Any way to only add 1 out of 2 GPUs to gigahorse #889
whattheflux
started this conversation in
General
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Hi team, I have dell r820 with 2x NVIDIA Tesla P4s. Unraid 6.11.5
I want to run one GPU attached to machinaris and leave one free for plotting direct from unraid CLI
Nvidia Driver Version:530.41.03
Open Source Kernel Module:No
Installed GPU(s):
0:
Tesla P4
04:00.0
GPU-d3fa90ce-8ecb-b9da-3ccb-0f38a4890172
1:
Tesla P4
44:00.0
GPU-99847a33-d393-b653-8eee-302e6399794d
In my docker configs I have extra parameters: --runtime=nvidia
Then in variables I have:
NVIDIA_DRIVER_CAPABILITIES: all
NVIDIA_VISIBLE_DEVICES:
Under NVIDIA VISIBLE DEVICES I've tried: 0 and 1 and GPU-d3fa90ce-8ecb-b9da-3ccb-0f38a4890172 but no matter what I do it always adds both to the container. I can see because when I run nvidia-smi from the CLI within the contianer both GPUs are visible and both are doing chia_harvester.
I disabled the CHIAPOS checking setting so at least it can farm now. Is there a bug that adds both or am I doing something wrong?
Thanks a bunch!
Beta Was this translation helpful? Give feedback.
All reactions