I don't mind cutting 5.1 from before 4.0 was added, but only if we have a high level of confidence that it won't take more than a couple weeks to approve and finalize; if we want 6.0 to stay on schedule as the "next" release then 5.1 shouldn't get in its way with a bunch of backports.
On Tue, Feb 2, 2021 at 11:24 AM Rawlin Peters <[email protected]> wrote: > I can get on board with a quick 5.1. > > However, API 4.x already exists in master, and I know there was at > least one more breaking change to the DS API that might be coming > soon. It would be nice to not have to turn around and immediately > create API 5.x for that, so maybe we could choose a branch point on > master from _before_ API 4.x was added. Then we could continue adding > stuff to 4.x in time for a 6.0 release in a couple months. > > - Rawlin > > On Tue, Feb 2, 2021 at 11:10 AM Jeremy Mitchell <[email protected]> > wrote: > > > > Now that TC 5.0 is official (thanks Brennan for seeing that thru as the > 5.0 > > release manager), we should think about our next TC release. Here are > some > > larger things in the works: > > > > 1. Perl removal > > 2. API 1.x removal > > 3. API 2.x deprecation > > 4. API 4.x addition > > 5. Go modules > > > > Those items make this feel like a larger release (6.0) is warranted > rather > > than a minor upgrade (5.1). > > > > However, there might be some value in cutting 5.1 ASAP (before those 5 > > things are merged) to pick up some topology improvements / bug fixes as > you > > can see here: > > https://github.com/apache/trafficcontrol/compare/5.0.x...master > > > > thoughts? > > > > Jeremy >
