karenbraganz opened a new issue, #58536:
URL: https://github.com/apache/airflow/issues/58536

   
   ### Discussed in https://github.com/apache/airflow/discussions/53868
   
   <div type='discussions-op-text'>
   
   <sup>Originally posted by **Szjs-z** July 29, 2025</sup>
   Hi Airflow community,
   
   I've encountered an issue where task duration keeps increasing for tasks 
that are skipped due to DAG timeout:
   
   **Problem Description:**
   When a DAG run times out due to `dagrun_timeout`, the unfinished tasks are 
marked as `SKIPPED`. However, in the Web UI, the task duration continues to 
increase even though the task state is `SKIPPED` :
   
   <img width="770" height="428" alt="Screenshot 2025-07-29 at 17 42 11" 
src="https://github.com/user-attachments/assets/720b243e-d915-4275-a615-b09a31bbbdb5";
 />
   
   
   **Root Cause Analysis:**
   From code inspection in `airflow/jobs/scheduler_job.py`, the 
`_schedule_dag_run` method handles DAG timeout by:
   
   ```python
           if (
               dag_run.start_date
               and dag.dagrun_timeout
               and dag_run.start_date < timezone.utcnow() - dag.dagrun_timeout
           ):
               dag_run.set_state(DagRunState.FAILED)
               unfinished_task_instances = session.scalars(
                   select(TI)
                   .where(TI.dag_id == dag_run.dag_id)
                   .where(TI.run_id == dag_run.run_id)
                   .where(TI.state.in_(State.unfinished))
               )
               for task_instance in unfinished_task_instances:
                   task_instance.state = TaskInstanceState.SKIPPED
                   session.merge(task_instance)
               session.flush()
               self.log.info("Run %s of %s has timed-out", dag_run.run_id, 
dag_run.dag_id)
   
   ```
   
   1. Setting DAG run state to FAILED
   2. Marking unfinished tasks as SKIPPED
   3. But it doesn't set `end_date` for these task instances
   
   I'm wondering if this is a known issue or if there's a recommended way to 
handle this.
   
   Thanks,
   Chris Zhao</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]

Reply via email to