You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@hbase.apache.org by Nick Dimiduk <nd...@apache.org> on 2020/08/25 17:36:53 UTC

[DISCUSS] A plan for 2.4

Heya!

Now that 2.3.x is in maintenance mode, I'd like to start a discussion on
2.4. We've long agreed that a more frequent cadence of minor releases is
superior to a long gap between them. I think we should target a 2.4.0
release sometime before the end of the calendar year, perhaps the first
weeks of December.

Hopefully the workload for the next release manager isn't as high as it was
for 2.3, given that much of the test stabilization effort happened on
branch-2. However, we are running in a new jenkins environment, and entropy
is nothing if not persistent, which puts that hard-won stability back into
doubt. Thus, I think it would be good if we can start sooner than later.

Would anyone like to volunteer to run the 2.4 release line? The level of
effort is admittedly non-negligible: (1) stabilize branch-2 (2) test
new-for-branch-2 features (3) run initial release (4) cut monthly patch
releases until 2.5 is out. I have a strong interest in seeing this release
through, so I volunteer to assist with these activities.

Thanks,
Nick

Re: [DISCUSS] A plan for 2.4

Posted by Nick Dimiduk <nd...@apache.org>.

> To meet Nick's objective of a 2.4 release in November

To be clear, my objective is a 2.4.0 release that lands roughly 6 months
after 2.3.0 landed, which was 13-07-2020. Given the extent of end-of-year
holidays, and how little feature development work we expect at that time, I
think it would be nice for 2.4.0 to land in early December. As I recall, we
needed about 6 weeks of RCs to settle out 2.3.0, so I'm trying to account
for a similar schedule for 2.4.0.

On Thu, Oct 15, 2020 at 12:00 PM Andrew Purtell <ap...@apache.org> wrote:

> Is it a blocker? Maybe mark it Critical? Either way, a week is well within
> the timeframe discussed earlier. Will come back here to check progress at
> the end of next week. In the meantime it's fine to wait in my opinion.
>
>
> On Wed, Oct 14, 2020 at 8:55 PM Stack <st...@duboce.net> wrote:
>
> > On Wed, Oct 14, 2020 at 6:58 PM 张铎(Duo Zhang) <pa...@gmail.com>
> > wrote:
> >
> > > I do not think HBASE-18070 is a blocker for 2.4 as it is still on a
> > feature
> > > branch.
> > > And in the title of several jiras we have something like 'root/meta',
> > which
> > > seems to me that the intention is to also solve something related to
> > > splittable meta, then I do not think the work can be done in the near
> > > future...
> > >
> > >
> > A few of us want it in 2.4. I've just made it a blocker (Hope that ok
> > Andrew). Should be done in the next week or so. Will shout if we run into
> > obstacles (and we can punt on it then). We'd like to deploy the feature
> to
> > address hbase:meta issues we are seeing in production.
> >
> > What we're building should work for root if needed but will worry about
> > that when root shows up -- if it shows up.
> >
> > Thanks,
> > S
> >
> >
> >
> > > Andrew Purtell <ap...@apache.org> 于2020年10月15日周四 上午1:53写道：
> > >
> > > > To meet Nick's objective of a 2.4 release in November, we could wait
> > two
> > > or
> > > > three weeks for the meta replica work to completely land. Will that
> be
> > > > sufficient?
> > > >
> > > > On Wed, Oct 14, 2020 at 10:03 AM Huaxiang Sun <huaxiangsun@gmail.com
> >
> > > > wrote:
> > > >
> > > > > For meta replica work, the task jira is HBASE-18070
> > > > > <https://issues.apache.org/jira/browse/HBASE-18070>. The source
> side
> > > > > change
> > > > > has been committed, Endpoint change is under review.
> > > > > Meta Replica load balance support at client side is in progress.
> > > > >
> > > > > Thanks,
> > > > > Huaxiang
> > > > >
> > > > > On Wed, Oct 14, 2020 at 9:42 AM Nick Dimiduk <nd...@apache.org>
> > > > wrote:
> > > > >
> > > > > > Thank you Andrew. I was of a similar mind -- if I'm doing 2.3,
> 2.4
> > > > would
> > > > > be
> > > > > > only of minimal incremental overhead. It would probably encourage
> > the
> > > > > early
> > > > > > retirement of 2.3 :)
> > > > > >
> > > > > > Does anyone else have outstanding work they're hoping to see in
> > 2.4?
> > > > I'd
> > > > > > love to see an initial RC posted in the first half of November.
> > > > > >
> > > > > > Thanks,
> > > > > > Nick
> > > > > >
> > > > > > On Fri, Oct 9, 2020 at 2:54 PM Andrew Purtell <
> apurtell@apache.org
> > >
> > > > > wrote:
> > > > > >
> > > > > > > If nobody else wants to take this, I will.
> > > > > > >
