Ubuntu 18.04安裝Docker CE及NVIDIA Container Toolkit步驟

Docker的好處之一，就是在Container裡面可以隨意瞎搞，不用擔心弄崩Host的環境。

為了在docker中支援GPU，NVidia之前是弄了個nvidia-docker2，現在升級為NVIDIA Container Toolkit了。官方說法是"Usage of nvidia-docker2 packages are deprecated since NVIDIA GPUs are now natively supported as devices in the Docker runtime"。

安裝環境

• OS：Ubuntu 18.04 64 bit
• 顯示卡：NVidia GTX 2080 Ti x 2
• CUDA：10.0
• cnDNN：7.4

任務：安裝Docker CE及NVIDIA Container Toolkit，以便後期開展深度學習

設定Docker源
# 更新源
$ sudo apt update

# 啟用HTTPS
$ sudo apt install -y
    apt-transport-https
    ca-certificates
    curl
    gnupg-agent
    software-properties-common

# 新增GPG key
$ curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo apt-key add -

# 新增穩定版的源
$ sudo add-apt-repository
  "deb [arch=amd64] https://download.docker.com/linux/ubuntu
  $(lsb_release -cs)
  stable"

安裝Docker CE

此刻Docker版本需要19.03，此後可能需要更新。
# 更新源
$ sudo apt update

# 安裝Docker CE
$ sudo apt install -y docker-ce

如果這種方式安裝失敗，也有解決方案。
 報錯時螢幕上會顯示下載失敗的deb檔案，想辦法下載下來，然後挨個手動安裝就好。

此刻我需要下載的是下面三個檔案，此後更新為當時最新版本即可：

• containerd.io_1.2.6-3_amd64.deb
• docker-ce-cli_19.03.0~3-0~ubuntu-bionic_amd64.deb
• docker-ce_19.03.0~3-0~ubuntu-bionic_amd64.deb

手動依次安裝：
$ sudo dpkg -i containerd.io_1.2.6-3_amd64.deb
$ sudo dpkg -i docker-ce-cli_19.03.0~3-0~ubuntu-bionic_amd64.deb
$ sudo dpkg -i docker-ce_19.03.0~3-0~ubuntu-bionic_amd64.deb

驗證Docker CE

如果出現下面的內容，說明安裝成功。
$ sudo docker run hello-world

Unable to find image 'hello-world:latest' locally
latest: Pulling from library/hello-world
1b930d010525: Pull complete
Digest: sha256:2557e3c07ed1e38f26e389462d03ed943586f744621577a99efb77324b0fe535
Status: Downloaded newer image for hello-world:latest

Hello from Docker!
This message shows that your installation appears to be working correctly.

To generate this message, Docker took the following steps:
 1. The Docker client contacted the Docker daemon.
 2. The Docker daemon pulled the "hello-world" image from the Docker Hub.
    (amd64)
 3. The Docker daemon created a new container from that image which runs the
    executable that produces the output you are currently reading.
 4. The Docker daemon streamed that output to the Docker client, which sent it
    to your terminal.

To try something more ambitious, you can run an Ubuntu container with:
 $ docker run -it ubuntu bash

Share images, automate workflows, and more with a free Docker ID:
 https://hub.docker.com/

For more examples and ideas, visit:
 https://docs.docker.com/get-started/

設定nvidia-docker源
# 新增源
$ 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

# 安裝並重新啟動docker
$ sudo apt update && sudo apt install -y nvidia-container-toolkit
$ sudo systemctl restart docker

使用
# 在官方CUDA映象上測試 nvidia-smi
$ sudo docker run --gpus all nvidia/cuda:9.0-base nvidia-smi

# 啟動支援雙GPU的容器
$ sudo docker run --gpus 2 nvidia/cuda:9.0-base nvidia-smi

# 指定GPU 1，執行容器
$ sudo docker run --gpus device=0 nvidia/cuda:9.0-base nvidia-smi

能看到顯示卡資訊就說明OK了，當前image是基於Ubuntu 16.04的。