potiuk opened a new issue, #30022: URL: https://github.com/apache/airflow/issues/30022
### Discussed in https://github.com/apache/airflow/discussions/29984 <div type='discussions-op-text'> <sup>Originally posted by **RNHTTR** March 8, 2023</sup> ### Apache Airflow version 2.5.1 ### What happened Today, if a DAG is running, and it has a top-level `Variable.get()` command, and that Variable is deleted while the DAG is still running, the DAG’s next task(s) will fail with no task logs (assuming the Variable eventually gets re-introduced), because the DAG cannot be parsed. I suspect this would happen with anything that can cause an import error, but I’m able to easily reproduce it using Variables. ### What you think should happen instead Once the DAG is able to be parsed/imported, any tasks that failed due to this problem should say something along the lines of "Task timed out due to a DAG import or parse error." ### How to reproduce 1. Create an Airflow Variable with the key "some_variable" 2. Trigger a DAG that makes a top-level call to `Variable.get("some_variable")`. While the DAG is running, delete the Airflow Variable. 3. Wait 10-15 minutes or so 4. Recreate an Airflow Variable with the key "some_variable" ### Operating System n/a ### Versions of Apache Airflow Providers _No response_ ### Deployment Astronomer ### Deployment details _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) </div> -- 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]
