Intro
Lately, I had to develop a demo working on a raspberrypi and send it to other project partners. However, we faced many difficulties in the beginning, since whenever I change something and push it to GitHub, they were not aware of that, and also faced problems with dependecies sometimes to make the demo work on their devices. So I decided to solve this problem and use a Docker container. I knew that with Docker you can run any application on any environemnt. But I was not aware it could also work on a Raspebrry Pi.
Moreover, Docker bind mounts allow for code developement. This is what I used to develop code on laptop first. Then when the demo is working, it is pushed to GitHub. A GitHub action workflow is automatically activated on every push action. This github worflow builds a Docker image and publish it in Docker Hub. This image can be then pulled by the rasperrypi and garenteed to work.
This was a high level description. Let’s get into the steps I followed.
1. Install Docker runtime
To install Docker runtime on a RaspberryPi, it’s straightforward by following this steps.
2. Write a DockerFile:
This Dockerfile sets up a Python environment, installs dependencies from requirements_pi.txt, and specifies the command to run the application. I intentinally used python:3.11-slim-bookworm for its low memory size since I am deploying on an edge device.
# Python runtime as the base image
FROM python:3.11-slim-bookworm
# Set the working directory in the container
WORKDIR ./demonstrator
# Copy the requirements file into the container
COPY requirements_pi.txt .
# Install any needed packages specified in requirements.txt
RUN pip install --upgrade pip
RUN pip install --no-cache-dir -r requirements_pi.txt
# Command to run your application
CMD ["python", "main.py"]
3. Set up compose.yml file
This Docker Compose file sets up a service named python_env to build an image using the Dockerfile in the current directory. It mounts the current directory to /demonstrator in the container, runs the command python main.py, and ensures the service restarts automatically.
version: '3.8'
services:
python_env:
build:
context: .
dockerfile: Dockerfile
working_dir: /demonstrator
volumes:
- .:/demonstrator
privileged: true
command: ["python", "main.py"]
restart: always
Once the Dockerfile and compose file are ready, it is possible to edit code on the host machine and run it in an Docker container since we are using a bind mount. we can now do code developement and test on the laptop before deploying on the RaspberryPi.
In other words, starting a container with bind mounts allows to use a container to run tools that we don’t want to install on our host, and yet still work with our host’s files.
4. Run
Finally, launching the demo is simply done with a single command: docker compose up
5. CI pipeline
GitHub Actions is a CI/CD tool that we’ll use to automate our pipeline. I first create a new file .github/workflows/ci.yml in the project repository with the following contents
name: ci
on:
push:
branches:
- main
jobs:
build:
runs-on: ubuntu-latest
steps:
-
name: Checkout
uses: actions/checkout@v4
-
name: Login to Docker Hub
uses: docker/login-action@v3
with:
username: ${{ secrets.DOCKER_USERNAME }}
password: ${{ secrets.DOCKERHUB_TOKEN }}
-
name: Set up Docker Buildx
uses: docker/setup-buildx-action@v3
-
name: Build and push
uses: docker/build-push-action@v5
with:
context: .
push: true
tags: ${{ secrets.DOCKER_USERNAME }}/${{ github.event.repository.name }}:latest
This workflow will run on every push to the main branch. It first checks the latest code version from the repository, then it logs in to Docker hub. Afterwards, it builds and pushs the Docker image to container registry DockerHub.