Windows 10: Full system freeze / lock-up at 11 hour uptime mark

Technically not a BSOD, but not sure where else I should post it.

At about 11 hours uptime (as reported by task manager, sometimes a few minutes early, sometimes a few minutes late) my system completely freezes. All activity stops, no mouse, no screen updates, and it doesn't recover if left alone for hours.

This started a couple of weeks ago. It was at first not obvious to me that it happened every 11 hours. I tried many things, nothing has helped in any way. When it started I did check if there were any Windows or driver updates in the surrounding days, and as I recall there weren't any. I also do not recall installing any software manually. As this is a Ryzen 2950x machine and Ryzen is known for stability issues, I first searched for Ryzen-specific issues and tried the obvious solutions to those.

Non-standard setup: my boot disk is a soft-mirrored SSD (Windows-managed), so Windows is set to Legacy rather than UEFI mode. Each crash my mirrors need to be resynced, which is hell on my disks, but it is complete before the next crash happens.

Device manager: "Corsair Link TM USB Dongle" has a warning triangle. This has always been this way since system build, and has never been an issue

Things tried:

- Reset BIOS

- Changed common Ryzen-related BIOS issues
--- Disabled C6 state
--- Normal current at idle rather than low
--- Disabled auto-power savings
--- Disabled all auto-(over/under-)clocking
--- RAM to slowest performance

- MemTest64: ran for 24 hours, no errors found

- Windows version: updated from Windows 10 Pro 18xx to 1903 18362.449

- Windows power management
--- Changed plan to full performance
--- Disabled drives and screens turning off
--- Disabled link state management

- Windows drivers: updated everything I could find to latest versions, including uninstalling and reinstalling chipset drivers

- scannow: no problems found

- dism: reimaged without issues

- netsh winsock reset: ok

- Drives: all drives chkdsk'd, no problems found

- Event viewer: I did not see any interesting events around freeze time

- Task scheduler: I did not see any tasks scheduled around crash time, but there are many tasks with custom triggers that I do not know how to check for when they will run

- Driver verifier: enabled just now as per forum instructions, not sure what to do with this after?

- V2 logs: attached as per forum instructions; not sure what is in these logs, but the last crash was about 2 hours before the time of log gathering

Once I figured out it happens every 11 hours, I tried catching the issue as it happened, opened up some diagnostic tools across my monitors to see what is going on at freeze time:

- Gigabyte SIV: temperatures and fan speeds normal

- Task manager: No major spikes in CPU usage (set to fastest refresh rate), mostly idle (<5%) as usual when I'm not doing heavy work or gaming

- Process Monitor: did not see anything interesting come by, but there's a lot of events so I might have missed something; also not sure if logging happens before event execution or after, if after, then the problematic event wouldn't show up, would it?

I manually killed many background processes and stopped many services a few minutes before the crash was expected to reduce ProcMon output; didn't stop the crash from occurring nor did it have ProcMon make more sense.

- - - Updated - - -

Update: Just now I decided to reboot the machine at about 10 hours uptime, to circumvent the crash. During the reboot, it BSOD'd with "unexpected kernel trap" in "storport.sys". Required restoring to previous restore point (luckily made yesterday before enabling driver verifier) to be able to boot. This is the first true BSOD I've seen in months, I guess driver verifier triggered it? Or some hardware broke even further

dumpchk (file also attached):

Code:

:)

 

Freezing, locking up... etc. ???

Try a different hard drive. I would suggest reformatting the one you have now, but it would suck to freeze in the middle of reformatting, leaving you with no operating system...

 

all games crash!

I suggest you run rthdribl > http://www.daionet.gr.jp/~masa/archives/rthdribl_1_2.zip
Increase the window size step by step and watch your vga temps permanently.
What happens?
If it crashes, run at stock speeds as suggested before and retry.

P.S.: When your system locks, does the display signal turn off after a time?
Stuttering sound or no sound at all when freezing?
What X1800XT model do you own exactly?

 

AMD FX OC'ers Club

That's what the auto setting will do since you're over the 1333 mark it jumps to the next step which isn't listed or officially supported I assume by that ram.

At this point though I would suggest dropping the multi one step or .5 so that we can get something stable and possibly take the heat issue out of the picture for now.