PC started crashing nightly while idle - No idea on what to do next!

[CorneliousJD]

I will try to keep this brief, while also giving relevant information... 

 

About a week ago my PC started crashing at night, I would wake up and the system would be unresponsive, but powered on, nothing would display on the monitors, nothing woke them up. Only option was to hard reboot.

BSODs logged in event viewer at random times over night. Sometimes 20 mins after I left my PC, sometimes 7 or 8 hours later. 

I was running a 2-year old install of Windows 10 and thought it was time to do a fresh wipe/reload and install Windows 11 anyways. 

 

I started with that, did BIOS update to latest version, installed Win 11, got drivers from Asus (mobo) website, and then the crashes continued overnight again.

Note that I work from home on my PC so I use this thing for 9+ hours a day, some days 12+ hours - no issues, it ONLY crashes overnight. WEIRD!

 

I did a Windows memory diag and it found problems. Cool, easy enough to troubleshoot and solve that.

Removed my DOCP RAM profile, used latest Memtest86+ to test individual DIMMs in different slots, found the bad one and removed it. Re-ran Memtest with 3 of 4 sticks and all came back as passed.

Bad RAM stick is awaiting warranty replacement now. 

 

I assumed things were solved when I removed that bad RAM stick, but once again this morning I woke up to an unresponsive PC.

Now I'm both frustrated and confused.

BSOD logged about 30 mins before I got to my PC. 

There was NOT a BSOD logged the night prior but I had experienced the same unresponsiveness all these previous nights, every single night.

 

I quickly ran Prime95 and Furmark just to make sure it wasn't CPU/GPU related - I didn't run them LONG yet but a quick 15 min test on each shows no issues.

 

TL;DR is

Crashes every night when idle, PC powered on still but unresponsive until hard reboot.

Fresh reload of Windows, latest drives, BIOS, etc. No Overclock whatsoever, not even DOCP profile on RAM at this point.

I assumed it was bad RAM when I found that, but now I have no idea what to check next... 

 

Hardware info below, along with whocrashed and bluescreenview results with DPC_WATCHDOG_VIOLATION 

 

 

 

HARDWARE

Windows version:	Windows 11, 10.0, version 2009, build: 22621 (x64)
Windows dir:	C:\Windows
Hardware:	ASUS, ASUSTeK COMPUTER INC., ROG CROSSHAIR VIII HERO (WI-FI)
CPU: GPU:	AMD AMD Ryzen 9 5950X 16-Core Processor 8664, level: 25 Nvidia RTX 3080 Founder's Edition
Processor count:	32 logical processors, active mask: 4294967295
RAM:	49065.2MB (This is with the 1 stick removed, normally 4x sticks of 16GB)

 

WHOCRASHED

On Thu 1/19/2023 7:39:16 AM your computer crashed or a problem was reported
 

Crash dump file:	C:\Windows\Minidump\011923-11515-01.dmp (Minidump)
Bugcheck code:	0x133(0x1, 0x1E00, 0xFFFFF8031A51C340, 0x0)
Bugcheck name:	DPC_WATCHDOG_VIOLATION
Bug check description:	The DPC watchdog detected a prolonged run time at an IRQL of DISPATCH_LEVEL or above. This could be caused by either a non-responding driver or non-responding hardware. This bug check can also occur because of overheated CPUs (thermal issue).
Analysis:	This is likely caused by a hardware problem, but there is a possibility that this is caused by a misbehaving driver. This bugcheck indicates that a timeout has occurred. This may be caused by a hardware failure such as a thermal issue or a bug in a driver for a hardware device. Read this article on thermal issues A full memory dump will likely provide more useful information on the cause of this particular bugcheck.

 

BlueScreenView is attached.

[CorneliousJD]

I should also note that Sleep mode and Hibernation are both disabled. PC should be on and running 24/7.

[191x7]

14 minutes ago, CorneliousJD said:

I will try to keep this brief, while also giving relevant information... 

 

About a week ago my PC started crashing at night, I would wake up and the system would be unresponsive, but powered on, nothing would display on the monitors, nothing woke them up. Only option was to hard reboot.

BSODs logged in event viewer at random times over night. Sometimes 20 mins after I left my PC, sometimes 7 or 8 hours later. 

I was running a 2-year old install of Windows 10 and thought it was time to do a fresh wipe/reload and install Windows 11 anyways. 

 

I started with that, did BIOS update to latest version, installed Win 11, got drivers from Asus (mobo) website, and then the crashes continued overnight again.

