and fix some problems

Reference:

https://www.lifewire.com/how-to-use-sfc-scannow-to-repair-windows-system-files-2626161

Forego DDU: Manually reinstall driver updates yourself via the manufacturer's website. Download, install, and reconfigure as necessary but only one driver at a time and test in between each reinstall.

 

[AO23]

The PSU is a EVGA 500 80+ Bronze. I've already done the SFC and it found no problems. The Reliability History says that it was a Hardware Error. So does that mean the GPU is definitely the problem?

How do I reinstall DirectX? Also, do you have a good guide on how to reinstall drivers manually if you're trying to roll them back?

If I shut down and turned it on right now there wouldn't be a problem. It's when I shut down and leave it for 10-15 hours that it fails to boot afterwards.

 

[Ralston18]

If the system boots while warm and fails to boot when "cold" (10-15 hours) then I would suspect that some contraction is creating a short that interferes with rebooting.

The startup process you go through eventually get things warmed up again and the system boots.

Open the case and reseat all cards, cables, jumpers, RAM, etc.. Ensure that all are fully in place and firmly seated.

Replace the CMOS battery just for good measure.

With respect to installing drivers:

DirectX - Read the following link:

https://support.microsoft.com/en-us/help/179113/how-to-install-the-latest-version-of-directx

Microsoft uses the update process to update Direct X.

Drivers in general:

Go to the applicable manufacturer's website and find the support webpages.

Those pages should provide additional links to 1) the necessary driver software, and 2) the driver installation process to be used.

Identifying the applicable driver is generally the most difficult: You must match OS, version, and possible other factors to ensure that the correct driver is downloaded, installed, and configurable to your requirements.

There may be options available. Some options may include other utilities and software beyond just the basic drivers.

Aka "PUPs" - Potentially Unwanted Programs.

Read the driver download pages carefully and be sure that you understand what is being installed. If printable instructions are available then print out those instructions as a checklist to work through.

If you are truly "rolling back" (installing an older driver version) it is likely that once you begin the driver installation process you will get some notice that the current version is newer. And then be asked "do you really want to install an older version driver?" prompt of some sort.

Click yes, accept, agree, etc... as applicable.

 

[AO23]

But how exactly do I reinstall DirectX? It already says I have the latest version of it.

When I first bought the PC, it would randomly shut down every day because the RAM stick wasn't locked in. It took me a week of using it to find that out, andso I've already made sure that everything was secured inside. I didn't touch the CMOS battery though. Is there anything I can check with that?

EDIT: And could I just replace it with any CMOS battery off of Amazon?

EDIT 2: I also thought I overclocked the RAM in the BIOS to 2667 MHz, but according to CPU-Z, it's only at 1333.

 

[AO23]

When I just tried to turn the computer, I got two BSODs. I downgraded to the 19.6.2 drivers. I also uninstalled Rivaturner Statistics.

Also, I clicked Install instead of Clean Install. Was that the right choice?

EDIT: One more thing I forgot to add. The seller did not ship the PC with the GPU and CPU separate. Could that have caused a problem?

EDIT 2: I just tried resetting Windows Update, but I came across two issues here.

https://docs.microsoft.com/en-us/windows/deployment/update/windows-update-resources

When I put in 'net stop wuauserv' it said 'The Windows Update Service was already stopped, but I was able to start it back up. I then tried Del "%ALLUSERSPROFILE%\Application Data\Microsoft\Network\Downloader\qmgr*.dat" but it said it couldn't find it, and when I tried registering the BITS files, it couldn't find like half of them.

 

[AO23]

Bump. Sorry about all the posts. I'm just trying to see if I can solve it on my own at the same time.

 

[AO23]

Forget everything above. I need to know how to reinstall DirectX. It took me one and a half hours to boot up today. I wasn't even able to use System Restore. I was forced to use chkdsk.

 

[xlokwinterx]

I would first look at this thread:

https://forums.tomshardware.com/faq...n-install-of-your-video-card-drivers.2402269/

See if you can do a full deletion and reinstall of your Display Drivers (not sure if DDU will do directx as well)

Then see what state you're in.

 

[xlokwinterx]

As for CMOS batteries, you need to take it out and get all the details that are written on it and find the closest match, especially the same voltage.

 

[AO23]

I would first look at this thread:

https://forums.tomshardware.com/faq...n-install-of-your-video-card-drivers.2402269/

See if you can do a full deletion and reinstall of your Display Drivers (not sure if DDU will do directx as well)

Then see what state you're in.

Thank you, but I've already done this. It worked for 2 days and then a Windows Update ruined it. I need to know which Windows Update included a DirectX update as well. There was a System Restore point created when the problem first started called 'Installed DirectX'.

 

[xlokwinterx]

So update me please, is it crashing while running or is the problem only when you shut it down and attempt a startup?

EDIT: does it run at all or do you just see the green and white lines and then immediately BSOD?

 

[AO23]

I've using it right now, but when I shut down and leave it alone for some time, it refuses to boot up. It sometimes takes only 1 or 2 restarts for the lines to go away and to go to the login page, but today it took over an hour. Even when I booted to the BIOS, I would still see the lines. When I tried to use System Restore, it said there was an error and told me to do chkdsk /r, but I did chkdsk and it found no errors.

It ended up doing a system restore point because everything I did yesterday was reversed, but I don't think it did it fully, and I didn't get to choose the restore point.

I took some pictures of the screens that I'll upload shortly.