> > > > > > > Reid is co-RMing the pending 1.7 release with me. We've been
> > going
> > > > slow
> > > > > > > with it waiting for backports but can move forward with it now
> > that
> > > > the
> > > > > > > desired work has landed. I am not sensing any particular
> urgency
> > > with
> > > > > 1.x
> > > > > > > releasing. Anyway, doing 2.4 as well would be only incremental
> > > > effort.
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > On Tue, Oct 6, 2020 at 3:38 PM Nick Dimiduk <
> ndimiduk@apache.org
> > >
> > > > > wrote:
> > > > > > >
> > > > > > > > Bump.
> > > > > > > >
> > > > > > > > We still need a volunteer for a release manager for 2.4. If
> we
> > > > delay
> > > > > > much
> > > > > > > > longer, we'll need another discussion, which is whether we
> > still
> > > > want
> > > > > > to
> > > > > > > > target the 6-month cadence on minor releases.
> > > > > > > >
> > > > > > > > In terms of outstanding work I'm hoping to see in 2.4, I'd
> like
> > > to
> > > > > get
> > > > > > a
> > > > > > > > number of these normalizer improvements in [0], or determine
> > that
> > > > > > they're
> > > > > > > > not a problem. I think there's work on meta region replicas
> > > that's
> > > > > also
> > > > > > > > targeting this release, but I don't have a JIRA handy. Are
> > there
> > > > > > > > other features folks are actively working on with 2.4 in
> mind?
> > > > Please
> > > > > > > keep
> > > > > > > > in mind that there's really only about 4 weeks worth of dev
> > time
> > > > left
> > > > > > > > before we'll need a feature freeze for stabilizing and the
> > first
> > > > RC.
> > > > > > > >
> > > > > > > > Thanks,
> > > > > > > > Nick
> > > > > > > >
> > > > > > > > [0]:
> > > > > > > >
> > > > > > > >
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20HBASE%20AND%20resolution%20%3D%20Unresolved%20AND%20component%20%3D%20Normalizer%20ORDER%20BY%20issuetype%20DESC
> > > > > > > >
> > > > > > > > On Tue, Aug 25, 2020 at 10:36 AM Nick Dimiduk <
> > > ndimiduk@apache.org
> > > > >
> > > > > > > wrote:
> > > > > > > >
> > > > > > > > > Heya!
> > > > > > > > >
> > > > > > > > > Now that 2.3.x is in maintenance mode, I'd like to start a
> > > > > discussion
> > > > > > > on
> > > > > > > > > 2.4. We've long agreed that a more frequent cadence of
> minor
> > > > > releases
> > > > > > > is
> > > > > > > > > superior to a long gap between them. I think we should
> > target a
> > > > > 2.4.0
> > > > > > > > > release sometime before the end of the calendar year,
> perhaps
> > > the
> > > > > > first
> > > > > > > > > weeks of December.
> > > > > > > > >
> > > > > > > > > Hopefully the workload for the next release manager isn't
> as
> > > high
> > > > > as
> > > > > > it
> > > > > > > > > was for 2.3, given that much of the test stabilization
> effort
> > > > > > happened
> > > > > > > on
> > > > > > > > > branch-2. However, we are running in a new jenkins
> > environment,
> > > > and
> > > > > > > > entropy
> > > > > > > > > is nothing if not persistent, which puts that hard-won
> > > stability
> > > > > back
> > > > > > > > into
> > > > > > > > > doubt. Thus, I think it would be good if we can start
> sooner
> > > than
> > > > > > > later.
> > > > > > > > >
> > > > > > > > > Would anyone like to volunteer to run the 2.4 release line?
> > The
> > > > > level
> > > > > > > of
> > > > > > > > > effort is admittedly non-negligible: (1) stabilize branch-2
> > (2)
> > > > > test
> > > > > > > > > new-for-branch-2 features (3) run initial release (4) cut
> > > monthly
> > > > > > patch
> > > > > > > > > releases until 2.5 is out. I have a strong interest in
> seeing
> > > > this
> > > > > > > > release
> > > > > > > > > through, so I volunteer to assist with these activities.
> > > > > > > > >
> > > > > > > > > Thanks,
> > > > > > > > > Nick
> > > > > > > > >
> > > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > --
> > > > > > > Best regards,
> > > > > > > Andrew
> > > > > > >
> > > > > > > Words like orphans lost among the crosstalk, meaning torn from
> > > > truth's
> > > > > > > decrepit hands
> > > > > > >    - A23, Crosstalk
> > > > > > >
> > > > > >
> > > > >
> > > >
> > > >
> > > > --
> > > > Best regards,
> > > > Andrew
> > > >
> > > > Words like orphans lost among the crosstalk, meaning torn from
> truth's
> > > > decrepit hands
> > > >    - A23, Crosstalk
> > > >
> > >
> >
>
>
> --
> Best regards,
> Andrew
>
> Words like orphans lost among the crosstalk, meaning torn from truth's
> decrepit hands
>    - A23, Crosstalk
>

