Miretpl commented on code in PR #76749:
URL: https://github.com/apache/airflow/pull/76749#discussion_r4576337490


##########
chart/values.yaml:
##########
@@ -646,6 +646,12 @@ workers:
       failureThreshold: 5
       periodSeconds: 60
       command: ~
+      # Grace period for a kill triggered by this probe. Unset, the pod's
+      # `terminationGracePeriodSeconds` applies, which workers often set to 
hours
+      # so that a rolling update drains running tasks. A worker that stopped
+      # answering can still wait on in-flight or hung tasks during its warm
+      # shutdown; a shorter period here bounds that wait. If set, minimum 
value is 1.

Review Comment:
   ```suggestion
   ```
   Most of this comment is about how this liveness works + consequences for the 
worker instance, which I think should be well known and rather obvious to 
anyone working with Kubernetes and Airflow.



##########
chart/newsfragments/76749.improvement.rst:
##########
@@ -0,0 +1 @@
+Add ``workers.celery.livenessProbe.terminationGracePeriodSeconds`` to bound 
how long a Celery worker killed by its liveness probe may take to stop (#76749)

Review Comment:
   Could you please remove it? I think that we don't need a newsfragment for it.



-- 
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