Azure App Service + PostgreSQL

I’m excited to share that I successfully connected my Azure Web App to an Azure Database for PostgreSQL - Flexible Server! This milestone marks a significant step in my development journey, and I want to walk you through the process to help anyone else attempting a similar setup.

Step 1: Create the Resources

To get started, I created a Django web application and an Azure Database for PostgreSQL - Flexible Server. During the setup, there are a few key options you need to configure:

1. Add Current Client IP Address: Ensure this option is selected to include your current IP in the PostgreSQL server’s firewall rules.
2. Allow Public Access: Enable the option to allow public access to the server using a public IP address.
3. Allow Access from Azure Services: Select the option to allow public access from any Azure Service within Azure to this server.

Once the PostgreSQL server was created, I navigated to:
Settings > Service Connector, and clicked +Create.

Service Connector Configuration

• Service Type: DB for PostgreSQL Flexible Server
• Connection Name: Choose any descriptive name for your connection.
• Subscription: Select your Azure subscription.

After creating the Service Connector, I viewed the Sample Code tab, which redirected me to the Microsoft Learn documentation. The documentation provides detailed instructions on how to use a connection string to connect your application to the database.

Important note: When you created the service connector, Azure already set up environment variables for you in your Azure Web App. To connect to it using django you need to edit your settings.py and add in those environment variables to the DATABASES config. For example:

host = os.getenv('AZURE_POSTGRESQL_HOST')
user = os.getenv('AZURE_POSTGRESQL_USER')
password = os.getenv('AZURE_POSTGRESQL_PASSWORD')
database = os.getenv('AZURE_POSTGRESQL_NAME')

DATABASES = {
    'default': {
        'ENGINE': 'django.db.backends.postgresql_psycopg2',
        'NAME': database,
        'USER': user,
        'PASSWORD': password,
        'HOST': host,
        'PORT': '5432',  # Port is 5432 by default 
        'OPTIONS': {'sslmode': 'require'},
    }
}

Pro Tip: Always follow security best practices by storing sensitive information, such as database credentials, in environment variables instead of hardcoding them into your application.

Step 2: Configure Django

Following the Django setup steps in the documentation, I updated my settings.py file to include the appropriate environment variables for the database connection. Additionally, I installed the required PostgreSQL driver by running:

pip install psycopg2

To Test the connection, I pushed all my Django models to the datatbase by running

python manage.py makemigrations
python manage.py migrate

And just like that… BAM! Pages loaded perfectly, and I was able to perform POST and GET requests without issues.

Step 3: Serve Static Files

One challenge I encountered was that Gunicorn does not serve static files. For less demanding websites (such as my development environment), I learned that the WhiteNoise middleware can be used to serve static files efficiently.

To implement WhiteNoise, I ran:

pip install whitenoise

Next, I updated my setting.py file to add whitenoise.middleware.WhiteNoiseMiddleware to the middleware list. It is important to place this middleware Immediately after django.middleware.security.securityMiddleware to ensure it functions correctly.

MIDDLEWARE = [
    'django.middleware.security.SecurityMiddleware',
    'whitenoise.middleware.WhiteNoiseMiddleware',
    # Other middleware...
]

With this setup, I was able to serve static files successfully

Conclusion

After testing and configuring everything, My development environment was finally in a stable and functional state. This project was a fantastic learning experience, and I am excited to explore more advanced deployments in the future.

If you are working on something similar, I hope this guide helps you along the way.