Re: [DISCUSS] A plan for 2.4

Posted by Andrew Purtell <ap...@apache.org>.

Sure, I'll try it out. My usual process is manual inspection and so a tool
to automate some of it sounds great.

On Wed, Oct 28, 2020 at 1:06 PM Nick Dimiduk <nd...@apache.org> wrote:

> Yikes! The time is nearly upon. Let me land my normalizer PRs as well:
> HBASE-25167, HBASE-24419.
>
> Andrew -- do you have any intention of trying out the git/jira auditor
> tool? You might want HBASE-24845 as well.
>
> Thanks,
> Nick
>
> On Wed, Oct 28, 2020 at 10:26 AM Andrew Purtell <ap...@apache.org>
> wrote:
>
> > That's fine, but after one more week let's roll the RC.
> >
> >
> > On Tue, Oct 27, 2020 at 3:52 PM Huaxiang Sun <hu...@gmail.com>
> > wrote:
> >
> > > Hi Andrew,
> > >
> > >     For meta replica work (HBASE-18070), we made some good progress.
> The
> > > current estimate is that it needs one more week to commit patches.
> > >     Hope the schedule works for your 2.4 release schedule. If not, it
> > will
> > > be pushed to 2.5 release.
> > >
> > >     Thanks,
> > >     Huaxiang
> > >
> > > On Fri, Oct 16, 2020 at 11:32 AM Andrew Purtell <ap...@apache.org>
> > > wrote:
> > >
> > > > Thank you, Critical seems more appropriate. As RM (assuming that's me
> > now
> > > > :-) ) I think it's fine to wait the estimated week for commit, then
> we
> > > can
> > > > proceed.
> > > >
> > > > On Thu, Oct 15, 2020 at 2:51 PM Stack <st...@duboce.net> wrote:
> > > >
> > > > > On Thu, Oct 15, 2020 at 12:00 PM Andrew Purtell <
> apurtell@apache.org
> > >
> > > > > wrote:
> > > > >
> > > > > > Is it a blocker? Maybe mark it Critical? Either way, a week is
> well
> > > > > within
> > > > > > the timeframe discussed earlier. Will come back here to check
> > > progress
> > > > at
> > > > > > the end of next week. In the meantime it's fine to wait in my
> > > opinion.
> > > > > >
> > > > > >
> > > > > Sorry. Critical yes, blocker no. My bad. Let me change priority.
> > > > > S
> > > > >
> > > > >
> > > > > >
> > > > > > On Wed, Oct 14, 2020 at 8:55 PM Stack <st...@duboce.net> wrote:
> > > > > >
> > > > > > > On Wed, Oct 14, 2020 at 6:58 PM 张铎(Duo Zhang) <
> > > palomino219@gmail.com
> > > > >
> > > > > > > wrote:
> > > > > > >
> > > > > > > > I do not think HBASE-18070 is a blocker for 2.4 as it is
> still
> > > on a
> > > > > > > feature
> > > > > > > > branch.
> > > > > > > > And in the title of several jiras we have something like
> > > > 'root/meta',
> > > > > > > which
> > > > > > > > seems to me that the intention is to also solve something
> > related
> > > > to
> > > > > > > > splittable meta, then I do not think the work can be done in
> > the
> > > > near
> > > > > > > > future...
> > > > > > > >
> > > > > > > >
> > > > > > > A few of us want it in 2.4. I've just made it a blocker (Hope
> > that
> > > ok
> > > > > > > Andrew). Should be done in the next week or so. Will shout if
> we
> > > run
> > > > > into
> > > > > > > obstacles (and we can punt on it then). We'd like to deploy the
> > > > feature
> > > > > > to
> > > > > > > address hbase:meta issues we are seeing in production.
> > > > > > >
> > > > > > > What we're building should work for root if needed but will
> worry
> > > > about
> > > > > > > that when root shows up -- if it shows up.
> > > > > > >
> > > > > > > Thanks,
> > > > > > > S
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > > Andrew Purtell <ap...@apache.org> 于2020年10月15日周四
> 上午1:53写道：
> > > > > > > >
> > > > > > > > > To meet Nick's objective of a 2.4 release in November, we
> > could
> > > > > wait
> > > > > > > two
> > > > > > > > or
> > > > > > > > > three weeks for the meta replica work to completely land.
> > Will
> > > > that
> > > > > > be
> > > > > > > > > sufficient?
> > > > > > > > >
> > > > > > > > > On Wed, Oct 14, 2020 at 10:03 AM Huaxiang Sun <
> > > > > huaxiangsun@gmail.com
> > > > > > >
> > > > > > > > > wrote:
> > > > > > > > >
> > > > > > > > > > For meta replica work, the task jira is HBASE-18070
> > > > > > > > > > <https://issues.apache.org/jira/browse/HBASE-18070>. The
> > > > source
> > > > > > side
> > > > > > > > > > change
> > > > > > > > > > has been committed, Endpoint change is under review.
> > > > > > > > > > Meta Replica load balance support at client side is in
> > > > progress.
> > > > > > > > > >
> > > > > > > > > > Thanks,
> > > > > > > > > > Huaxiang
> > > > > > > > > >
> > > > > > > > > > On Wed, Oct 14, 2020 at 9:42 AM Nick Dimiduk <
> > > > > ndimiduk@apache.org>
> > > > > > > > > wrote:
> > > > > > > > > >
> > > > > > > > > > > Thank you Andrew. I was of a similar mind -- if I'm
> doing
> > > > 2.3,
> > > > > > 2.4
> > > > > > > > > would
> > > > > > > > > > be
> > > > > > > > > > > only of minimal incremental overhead. It would probably
> > > > > encourage
> > > > > > > the
> > > > > > > > > > early
> > > > > > > > > > > retirement of 2.3 :)
> > > > > > > > > > >
> > > > > > > > > > > Does anyone else have outstanding work they're hoping
> to
> > > see
> > > > in
> > > > > > > 2.4?
> > > > > > > > > I'd
> > > > > > > > > > > love to see an initial RC posted in the first half of
> > > > November.
> > > > > > > > > > >
> > > > > > > > > > > Thanks,
> > > > > > > > > > > Nick
> > > > > > > > > > >
> > > > > > > > > > > On Fri, Oct 9, 2020 at 2:54 PM Andrew Purtell <
> > > > > > apurtell@apache.org
> > > > > > > >
> > > > > > > > > > wrote:
> > > > > > > > > > >
> > > > > > > > > > > > If nobody else wants to take this, I will.
> > > > > > > > > > > >
> > > > > > > > > > > > Reid is co-RMing the pending 1.7 release with me.
> We've
> > > > been
> > > > > > > going
> > > > > > > > > slow
> > > > > > > > > > > > with it waiting for backports but can move forward
> with
> > > it
> > > > > now
> > > > > > > that
> > > > > > > > > the
> > > > > > > > > > > > desired work has landed. I am not sensing any
> > particular
> > > > > > urgency
> > > > > > > > with
> > > > > > > > > > 1.x
> > > > > > > > > > > > releasing. Anyway, doing 2.4 as well would be only
> > > > > incremental
> > > > > > > > > effort.
> > > > > > > > > > > >
> > > > > > > > > > > >
> > > > > > > > > > > >
> > > > > > > > > > > > On Tue, Oct 6, 2020 at 3:38 PM Nick Dimiduk <
> > > > > > ndimiduk@apache.org
