ashb commented on code in PR #27725:
URL: https://github.com/apache/airflow/pull/27725#discussion_r1028066724
##########
airflow/jobs/scheduler_job.py:
##########
@@ -774,25 +775,30 @@ def _execute(self) -> None:
@provide_session
def _update_dag_run_state_for_paused_dags(self, session: Session =
NEW_SESSION) -> None:
try:
- paused_dag_ids = DagModel.get_all_paused_dag_ids()
- for dag_id in paused_dag_ids:
- if dag_id in self._paused_dag_without_running_dagruns:
- continue
-
- dag = SerializedDagModel.get_dag(dag_id)
- if dag is None:
- continue
- dag_runs = session.query(DagRun).filter(
- DagRun.dag_id == dag_id,
+ dag = None
+ paused_runs = (
+ session.query(DagRun)
+ .join(DagRun.dag_model)
+ .join(TaskInstance)
+ .filter(
+ DagModel.is_paused == expression.true(),
DagRun.state == DagRunState.RUNNING,
DagRun.run_type != DagRunType.BACKFILL_JOB,
)
- for dag_run in dag_runs:
- dag_run.dag = dag
- _, callback_to_run =
dag_run.update_state(execute_callbacks=False)
- if callback_to_run:
- self._send_dag_callbacks_to_processor(dag,
callback_to_run)
- self._paused_dag_without_running_dagruns.add(dag_id)
+ .having(DagRun.last_scheduling_decision <=
func.max(TaskInstance.updated_at))
+ .group_by(DagRun)
+ .order_by(DagRun.dag_id)
+ )
+ for dag_run in paused_runs:
+ if dag is None or dag.dag_id != dag_run.dag_id:
+ dag = SerializedDagModel.get_dag(dag_run.dag_id)
Review Comment:
Rather than using SerializedDagModel directly we should use ` DagBag` as
that has caching built in.
```suggestion
dag = self.dagbag.get_dag(dag_run.dag_id,
session=session)
```
--
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]
