Only the May 10 input is missing; processing after May 11 is complete.
Backfill that date and update downstream aggregates without rerunning the entire period.
Which configuration is most appropriate?
Run the Airflow DAG for the selected logical date using a targeted backfill.
Detailed explanation
Changing permissions does not regenerate a missing date.
Changing permissions does not regenerate a missing date.
A logical-date backfill reruns only the needed tasks instead of rebuilding every period.
A logical-date backfill reruns only the needed tasks instead of rebuilding every period.
This risks correct data and adds unnecessary processing time and cost.
This risks correct data and adds unnecessary processing time and cost.
Audit events cannot be edited and editing them does not rerun ETL.
Audit events cannot be edited and editing them does not rerun ETL.
Check official documentation
Compare the following points with the official documentation.
Amazon MWAA・Apache Airflowのbackfill、logical date、catchupの公式説明を確認する。
What to confirm
全量再構築と、対象日を限定した再実行の違いを説明できる。
Key points
- MWAA
- Airflow
- Backfill
- Logical date
Notes
- Environment: AWS公式DEA-C01試験ガイドと各サービスの公式ドキュメントの確認
- Confirm current AWS service behavior in the official DEA-C01 documentation.
Foundation review
Targeted replay
Partition- or date-level replay limits duplicate work and downstream impact.
DAG idempotency
Outputs should be safely overwritten or merged when a date is run again.