DonHaul opened a new issue, #67695: URL: https://github.com/apache/airflow/issues/67695
### Under which category would you file this issue? Providers ### Apache Airflow version 3.2.1 ### What happened and how to reproduce it? Issue 1: In the ui. If a dagrun has already succeeded we cannot "remark" it as succeeded. However through the API [`patch_dag_run`](https://airflow.apache.org/docs/apache-airflow/3.2.1/stable-rest-api-ref.html#operation/patch_dag_run), when sending a payload such as ``` { "state": "success", "note": note, } ``` It allows dag to be "remarked" as success. This has the unwanted effect of marking tasks that in the initial dag run were skipped or failed, to now be shown as `success`. Issue 2: When marking through the ui a dagrun that is currently `running` as `success` and `failed` we get different results. - When doing `mark as failed` all the tasks that haven't yet been executed get skipped and the dag gets failed as `failed` - OK - See Image 2 - When doing `mark as success`, all the tasks, even the ones that have been previously executed and succeeded/ failed or got skipped change their status to success. - This is misleading, as a task that has already happened should not have its status changed - See Image 3 Dag example. The ui is pressed once the dag is on `task_mark_dag_here` ```python from airflow.sdk import dag, task import time @dag def test_short_circuit_dag(): @task.branch def task_branch(): return "task_succeeds" @task def task_succeeds(): pass @task def task_skipped(): pass @task(trigger_rule="all_done") def task_fails_successfully(): raise Exception @task(trigger_rule="one_failed") def task_succeeds2(): pass @task def task_mark_dag_here(): time.sleep(5) @task def task_not_executed(): pass task_succeeds_1 = task_succeeds() task_branch() >> [task_succeeds_1, task_skipped()] task_succeeds_1 >> task_fails_successfully() >> task_succeeds2() >> task_mark_dag_here() >> task_not_executed() >> task_not_executed() test_short_circuit_dag() ``` Image 1 - "Initial State" <img width="1460" height="245" alt="Image" src="https://github.com/user-attachments/assets/46823e11-055d-4e52-b3b6-e6797d8e0614"; /> Image 2 - Marking as Failed <img width="1458" height="238" alt="Image" src="https://github.com/user-attachments/assets/53b88f4f-666b-4a66-8333-ae374c12500a"; /> Image 3 - Marking as Success <img width="1440" height="219" alt="Image" src="https://github.com/user-attachments/assets/8c80fcb0-63f6-4c29-b6b9-4d6b35a65c96"; /> ### What you think should happen instead? Issue 1: if a dag has already `succeeded` sending a `patch_dag_rung` with the `"state": "success"` in the payload should not do anything. Issue 2: If the dag is `running`, pressing mark as `success` should: - mark all next tasks as `skipped` - not alter states of tasks that already executed - mark dag as `success` ### Operating System _No response_ ### Deployment None ### Apache Airflow Provider(s) _No response_ ### Versions of Apache Airflow Providers _No response_ ### Official Helm Chart version Not Applicable ### Kubernetes Version _No response_ ### Helm Chart configuration _No response_ ### Docker Image customizations _No response_ ### 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]
