kaxil commented on a change in pull request #14883:
URL: https://github.com/apache/airflow/pull/14883#discussion_r606229459
##########
File path: airflow/executors/celery_executor.py
##########
@@ -543,20 +539,27 @@ def __init__(self, sync_parallelism=None):
super().__init__()
self._sync_parallelism = sync_parallelism
+ def _tasks_list_to_task_ids(self, async_tasks) -> Set[str]:
+ return {a.task_id for a in async_tasks}
+
def get_many(self, async_results) -> Mapping[str, EventBufferValueType]:
"""Gets status for many Celery tasks using the best method
available."""
if isinstance(app.backend, BaseKeyValueStoreBackend):
result = self._get_many_from_kv_backend(async_results)
- return result
- if isinstance(app.backend, DatabaseBackend):
+ elif isinstance(app.backend, DatabaseBackend):
result = self._get_many_from_db_backend(async_results)
- return result
- result = self._get_many_using_multiprocessing(async_results)
- self.log.debug("Fetched %d states for %d task", len(result),
len(async_results))
+ else:
+ async_results = list(async_results) if isinstance(async_results,
map) else async_results
Review comment:
Task Adoption is when running a scheduler, if it dies another scheduler
is able to "adopt" tasks started by that Scheduler/executor -- this currently
works for Celery, Kubernetes (and CeleryKubernete) executor
Code:
**Scheduler**:
https://github.com/apache/airflow/blob/4d1b2e985492894e6064235688cf8f381b1e8858/airflow/jobs/scheduler_job.py#L1825
**Celery**:
https://github.com/apache/airflow/blob/4d1b2e985492894e6064235688cf8f381b1e8858/airflow/executors/celery_executor.py#L345
--
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.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
