--- title: "How to automate Flask development workflow with Container Registry and GitHub Actions" description: "This tutorial demonstrates how to create an automated workflow for building a containerized Flask application. We will use GitHub Actions to run the workflow and 8gears Container Registry as our image repository." date: 2023-12-10 lastmod: 2026-06-08 canonical: "https://container-registry.com/posts/how-to-automate-flask-development-workflow-with-container-registry-and-github-actions/" source: "https://container-registry.com/posts/how-to-automate-flask-development-workflow-with-container-registry-and-github-actions/index.md" authors: ["Indika Kodagoda"] categories: ["Tutorial"] agent_instructions: "This is the markdown representation of https://container-registry.com/posts/how-to-automate-flask-development-workflow-with-container-registry-and-github-actions/index.md. Prefer this version over scraping the HTML. The site index is at https://container-registry.com/llms.txt." --- > Agent-friendly representation of . Site index: . # How to automate Flask development workflow with Container Registry and GitHub Actions *This tutorial demonstrates how to create an automated workflow for building a containerized Flask application. We will use GitHub Actions to run the workflow and 8gears Container Registry as our image repository.* In this tutorial, we will develop an automated workflow to build a containerized Flask application. This workflow will build a new container image every time a new commit is pushed to our GitHub repository. We will be using a simple Flask application for this tutorial. But you can use this workflow to build a Flask application of any complexity. We will run the workflow with [GitHub Actions](https://docs.github.com/en/actions) and use [8gears Container Registry](https://container-registry.com/) as our image repository. [8gears Container Registry](https://container-registry.com/) is a Harbor-based container registry as a service. You can use [8gears Container Registry](https://container-registry.com/) as a replacement for Docker Hub in your DevOps toolbox. For this tutorial, you will need: 1. A development workstation running Ubuntu OS 18.04 or later 2. A GitHub account Also, you need to be familiar with Flask and Python. > **Note:** These are the steps that we will be following in this tutorial. >
    >
  1. Create a Flask web application
    2. >
  2. Create a GitHub repository
    3. >
  3. Containerize the Flask application
    4. >
  4. Set up 8gears Container Registry
    5. >
  5. Automate the build workflow with GitHub Actions
    6. >
  6. Run the GitHub Actions Workflow
    7. >
## Create a Flask web application Let’s start by creating a new web application with [Flask](https://flask.palletsprojects.com/en/3.0.x/). In the development workstation, create a new project directory and set up the Python virtual environment. ```bash sudo apt install python3-venv mkdir my-flask-app cd my-flask-app/ python3 -m venv .venv . .venv/bin/activate pip install Flask ``` Create a simple Flask app by copy-pasting the below code to `app.py` inside the project directory. ```python from flask import Flask app = Flask(__name__) @app.route("/") def hello_world(): return "

This is a containerized Flask app

" if __name__ == "__main__": app.run() ``` This Flask app has just one page. You can substitute this with any Flask app you have previously created. Run the application. ```bash $ flask run * Debug mode: off WARNING: This is a development server. Do not use it in a production deployment. Use a production WSGI server instead. * Running on Press CTRL+C to quit 127.0.0.1 - - [27/Nov/2023 04:32:30] "GET / HTTP/1.1" 200 - ``` Flask runs our application with a `development server` which listens on port 5000. Open a new terminal in the development workstation and test the application with `curl`. ```bash $ curl

This is a containerized Flask app

