Jarek Potiuk
Jarek Potiuk
Hey @ashb @yuqian90 ? @uranusjr @ephraimbuddy @jedcunningham @RNHTTR - tagging those who I know were somewhat involved with Celery and maybe you have more of an experience/insight. I think this...
Those are the two experimental PRs that will hopefully help to validate the hypothesis about "moving test being trigger": https://github.com/apache/airflow/pull/33001: [EXPERIMENT] Tests for Celery executor moved back to core https://github.com/apache/airflow/pull/33002:...
> #33001 succeeds in AWS team test environment Wonderful. Thank you !
Sure.
Another error that seems to be a duplicate in #30898 - @parth166 if you are still working on it, that might give you more clues on reproducibility
deassigned you. Hopefully someone else will look at it
Assigned you - reopened the other one then, it `looked` the same but maybe the reason is different.
Then we will have to wait for 3.0.0 of connexion https://pypi.org/project/connexion/3.0.0a6/ released 24 April with werkzeug
> This seems like a related werkzeug issue : [pallets/werkzeug#2657](https://github.com/pallets/werkzeug/issues/2657) . @fokmess Can you please confirm the version of werkzeug used and the URL before and after you hit search...
> @potiuk My question was since the bug is present only on 2.3.0 (fixed in 2.3.1) and there is "connexion 2.14.2 requires werkzeug=1.0, but you have werkzeug 2.3.0 which is...