Serving your application

Your deployment web server should have skipole installed.

Once you have developed a project inside a 'projectfiles' directory, copy projectfiles to your target web server. Delete the skiadmin project - you will not be serving that. Edit 'myproj.py' so it creates the application, and adds the 'skis' application, but does not create the Python web server or adds skiadmin.

For test purposes, this could be served by some web servers straight from the command line. Though you would have to put your ../projectfiles onto the Python path - something like:

export PYTHONPATH=${PYTHONPATH}:$HOME/projectfiles

And then the application created in myproj could be served, for example using gunicorn:

gunicorn3 -w 4 myproj:application

Where gunicorn3 is the python3 version of gunicorn

Or similarly using Waitress (python3-waitress on Debian based distro's)

waitress-serve --port=8000 myproj:application

In general however, you are likely to want the application to start on boot up, examples of running the wsgi app with a debian based server are given below:

Simple serving for a Raspberry Pi

Further suggestions for Pi

nginx and uwsgi

Pound reverse proxy and Waitress

nginx load balancing and Waitress