vbottu commented on code in PR #61798:
URL: https://github.com/apache/airflow/pull/61798#discussion_r2807874030
##########
providers/cncf/kubernetes/src/airflow/providers/cncf/kubernetes/executors/kubernetes_executor.py:
##########
@@ -79,14 +79,29 @@ class KubernetesExecutor(BaseExecutor):
RUNNING_POD_LOG_LINES = 100
supports_ad_hoc_ti_run: bool = True
+ supports_multi_team: bool = True
if TYPE_CHECKING and AIRFLOW_V_3_0_PLUS:
# In the v3 path, we store workloads, not commands as strings.
# TODO: TaskSDK: move this type change into BaseExecutor
queued_tasks: dict[TaskInstanceKey, workloads.All] # type:
ignore[assignment]
- def __init__(self):
- self.kube_config = KubeConfig()
+ def __init__(self, *args, **kwargs):
+ super().__init__(*args, **kwargs)
Review Comment:
@jscheffl Per my understanding--
The original code called
`super().__init__(parallelism=self.kube_config.parallelism)` at the end. My
change calls `super().__init__(*args, **kwargs)` first, then overrides with
`self.parallelism = self.kube_config.parallelism` .The end result is the same
`self.parallelism` gets the KubeConfig value either way.
This is the same pattern CeleryExecutor and LocalExecutor follow — both call
`super().__init__(*args, **kwargs)` without passing parallelism explicitly. The
AIP-67 multi-team pattern requires [`args, **kwargs` to forward `team_name`
from the scheduler, so we can't use the old explicit-parameter approach anymore.
Please let me know your thoughts
--
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]