EDIT: Here they are.
https://i.ibb.co/xqn37g6/20191119-184922.jpg
https://i.ibb.co/1v3MR4S/20191119-190550.jpg
https://i.ibb.co/5n6Yfqt/20191119-185316.jpg


EDIT 2: I still can't use System Restore. It keeps telling me to try chkdsk even though I just did it. So the problem is the SSD.

What should I do now? I already filed a claim on eBay to get a new SSD.

 

[xlokwinterx]

Hmm it might be the SSD but both of those crash dump codes you posted are GPU related, perhaps its just that the drive is corrupt and is making the drivers fail, but i would assume that a graphical glitch like the lines would also point to a display related issue.

 

[xlokwinterx]

When are the BSOD's happening? Is it during the boot up sequence or after it has booted?

 

[AO23]

When are the BSOD's happening? Is it during the boot up sequence or after it has booted?

Right when I boot up I see black and white lines, and then during every screen afterwards I see what you see in the images. They'll randomly go away after a restart or shutdown after some time, but the computer usually crashes the first time it happens while I'm using it.

The seller shipped the computer without the GPU separate. Could that be the problem? I noticed that the first HDMI slot on the GPU doesn't even work.

EDIT: Also, if the SSD isn't the issue, then what's going on with my System Restore? Could both be defective?

EDIT2: One thing I didn't do after using DDU was installing the chipset drivers. How do I know which ones to download? And how do I install them?

EDIT3: Sorry for all the edits. I decided to use DDU to uninstall all the drivers and then skip on reinstalling the drivers. So I'm currently using the computer with no display drivers, but it still works. Would you advise against this?

 

[AO23]

Bump. Should I just make new posts instead of editing my posts?

 

[xlokwinterx]

Tbh with you i defiantly think its a hardware issue by the fact that you are still getting the lines in the BIOS menu and my bet would be on it being GPU related. As for the system restore im unsure to be honest it may be that the SSD is at the end of its life but there are ways to check its health.

I would do the following:

• Check the manufacturer of the SSD and HDD(if you have one), google to see if the manufacturer distributes a free diagnostic software to check the status of the drives integrity.
• Do a complete reinstall of windows within windows using the first option on this website: https://www.easeus.com/todo-backup-resource/reinstall-windows-10-without-cd.html
• Re-seat the GPU(make sure its firmly placed in the slot) and check to see if there is any scorching on any of the pins on the GPU its self or within the PCIE slot. EDIT: Make sure first tho that you switch the power off on the outside of the PSU and even dissconnect it from power if youre unsure.

Do these things and see how you go.

 

[xlokwinterx]

Bump. Should I just make new posts instead of editing my posts?

No its good to keep the whole issue within one thread so that anyone else that experiences this problem can work through the steps we take.

 

[AO23]

Tbh with you i defiantly think its a hardware issue by the fact that you are still getting the lines in the BIOS menu and my bet would be on it being GPU related. As for the system restore im unsure to be honest it may be that the SSD is at the end of its life but there are ways to check its health.

I would do the following:

• Check the manufacturer of the SSD and HDD(if you have one), google to see if the manufacturer distributes a free diagnostic software to check the status of the drives integrity.
• Do a complete reinstall of windows within windows using the first option on this website: https://www.easeus.com/todo-backup-resource/reinstall-windows-10-without-cd.html
• Re-seat the GPU(make sure its firmly placed in the slot) and check to see if there is any scorching on any of the pins on the GPU its self or within the PCIE slot. EDIT: Make sure first tho that you switch the power off on the outside of the PSU and even dissconnect it from power if youre unsure.

Do these things and see how you go.

What are the pins on a GPU? What would scorching look like. I made sure it was seated properly yesterday, but I noticed that the click underneath was moving weirdly. It did click, but it barely moved.

There was a firmware upgrade program on the PNY website. It said to do it in AHCI Mode. How would I do that? Or could I just do it in Normal mode? Also, it looks like it's telling me to downgrade. Am I looking at this right?

Pic:
https://i.ibb.co/Y8JW1HG/ssd.png

 

[xlokwinterx]

The "pins" are the gold surfaces where the GPU inserts into the motherboard, look for any discolouration and signs that there has been burning or damage on those surfaces and also shine a light down into the port that the GPU sits in and see what they look like as well.

 

[AO23]

I'll do that later today. I'll do the Windows 10 reinstall on Friday when I have time. Do you know how long it will last? What should I expect once it's done?

Also, my new SSD will be here on Friday. If I clone it, I won't have to install Windows all over again, right?

 

[xlokwinterx]

It takes about an hour to reinstall (or at least is does for me) and fingers crossed after it you stop having troubles but i wouldn't get your hopes up, basically you want to try and eliminate the possibility of it being a software problem by doing things like fresh installs and booting in safe mode before you have to start buying new parts to figure out what part is broken.

As for the SSD if you know how to clone the drive i would do that but i would still suggest you do a fresh reinstall afterwards just to ensure that none of your windows files are corrupt, which allows you to rule out that as being the possible reason for the problem. If you're simply doing system restores and there is a corrupt file then you would just be reinstalling that corrupt file back onto the computer.

 

[AO23]

When I booted up today, I saw a blue screen with an option to use System Restore after 2 BSODs and it actually worked. I installed the 19.11.2 drivers last night, and it restored back to when I was using the computer without display drivers.

So I'm able to use System Restore then but not in normal and safe mode. What could this mean?

EDIT: Also, what would reinstalling Windows do that the System File Checker couldn't?

 

[AO23]

Bump.