potiuk commented on issue #30146:
URL: https://github.com/apache/airflow/issues/30146#issuecomment-1473460460

   I do not have a lot of time to look at this in detail but from the quick 
view and speed with which we got the report (thanks @eejbyfeldt ) I think we 
should revert it for 2.5.3. And think about potentially improving the callback 
mechanism in 2.6 or beyond. Part of discussion on the new SLA proposal  
(https://lists.apache.org/thread/kn1rb9l70jk16j83fycp68l2g70wnhzl) is about 
potential of making changes for all the callbacks (including the proposed sla 
callback) following much more "at least one" [delivery 
semantics](https://wiki.cantara.no/display/architecture/At-least+once+semantics)
 in a more "systematic" way - because we have loosly "at-most-once" now it 
seems coupled with some buggy cases where we have "not-even-once" semantics :D. 
   
   Also dragging attention of @pierrejeambrun and @jedcunningham.
   
   We already have quite a few fixes merged for 2.5.3 it looks, so I would be 
personally (also following our "release-early/release-often" discussion) be for 
reverting that one and releasing 2.5.3 - maybe even in a week or so. Possibly 
with some more scrutiny and attempts to better diagnose some of those issues 
that we've been regularly moving from milestone-to-milestone - there are a few 
of those - likely important - that seem to be perpetually in the "next" bugfix 
release.  Happy to join the efforts :). 


-- 
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