> > > > > > > >
> > > > > > > > > > wrote:
> > > > > > > > > > > >
> > > > > > > > > > > > > Bump.
> > > > > > > > > > > > >
> > > > > > > > > > > > > We still need a volunteer for a release manager for
> > > 2.4.
> > > > If
> > > > > > we
> > > > > > > > > delay
> > > > > > > > > > > much
> > > > > > > > > > > > > longer, we'll need another discussion, which is
> > whether
> > > > we
> > > > > > > still
> > > > > > > > > want
> > > > > > > > > > > to
> > > > > > > > > > > > > target the 6-month cadence on minor releases.
> > > > > > > > > > > > >
> > > > > > > > > > > > > In terms of outstanding work I'm hoping to see in
> > 2.4,
> > > > I'd
> > > > > > like
> > > > > > > > to
> > > > > > > > > > get
> > > > > > > > > > > a
> > > > > > > > > > > > > number of these normalizer improvements in [0], or
> > > > > determine
> > > > > > > that
> > > > > > > > > > > they're
> > > > > > > > > > > > > not a problem. I think there's work on meta region
> > > > replicas
> > > > > > > > that's
> > > > > > > > > > also
> > > > > > > > > > > > > targeting this release, but I don't have a JIRA
> > handy.
> > > > Are
> > > > > > > there
> > > > > > > > > > > > > other features folks are actively working on with
> 2.4
> > > in
> > > > > > mind?
> > > > > > > > > Please
> > > > > > > > > > > > keep
> > > > > > > > > > > > > in mind that there's really only about 4 weeks
> worth
> > of
> > > > dev
> > > > > > > time
> > > > > > > > > left
> > > > > > > > > > > > > before we'll need a feature freeze for stabilizing
> > and
> > > > the
> > > > > > > first
> > > > > > > > > RC.
> > > > > > > > > > > > >
> > > > > > > > > > > > > Thanks,
> > > > > > > > > > > > > Nick
> > > > > > > > > > > > >
> > > > > > > > > > > > > [0]:
> > > > > > > > > > > > >
> > > > > > > > > > > > >
> > > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > >
> > > > > > > > >
> > > > > > > >
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20HBASE%20AND%20resolution%20%3D%20Unresolved%20AND%20component%20%3D%20Normalizer%20ORDER%20BY%20issuetype%20DESC
> > > > > > > > > > > > >
> > > > > > > > > > > > > On Tue, Aug 25, 2020 at 10:36 AM Nick Dimiduk <
> > > > > > > > ndimiduk@apache.org
> > > > > > > > > >
> > > > > > > > > > > > wrote:
> > > > > > > > > > > > >
> > > > > > > > > > > > > > Heya!
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Now that 2.3.x is in maintenance mode, I'd like
> to
> > > > start
> > > > > a
> > > > > > > > > > discussion
> > > > > > > > > > > > on
> > > > > > > > > > > > > > 2.4. We've long agreed that a more frequent
> cadence
> > > of
> > > > > > minor
> > > > > > > > > > releases
> > > > > > > > > > > > is
> > > > > > > > > > > > > > superior to a long gap between them. I think we
> > > should
> > > > > > > target a
> > > > > > > > > > 2.4.0
> > > > > > > > > > > > > > release sometime before the end of the calendar
> > year,
> > > > > > perhaps
> > > > > > > > the
> > > > > > > > > > > first
> > > > > > > > > > > > > > weeks of December.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Hopefully the workload for the next release
> manager
> > > > isn't
> > > > > > as
> > > > > > > > high
> > > > > > > > > > as
> > > > > > > > > > > it
> > > > > > > > > > > > > > was for 2.3, given that much of the test
> > > stabilization
> > > > > > effort
> > > > > > > > > > > happened
> > > > > > > > > > > > on
> > > > > > > > > > > > > > branch-2. However, we are running in a new
> jenkins
> > > > > > > environment,
> > > > > > > > > and
> > > > > > > > > > > > > entropy
> > > > > > > > > > > > > > is nothing if not persistent, which puts that
> > > hard-won
> > > > > > > > stability
> > > > > > > > > > back
> > > > > > > > > > > > > into
> > > > > > > > > > > > > > doubt. Thus, I think it would be good if we can
> > start
> > > > > > sooner
> > > > > > > > than
> > > > > > > > > > > > later.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Would anyone like to volunteer to run the 2.4
> > release
> > > > > line?
> > > > > > > The
> > > > > > > > > > level
> > > > > > > > > > > > of
> > > > > > > > > > > > > > effort is admittedly non-negligible: (1)
> stabilize
> > > > > branch-2
> > > > > > > (2)
> > > > > > > > > > test
> > > > > > > > > > > > > > new-for-branch-2 features (3) run initial release
> > (4)
> > > > cut
> > > > > > > > monthly
> > > > > > > > > > > patch
> > > > > > > > > > > > > > releases until 2.5 is out. I have a strong
> interest
> > > in
> > > > > > seeing
> > > > > > > > > this
> > > > > > > > > > > > > release
> > > > > > > > > > > > > > through, so I volunteer to assist with these
> > > > activities.
> > > > > > > > > > > > > >
> > > > > > > > > > > > > > Thanks,
> > > > > > > > > > > > > > Nick
> > > > > > > > > > > > > >
> > > > > > > > > > > > >
> > > > > > > > > > > >
> > > > > > > > > > > >
> > > > > > > > > > > > --
> > > > > > > > > > > > Best regards,
> > > > > > > > > > > > Andrew
> > > > > > > > > > > >
> > > > > > > > > > > > Words like orphans lost among the crosstalk, meaning
> > torn
> > > > > from
> > > > > > > > > truth's
> > > > > > > > > > > > decrepit hands
> > > > > > > > > > > >    - A23, Crosstalk
> > > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > >
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > --
> > > > > > > > > Best regards,
> > > > > > > > > Andrew
> > > > > > > > >
> > > > > > > > > Words like orphans lost among the crosstalk, meaning torn
> > from
> > > > > > truth's
> > > > > > > > > decrepit hands
> > > > > > > > >    - A23, Crosstalk
> > > > > > > > >
> > > > > > > >
> > > > > > >
> > > > > >
> > > > > >
> > > > > > --
> > > > > > Best regards,
> > > > > > Andrew
> > > > > >
> > > > > > Words like orphans lost among the crosstalk, meaning torn from
> > > truth's
> > > > > > decrepit hands
> > > > > >    - A23, Crosstalk
> > > > > >
> > > > >
> > > >
> > > >
> > > > --
> > > > Best regards,
> > > > Andrew
> > > >
> > > > Words like orphans lost among the crosstalk, meaning torn from
> truth's
> > > > decrepit hands
> > > >    - A23, Crosstalk
> > > >
> > >
> >
> >
> > --
> > Best regards,
> > Andrew
> >
> > Words like orphans lost among the crosstalk, meaning torn from truth's
> > decrepit hands
> >    - A23, Crosstalk
> >
>


