potiuk commented on issue #59043:
URL: https://github.com/apache/airflow/issues/59043#issuecomment-3775794337

   > With "all runs" option the existing runs are cleared without passing 
params [#51439 
(comment)](https://github.com/apache/airflow/issues/51439#issuecomment-2949835692)
 .
   > 
   > [@dstandish](https://github.com/dstandish) Is it intentional to not pass 
params to existing dagruns during backfill?
   > 
   > 
[airflow/airflow-core/src/airflow/models/backfill.py](https://github.com/apache/airflow/blob/0a1e798fb71520f5848583f63c006c97c6ea7065/airflow-core/src/airflow/models/backfill.py#L355-L364)
   > 
   > Lines 355 to 364 in 
[0a1e798](/apache/airflow/commit/0a1e798fb71520f5848583f63c006c97c6ea7065)
   > 
   >  if lock: 
   >      _handle_clear_run( 
   >          session=session, 
   >          dag=dag, 
   >          dr=dr, 
   >          info=info, 
   >          backfill_id=backfill_id, 
   >          sort_ordinal=backfill_sort_ordinal, 
   >          run_on_latest=run_on_latest_version, 
   >      )
   
   I think this is a question to @dstandish  mostly - to see if that behaviour 
is intentional (and why - possibly there is a good reason) - I can imagine 
changing the dag run parameters for existing DagRuns could be somewhat 
problematic - but I am not sure if it was a deliberate decision or sometihing 
that should be fixed (seems like fix might be easy, but maybe also this might 
require a whole new feature wher user might be able to choose whether to 
override the params or not for existing DagRuns. The original AIP does not 
mention it 
https://cwiki.apache.org/confluence/display/AIRFLOW/AIP-78+Scheduler-managed+backfill
 but maybe this has been discussed somewhere?


-- 
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