Airflow in local

Example of how to run Airflow in local, without Docker.

Installation

(Optionnal) Install conda and create a conda environment. It will help you to handle several versions of Python on your computer:

# Example here for python 3.8
conda create -y -q --no-default-packages --name airflow_in_local_3.8 python=3.8 virtualenv invoke

conda activate airflow_in_local_3.8

Install python invoke lib:

pip install invoke

Install the project:

invoke install

How to use

Run this command to run Airflow (scheduler + web server). The web server will be available on the address http://127.0.0.1:8080. The login/password is admin/admin. To stop the server, press Ctrl+c or Command+c.

invoke start-airflow

Now you can start to create some DAGs in src/ folder.