Deploy a Python (Flask) app to Azure to Azure Containers Apps

This Python web app is a simple restaurant review application built with the Flask framework. The web app stores application data in PostgreSQL with environment variables defining the connection info.

This repo was created to be built to a Docker image and run as a container instance in Azure Container Apps. For more information, see the tutorial Deploy a Python web app on Azure Container Apps with GitHub Actions.

This Python web app repo can also be used in other ways:

• You can run the web app locally in a virtual environment. Make sure to define .env file with environment settings.

  • In the .env file, fill in a secret value for SECRET_KEY. You can use this command to generate an appropriate value for your product deployment:

    python -c 'import secrets; print(secrets.token_hex())'

• You can create a container locally and run it in Docker locally. You need Docker Desktop installed. For this scenario, set REMOTE_POSTGRESQL=1 and other environment variables in a .env file to point to a PostgreSQL server instance in Azure. You also need to create an application service principal, add it as a PostgreSQL ROLE to the server, and grant it necessary permissions on the database. See the .env.example file for details.

  docker build --file Dockerfile --tag pythoncontainer:latest .
  docker run -it --env-file .env --publish 5000:5000/tcp pythoncontainer:latest

  If you want to use PostgreSQL instance locally, add --add-host to the Docker command. For more information, see the Docker run command. For an example of how to do this with MongoDB, see Build and test a containerized Python web app locally.

• You can deploy the code (not a container) to App Service. For guidance on how to deploy code, see Quickstart: Deploy a Python (Django or Flask) web app to Azure App Service and Overview: Deploy a Python web app to Azure with managed identity.

• You can create a Docker image from this repo and host the container instance in Web Apps for Containers (App Service). See Overview: Containerized Python web app on Azure.

If you need an Azure account, you can create on for free.

A Django sample application with similar functionality is at https://github.com/Azure-Samples/msdocs-python-django-azure-container-apps.

Requirements

The requirements.txt has the following packages:

Package	Description
Flask	Web application framework.
SQLAlchemy	Provides a database abstraction layer to communicate with PostgreSQL.
Flask-SQLAlchemy	Adds SQLAlchemy support to Flask application by simplifying using SQLAlchemy. Requires SQLAlchemy.
Flask-Migrate	SQLAlchemy database migrations for Flask applications using Alembic. Allows functionality parity with Django version of this sample app.
pyscopg2-binary	PostgreSQL database adapter for Python.
gunicorn	WSGI HTTP Server for UNIX. Required for running containers locally in VS Code.
python-dotenv	Read key-value pairs from .env file and set them as environment variables. In this sample app, environment variables describe how to connect to the database and storage resources. Because managed identity is used no sensitive information is included in environment variables. Flask's dotenv support sets environment variables automatically from an .env file.
flask_wtf	Form rendering, validation, and CSRF protection for Flask with WTForms. Uses CSRFProtect extension.

The steps to do this are covered more completely in the tutorial Deploy a Python web app on Azure Container Apps with GitHub Actions. Briefly, here are the steps:

1. Fork and then clone locally.
2. Create an Azure Container Registry.
3. Build a container image from the repo.
4. Create an Azure Database for PostgreSQL Flexible Server instance.
5. Make your Azure user account a Microsoft Entra administrator on the server.
6. Create a database on the server.
7. Create a user-assigned managed identity.
8. Add the managed identity as a ROLE on the PostgreSQL server.
9. Assign the managed identity PostgreSQL permissions on the database.
10. Deploy the web app container to Azure Container Apps.
11. Configure continuous deployment.

See the create_resources.sh Bash script for automating the creation of these steps. For best results, run the script in Azure Cloud Shell.