Is this because the scheduler can't understand that a month has passed and kick off the DAG? Should I just set the start date to some past date like datetime(2017,1,1)?
On Sat, Sep 2, 2017 at 5:16 PM, Maxime Beauchemin < [email protected]> wrote: > We advise against using dynamic start_date. > > Keep in mind that we use use the left bound of the period covered, meaning > the monthly DAG run labelled 2017-09-01 will be kicked off soon after > 2017-10-01 assuming it processes the data accumulated over the month of > September. > > On Sep 2, 2017 6:27 AM, "Ali Uz" <[email protected]> wrote: > > > I saw in some tutorial that they set the start date as following: > > > > 'start_date': datetime_obj.combine( > > datetime_obj.today() - datetime.timedelta(1), > > datetime_obj.min.time()) > > > > I guess this is not the correct way to do it for a monthly schedule? > > > > > > On Sat, Sep 2, 2017 at 4:25 PM, <[email protected]> wrote: > > > > > Hi Ali, > > > What is the 'start date' of your dag? > > > > > > > > > ב-2 בספט׳ 2017, בשעה 13:54, Ali Uz <[email protected]> כתב/ה: > > > > > > > Hello, > > > > > > > > I have a couple of DAGs that are set to run @monthly. They did not > run > > on > > > > September 1st 00:00 nor did they run after that period had ended > > > (September > > > > 2nd 00:00). I'm not sure why, and hope you guys can help me out in > > > > troubleshooting it. > > > > > > > > Thanks in advance! > > > > > > > > -Ali > > > > > >
