kaxil opened a new pull request, #60957:
URL: https://github.com/apache/airflow/pull/60957
## Summary
In `dags_needing_dagruns()`, the query loads `AssetDagRunQueue` records with
`joinedload(dag_model)`, but then accesses `adrq.asset` without eager loading
when building the `dag_statuses` dict. This triggers a separate query for each
ADRQ record (N+1).
```python
# Line 643 accesses adrq.asset without eager loading
dag_statuses: dict[str, dict[UKey, bool]] = {
dag_id: {SerializedAssetUniqueKey.from_asset(adrq.asset): True for adrq
in adrqs}
for dag_id, adrqs in adrq_by_dag.items()
}
```
## Fix
Add `joinedload(AssetDagRunQueue.asset)` to load assets in a single joined
query:
```python
for adrq in session.scalars(
select(AssetDagRunQueue).options(
joinedload(AssetDagRunQueue.dag_model),
joinedload(AssetDagRunQueue.asset), # <-- Added
)
):
```
## Test
Added `test_dags_needing_dagruns_query_count` to verify query count stays
constant regardless of ADRQ record count, preventing N+1 regression.
--
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]
