mwaa-dr

What it is and what it does

mwaa-dr is a Python library that generates DAGs to automate backup, restore, and cleanup of Amazon MWAA metadata stores. It addresses the constraint that MWAA metadata access is only available through DAGs by providing factory classes that generate these DAGs for you, eliminating manual DAG creation.

The library handles metadata export to S3 or local filesystem, supports configurable restore strategies for variables and connections (DO_NOTHING, APPEND, REPLACE), and allows customization through inheritance to include or exclude specific tables. It depends on smart-open for storage abstraction and requires Python 3.7 or later.

Use it for:

• Automate regular metadata backups of your MWAA environment to S3 for disaster recovery
• Restore a MWAA metadata store from backup after environment failure or data loss
• Exclude sensitive tables from backup when using AWS Secrets Manager for variables and connections
• Customize backup scope by extending factory classes to add or remove specific metadata tables
• Test configurations locally using the LOCAL_FS storage type

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Generates backup, restore, and cleanup DAGs for Amazon MWAA metadata stores by providing factory classes that create reusable workflows without manual DAG authoring.

Yes. If you operate Amazon MWAA and need disaster recovery for metadata, this library eliminates boilerplate code and provides a tested, actively maintained solution. Install friction is minimal, and the package is production-stable. The unclear license requires review, but the permissive grant language suggests low risk for most use cases.

Install

mwaa-dr on PyPI

pip install mwaa-dr

uv add mwaa-dr

poetry add mwaa-dr

Installing mwaa-dr

Before you install

Low friction: pure Python wheel with a single runtime dependency (smart-open). Actively maintained with recent commits and production-stable classification.

License in practice

License treatment is unclear—the package carries an Amazon copyright with a permissive grant but lacks an SPDX identifier. Review the full license text before use in proprietary or commercial contexts.

Quickstart

from mwaa_dr.v_2_10.dr_factory import DRFactory_2_10

factory = DRFactory_2_10(
    dag_id='backup',
    path_prefix='data',
    storage_type='S3'
)
dag = factory.create_backup_dag()

Requires MWAA execution role to have read/write permissions on the target S3 bucket; an Airflow variable DR_BACKUP_BUCKET must be set with the bucket name (not ARN)

Verify before relying

• Whether the unclear license is compatible with your organization's licensing policy
• Minimum MWAA version compatibility beyond the documented examples (2.7.2, 2.10.3)
• Whether smart-open's dependencies introduce additional install friction or licensing constraints

Package facts

Evidence: mwaa_dr-2.2.0-py3-none-any.whl

Keywords: MWAA, airflow, disaster, recovery, DR

Tags

More Distributed Computing packages