saadbelgi commented on issue #73658: URL: https://github.com/apache/airflow/issues/73658#issuecomment-5860216275
Verified #71652 against the PyPI-installed `apache-airflow-providers-amazon==9.37.0rc1` with Airflow 3.3.2 and Python 3.12. Verified the following behaviour for `EmrServerlessJobSensor`: - `deferrable=False`, with the default `target_states` (`SUCCESS`): `PENDING`, `RUNNING`, `SCHEDULED`, and `SUBMITTED` keep the sensor waiting, `SUCCESS` completes it, and `FAILED`, `CANCELLING`, and `CANCELLED` fail it. - `deferrable=True` while the job is not in a target state: the task defers with `EmrServerlessJobSensorTrigger`, and the caller-supplied `target_states` are passed through with the application id, job run id, poke interval, and AWS connection settings. - `deferrable=True` when the job is already in a target state: it does not defer. - A success event completes the task. An error event fails it. - The trigger round-trips through serialization, and failure states are matched before success states on `jobRun.state`. AWS calls were mocked. I did not run a triggerer or a live EMR Serverless job. --- Drafted-by: Grok 4.7; reviewed by @saadbelgi before posting -- 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]
