AWS SSM Parameter Store Secrets Backend

To enable SSM parameter store, specify SystemsManagerParameterStoreBackend as the backend in [secrets] section of airflow.cfg.

Here is a sample configuration:

backend =
backend_kwargs = {"connections_prefix": "/airflow/connections", "variables_prefix": "/airflow/variables", "profile_name": "default"}

To authenticate you can either supply arguments listed in Amazon Webservices Connection Extra config or set environment variables.

backend =
backend_kwargs = {"connections_prefix": "airflow/connections", "variables_prefix": "airflow/variables", "role_arn": "arn:aws:iam::123456789098:role/role-name"}

Optional lookup

Optionally connections, variables, or config may be looked up exclusive of each other or in any combination. This will prevent requests being sent to AWS SSM Parameter Store for the excluded type.

If you want to look up some and not others in AWS SSM Parameter Store you may do so by setting the relevant *_prefix parameter of the ones to be excluded as null.

For example, if you want to set parameter connections_prefix to "/airflow/connections" and not look up variables, your configuration file should look like this:

backend =
backend_kwargs = {"connections_prefix": "/airflow/connections", "variables_prefix": null, "profile_name": "default"}

Storing and Retrieving Connections

If you have set connections_prefix as /airflow/connections, then for a connection id of smtp_default, you would want to store your connection at /airflow/connections/smtp_default.

Optionally you can supply a profile name to reference aws profile, e.g. defined in ~/.aws/config.

The value of the SSM parameter must be the connection URI representation or in the JSON Format of the connection object.

In some cases, URI’s that you will need to store in AWS SSM Parameter Store may not be intuitive, for example when using HTTP / HTTPS or SPARK, you may need URI’s that will look like this:


This is a known situation, where schema and protocol parts of the URI are independent and in some cases, need to be specified explicitly.

See GitHub issue #10256 and #10913 for more detailed discussion that led to this documentation update. This may get resolved in the future.

The same connections could be represented in AWS SSM Parameter Store as a JSON Object

{"conn_type": "http", "host": ""}

{"conn_type": "spark", "host": "spark://spark-master-0.spark-master.spark", "port": 7077}

Storing and Retrieving Variables

If you have set variables_prefix as /airflow/variables, then for an Variable key of hello, you would want to store your Variable at /airflow/variables/hello.

Optionally you can supply a profile name to reference aws profile, e.g. defined in ~/.aws/config.

Was this entry helpful?