Replies: 1 comment
-
|
I think this is a useful idea, but I would be careful about changing the existing semantics directly. Today, a DAG run is considered successful when the leaf tasks are success or skipped. A new “skipped DAG run” state would be more expressive, but it could also break existing dashboards, alerts, SLAs, and automations that rely on the current success behavior. A safer approach would be to keep the current state model and add a derived status like “all tasks skipped” or “effectively skipped”, either in the UI or as an opt-in feature. That would give better observability without changing behavior for existing users. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
We have recently migrated from Airflow v2.11 to v3.2.2 - as part of this upgrade we have deprecated the CeleryExecutor in favour of the KubernetesExecutor. We have finally resolved the OOM and Pod eviction issues and the environment is running stable but we have moved all of DAGs from using Sensors to running on a continued Schedule.
A feature that would make a good improvement in observability is to be able to filter DAG runs that have succesfully run versus ones that have skipped.
Current behaviour:
Proposed Change:
Beta Was this translation helpful? Give feedback.
All reactions