> If this is true, why do we even bother running any CI before the CEP-21 > merge? It will all be invalidated anyway, right? I'm referring to manual validation or soak testing in qa environments rather than automated. Just because a soft-frozen branch without those features works in QA doesn't mean a branch after those features merge will be equally qualified.
There's certainly value in rolling out a soft-frozen branch before those features merge, but we will also need to be very deliberate and clear about having a minimum bound of time for people to perform testing *after* these two features merge as well if we go the "freeze but let them merge after" route. On Mon, Apr 17, 2023, at 5:12 PM, Mick Semb Wever wrote: >> My personal .02: I think we should consider branching 5.0 September 1st. >> That gives us basically 12 weeks for folks to do their testing and for us to >> stabilize anything that's flaky in circle or regressed in ASF CI. > > > I'm not for this, sorry. I see the real risk here of there being no GA > release this year. > > My proposal was based on reading through the thread and gathering what I saw > to be the best middle ground for everyone. It's not my first choice, but as a > middle ground I can accept it. > > Caleb, you appear to be the only one objecting, and it does not appear that > you have made any compromises in this thread. Can I ask that you do? I (and > others) do see that letting testing start as soon as it can, where they can, > as an important tactic to de-risking an important 5.0 release. > >
