airflow-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Greg Lira <>
Subject BranchPythonOperator and wait_for_downstream
Date Thu, 01 Sep 2016 08:24:56 GMT

For some of our DAGs [where we clear and re-import the staging tables], having wait_for_downstream
on the tasks is very important.
However, we have a BranchPythonOperator for happy and error path, so if the flow has successfully
completed, the tasks of the error path as marked as skipped.
It seems that it doesn't work well with  wait_for_downstream, where airflow checks for the
task instances with SUCCESS state only.
Is this 'by design'?
How can we handle sequential execution of the DAGs then? Should we completely re-design the
workflow or there is a way to do that with airflow configuration settings?


View raw message