-- 
Best regards,
Andrew

Words like orphans lost among the crosstalk, meaning torn from truth's
decrepit hands
   - A23, Crosstalk

Re: [DISCUSS] A plan for 2.4

Posted by Nick Dimiduk <nd...@apache.org>.

Yikes! The time is nearly upon. Let me land my normalizer PRs as well:
HBASE-25167, HBASE-24419.

Andrew -- do you have any intention of trying out the git/jira auditor
tool? You might want HBASE-24845 as well.

Thanks,
Nick

On Wed, Oct 28, 2020 at 10:26 AM Andrew Purtell <ap...@apache.org> wrote:

> That's fine, but after one more week let's roll the RC.
>
>
> On Tue, Oct 27, 2020 at 3:52 PM Huaxiang Sun <hu...@gmail.com>
> wrote:
>
> > Hi Andrew,
> >
> >     For meta replica work (HBASE-18070), we made some good progress. The
> > current estimate is that it needs one more week to commit patches.
> >     Hope the schedule works for your 2.4 release schedule. If not, it
> will
> > be pushed to 2.5 release.
> >
> >     Thanks,
> >     Huaxiang
> >
> > On Fri, Oct 16, 2020 at 11:32 AM Andrew Purtell <ap...@apache.org>
> > wrote:
> >
> > > Thank you, Critical seems more appropriate. As RM (assuming that's me
> now
> > > :-) ) I think it's fine to wait the estimated week for commit, then we
> > can
> > > proceed.
> > >
> > > On Thu, Oct 15, 2020 at 2:51 PM Stack <st...@duboce.net> wrote:
> > >
> > > > On Thu, Oct 15, 2020 at 12:00 PM Andrew Purtell <apurtell@apache.org
> >
> > > > wrote:
> > > >
> > > > > Is it a blocker? Maybe mark it Critical? Either way, a week is well
> > > > within
> > > > > the timeframe discussed earlier. Will come back here to check
> > progress
> > > at
> > > > > the end of next week. In the meantime it's fine to wait in my
> > opinion.
> > > > >
> > > > >
> > > > Sorry. Critical yes, blocker no. My bad. Let me change priority.
> > > > S
> > > >
> > > >
> > > > >
> > > > > On Wed, Oct 14, 2020 at 8:55 PM Stack <st...@duboce.net> wrote:
> > > > >
> > > > > > On Wed, Oct 14, 2020 at 6:58 PM 张铎(Duo Zhang) <
> > palomino219@gmail.com
> > > >
> > > > > > wrote:
> > > > > >
> > > > > > > I do not think HBASE-18070 is a blocker for 2.4 as it is still
> > on a
> > > > > > feature
> > > > > > > branch.
> > > > > > > And in the title of several jiras we have something like
> > > 'root/meta',
> > > > > > which
> > > > > > > seems to me that the intention is to also solve something
> related
> > > to
> > > > > > > splittable meta, then I do not think the work can be done in
> the
> > > near
> > > > > > > future...
> > > > > > >
> > > > > > >
> > > > > > A few of us want it in 2.4. I've just made it a blocker (Hope
> that
> > ok
> > > > > > Andrew). Should be done in the next week or so. Will shout if we
> > run
> > > > into
> > > > > > obstacles (and we can punt on it then). We'd like to deploy the
> > > feature
> > > > > to
> > > > > > address hbase:meta issues we are seeing in production.
> > > > > >
> > > > > > What we're building should work for root if needed but will worry
> > > about
> > > > > > that when root shows up -- if it shows up.
> > > > > >
> > > > > > Thanks,
> > > > > > S
> > > > > >
> > > > > >
> > > > > >
> > > > > > > Andrew Purtell <ap...@apache.org> 于2020年10月15日周四 上午1:53写道：
> > > > > > >
> > > > > > > > To meet Nick's objective of a 2.4 release in November, we
> could
> > > > wait
> > > > > > two
> > > > > > > or
> > > > > > > > three weeks for the meta replica work to completely land.
> Will
> > > that
> > > > > be
> > > > > > > > sufficient?
> > > > > > > >
> > > > > > > > On Wed, Oct 14, 2020 at 10:03 AM Huaxiang Sun <
> > > > huaxiangsun@gmail.com
> > > > > >
> > > > > > > > wrote:
> > > > > > > >
> > > > > > > > > For meta replica work, the task jira is HBASE-18070
> > > > > > > > > <https://issues.apache.org/jira/browse/HBASE-18070>. The
> > > source
> > > > > side
> > > > > > > > > change
> > > > > > > > > has been committed, Endpoint change is under review.
> > > > > > > > > Meta Replica load balance support at client side is in
> > > progress.
> > > > > > > > >
> > > > > > > > > Thanks,
> > > > > > > > > Huaxiang
> > > > > > > > >
> > > > > > > > > On Wed, Oct 14, 2020 at 9:42 AM Nick Dimiduk <
> > > > ndimiduk@apache.org>
> > > > > > > > wrote:
> > > > > > > > >
> > > > > > > > > > Thank you Andrew. I was of a similar mind -- if I'm doing
> > > 2.3,
> > > > > 2.4
> > > > > > > > would
> > > > > > > > > be
> > > > > > > > > > only of minimal incremental overhead. It would probably
> > > > encourage
> > > > > > the
> > > > > > > > > early
> > > > > > > > > > retirement of 2.3 :)
> > > > > > > > > >
> > > > > > > > > > Does anyone else have outstanding work they're hoping to
> > see
> > > in
> > > > > > 2.4?
> > > > > > > > I'd
> > > > > > > > > > love to see an initial RC posted in the first half of
> > > November.
> > > > > > > > > >
> > > > > > > > > > Thanks,
> > > > > > > > > > Nick
> > > > > > > > > >
> > > > > > > > > > On Fri, Oct 9, 2020 at 2:54 PM Andrew Purtell <
> > > > > apurtell@apache.org
> > > > > > >
> > > > > > > > > wrote:
> > > > > > > > > >
> > > > > > > > > > > If nobody else wants to take this, I will.
> > > > > > > > > > >
> > > > > > > > > > > Reid is co-RMing the pending 1.7 release with me. We've
> > > been
> > > > > > going
> > > > > > > > slow
> > > > > > > > > > > with it waiting for backports but can move forward with
> > it
> > > > now
> > > > > > that
> > > > > > > > the
> > > > > > > > > > > desired work has landed. I am not sensing any
> particular
> > > > > urgency
> > > > > > > with
> > > > > > > > > 1.x
> > > > > > > > > > > releasing. Anyway, doing 2.4 as well would be only
> > > > incremental
> > > > > > > > effort.
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > > On Tue, Oct 6, 2020 at 3:38 PM Nick Dimiduk <
> > > > > ndimiduk@apache.org
