Docker on Jetson

Back to Module 3 | Back to Table of Contents

13. Install Docker and Basic Use

Introduction

Docker is a lightweight container platform that packages an application and all its dependencies into portable containers. It helps you run the same environment in development, testing, and production with much less configuration drift.

Docker improves consistency by using process isolation, fast startup, and efficient resource usage. It is well suited for local development, server deployment, and multi-service AI pipelines.

Install Docker on Jetson

1. Install Docker CE

sudo apt update
sudo apt install -y apt-transport-https ca-certificates curl software-properties-common

Add Docker GPG key and repository (example below uses Aliyun mirror):

curl -fsSL https://mirrors.aliyun.com/docker-ce/linux/ubuntu/gpg \
  | sudo gpg --dearmor -o /usr/share/keyrings/docker-ce.gpg

echo \
"deb [arch=$(dpkg --print-architecture) signed-by=/usr/share/keyrings/docker-ce.gpg] \
https://mirrors.aliyun.com/docker-ce/linux/ubuntu \
$(lsb_release -cs) stable" \
| sudo tee /etc/apt/sources.list.d/docker.list

sudo apt update
sudo apt install -y docker-ce docker-ce-cli containerd.io

Verify Docker installation:

docker --version

Add your user to the docker group (restart shell or run newgrp):

sudo usermod -aG docker "$USER"
newgrp docker

After this, you can run Docker commands without sudo.

2. Install NVIDIA Container Toolkit

distribution=$(. /etc/os-release; echo "$ID$VERSION_ID")

curl -s -L https://nvidia.github.io/libnvidia-container/gpgkey \
  | sudo gpg --dearmor -o /usr/share/keyrings/nvidia-container-toolkit-keyring.gpg

curl -s -L https://nvidia.github.io/libnvidia-container/stable/deb/nvidia-container-toolkit.list \
  | sed 's#deb https://#deb [signed-by=/usr/share/keyrings/nvidia-container-toolkit-keyring.gpg] https://#g' \
  | sudo tee /etc/apt/sources.list.d/nvidia-container-toolkit.list

Install toolkit:

sudo apt update
sudo apt install -y nvidia-container-toolkit

Enable GPU runtime for Docker:

sudo nvidia-ctk runtime configure --runtime=docker
sudo systemctl restart docker

Verify GPU access inside a container:

sudo docker run --rm --runtime=nvidia --gpus all --network host ubuntu nvidia-smi

If network restrictions exist in your environment, use a mirror/proxy when pulling images or packages.

3. Manual installation (APT unavailable)

If Docker CE cannot be installed from the Ubuntu repository, download .deb packages manually from:

https://download.docker.com/linux/ubuntu/dists/

Choose your Ubuntu version:

Then select your architecture folder:

• amd64: x86_64 (Server / PC)
• arm64: ARM 64-bit, including Jetson
• armhf: ARM 32-bit
• s390x: IBM z systems

Download the five matching packages:

• containerd.io_<version>_<arch>.deb
• docker-ce_<version>_<arch>.deb
• docker-ce-cli_<version>_<arch>.deb
• docker-buildx-plugin_<version>_<arch>.deb
• docker-compose-plugin_<version>_<arch>.deb

Install them from the folder where you downloaded the files:

sudo dpkg -i ./containerd.io_<version>_<arch>.deb \
  ./docker-ce_<version>_<arch>.deb \
  ./docker-ce-cli_<version>_<arch>.deb \
  ./docker-buildx-plugin_<version>_<arch>.deb \
  ./docker-compose-plugin_<version>_<arch>.deb

If dependencies are missing, run:

sudo apt -f install

Check Docker service status:

sudo systemctl status docker

If Docker is not running:

sudo systemctl start docker

Docker Basics

The Docker Engine provides the Docker CLI (docker), which lets you manage images, containers, networks, and storage from the command line.

Before learning common commands, review these two core concepts:

• Image: A read-only template that contains the base OS, runtime, libraries, and app code needed to run a container.
• Container: A running instance created from an image; it has its own process and environment, and can be started/stopped independently.

In short:

A Docker image is like an installation package, while a container is a running program instance.

Common commands

1) Show Docker info

docker info

2) Show Docker version

docker --version

3) Pull image

docker pull <image_name>

If no tag is specified, Docker pulls latest by default.

docker pull <image_name>:<tag>

4) Run image

docker run <image_name>

Run a specific image interactively:

docker run -it ubuntu:18.04 /bin/bash

Exit a running container by typing exit.

4.1) List running containers

docker ps

4.2) List all containers

docker ps -a

5) Remove stopped containers

docker container prune

6) List local images

docker images

7) Remove an image

Remove only after stopping and deleting related containers.

docker rmi <image_name>

8) Save a container as a new image

docker commit <container_id> <new_image_name>:<tag>

Choose a name and tag according to the actual container ID.

9) Stop a container

docker stop <container_id_or_name>

Back to Module 3