WSL2 설치 및 Docker & CUDA 사용 방법

1. Docker Install 

WSL2 설치 및 사용 방법

https://www.44bits.io/ko/post/wsl2-install-and-basic-usage

WSL2(Windows Subsystem for Linux 2) 설치 및 사용 방법

 

- 참고 사이트

https://bbongcol.github.io/wsl/2020/07/30/wsl2_performance_configuration.html

TECHJUN

https://muritzy.tistory.com/2409

Windows 10 프리뷰 버전 "Build 20150" - WSL 2의 Ubuntu에서 CUDA 시작하기

2. Cuda Install 

https://docs.nvidia.com/cuda/wsl-user-guide/index.html#installing-nvidia-drivers

CUDA on WSL :: CUDA Toolkit Documentation

 

- 설치하기전 본인의 컴퓨터의 GPU가 Radeon 인지 CUDA 인지 잘 확인할 것 

- stackoverflow에서 Is it possible to run CUDA on AMD GPUs?  이란 질문이 나왔을때 단호화게 Nope.. 이라고 적혀 있는걸 보니 확실히 안되는듯 

 

 

 

Error1) 

 

 

$ docker version 

$ docker version Client: Version: 19.03.8 API version: 1.40 Go version: go1.13.8
Git commit: afacb8b7f0
Built: Tue Jun 23 22:26:12 2020
OS/Arch: linux/amd64 Experimental: false Cannot connect to the Docker daemon at tcp://0.0.0.0:2375. Is the docker daemon running?

 

 

 

$ export PATH= "$HOME/bin:$HOME/.local/bin:$PATH"
$ export PATH= "$PATH:/mnt/c/Program\ Files/Docker/Docker/resources/bin"
$ alias docker=docker.exe
$ alias docker-compose=docker-compose.exe

 

$ docker -H tcp://0.0.0.0:2375 images

$  docker images

 

$ docker version 

 

Error2)

(base) root@DESKTOP-VOC4MP7:/mnt/c/Users/user# sudo docker ps 
Cannot connect to the Docker daemon at unix:///var/run/docker.sock. Is the docker daemon running?

원인

(1) docker service가 실행이 안되어 있는것 

 

$sudo systemctl start docker

$sudo systemctl enable docker

 

Error3)

(base) root@DESKTOP-VOC4MP7:/mnt/c/Users/user# sudo systemctl status docker
System has not been booted with systemd as init system (PID 1). Can't operate.
Failed to connect to bus: Host is down

(1) wsl를 쓸때마다 docker 관련 환경 관련 변수를 항상 바인딩시켜주어야 한다. 

 

 

Error4)

(base) root@DESKTOP-VOC4MP7:/mnt/c/Users/user# sudo service docker stop docker: unrecognized service (base) root@DESKTOP-VOC4MP7:/mnt/c/Users/user# sudo service docker start docker: unrecognized service

 

<출처> 

1, https://docs.nvidia.com/cuda/wsl-user-guide/index.html

2. https://m.blog.naver.com/PostView.nhn?blogId=hahaysh&logNo=221233313670&proxyReferer=https:%2F%2Fwww.google.com%2F

3.