ExecuteFlaskAppOnLinux

Install and Execute a Flask App on Linux

Assumptions:

• You develop on Windows but deploy on Linux.
• You have a Flask project in GitHub.
• You project has a Python Virtual Environment and a requirements.text file for the dependencies.
• You have python 3.6+ installed (we had to upgrade our Linux 16.04 to python 3.7 as e.g. described here, or here for python 3.8 (note: after step 2 run sudo apt update).
• You have sudo permissions on the machine.
• For DMZ machines open ports are 80 and 443. Set your Flask app to run on 443.

Now follow these steps (for python3.7 that we use):

• Clone the project and change to project directory.
• Create a Virtual Python Environment (You may need to sudo apt-get install python3.7-venv)

python3.7 -m venv env

(This will create the env/ directory.)

• Activate the environment:

screen // from this point we work in a dedicated screen for a long-running process.
source env/bin/activate

(Windows: source env/Scripts/activate.bat // use Git for Windows SHELL).

• Install dependencies:

pip install -r requirements.txt

• Run the server:

python app.py // if PermissionDenied see notes

Notes:

• For the machine to be accesses from outside, use app.run(host='0.0.0.0', port=port).
• For ports such as 443/80 you need sudo, use sudo env/bin/python app.py (and not sudo python app.py since sudo does not recognize environment vars).
• A better and more secure option is to run on port 80 as a regular user:

sudo touch /etc/authbind/byport/80
sudo chmod 777 /etc/authbind/byport/80
authbind --deep python app.py

(Taken from this link).

• The server is run in development mode, deployment mode is not covered here).