Stabilizing the branch for 2.0 should take at least that long I would expect.
I think we can cut a branch now. Seems reasonable. Start work on stabilizing what's there. Wrap up work or land new things in the new master aka 3.0 and bring into branch 2.0 as reasonable under supervision of the co RMs. No time like the present to begin. > On Oct 14, 2016, at 12:09 PM, Ted Yu <yuzhih...@gmail.com> wrote: > > I am in support of getting more JIRAs resolved before branch-2 is created. > > However, we should keep high bar for new features. > > Mikhail et al spent 5 months stabilizing branch-1.3 > > Hopefully stabilizing branch-2 doesn't take that long. > > Cheers > > On Fri, Oct 14, 2016 at 12:04 PM, Stephen Jiang <syuanjiang...@gmail.com> > wrote: > >> I did some search on how big HBASE 2.0 is. Here is some interesting >> numbers: >> >> (1). How many resolved as fixed JIRAs are for 2.0 only? >> As of today morning (10/14), we have *875*! >> >> I used the following query: " >> project = HBase AND fixVersion = 2.0.0 AND fixVersion not in >> releasedVersions() AND fixVersion not in (1.3.0, 1.4.0) AND resolution = >> Fixed ORDER BY fixVersion ASC >> " - if we think 1.4.0 release would not happen sooner than 2.0.0, then the >> total number is bigger (*970*!) >> >> (2). How many JIRAs are not resolved, but marked for 2.0.0? >> >> As of today morning (10/14), we have *531* tickets. *132* of the 531 >> tickets are Patch Available. >> >> Of the 531 JIRAs, *352* are non-BUG/Test; so we have a lot of new >> features/improvement/tasks going on for 2.0.0. >> >> I used the following query: " >> project = HBase AND (fixVersion = 2.0.0 OR affectedVersion = 2.0.0) AND >> resolution is EMPTY ORDER BY fixVersion DESC >> " - this does not include any features that not marked any version; so the >> actual number would be bigger. >> >> >> If we can get more JIRAs resolved before branch-2, that would help reduce >> development/maintenance cost. After branch-2 is created, we need to trim >> down moving parts for sure. >> >> Thanks >> Stephen >>