Note that I work from home on my PC so I use this thing for 9+ hours a day, some days 12+ hours - no issues, it ONLY crashes overnight. WEIRD!

 

I did a Windows memory diag and it found problems. Cool, easy enough to troubleshoot and solve that.

Removed my DOCP RAM profile, used latest Memtest86+ to test individual DIMMs in different slots, found the bad one and removed it. Re-ran Memtest with 3 of 4 sticks and all came back as passed.

Bad RAM stick is awaiting warranty replacement now. 

 

I assumed things were solved when I removed that bad RAM stick, but once again this morning I woke up to an unresponsive PC.

Now I'm both frustrated and confused.

BSOD logged about 30 mins before I got to my PC. 

There was NOT a BSOD logged the night prior but I had experienced the same unresponsiveness all these previous nights, every single night.

 

I quickly ran Prime95 and Furmark just to make sure it wasn't CPU/GPU related - I didn't run them LONG yet but a quick 15 min test on each shows no issues.

 

TL;DR is

Crashes every night when idle, PC powered on still but unresponsive until hard reboot.

Fresh reload of Windows, latest drives, BIOS, etc. No Overclock whatsoever, not even DOCP profile on RAM at this point.

I assumed it was bad RAM when I found that, but now I have no idea what to check next... 

 

Hardware info below, along with whocrashed and bluescreenview results with DPC_WATCHDOG_VIOLATION 

 

 

 

HARDWARE

Windows version:	Windows 11, 10.0, version 2009, build: 22621 (x64)
Windows dir:	C:\Windows
Hardware:	ASUS, ASUSTeK COMPUTER INC., ROG CROSSHAIR VIII HERO (WI-FI)
CPU: GPU:	AMD AMD Ryzen 9 5950X 16-Core Processor 8664, level: 25 Nvidia RTX 3080 Founder's Edition
Processor count:	32 logical processors, active mask: 4294967295
RAM:	49065.2MB (This is with the 1 stick removed, normally 4x sticks of 16GB)

 

WHOCRASHED

On Thu 1/19/2023 7:39:16 AM your computer crashed or a problem was reported
 

Crash dump file:	C:\Windows\Minidump\011923-11515-01.dmp (Minidump)
Bugcheck code:	0x133(0x1, 0x1E00, 0xFFFFF8031A51C340, 0x0)
Bugcheck name:	DPC_WATCHDOG_VIOLATION
Bug check description:	The DPC watchdog detected a prolonged run time at an IRQL of DISPATCH_LEVEL or above. This could be caused by either a non-responding driver or non-responding hardware. This bug check can also occur because of overheated CPUs (thermal issue).
Analysis:	This is likely caused by a hardware problem, but there is a possibility that this is caused by a misbehaving driver. This bugcheck indicates that a timeout has occurred. This may be caused by a hardware failure such as a thermal issue or a bug in a driver for a hardware device. Read this article on thermal issues A full memory dump will likely provide more useful information on the cause of this particular bugcheck.

 

BlueScreenView is attached.

Get the latest chipset drivers;

https://www.amd.com/en/support/chipsets/amd-socket-am4/x570

Those on Asuses website aren't the latest, they are a few revisions older.

The latest AMD chipset drivers fix some BlueScreen and Black screen issues, that's stated in the release notes and online info about the release.

I am not claiming these will fix, but they might.

 

Also, if you had a bad stick of RAM during all the reinstall, the installations of programs and so on - your Windows install might be bad.

I'd recommend at least doing an sfc /scannow check.

 

Do you have the latest vBios on the 3080? Check both links.

https://nvidia.custhelp.com/app/answers/detail/a_id/5165/~/nvidia-resizable-bar-firmware-update-tool

https://nvidia.custhelp.com/app/answers/detail/a_id/5233/~/nvidia-gpu-firmware-update-tool-for-displayid

 

[CorneliousJD]

8 minutes ago, 191x7 said:

Get the latest chipset drivers;

https://www.amd.com/en/support/chipsets/amd-socket-am4/x570

Those on Asuses website aren't the latest, they are a few revisions older.

The latest AMD chipset drivers fix some BlueScreen and Black screen issues, that's stated in the release notes and online info about the release.

I am not claiming these will fix, but they might.

 

Also, if you had a bad stick of RAM during all the reinstall, the installations of programs and so on - your Windows install might be bad.

I'd recommend at least doing an sfc /scannow check.

 

Do you have the latest vBios on the 3080? Check both links.

https://nvidia.custhelp.com/app/answers/detail/a_id/5165/~/nvidia-resizable-bar-firmware-update-tool

https://nvidia.custhelp.com/app/answers/detail/a_id/5233/~/nvidia-gpu-firmware-update-tool-for-displayid

 

Thanks for the helpful hints.

