Absolutely - it's great that we have the checks in master :). But I am thinking about some solutions already. We actually can do something using Kaxil's probot. While preventing merges for all PRs that have not been rebased to latest master is quite an overkill, but I think we can prevent merging PRs that have not been rebased AND have some modifications in "airflow/migrations".
It's rather easy to add such features to probot I'd say. Maybe someone would like to learn how to write probots and would like to add such check? I am happy to provide mentoring and guidance on how to build/test it (it's rather easy once you have it setup). J. On Tue, Jan 14, 2020 at 9:51 AM Driesprong, Fokko <fo...@driesprong.frl> wrote: > Hi Jarek, > > Hopefully, this doesn't happen too often since the tables should be slowly > changing. The solution here is to ask the contributors to rebase their PR > more often. > > I'm happy that this is being caught now, and we don't have multiple heads > as before :-) > > Cheers, Fokko > > Op ma 13 jan. 2020 om 22:23 schreef Jarek Potiuk <jarek.pot...@polidea.com > >: > > > BTW. We'll have to find a better solution to prevent it as it has > happened > > in the past. I will think about it :). Any ideas are welcome :). > > > > On Mon, Jan 13, 2020 at 10:21 PM Jarek Potiuk <jarek.pot...@polidea.com> > > wrote: > > > > > Hello everyone, > > > > > > Without asking I quickly reverted the > > > https://github.com/apache/airflow/pull/6975 ([AIRFLOW-1467 > > > <https://issues.apache.org/jira/browse/AIRFLOW-1467>] Dynamic pooling > > via > > > allowing tasks to use more than one pool slot (depending upon the > need)) > > > as it had duplicated heads in sqlite migrations - clashing with an > > earlier > > > change > > > > > > https://github.com/apache/airflow/commit/a7cacf593f5cf4bfc8b192b799aa2b14c96eac5b > > added in > > > this PR https://github.com/apache/airflow/pull/6489: [AIRFLOW-4026] > > > <https://issues.apache.org/jira/browse/AIRFLOW-4026> > > > > > > The problem was that both created independently SQL Alchemy migration > > > heads and we had a test failing because of that in master. The problem > > was > > > that the AIRFLOW-1467 was not rebased to the latest master before > merging > > > (otherwise our tests detect this problem). > > > > > > I will let the author of AIRFLOW-1467 how to fix it. > > > > > > J. > > > > > > -- > > > > > > Jarek Potiuk > > > Polidea <https://www.polidea.com/> | Principal Software Engineer > > > > > > M: +48 660 796 129 <+48660796129> > > > [image: Polidea] <https://www.polidea.com/> > > > > > > > > > > -- > > > > Jarek Potiuk > > Polidea <https://www.polidea.com/> | Principal Software Engineer > > > > M: +48 660 796 129 <+48660796129> > > [image: Polidea] <https://www.polidea.com/> > > > -- Jarek Potiuk Polidea <https://www.polidea.com/> | Principal Software Engineer M: +48 660 796 129 <+48660796129> [image: Polidea] <https://www.polidea.com/>
