The easiest way to install the latest stable version of Airflow is with
pip install airflow
You can also install Airflow with support for extra features like
pip install "airflow[s3, postgres]"
airflow PyPI basic package only installs what’s needed to get started.
Subpackages can be installed depending on what will be useful in your
environment. For instance, if you don’t need connectivity with Postgres,
you won’t have to go through the trouble of installing the
yum package, or whatever equivalent applies on the distribution you are using.
Behind the scenes, Airflow does conditional imports of operators that require these extra dependencies.
Here’s the list of the subpackages and what they enable:
||All Airflow features known to man|
||All databases integrations|
||Async worker classes for gunicorn|
||Minimum dev tools requirements|
||Airflow + dependencies on the Hadoop stack|
||Encrypt connection passwords in metadata db|
||Druid.io related operators & hooks|
||Google Cloud Platform hooks and operators
||JDBC hooks and operators|
||HDFS hooks and operators|
||All Hive related operators|
||kerberos integration for kerberized hadoop|
||ldap authentication for users|
||Microsoft SQL operators and hook, support as an Airflow backend|
||MySQL operators and hook, support as an Airflow backend|
||Password Authentication for users|
||Postgres operators and hook, support as an Airflow backend|
||Enable QDS (qubole data services) support|
||Rabbitmq support as a Celery backend|
||Vertica hook support as an Airflow backend|
||Redis hooks and sensors|