molcay commented on PR #37087: URL: https://github.com/apache/airflow/pull/37087#issuecomment-1921398205
> Not sure if it belongs to this PR but if it's not too much work can u add it to the cluster activity? Hi @romsharon98, Thank you for pointing out, it looks like it is an rather small touch. I can definitely add it. If you think any other places that also needs to update please feel free to point them out. --- > I am worried this may cause extra confusion. For example with triggering DAGRun via the rest API. WDYT? Hi @eladkal, Currently we are using `manual` run type for almost everything (except the other run types: `backfill`, `scheduled`, `dataset_triggered`) like; , - manually (clicked the Run button on UI) triggered DAG runs - programmatically (using `TriggerDagRunOperator`) triggered DAG runs - (as you said) REST API calls triggered DAG runs - by running [a CLI command](https://airflow.apache.org/docs/apache-airflow/stable/core-concepts/dag-run.html#external-triggers): `airflow dags trigger --exec-date logical_date run_id` - and maybe others that we cannot remember/know right now and we do not have a way to separate which runs are triggered by which way. Right now, on the UI, when I look for the run execution I cannot distinguish the manually triggered and programmatically triggered runs (`TriggerDagRunOperator`). Also, same goes for the logs of tasks. I think this PR will help us to distinguish **manually triggered** and **programmatically triggered** DAG runs as a beginning. If we want, we can add a new type for REST API calls or we can just assume that if a DAG run started with REST API call, then it is manual. To make it more clear, maybe we can add a section about run types in this page: https://airflow.apache.org/docs/apache-airflow/stable/core-concepts/dag-run.html and explain each DAG Run Type. > Btw, I think we definitely need to document run types no matter what we decide to do on this matter. -- 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]