 

I downloaded and installed latest chipset drivers you suggested. Fingers crossed for something.

In terms of GPU, both of those updates were already installed. Downloaded and tried again for good measure but it did confirm both were already installed.

 

I did also just run the DDU to fully remove nvidia driver and reinstall it fresh for good measure. 

 

I did run SFC /scannow and it found (and fixed) corrupt files 

Re-run of sfc/scannow shows no integrity violations.

 

Rebooting again after this and we will see.

Sadly I have to wait overnight to really know what's going on.

 

If you (or anyone else) has any other suggestions to try, I'm all ears, I can throw things at it during the day and hopefully something will stick by nighttime! 

[191x7]

27 minutes ago, CorneliousJD said:

Thanks for the helpful hints.

...

 

If you (or anyone else) has any other suggestions to try, I'm all ears, I can throw things at it during the day and hopefully something will stick by nighttime! 

You can use Hard Disk Sentinel to check the drives and you can check the Windows Reliability Monitor to see if there are issues.

 

And there are OCCT (CPU, Linpack, GPU, RAM, ...), MSI Kombustor (Furmark), 3DMark (Stability test in the advanced version) and other stress tests you could try.

I've seen "perfectly stable" systems having BSOD-s while starting simple tools like CPU-Z or HWMonitor.

[RiffTheRaff]

Try do disable "Global C-State Control" in BIOS (Advanced\AMD CBS).

 

And check your "Minimum processor state" in the "Power Options". AMD set it to 99% on my machine. I think this power profile is installed with the chipset drives.

 

[CorneliousJD]

2 hours ago, RiffTheRaff said:

Try do disable "Global C-State Control" in BIOS (Advanced\AMD CBS).

 

And check your "Minimum processor state" in the "Power Options". AMD set it to 99% on my machine. I think this power profile is installed with the chipset drives.

I disabled C-States in BIOS (it was set to AUTO before) -- power options I don't have that option for AMD Ryzen, looks like that might be laptop specific for AMD Ryzens.

I do know for a fact when these issues started on Win 10 that I had min/max both at 100% for sure.

[RiffTheRaff]

You have that option too. The image is from my desktop PC with an UPS.

Open the start menu and type "Edit power plan".

[CorneliousJD]

Woke up to NO crash today. Woo! It's too early to tell if it's actually fixed with just one day, but this is the first day w/out a crash overnight in a week or so.

 

Summary of changes.

• got the latest chipset drivers from AMD instead of Asus (mobo),
• ran DDU (display driver uninstaller) and pushed a totally fresh copy of nvidia drivers (same version I had tho)
• disabled C-States in bios.
• Ran SFC /scannow and it fixed things, although i think this was likely caused by crashes?

 

-- no idea which of these fixed it. or if today was just a lucky day. 

[CorneliousJD]

No crashes again overnight, which is good, but I do get things like this every now and then on shutdown, sometime's it's my Jabra Direct, othertimes it's BitWarden.

 

Not sure if it's worth reloading when sfc comes back as clean and DISM health checks come back normal too...

 

[DekaWar]

This problem is back with the New Nvidia Drivers was trying to trouble shoot this. and a another problem and had updated the drivers again and saw this in the notes update notes said it fixed this problem.  Just update it to fix it. 

[CorneliousJD]

5 hours ago, DekaWar said:

This problem is back with the New Nvidia Drivers was trying to trouble shoot this. and a another problem and had updated the drivers again and saw this in the notes update notes said it fixed this problem.  Just update it to fix it. 

Do you mean the exception breakpoint on shutdowns is caused by Nvidia drivers? 

 

Do you know which version fixed it? I just installed new drivers today and already saw it again. I just want to review their release notes. 

[bit42]

I have the exact same problem as this.  Did you manage to find out what was causing your problem?

[CorneliousJD]

12 hours ago, bit42 said:

I have the exact same problem as this.  Did you manage to find out what was causing your problem?

For me the root cause ended up being a dying 5950X - I had it warrantied by AMD and the problem has not resurfaced.

 

While I was searching I did disable c-states on my processor which HELPED reduce the crashing but didn't eliminate it. 

 

Separately I ALSO had a kit of bad RAM which corsair RMA'd. The RAM wont play nice still with any XMP/DOCP profile, so it runs at stock speeds sadly but the system is stable and passes 12+ hour AIDA64 stress tests now when before it would fail and/or crash within an hour every time.