安装
Docker-CE
安装过程参考官网，Ubuntu中如下：
## 1. 从仓库安装
$ sudo apt-get update
# 使apt使用https的仓库
$ sudo apt-get install apt-transport-https ca-certificates curl software-properties-common
# 添加docker的官方GPG Key
$ curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo apt-key add -
# 验证
$ sudo apt-key fingerprint 0EBFCD88
pub 4096R/0EBFCD88 2017-02-22
Key fingerprint = 9DC8 5822 9FC7 DD38 854A E2D8 8D81 803C 0EBF CD88
uid Docker Release (CE deb) <docker@docker.com>
sub 4096R/F273FCD8 2017-02-22
# 添加稳定版仓库
$ sudo add-apt-repository "deb [arch=amd64] https://download.docker.com/linux/ubuntu \
$(lsb_release -cs) \
stable"
# 再次update
$ sudo apt-get update
# 安装docker-ce，这一步比较慢，清华的源也是在官网下载安装的
$ sudo apt-get install docker-ce
# hello world
$ sudo docker run hello-world
## 1.1 安装其他可用版本
# 显示可用版本
$ apt-cache madison docker-ce
# 指定版本
$ sudo apt-get install docker-ce=<VERSION>
## 2. 从deb包安装
# 从https://download.docker.com/linux/ubuntu/dists/选择对应的发行版本，进入pool/stable/amd64，下载一个合适的版本
$ sudo dpkg -i /path/to/package.debnvidia-docker
参考github
# 删除nvidia-docker1.0的容器，卸载nvidia-docker1.0
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 -y nvidia-docker
# 添加仓库
curl -s -L https://nvidia.github.io/nvidia-docker/gpgkey | \
sudo apt-key add -
curl -s -L https://nvidia.github.io/nvidia-docker/ubuntu16.04/amd64/nvidia-docker.list | \
sudo tee /etc/apt/sources.list.d/nvidia-docker.list
sudo apt-get update
# 安装nvidia-docker2，重新加载docker daemon的配置
sudo apt-get install -y nvidia-docker2
sudo pkill -SIGHUP dockerd
# 测试nvidia-smi命令
docker run --runtime=nvidia --rm nvidia/cuda nvidia-smi下面是安装nvidia-docker2之后的/etc/docker/daemon.json
{
"runtimes": {
"nvidia": {
"path": "/usr/bin/nvidia-container-runtime",
"runtimeArgs": []
}
}
}ustc mirror
在/etc/docker/daemon.json中添加新的项，使用中科大的docker镜像，多个镜像可以同时添加到列表中：
{
"registry-mirrors": ["https://docker.mirrors.ustc.edu.cn/"]
}然后重启docker服务
sudo service docker restartservice failed
Fixing the Job for docker.service failed because the control process exited with error code. See "systemctl status docker.service" and "journalctl -xe" for details. problem, for me this worked:
- sudo rm -rf /var/lib/docker
- create daemon.json in /etc/docker/
- put this in it:
{
"exec-root": "/path/to/docker/run",
"storage-driver": "overlay",
"graph": "/path/to/docker/lib"
} - systemctl start docker
- docker run --rm hello-world