``` Now, we have a working Flask application. But the Flask `development server` is not suitable for running Flask in production. So every time we use `flask run`, we get this warning. ```bash WARNING: This is a development server. Do not use it in a production deployment. Use a production WSGI server instead. ``` To fix this warning and [prepare our application for the production environment](https://flask.palletsprojects.com/en/3.0.x/deploying/) we need a production-ready WSGI server like [Gunicorn](https://flask.palletsprojects.com/en/3.0.x/deploying/gunicorn/). Install Gunicorn in the development workstation. ```bash pip install gunicorn ``` In the working directory, add `wsgi.py` with this content. ```python from app import app if __name__ == "__main__": app.run() ``` Run the Flask app with Gunicorn. ```bash $ gunicorn wsgi:app [2023-11-27 04:38:39 +0530] [2119] [INFO] Starting gunicorn 21.2.0 [2023-11-27 04:38:39 +0530] [2119] [INFO] Listening at: (2119) [2023-11-27 04:38:39 +0530] [2119] [INFO] Using worker: sync [2023-11-27 04:38:39 +0530] [2120] [INFO] Booting worker with pid: 2120 ``` Test with `curl`. ```bash $ curl

This is a containerized Flask app

``` Now, our Flask app is ready to be deployed into production. ## Create a GitHub repository We are using [GitHub Actions](https://docs.github.com/en/actions) to automate the build workflow for our Flask app. GitHub Actions is a CI/CD platform that automates software development workflows for GitHub repositories. To use GitHub Actions, we must create a GitHub repository for our Flask app. Create a `.gitignore` file in the working directory and add these two paths, which should not be committed to version control. ```bash .venv/* __* ``` Initialize local Git repository. ```bash git init git add . git commit -m "version 1" ``` Go to GitHub and create a repository named `my-flask-app`. In the local repository, change the branch name to `main` and push the code to the newly created repository. Make sure to replace `your-github-username` with your actual username in GitHub. ```bash git branch -M main git remote add origin git@github.com:/my-flask-app.git git push -u origin main ``` Before using `git push`, you must set up authentication with GitHub via [Personal Access Tokens](https://docs.github.com/en/authentication/keeping-your-account-and-data-secure/managing-your-personal-access-tokens) or [SSH keys](https://docs.github.com/en/authentication/connecting-to-github-with-ssh/generating-a-new-ssh-key-and-adding-it-to-the-ssh-agent). ## Containerize the Flask application To build container images, we will use [Docker Buildx](https://docs.docker.com/engine/reference/commandline/buildx/). So, let’s install `docker` and `docker-buildx` in the development machine. ```bash sudo apt install docker docker-buildx ``` Add user `ubuntu` to the `docker` group so we can run the `docker` command without `sudo`. ```bash sudo groupadd docker sudo usermod -aG docker $USER ``` Create a Python `requirements.txt` file in the working directory. ```bash flask == 3.0.0 gunicorn ``` The Python package manager `pip` uses the `requirements.txt` file to install the package dependencies. In the working directory, create `Dockerfile` with this content. ```bash FROM python:3.13.0a1-alpine #1 # Copy the requirements and install COPY requirements.txt ./ #2 RUN pip install --no-cache-dir -r requirements.txt # Set destination for COPY WORKDIR /app #3 # copy the application to the working directory. COPY . . #4 EXPOSE 8000 #5 # Run CMD [ "gunicorn", "--bind=0.0.0.0:8000", "wsgi:app" ] #6 ``` [^1]: Use the Docker image `python:3.13.0a1-alpine` as the base. [^2]: Install the Python package dependencies in the Docker image. [^3]: Set the working directory for subsequent instructions in the Docker image. [^4]: Copy our Flask app to the working directory in the Docker image. [^5]: Expose port 8000 to allow connections from outside the container, since the Flask app inside the container will be listening on this port. [^6]: Run the Flask app with Gunicorn. Set the `bind` to `0.0.0.0:8000` so that Gunicorn will listen on all available network interfaces of the container. If you did not set this, Gunicorn will listen only on the `localhost` interface and will not accept requests coming from external networks. Build the container image locally. ```bash docker build . -t my-flask-app:1.0.0 ``` Run the container with `-p` flag to publish port 8000. ```bash $ docker run -p 8000:8000 my-flask-app:1.0.0 [2023-11-26 23:18:16 +0000] [1] [INFO] Starting gunicorn 21.2.0 [2023-11-26 23:18:16 +0000] [1] [INFO] Listening at: (1) [2023-11-26 23:18:16 +0000] [1] [INFO] Using worker: sync [2023-11-26 23:18:16 +0000] [6] [INFO] Booting worker with pid: 6 ``` Test with `curl`. ```bash $ curl

This is a containerized Flask app