> > > > > > >
> > > > > > > > > wrote:
> > > > > > > > > > >
> > > > > > > > > > > > Bump.
> > > > > > > > > > > >
> > > > > > > > > > > > We still need a volunteer for a release manager for
> > 2.4.
> > > If
> > > > > we
> > > > > > > > delay
> > > > > > > > > > much
> > > > > > > > > > > > longer, we'll need another discussion, which is
> whether
> > > we
> > > > > > still
> > > > > > > > want
> > > > > > > > > > to
> > > > > > > > > > > > target the 6-month cadence on minor releases.
> > > > > > > > > > > >
> > > > > > > > > > > > In terms of outstanding work I'm hoping to see in
> 2.4,
> > > I'd
> > > > > like
> > > > > > > to
> > > > > > > > > get
> > > > > > > > > > a
> > > > > > > > > > > > number of these normalizer improvements in [0], or
> > > > determine
> > > > > > that
> > > > > > > > > > they're
> > > > > > > > > > > > not a problem. I think there's work on meta region
> > > replicas
> > > > > > > that's
> > > > > > > > > also
> > > > > > > > > > > > targeting this release, but I don't have a JIRA
> handy.
> > > Are
> > > > > > there
> > > > > > > > > > > > other features folks are actively working on with 2.4
> > in
> > > > > mind?
> > > > > > > > Please
> > > > > > > > > > > keep
> > > > > > > > > > > > in mind that there's really only about 4 weeks worth
> of
> > > dev
> > > > > > time
> > > > > > > > left
> > > > > > > > > > > > before we'll need a feature freeze for stabilizing
> and
> > > the
> > > > > > first
> > > > > > > > RC.
> > > > > > > > > > > >
> > > > > > > > > > > > Thanks,
> > > > > > > > > > > > Nick
> > > > > > > > > > > >
> > > > > > > > > > > > [0]:
> > > > > > > > > > > >
> > > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > >
> > > > > > > > >
> > > > > > > >
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20HBASE%20AND%20resolution%20%3D%20Unresolved%20AND%20component%20%3D%20Normalizer%20ORDER%20BY%20issuetype%20DESC
> > > > > > > > > > > >
> > > > > > > > > > > > On Tue, Aug 25, 2020 at 10:36 AM Nick Dimiduk <
> > > > > > > ndimiduk@apache.org
> > > > > > > > >
> > > > > > > > > > > wrote:
> > > > > > > > > > > >
> > > > > > > > > > > > > Heya!
> > > > > > > > > > > > >
> > > > > > > > > > > > > Now that 2.3.x is in maintenance mode, I'd like to
> > > start
> > > > a
> > > > > > > > > discussion
> > > > > > > > > > > on
> > > > > > > > > > > > > 2.4. We've long agreed that a more frequent cadence
> > of
> > > > > minor
> > > > > > > > > releases
> > > > > > > > > > > is
> > > > > > > > > > > > > superior to a long gap between them. I think we
> > should
> > > > > > target a
> > > > > > > > > 2.4.0
> > > > > > > > > > > > > release sometime before the end of the calendar
> year,
> > > > > perhaps
> > > > > > > the
> > > > > > > > > > first
> > > > > > > > > > > > > weeks of December.
> > > > > > > > > > > > >
> > > > > > > > > > > > > Hopefully the workload for the next release manager
> > > isn't
> > > > > as
> > > > > > > high
> > > > > > > > > as
> > > > > > > > > > it
> > > > > > > > > > > > > was for 2.3, given that much of the test
> > stabilization
> > > > > effort
> > > > > > > > > > happened
> > > > > > > > > > > on
> > > > > > > > > > > > > branch-2. However, we are running in a new jenkins
> > > > > > environment,
> > > > > > > > and
> > > > > > > > > > > > entropy
> > > > > > > > > > > > > is nothing if not persistent, which puts that
> > hard-won
> > > > > > > stability
> > > > > > > > > back
> > > > > > > > > > > > into
> > > > > > > > > > > > > doubt. Thus, I think it would be good if we can
> start
> > > > > sooner
> > > > > > > than
> > > > > > > > > > > later.
> > > > > > > > > > > > >
> > > > > > > > > > > > > Would anyone like to volunteer to run the 2.4
> release
> > > > line?
> > > > > > The
> > > > > > > > > level
> > > > > > > > > > > of
> > > > > > > > > > > > > effort is admittedly non-negligible: (1) stabilize
> > > > branch-2
> > > > > > (2)
> > > > > > > > > test
> > > > > > > > > > > > > new-for-branch-2 features (3) run initial release
> (4)
> > > cut
> > > > > > > monthly
> > > > > > > > > > patch
> > > > > > > > > > > > > releases until 2.5 is out. I have a strong interest
> > in
> > > > > seeing
> > > > > > > > this
> > > > > > > > > > > > release
> > > > > > > > > > > > > through, so I volunteer to assist with these
> > > activities.
> > > > > > > > > > > > >
> > > > > > > > > > > > > Thanks,
> > > > > > > > > > > > > Nick
> > > > > > > > > > > > >
> > > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > > --
> > > > > > > > > > > Best regards,
> > > > > > > > > > > Andrew
> > > > > > > > > > >
> > > > > > > > > > > Words like orphans lost among the crosstalk, meaning
> torn
> > > > from
> > > > > > > > truth's
> > > > > > > > > > > decrepit hands
> > > > > > > > > > >    - A23, Crosstalk
> > > > > > > > > > >
> > > > > > > > > >
> > > > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > > > --
> > > > > > > > Best regards,
> > > > > > > > Andrew
> > > > > > > >
> > > > > > > > Words like orphans lost among the crosstalk, meaning torn
> from
> > > > > truth's
> > > > > > > > decrepit hands
> > > > > > > >    - A23, Crosstalk
> > > > > > > >
> > > > > > >
> > > > > >
> > > > >
> > > > >
> > > > > --
> > > > > Best regards,
> > > > > Andrew
> > > > >
> > > > > Words like orphans lost among the crosstalk, meaning torn from
> > truth's
> > > > > decrepit hands
> > > > >    - A23, Crosstalk
> > > > >
> > > >
> > >
> > >
> > > --
> > > Best regards,
> > > Andrew
> > >
> > > Words like orphans lost among the crosstalk, meaning torn from truth's
> > > decrepit hands
> > >    - A23, Crosstalk
> > >
> >
>
>
> --
> Best regards,
> Andrew
>
> Words like orphans lost among the crosstalk, meaning torn from truth's
> decrepit hands
>    - A23, Crosstalk
>

