nvlddmkm stopped responding, Windows 10

[joshuajones02]

I'm currently using Windows 10, and after several hours of mining my rig will usually go offline for some nvlddmkm related issue. It will either stops responding, recover and reboot itself, or just freeze up (not 100% sure if its nvlddmkm, event viewer just says it recovered from a BugCheck). I've tried reinstalling the drivers - version 385.69 .. Getting really frustrated as I stay up most of the night seeing if the fix i applied changes anything and then a few hours later it crashes but I wake up several hours after that.. Getting really annoyed, any suggestions would be appreciated.

[gt_addict]

What cards are you running?

[joshuajones02]

5x 1080 ti on Windows 10

[joshuajones02]

Changed the drivers to 385.28 and then this happened about 30 mins later

https://ibb.co/geT8tk

another nvlddmkm error that caused a reboot

[joshuajones02]

This time I received a blue screen with CLOCK_WATCHDOG_TIMEOUT as the stop code

[gt_addict]

Change to 372.70 and see how you get on. Ive found those drivers the most stable.

[joshuajones02]

Change to 372.70 and see how you get on. Ive found those drivers the most stable.

Changing them now

[joshuajones02]

Its telling me that those drivers are not compatible with the 1080 ti's

[joshuajones02]

Trying 385.41

[dragonmike]

Are you overclocking your cpu by any chance? The only times I was having BSODs with clock watchdog timeouts was when I was trying to push my cpu too much.

[joshuajones02]

nope, no oc on the cards either, just lowered TDP to 85% , it just went offline again after running for 2 hours with the 385.41 drivers, haven't seen the reason yet going to be home in a hour

[joshuajones02]

Found this article, going to go through it when I get home... sounds like my issue https://forums.geforce.com/default/topic/413110/the-nvlddmkm-error-what-is-it-an-fyi-for-those-seeing-this-issue/

[GaryH]

I'm using Win 10 also and occasionally I see the nvlddmkm error, but I've found either Claymore or EWBF miners seem the run the best for my rig.  EWBF I have a watchdog script for that will autorestart EWBF, and Claymore seems to behave.  CCMiner will eventually crash so I stay away from that.

What algos are you trying to mine?  I also found Awesome Miner to be a very nice program (free for 2 computers) as you can set whatever algo and it will restart the miner if it drops to 0 or crashes.  BSODs I don't think anything can help though.

Also what motherboard you have and did you configure the BIOS to use the correct PCI Gen, etc?

[joshuajones02]

I mine Equihash with the latest EWBF, mobo is a ASRock H110 Pro BTC+ with 5 1080 ti's and 2 EVGA PSUs (1000W P2 and 850W G2)

[joshuajones02]

It runs great, but then crashes after 2 hours... have mined up to 6 hours one time without crashing but that was once, it's usually only a couple hours and then it crashes. I'm thinking about going back to Windows 7

[GaryH]

It runs great, but then crashes after 2 hours... have mined up to 6 hours one time without crashing but that was once, it's usually only a couple hours and then it crashes. I'm thinking about going back to Windows 7

are you saying that Windows crashes or EWBF crashes and stops running?

Device manager shows my driver version at:  22.21.13.8494

Driver version filename is:  385.41-desktop-win10-64bit-international-whql.exe

[joshuajones02]

Windows freezes, crashes, and or reboots itself

[GaryH]

If you have all the latest Windows updates then I believe you will have to half-split the problem which is either the OS or the hardware.  Easiest way I can think of is to download the free nvOC software and give it a go.  See if things improve or not.  I just don't know a good way to figure out if you are dealing with software or hardware.

The nvOC the last time I checked, you hook your monitor to the GPU that's in the first x16 slot--apparently it doesn't like onboard video being used.

Perhaps you can remove the GPUs and start with one and isolate some sort of pattern to the crashing.  Like perhaps it works fine with 3, or maybe it crashes with one particular card in the mix?

[joshuajones02]

yeah, I went from one problem to another.. I had everything working great on my last motherboard except that I could not get a fifth GPU working.. took days of troubleshooting to figure out it was the mobo... While troubleshooting the last problem I tried going from Windows 7 (which did work with the 4 GPUs) to Windows 10, which if I remember right I think I had it running with 4 GPUs with the last mobo until the new mobo came in..

Anyways, now I have a new mobo that recognizes and all 5 GPUs and they work but now they are crashing after running for so many hours (2-6). I installed a fresh copy of Windows 10 and that did not fix the problem. Right now I'm trying to go back to windows 7 but running some updates on the operating system (its an old copy)... Hopefully this works, if not theres an article of someone with a similar problem they were able to fix it, so I'll try that unless someone else has something else to suggest

[joshuajones02]

the nvOC os looks cool, I don't have the time to set it up, leaving soon for a couple weeks.. Hoping I can get something working or at least 4 of the cards