Docker

Docker is a platform for developing, shipping, and running applications in containers. It enables developers to package an application with all of its dependencies into a standardized unit for software development. Docker containers are lightweight and contain everything needed to run the application, including code, runtime, system tools, system libraries, and settings.

Docker Installations

There are two main Docker installations:

  1. Docker Desktop for Linux
  2. Docker Engine

Docker Desktop is recommended for beginners as it provides a user-friendly interface and additional features. Source

Installation

Caution

Always check the official Docker documentation for the most up-to-date instructions. The following steps are for Ubuntu 22.04.

Preparing for Installation

First, remove any old or unofficial versions of Docker:

for pkg in docker.io docker-doc docker-compose docker-compose-v2 podman-docker containerd runc; do sudo apt-get remove $pkg; done

Ubuntu Installation

  1. Set up Docker's apt repository: Follow the instructions in the official guide.

  2. Install Docker Desktop: Follow the steps outlined here.

Note

You may need to enable virtualization in your system's BIOS settings for Docker Desktop to work properly.

Setting up pass for Credential Storage

Docker uses pass to store credentials securely. Initialize it with:

gpg --generate-key
pass init "Your GPG ID or email"

Installing RStudio Server via Docker

To run RStudio Server in a Docker container:

  1. Pull the RStudio Server image:
docker pull rocker/rstudio
  1. Run the container:
docker run -d -p 8787:8787 -e PASSWORD=yourpassword --name rstudio rocker/rstudio

This command:

  • Runs the container in detached mode (-d)
  • Maps port 8787 on the host to port 8787 in the container (-p 8787:8787)
  • Sets a password for the RStudio user (-e PASSWORD=yourpassword)

  • Names the container "rstudio" (--name rstudio)

  • Access RStudio Server by navigating to http://localhost:8787 in your web browser.

Docker Engine Examples

Building Multi-Architecture Images

To build images for different architectures:

  1. Set up buildx:
docker buildx create --use
  1. Build and push an amd64 image: 
    docker buildx build --platform linux/amd64 -t yourusername/imagename:tag . --push

This process allows you to create images that can run on different CPU architectures, improving portability across various systems.

Best Practices

  1. Use official base images when possible.
  2. Minimize the number of layers in your Dockerfile.
  3. Use multi-stage builds to reduce final image size.
  4. Regularly update your Docker images to include security patches.
  5. Use Docker Compose for managing multi-container applications.

For more advanced usage and best practices, refer to the Docker documentation.

Created by Ryan D. Najac for the Palomero Lab at the Institute for Cancer Genetics.
Page last updated on 2024-10-17.