AddAirflowConnectionDag.py

from airflow import DAG, settings
from airflow.operators import PythonOperator
from datetime import datetime, timedelta
from airflow.models import Connection

import json

def add_gcp_connection(**kwargs):
    new_conn = Connection(
            conn_id="google_cloud_default",
            conn_type='google_cloud_platform',
    )
    extra_field = {
        "extra__google_cloud_platform__scope": "https://www.googleapis.com/auth/cloud-platform",
        "extra__google_cloud_platform__project": "strictlypossession",
        "extra__google_cloud_platform__key_path": '/usr/local/airflow/dags/gcpLicense.json'
    }

    session = settings.Session()

    #checking if connection exist
    if session.query(Connection).filter(Connection.conn_id == new_conn.conn_id).first():
        my_connection = session.query(Connection).filter(Connection.conn_id == new_conn.conn_id).one()
        my_connection.set_extra(json.dumps(extra_field))
        session.add(my_connection)
        session.commit()
    else: #if it doesn't exit create one
        new_conn.set_extra(json.dumps(extra_field))
        session.add(new_conn)
        session.commit()

        
default_args = {
    "owner": "airflow",
    "depends_on_past": False,
    "start_date": datetime(2020, 5, 9),
    "email": ["airflow@airflow.com"],
    "email_on_failure": False,
    "email_on_retry": False,
    "retries": 3,
    "retry_delay": timedelta(minutes=5)
}

dag = DAG("AddGCPConnection", default_args=default_args, schedule_interval="@once")

with dag:
    activateGCP = PythonOperator(
        task_id='add_gcp_connection_python',
        python_callable=add_gcp_connection,
        provide_context=True,
    )

    activateGCP

The script should be run from the virtual machine, I either use the airflow UI to run the dag or I use the airflow command in the Docker container to run the job.

Thank for the help