Re: [DISCUSS] A plan for 2.4

Posted by Andrew Purtell <ap...@apache.org>.

That's fine, but after one more week let's roll the RC.


On Tue, Oct 27, 2020 at 3:52 PM Huaxiang Sun <hu...@gmail.com> wrote:

> Hi Andrew,
>
>     For meta replica work (HBASE-18070), we made some good progress. The
> current estimate is that it needs one more week to commit patches.
>     Hope the schedule works for your 2.4 release schedule. If not, it will
> be pushed to 2.5 release.
>
>     Thanks,
>     Huaxiang
>
> On Fri, Oct 16, 2020 at 11:32 AM Andrew Purtell <ap...@apache.org>
> wrote:
>
> > Thank you, Critical seems more appropriate. As RM (assuming that's me now
> > :-) ) I think it's fine to wait the estimated week for commit, then we
> can
> > proceed.
> >
> > On Thu, Oct 15, 2020 at 2:51 PM Stack <st...@duboce.net> wrote:
> >
> > > On Thu, Oct 15, 2020 at 12:00 PM Andrew Purtell <ap...@apache.org>
> > > wrote:
> > >
> > > > Is it a blocker? Maybe mark it Critical? Either way, a week is well
> > > within
> > > > the timeframe discussed earlier. Will come back here to check
> progress
> > at
> > > > the end of next week. In the meantime it's fine to wait in my
> opinion.
> > > >
> > > >
> > > Sorry. Critical yes, blocker no. My bad. Let me change priority.
> > > S
> > >
> > >
> > > >
> > > > On Wed, Oct 14, 2020 at 8:55 PM Stack <st...@duboce.net> wrote:
> > > >
> > > > > On Wed, Oct 14, 2020 at 6:58 PM 张铎(Duo Zhang) <
> palomino219@gmail.com
> > >
> > > > > wrote:
> > > > >
> > > > > > I do not think HBASE-18070 is a blocker for 2.4 as it is still
> on a
> > > > > feature
> > > > > > branch.
> > > > > > And in the title of several jiras we have something like
> > 'root/meta',
> > > > > which
> > > > > > seems to me that the intention is to also solve something related
> > to
> > > > > > splittable meta, then I do not think the work can be done in the
> > near
> > > > > > future...
> > > > > >
> > > > > >
> > > > > A few of us want it in 2.4. I've just made it a blocker (Hope that
> ok
> > > > > Andrew). Should be done in the next week or so. Will shout if we
> run
> > > into
> > > > > obstacles (and we can punt on it then). We'd like to deploy the
> > feature
> > > > to
> > > > > address hbase:meta issues we are seeing in production.
> > > > >
> > > > > What we're building should work for root if needed but will worry
> > about
> > > > > that when root shows up -- if it shows up.
> > > > >
> > > > > Thanks,
> > > > > S
> > > > >
> > > > >
> > > > >
> > > > > > Andrew Purtell <ap...@apache.org> 于2020年10月15日周四 上午1:53写道：
> > > > > >
> > > > > > > To meet Nick's objective of a 2.4 release in November, we could
> > > wait
> > > > > two
> > > > > > or
> > > > > > > three weeks for the meta replica work to completely land. Will
> > that
> > > > be
> > > > > > > sufficient?
> > > > > > >
> > > > > > > On Wed, Oct 14, 2020 at 10:03 AM Huaxiang Sun <
> > > huaxiangsun@gmail.com
> > > > >
> > > > > > > wrote:
> > > > > > >
> > > > > > > > For meta replica work, the task jira is HBASE-18070
> > > > > > > > <https://issues.apache.org/jira/browse/HBASE-18070>. The
> > source
> > > > side
> > > > > > > > change
> > > > > > > > has been committed, Endpoint change is under review.
> > > > > > > > Meta Replica load balance support at client side is in
> > progress.
> > > > > > > >
> > > > > > > > Thanks,
> > > > > > > > Huaxiang
> > > > > > > >
> > > > > > > > On Wed, Oct 14, 2020 at 9:42 AM Nick Dimiduk <
> > > ndimiduk@apache.org>
> > > > > > > wrote:
> > > > > > > >
> > > > > > > > > Thank you Andrew. I was of a similar mind -- if I'm doing
> > 2.3,
> > > > 2.4
> > > > > > > would
> > > > > > > > be
> > > > > > > > > only of minimal incremental overhead. It would probably
> > > encourage
> > > > > the
> > > > > > > > early
> > > > > > > > > retirement of 2.3 :)
> > > > > > > > >
> > > > > > > > > Does anyone else have outstanding work they're hoping to
> see
> > in
> > > > > 2.4?
> > > > > > > I'd
> > > > > > > > > love to see an initial RC posted in the first half of
> > November.
> > > > > > > > >
> > > > > > > > > Thanks,
> > > > > > > > > Nick
> > > > > > > > >
> > > > > > > > > On Fri, Oct 9, 2020 at 2:54 PM Andrew Purtell <
> > > > apurtell@apache.org
> > > > > >
> > > > > > > > wrote:
> > > > > > > > >
> > > > > > > > > > If nobody else wants to take this, I will.
> > > > > > > > > >
> > > > > > > > > > Reid is co-RMing the pending 1.7 release with me. We've
> > been
> > > > > going
> > > > > > > slow
> > > > > > > > > > with it waiting for backports but can move forward with
> it
> > > now
> > > > > that
