> So we need to roll new RCs for both 2.1.5 and 2.2.0? Yes, both RCs have missing commits and incomplete Release notes.
> The base patch didn't commit to branch-2.2? The committer missed to push the change to branch-2.2 even though there was a discussion HBASE-21991 should land there. Fix version also included 2.2.0 but not 2.1.5 where it was actually committed to. On Sun, May 26, 2019 at 4:56 AM Guanghao Zhang <[email protected]> wrote: > The base patch didn't commit to branch-2.2? I found HBASE-21991 because it > was a reopened issue and fix version was 2.2.0. Then I helped to commit the > addendum to branch-2.2 and rolled 2.2.0RC4...... > > 张铎(Duo Zhang) <[email protected]> 于2019年5月26日周日 上午10:39写道： > > > So we need to roll new RCs for both 2.1.5 and 2.2.0? > > > > Peter Somogyi <[email protected]>于2019年5月26日 周日05:46写道： > > > > > Apologies, I misinterpreted git log and JIRA previously. There is no > need > > > for a new issue to move the LossyCounting class to IA.Private. > > > > > > What is needed: > > > branch-2.1: commit missing addendum and add 2.1.5 to fixed versions > > > branch-2.2: commit missing base patch > > > > > > On Sat, May 25, 2019 at 10:30 PM Andrew Purtell < > > [email protected]> > > > wrote: > > > > > > > We have decided this is not part of the public API so all that is > > needed > > > > is to change the annotation and post new RCs with that change with an > > > > update to release notes. It doesn’t matter if there was an > incompatible > > > > change to the class made or not. A simple audience annotation mistake > > is > > > > taking disproportionate attention away from more important efforts. > > > > > > > > If an annotation change to one class is the only update in a new RC > you > > > > can have confidence in porting your votes from the last RC to the new > > one > > > > after confirming sums and signatures. For your consideration. > > > > > > > > > On May 25, 2019, at 12:37 PM, Peter Somogyi > > > > <[email protected]> wrote: > > > > > > > > > > Unfortunately, that would require a new RC for 2.1.5 and I'm afraid > > > > > Guanghao already started the process for 2.0.0. > > > > > > > > > > On Sat, May 25, 2019 at 12:21 PM 张铎(Duo Zhang) < > > [email protected]> > > > > > wrote: > > > > > > > > > >> A new issue can always solve the problem, I believe. I mean, > revert > > > the > > > > >> addendum from branch-2.2-, and open a new issue, which just > changes > > > the > > > > >> annotation for branch-2.2-, and commit the addendum again, with a > > new > > > > >> commit message. > > > > >> > > > > >> Peter Somogyi <[email protected]>于2019年5月25日 周六16:42写道： > > > > >> > > > > >>> On the 2.1.5RC0 testing I noticed that the release notes do not > > state > > > > >> that > > > > >>> LossyCounting was moved to IA.Private. This is tricky since the > > > > original > > > > >>> commit which introduced the incompatibility was not committed to > > > > >>> branch-2.1, only the addendum which only modified the IA > > annotation. > > > > For > > > > >>> 2.2.0 we have the same problem, however, 2.2.0 was added as fixed > > > > version > > > > >>> to HBASE-21991 even though only the addendum was committed to > that > > > > branch > > > > >>> as well. > > > > >>> > > > > >>> Do we have any best practices for such a case? > > > > >>> > > > > >>> Thanks, > > > > >>> Peter > > > > >>> > > > > >>> On Fri, May 10, 2019 at 6:57 PM Sakthi < > [email protected] > > > > > > > >> wrote: > > > > >>> > > > > >>>> Looks good to me. > > > > >>>> > > > > >>>> Sakthi > > > > >>>> > > > > >>>> On Fri, May 10, 2019 at 6:18 AM 张铎(Duo Zhang) < > > > [email protected]> > > > > >>>> wrote: > > > > >>>> > > > > >>>>> +1. > > > > >>>>> > > > > >>>>> Jan Hentschel <[email protected]> 于2019年5月10日周五 > > > > >>> 下午9:08写道： > > > > >>>>> > > > > >>>>>> Also +1 for making it IA.Private. > > > > >>>>>> > > > > >>>>>> From: Peter Somogyi <[email protected]> > > > > >>>>>> Reply-To: "[email protected]" <[email protected]> > > > > >>>>>> Date: Friday, May 10, 2019 at 1:41 PM > > > > >>>>>> To: "[email protected]" <[email protected]> > > > > >>>>>> Subject: Re: [DISCUSS] Moving IA.Public class LossyCounting to > > > > >>>> IA.Private > > > > >>>>>> in all maintenance branches > > > > >>>>>> > > > > >>>>>> +1 on moving LossyCounting to IA.Private > > > > >>>>>> > > > > >>>>>> On Fri, May 10, 2019 at 7:54 AM Stack <[email protected] > <mailto: > > > > >>>>>> [email protected]>> wrote: > > > > >>>>>> > > > > >>>>>> Looks good to me. > > > > >>>>>> S > > > > >>>>>> > > > > >>>>>> On Thu, May 9, 2019 at 7:02 PM Sean Busbey <[email protected] > > > > >>> <mailto: > > > > >>>>>> [email protected]>> wrote: > > > > >>>>>> > > > > >>>>>>> Hi folks! > > > > >>>>>>> > > > > >>>>>>> just a heads up that a few of us are planning to move a class > > out > > > > >>> of > > > > >>>>>>> the public API without a deprecation cycle. > > > > >>>>>>> > > > > >>>>>>> From the planned release note: > > > > >>>>>>> > > > > >>>>>>>> The class LossyCounting was unintentionally marked Public > but > > > > >> was > > > > >>>>> never > > > > >>>>>>>> intended to be part of our public API. This oversight has > been > > > > >>>>>> corrected > > > > >>>>>>>> and LossyCounting is now marked as Private and going forward > > > > >> may > > > > >>> be > > > > >>>>>>>> subject to additional breaking changes or removal without > > > > >> notice. > > > > >>>> If > > > > >>>>>> you > > > > >>>>>>>> have taken a dependency on this class we recommend cloning > it > > > > >>>> locally > > > > >>>>>>> into > > > > >>>>>>>> your project before upgrading to this release. > > > > >>>>>>> > > > > >>>>>>> This class was came in via HBASE-19722 and was published in > > HBase > > > > >>>>>>> 1.4.6, 2.0.2, and 2.1.3 (and depending on RC timing might be > in > > > > >>>>>>> 2.2.0). > > > > >>>>>>> > > > > >>>>>>> It will move to IA.Private as of 1.4.10, 1.5.0, 2.0.6, 2.1.5 > > and > > > > >>>> later > > > > >>>>>>> (maybe 2.2.1 depending on RC timing). The class already has > > > > >>>>>>> backwards-incompatible changes set to happen in upcoming > > releases > > > > >>>>>>> 1.4.10, 1.5.0, and 2.2.0. > > > > >>>>>>> > > > > >>>>>>> Please speak up sooner rather than later if you'll have a > > problem > > > > >>>>>>> voting on RCs that include this change, either here or on > > > > >>>> HBASE-21991. > > > > >>>>>>> > > > > >>>>>> > > > > >>>>>> > > > > >>>>>> > > > > >>>>> > > > > >>>> > > > > >>> > > > > >> > > > > > > > > > >
