sideef5ect opened a new issue, #41522:
URL: https://github.com/apache/airflow/issues/41522

   ### Description
   
   Currently, Apache Airflow enforces a unique constraint on the combination of 
dag_id and execution_date for identifying DAG runs. This constraint ensures 
that each DAG run is uniquely identified by its DAG ID and execution date. 
However, there are scenarios where this constraint can be limiting and can not 
schedule multiple DAG run at the same execution_date. Additionally, recent 
versions of Airflow have introduced another unique key (dag_id, run_id), which 
should be sufficient to uniquely identify DAG runs.
   
   ### Use case/motivation
   
   Use Case Example: Cannot Schedule DAG Run at the Same Execution Date:
   Scenario: A data engineering team needs to reprocess a daily ETL pipeline 
due to an upstream data issue. The pipeline is scheduled to run at midnight 
every day.
   Problem: Due to the unique constraint on (dag_id, execution_date), the team 
cannot schedule a run of the DAG for the same execution date (midnight of the 
same day) without changing the execution date or creating a new DAG.
   Impact: This limitation forces the team to implement workarounds, such as 
modifying the execution date or duplicating the DAG, which complicates workflow 
management and increases the risk of errors.
   Solution: Removing the unique constraint on (dag_id, execution_date) would 
allow the team to schedule multiple runs of the DAG for the same execution 
date, simplifying reprocessing and improving workflow efficiency.
   
   ### Related issues
   
   https://github.com/apache/airflow/issues/15150
   
   ### Are you willing to submit a PR?
   
   - [ ] Yes I am willing to submit a PR!
   
   ### Code of Conduct
   
   - [X] I agree to follow this project's [Code of 
Conduct](https://github.com/apache/airflow/blob/main/CODE_OF_CONDUCT.md)
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to