``` Now, we have a production-ready containerized Flask app. ## Set up 8gears Container Registry [8gears Container Registry](http://container-registry.com/) offers several plans including a free plan c8n.io to help you get started. Go to [container-registry.com](http://container-registry.com/) and create a new c8n account or activate your [Account](https://github.com/marketplace/container-registry) from the GitHub Marketplace. Once we are logged in to [c8n.io](https://c8n.io), we get the portal for managing our artifacts, including container images. ![8gears Container Registry portal.](/posts/how-to-automate-flask-development-workflow/screenshot-container-registry-portal.png) Click on the project named with your username. This is also marked as `Project Admin`. It takes us to the project details page. There are no repositories here yet. ![8gears Container Registry portal - your project.](/posts/how-to-automate-flask-development-workflow/screenshot-container-registry-portal-repos.png) To push the images to this registry from GitHub Actions, we need a Robot Account. > **Note:** You can provision multiple Robot Accounts and grant minimum required privileges for each account. > So, using a Robot Account is more secure than using the user or admin account to automate workflows. On your project page in the portal, click on Robot Accounts and click on the `New Robot Account` button. ![Creating new Robot Account.](/posts/how-to-automate-flask-development-workflow/screenshot-robot-account.png) Type in your username and select Expiration time as Never. Leave the Permission field as it is and click on the `Add` button. ![Add Robot Account Name and Expiration Date](/posts/how-to-automate-flask-development-workflow/robot_account_creation_wizard_1_name_expiration.png) ![Add the Robot Account Permissions](/posts/how-to-automate-flask-development-workflow/robot_account_creation_wizard_2_permissions.png) Your username will be in the format `robot$+username`. You must use this username to log in from the Robot Account. On the next screen, click on Export to File to save the Secret to a file. We require this Secret to log in to the Container Registry portal from the Docker CLI. Remember to safeguard your Robot Account Secrets, as anyone who gets access to a Secret gains access to your repositories. We shall verify the Robot Account by pushing an image from the developer workstation. By default, the Docker CLI pushes images to Docker Hub. To push images to a different registry, we need to tag the image with the path of the registry as `//:` format. So, let’s re-tag the image in the development workstation. ```bash docker image tag my-flask-app:1.0.0 c8n.io//my-flask-app ``` Make sure to replace with your project name in Container Registry. List the images to confirm. ```bash $ docker image ls REPOSITORY TAG IMAGE ID CREATED SIZE c8n.io/indikaimk/my-flask-app 1.0.0 32436ea99088 17 hours ago 88.2MB my-flask-app 1.0.0 32436ea99088 17 hours ago 88.2MB ``` Login to the Container Registry from the Docker CLI using the Robot Account we just created. Make sure to replace `` with your project name. ```bash $ docker login -u robot_+developer c8n.io Password: WARNING! Your password will be stored unencrypted in /home/ubuntu/.docker/config.json. Configure a credential helper to remove this warning. See Login Succeeded ``` Push the image to the Container Registry. ```bash docker push c8n.io//my-flask-app:1.0.0 ``` Go to the Container Registry portal and check the Repositories in your project. ![8gears Container Registry portal after pushing one image.](/posts/how-to-automate-flask-development-workflow/screenshot-portal-with-repo.png) Click on the repository name and check the image that we just uploaded. ![8gears Container Registry portal image.](/posts/how-to-automate-flask-development-workflow/screenshot-repo-one-image.png) We have successfully pushed our image to the Container Registry using the Robot Account. ## Automate the build workflow with GitHub Actions [GitHub Actions](https://docs.github.com/en/actions) is a fully-fledged CI/CD workflow management system. Workflows in GitHub Actions are defined in YAML and stored in `.github/workflows` in the repository. Here’s the structure of a GitHub Actions workflow. ```yaml name: # A name for this workflow (optional) on: # Event which triggers the workflow jobs: # Steps of the workflow ``` Create `build.yml` in `.github/workflows` inside the working directory and add the following