[GitHub] [airflow] potiuk edited a comment on pull request #19862: Clarify behaviour of test_backfill_depends_on_past

[GitBox <gi...@apache.org>]

potiuk edited a comment on pull request #19862:
URL: https://github.com/apache/airflow/pull/19862#issuecomment-981097308


   I am not 100% - but from what I debugged, the behaviour of backfill in case `depends_on_past==Truw` in a task have been fixed in the meantime (likely during timeteable implementation):
   
   From what I see the backfil job will not deadlock any more but it will **just work** even if "ignore_first_depends_on_past" is set to true, simply because in prev_dagrun_dep, the prev_dagrun_dep is None (as it should be IMHO):
   
   in  https://github.com/apache/airflow/blob/846586eb86cfa927eaebbba7124e8cd34c0bebce/airflow/ti_deps/deps/prev_dagrun_dep.py#L52
   
   No matter if catchup is True or false, previous DagRun is None  (as I think it should be for backfill job).
   ```
   
           if catchup:
               last_dagrun = dr.get_previous_scheduled_dagrun(session)
           else:
               last_dagrun = dr.get_previous_dagrun(session=session)
   ```
   
   However I'd love to hear other's comment if understand it right. I am not sure if there is other case where we actually still need `ignore_first_depends_on_past` in this case or should we depreacate 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: commits-unsubscribe@airflow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org