Introduction
In this lab, we will learn how to deploy a Flask application to a server. We will create a distribution file for our application and install it on the server. The lab assumes you have a basic understanding of Flask, Python's virtual environments, and the command line.
Note: You need to create the code file yourself and run it in the environment. You can preview the Flask service status on Web 5000.
Build the Application
First, we need to create a wheel file for our application. We will use the build tool for this. Install the build tool using pip if you haven't already:
## Install the build tool
pip install build
Now, use the build tool to create the wheel file:
## Build the wheel file
python -m build --wheel
The wheel file should be in the dist directory with a name like flaskr-1.0.0-py3-none-any.whl.
Install the Application on the Server
Copy the wheel file to your server. Once it's there, set up a new Python virtual environment and install the wheel file using pip:
## Install the wheel file
pip install flaskr-1.0.0-py3-none-any.whl
Since this is a new environment, you need to initialize the database again:
## Initialize the database
flask --app flaskr init-db
Configure the Secret Key
In a production environment, you should change the secret key to a random value. To generate a random secret key, run the following command:
## Generate a random secret key
python -c 'import secrets; print(secrets.token_hex())'
Create a config.py file in the instance folder and set SECRET_KEY to the generated value.
## .venv/var/flaskr-instance/config.py
SECRET_KEY = 'your_generated_secret_key'
Run the Application with a Production Server
For a production environment, you should use a WSGI server instead of the built-in development server. We will use Waitress as our WSGI server.
First, install Waitress:
## Install Waitress
pip install waitress
Now, tell Waitress to serve your application:
## Run the application with Waitress
waitress-serve --call 'flaskr:create_app'
Summary
In this lab, we learned how to deploy a Flask application to a production server. We built our application into a wheel file, installed it on the server, configured the secret key, and ran the application with a production WSGI server.