I’ve been working with 2.4 for months and it is stable enough for me to +1 moving the pointer at this time. This work includes many tests with 2.4 using ITBLL with slowDeterministic and serverKilling policies and there has not been an unstable result, as defined by crashes/loss of daemons by the IT framework, or data issues requiring hbck.
> On Aug 2, 2021, at 2:27 PM, Nick Dimiduk <ndimi...@apache.org> wrote: > > On Mon, Aug 2, 2021 at 1:22 PM Andrew Purtell <andrew.purt...@gmail.com> > wrote: > >> EOL of 2.3 seems premature. >> >> The stable pointer is still pointing to 2.3. >> >> Advancing the stable pointer is an obvious prerequisite. If there are any >> concerns blocking advancing the pointer to 2.4, this is where we should >> start the discussion. >> > > Point taken. I see the only conclusion of our thread [0] was to agree that > we should define "stable" criteria. I thought we'd already advanced the > marker. > > [0]: > https://lists.apache.org/thread.html/r1cc4528a6a35cd1b0d38398aa61ad642a368901795d6970544d0a0a9%40%3Cdev.hbase.apache.org%3E > >>> On Aug 2, 2021, at 11:23 AM, Nick Dimiduk <ndimi...@apache.org> wrote: >>> >>> Heya, >>> >>> I'd like to start a discussion regarding the declaration of the end of >> life >>> for branch-2.3. This release line has matured nicely, but with the stable >>> pointer moving forward, and our community interest in pushing on 2.5 and >>> 3.0, I think it's worth considering whether we want to continue putting >>> resources into this release line. >>> >>> I propose we make one final release, 2.3.7, whenever we have enough >> commits >>> to the branch, or the beginning of October, whichever comes first. >>> >>> Thoughts? >>> >>> Thanks, >>> Nick >>