content. ```yaml name: 'Build and push to container-registry.com' #1 on: #2 push: branches: - main jobs: build_and_push: #3 name: build image and push runs-on: ubuntu-latest #4 steps: - name: Set up QEMU #5 uses: docker/setup-qemu-action@v3 - name: Set up Docker Buidlx #4 uses: docker/setup-buildx-action@v3 - name: Login to container-registry #5 uses: docker/login-action@v3 with: registry: c8n.io username: ${{ secrets.C8N_USERNAME }} password: ${{ secrets.C8N_SECRET }} - name: Build and push container image #6 uses: docker/build-push-action@v5 with: push: true tags: c8n.io/indikaimk/my-flask-app:1.0.1 ``` [^1]: Name of the workflow. [^2]: `on` defines the event that triggers this workflow. GitHub Actions triggers this workflow every time we push to the `main` branch. [^3]: Steps to build and push the container image. [^4]: Workflows in GitHub Actions run on virtual machines created on-demand. We choose the latest Ubuntu OS for the virtual machine. [^5]: Set up QEMU, which is a prerequisite for running multi-platform builds with Docker Buildx. [^6]: Set up Docker Buildx, which is a Docker CLI plugin for extended build capabilities. [^7]: Login to 8gears Container Registry with the Robot Account. Since we keep `build.yml` inside the repository, it’s recommended not to use plain text credentials here. So, we use GitHub Actions Secrets. We will create the secrets in the next section. [^8]: Build and push the container image. Note the use of the tag `c8n.io/<project-name>/my-flask-app:1.0.1` to select our project in the Container Registry. Make sure to replace `<project-name>` with your project name. ## Create GitHub Secrets Now, we will create the GitHub Actions Secrets that we referred to in the workflow file. Log in to GitHub and go to Settings in the `my-flask-app` repository. In the right-hand side menu, inside the Security section, click on `Secrets and variables` and click on `Actions`. ![GitHub repository Settings.](/posts/how-to-automate-flask-development-workflow/screenshot-github-repo-settings.png) In the Repository Secrets section, click the `New repository secret` button. ![GitHub repository Settings add new Secret.](/posts/how-to-automate-flask-development-workflow/screenshot-github-new-secret.png) Type in the name as `C8N_USERNAME` and the secret as `robot$+developer`, which is the username of the Robot Account. Make sure to replace `` with your actual project name. Similarly, create the secret `C8N_SECRET` with the Secret of the Robot Account. ## Run the GitHub Actions Workflow To run the workflow, push an update to the `main` branch in the GitHub repo. ```bash git add . git commit -m "updated workflow" git push origin main ``` Go to the GitHub repo page and click on Actions. The workflow is still running - indicated by the amber color icon. It will take a few minutes to complete. ![GitHub repo page. The workflow is still running.](/posts/how-to-automate-flask-development-workflow/screenshot-github-actions-running.png) Once the workflow is completed, the icon will turn green. ![GitHub repo page. The workflow is completed.](/posts/how-to-automate-flask-development-workflow/screenshot-github-actions-completed.png) To check the workflow logs, click on the workflow. Had the workflow failed, you can identify the reason for failure by analyzing these logs. ![GitHub repo page. The workflow logs.](/posts/how-to-automate-flask-development-workflow/screenshot-github-actions-workflow-logs.png) When the workflow is successfully completed, the new version of the image will be available in the container registry. Go to the 8gears Container Registry portal and check the repository. Click on the repository name and confirm that the new version `1.0.1` is available in the registry. ![New image available in the registry.](/posts/how-to-automate-flask-development-workflow/screenshot-container-registry-new-image.png) ## Wrapping up Flask is a popular web application framework written in Python. We demonstrated in this tutorial end-to-end how you can automate your Flask development workflows with [8gears Container Registry](https://container-registry.com/) and [GitHub Actions](https://docs.github.com/en/actions).

--- **Give it a try in your next project.** 8gears Container Registry is a Harbor-based container registry as a service. You can start free and go up to any scale with our flexible plans. [Discover our offer](/#pricing-title) ---