Re: Idea for Discussion: custom TI dependencies

I already like the idea. It can indeed offer better flexibility and control over the scheduling process beyond the default checks My main concern with the approach is in line with what Jens mentioned, performance but I like yout mitigation through a metric. I'd suggest that the metric should have

Re: [VOTE] AIP 61 - Hybrid Executors

+1 but may I make a small(?) suggestion: If we also swap the alias and class name order around: From: 'LocalExecutor,my.custom.module.Executor' To: 'default=LocalExecutor, shortname=my.custom.module.Executor' This way opens up the option to have two different instances of the same executor clas

Re: [RESULT][VOTE] January 2024 PR of the Month

Ohh haha. Mail was trickling on slowly. Well deserved Jarek. Sent from my iPhone > On 30 Jan 2024, at 20:07, Briana Okyere > wrote: > > Hey All, > > Congratulations to Jarek on winning PR of the Month for January 2044 with PR > #36537: Standardize airflow build process and switch to Hatchlin

Re: [VOTE] January 2024 PR of the Month

I'll give 22253 my vote. That one was from outside and resurrected from the dead. Jarek's is also great. Bolke Sent from my iPhone > On 25 Jan 2024, at 01:52, Mehta, Shubham wrote: > > +1 for #36537. > > Shubham > > > On 2024-01-23, 3:03 PM, "Scheffler Jens (XC-AS/EAE-ADA-T)" >

Re: [VOTE] AIP 61 - Hybrid Executors

+1 (non-binding) - Eugene Kostieiev On Sat, Feb 3, 2024 at 4:43 PM Andrey Anshin wrote: > +1 binding > > > On 1 Feb 2024, at 09:12, Oliveira, Niko > wrote: > > > > Hey folks, > > > > > > The AIP for Hybrid Executors has been out for a few weeks now. Some > great feedback came in and some chal