m0thm/monero_mine_nvidia repository overview

The Docker checks for this on run and builds it if it is not where it is supposed to be.... this has to be done on run as without running it in the --runtime=nvidia the cmake fails... I'm new to this if anyone knows how to set the runtime in my Dockerfile please comment :) I will also get my dockerfiles etc up on github in the next day or so.

XMRig built to use Nvidia CUDA based on https://hub.docker.com/r/nvidia/cuda⁠ This requires some work BEFORE you can use CUDA in your docker.

I will write more on the config later but I with great apreachation used the following sites to get CUDA running in my docker

https://marmelab.com/blog/2018/03/21/using-nvidia-gpu-within-docker-container.html⁠ https://www.pugetsystems.com/labs/hpc/How-To-Install-CUDA-10-together-with-9-2-on-Ubuntu-18-04-with-support-for-NVIDIA-20XX-Turing-GPUs-1236/⁠ https://developer.nvidia.com/cuda-downloads?target_os=Linux&target_arch=x86_64&target_distro=Ubuntu&target_version=1804&target_type=deblocal⁠

set this in your config file

"cuda": { "enabled": true, "loader": "/xmrig-cuda//build/libxmrig-cuda.so", "nvml": true },

Sugessted docker run

docker run -dit --restart unless-stopped --runtime=nvidia --name monero_mine_nvidia -v {Path to your config file}:/xmrig/build/config monero_mine_nvidia