Quote from: papampi on February 10, 2018, 08:03:43 AM

Quote from: joshuajones02 on February 09, 2018, 11:40:35 PM

Having an issue, not sure how to troubleshoot it. I've formatted a new rig with 1.4 and updated it with 2.0, its loaded with EVGA Hybrid SC2 cards and I keep getting this error every few hours

Code:

Unable to determine the device handle for GPU 0000:02:00.0: GPU is lost. Reboot the system to recover this GPU

Fri Feb 9 03:49:35 EST 2018 - reboot in 10 seconds
Fri Feb 9 12:03:26 EST 2018 - Lost GPU so restarting system. Found GPU's:
pci.bus_id
00000000:01:00.0
00000000:02:00.0
00000000:03:00.0
00000000:04:00.0
00000000:05:00.0
00000000:06:00.0

Fri Feb 9 12:03:26 EST 2018 - reboot in 10 seconds

[31m [1mWARNING: Thu Feb 8 21:54:34 EST 2018 - Problem detected! GPU1 is not responding. Will give watchdog 60 seconds to react, if not we will reboot! [0m
[31m [1mWARNING: Fri Feb 9 03:49:35 EST 2018 - Problem detected! GPU1 is not responding. Will give watchdog 60 seconds to react, if not we will reboot! [0m
[31m [1mWARNING: Fri Feb 9 12:03:26 EST 2018 - Problem detected! GPU1 is not responding. Will give watchdog 60 seconds to react, if not we will reboot! [0m

Every time watchdog fails to complete the boot and when I turn the screen on it just shows a thin purple line at the top and a black/blank screen

Thats a frozen rig with failed GPU.
Could be riser problem, Power or high OC.

I had the exact same issue. I was getting where it couldnt find the device every 36-48 hrs. It wasnt an overclock issue or a bad riser. Once I updated to the newest miners the issue seemed to have worked itself out.