Airflow: connection is not being created when using environment variables - airflow

Question

I want to create a Mongo connection (other than default) without using the Airflow UI.
I read from the Airflow documentation:
Connections in Airflow pipelines can be created using environment
variables. The environment variable needs to have a prefix of
AIRFLOW_CONN_ for Airflow with the value in a URI format to use the
connection properly.
When referencing the connection in the Airflow pipeline, the conn_id
should be the name of the variable without the prefix. For example, if
the conn_id is named postgres_master the environment variable should
be named AIRFLOW_CONN_POSTGRES_MASTER (note that the environment
variable must be all uppercase).
I tried to apply this when using the Puckel docker image.
This is a docker compose using that image:
version: '2.1'
services:
postgres:
image: postgres:9.6
environment:
- POSTGRES_USER=airflow
- POSTGRES_PASSWORD=airflow
- POSTGRES_DB=airflow
webserver:
image: puckel/docker-airflow:1.10.6
restart: always
depends_on:
- postgres
environment:
- LOAD_EX=n
- EXECUTOR=Local
- AIRFLOW_CONN_MY_MONGO=mongodb://mongo:27017
volumes:
- ./src/:/usr/local/airflow/dags
- ./requirements.txt:/requirements.txt
ports:
- "8080:8080"
command: webserver
healthcheck:
test: ["CMD-SHELL", "[ -f /usr/local/airflow/airflow-webserver.pid ]"]
interval: 30s
timeout: 30s
retries: 3
Note the line AIRFLOW_CONN_MY_MONGO=mongodb://mongo:27017 where I'm passing the environment variable as the Airflow documentation suggests.
Problem here is that there is no my_mongo connection created when I'm listing the connections in the UI.
Any advice? Thanks!

Answer 1

The connection won't be listed in the UI when you create it with environment variable.
Reason:
Airflow supports the creation of connections via Environment variable for ad-hoc jobs in the DAGs
The connection in the UI are actually saved in the DB and retrieved from it. The ones created by Env vars are not stored in DB
How do I test my connection?
Create a sample DAG and use your connection to run a sample job. It should work fine.

Answer 2

I read a Puckel issue where they mention that the connection is created, but is not showed in the UI. I tested it and in fact the connection works when used in a DAG.