[
https://issues.apache.org/jira/browse/AIRFLOW-957?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16601543#comment-16601543
]
Apache Spark commented on AIRFLOW-957:
--------------------------------------
User 'anxodio' has created a pull request for this issue:
https://github.com/apache/incubator-airflow/pull/2238
> the execution_date of dagrun that is created by TriggerDagRunOperator is not
> euqal the execution_date of TriggerDagRunOperator's task instance
> ----------------------------------------------------------------------------------------------------------------------------------------------
>
> Key: AIRFLOW-957
> URL: https://issues.apache.org/jira/browse/AIRFLOW-957
> Project: Apache Airflow
> Issue Type: Improvement
> Components: operators
> Reporter: Calvin Wang
> Assignee: Angel Fernandez
> Priority: Major
>
> The execution_date of Dagrun that be created by TriggerDagRunOperator is
> triggered time. When I rerun a dag that execution_date is a past time(e.g.
> 20160101), the execution_date of external dag that trigger by
> TriggerDagRunOperator is current time(e.g. 20170309). I expect
> execution_date is a past time(e.g. 20160101) . Why TriggerDagRunOperator
> doesn't provide the exection_date option at the beginning? Is there anything
> wrong in this way?
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
