MatrixManAtYrService opened a new issue #18473: URL: https://github.com/apache/airflow/issues/18473
### Apache Airflow version 2.2.0b2 (beta snapshot) ### Operating System ubuntu 20.04 ### Versions of Apache Airflow Providers n/a ### Deployment Virtualenv installation ### Deployment details pip install /path/to/airflow/src ### What happened Given any DAG initialized with: `schedule_interval=None` Run `airflow dags test mydagname $(date +%Y-%m-%d)` and get an error: ``` INFO - No run dates were found for the given dates and dag interval. ``` This behavior changed in https://github.com/apache/airflow/pull/15397, it used to trigger a backfill dagrun at the given date. ### What you expected to happen I expected a backfill dagrun with the given date, regardless of whether it fit into the `schedule_interval`. If AIP-39 made that an unrealistic expectation, then I'd hope for some way to define unscheduled dags which can still be tested from the command line (which, so far as I know, is the fastest way to iterate on a DAG.). As it is, I keep changing `schedule_interval` back and forth depending on whether I want to iterate via `astro dev start` (which tolerates `None` but does superfluous work if the dag is scheduled) or via `airflow dags test ...` (which doesn't tolerate `None`). ### How to reproduce Initialize a DAG with: `schedule_interval=None` and run it via `airflow dags test mydagname $(date +%Y-%m-%d)` ### Anything else _No response_ ### Are you willing to submit 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]
