Ubuntu16.04+CUDA8.0环境下安装nvidia-docker

安装docker和nvidia-docker的过程中经历了一些坑，记录下来。

最大的坑源：安装nvidia-docker版本错误

按照下面的命令默认安装的nvidia-docker1，在执行第二条命令时报错

wget -P /tmp https://github.com/NVIDIA/nvidia-docker/releases/download/v1.0.1/nvidia-docker_1.0.1-1_amd64.deb
 
sudo dpkg -i /tmp/nvidia-docker*.deb && rm /tmp/nvidia-docker*.deb

卸载docker和nvidia-docker，重新安装

卸载：

1.卸载docker

sudo apt-get remove docker docker-engine docker.io

2.卸载nvidia-docker

docker volume ls -q -f driver=nvidia-docker | xargs -r -I{} -n1 docker ps -q -a -f volume={} | xargs -r docker rm -f

sudo apt-get purge nvidia-docker

安装：

1.安装docker

参考https://yeasy.gitbooks.io/docker_practice/install/ubuntu.html

$ sudo apt-get update

$ sudo apt-get install \
    apt-transport-https \
    ca-certificates \
    curl \
    software-properties-common

$ curl -fsSL https://mirrors.ustc.edu.cn/docker-ce/linux/ubuntu/gpg | sudo apt-key add -

$ sudo add-apt-repository \
    "deb [arch=amd64] https://mirrors.ustc.edu.cn/docker-ce/linux/ubuntu \
    $(lsb_release -cs) \
    stable"

$ sudo apt-get update

$ sudo apt-get install docker-ce

启动docker

$ sudo systemctl enable docker
$ sudo systemctl start docker

测试docker

$ docker run hello-world

2.安装nvidia-docker2

参考https://github.com/NVIDIA/nvidia-docker和https://blog.csdn.net/zh_jessica/article/details/79644544

# Add the package repositories
$ distribution=$(. /etc/os-release;echo $ID$VERSION_ID)
$ curl -s -L https://nvidia.github.io/nvidia-docker/gpgkey | sudo apt-key add -
$ curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.list | sudo tee /etc/apt/sources.list.d/nvidia-docker.list

  $ sudo apt-get update 
  $ sudo apt-get install -y nvidia-docker2 
  $ sudo pkill -SIGHUP dockerd 

测试nvidia-docker

docker run --runtime=nvidia --rm nvidia/cuda:8.0-devel nvidia-smi

成功显示nvidia-smi显示的驱动信息。