> > > > > > > the
> > > > > > > > > > desired work has landed. I am not sensing any particular
> > > > urgency
> > > > > > with
> > > > > > > > 1.x
> > > > > > > > > > releasing. Anyway, doing 2.4 as well would be only
> > > incremental
> > > > > > > effort.
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > > On Tue, Oct 6, 2020 at 3:38 PM Nick Dimiduk <
> > > > ndimiduk@apache.org
> > > > > >
> > > > > > > > wrote:
> > > > > > > > > >
> > > > > > > > > > > Bump.
> > > > > > > > > > >
> > > > > > > > > > > We still need a volunteer for a release manager for
> 2.4.
> > If
> > > > we
> > > > > > > delay
> > > > > > > > > much
> > > > > > > > > > > longer, we'll need another discussion, which is whether
> > we
> > > > > still
> > > > > > > want
> > > > > > > > > to
> > > > > > > > > > > target the 6-month cadence on minor releases.
> > > > > > > > > > >
> > > > > > > > > > > In terms of outstanding work I'm hoping to see in 2.4,
> > I'd
> > > > like
> > > > > > to
> > > > > > > > get
> > > > > > > > > a
> > > > > > > > > > > number of these normalizer improvements in [0], or
> > > determine
> > > > > that
> > > > > > > > > they're
> > > > > > > > > > > not a problem. I think there's work on meta region
> > replicas
> > > > > > that's
> > > > > > > > also
> > > > > > > > > > > targeting this release, but I don't have a JIRA handy.
> > Are
> > > > > there
> > > > > > > > > > > other features folks are actively working on with 2.4
> in
> > > > mind?
> > > > > > > Please
> > > > > > > > > > keep
> > > > > > > > > > > in mind that there's really only about 4 weeks worth of
> > dev
> > > > > time
> > > > > > > left
> > > > > > > > > > > before we'll need a feature freeze for stabilizing and
> > the
> > > > > first
> > > > > > > RC.
> > > > > > > > > > >
> > > > > > > > > > > Thanks,
> > > > > > > > > > > Nick
> > > > > > > > > > >
> > > > > > > > > > > [0]:
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > >
> > > > > > > > >
> > > > > > > >
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20HBASE%20AND%20resolution%20%3D%20Unresolved%20AND%20component%20%3D%20Normalizer%20ORDER%20BY%20issuetype%20DESC
> > > > > > > > > > >
> > > > > > > > > > > On Tue, Aug 25, 2020 at 10:36 AM Nick Dimiduk <
> > > > > > ndimiduk@apache.org
> > > > > > > >
> > > > > > > > > > wrote:
> > > > > > > > > > >
> > > > > > > > > > > > Heya!
> > > > > > > > > > > >
> > > > > > > > > > > > Now that 2.3.x is in maintenance mode, I'd like to
> > start
> > > a
> > > > > > > > discussion
> > > > > > > > > > on
> > > > > > > > > > > > 2.4. We've long agreed that a more frequent cadence
> of
> > > > minor
> > > > > > > > releases
> > > > > > > > > > is
> > > > > > > > > > > > superior to a long gap between them. I think we
> should
> > > > > target a
> > > > > > > > 2.4.0
> > > > > > > > > > > > release sometime before the end of the calendar year,
> > > > perhaps
> > > > > > the
> > > > > > > > > first
> > > > > > > > > > > > weeks of December.
> > > > > > > > > > > >
> > > > > > > > > > > > Hopefully the workload for the next release manager
> > isn't
> > > > as
> > > > > > high
> > > > > > > > as
> > > > > > > > > it
> > > > > > > > > > > > was for 2.3, given that much of the test
> stabilization
> > > > effort
> > > > > > > > > happened
> > > > > > > > > > on
> > > > > > > > > > > > branch-2. However, we are running in a new jenkins
> > > > > environment,
> > > > > > > and
> > > > > > > > > > > entropy
> > > > > > > > > > > > is nothing if not persistent, which puts that
> hard-won
> > > > > > stability
> > > > > > > > back
> > > > > > > > > > > into
> > > > > > > > > > > > doubt. Thus, I think it would be good if we can start
> > > > sooner
> > > > > > than
> > > > > > > > > > later.
> > > > > > > > > > > >
> > > > > > > > > > > > Would anyone like to volunteer to run the 2.4 release
> > > line?
> > > > > The
> > > > > > > > level
> > > > > > > > > > of
> > > > > > > > > > > > effort is admittedly non-negligible: (1) stabilize
> > > branch-2
> > > > > (2)
> > > > > > > > test
> > > > > > > > > > > > new-for-branch-2 features (3) run initial release (4)
> > cut
> > > > > > monthly
> > > > > > > > > patch
> > > > > > > > > > > > releases until 2.5 is out. I have a strong interest
> in
> > > > seeing
> > > > > > > this
> > > > > > > > > > > release
> > > > > > > > > > > > through, so I volunteer to assist with these
> > activities.
> > > > > > > > > > > >
> > > > > > > > > > > > Thanks,
> > > > > > > > > > > > Nick
> > > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > > --
> > > > > > > > > > Best regards,
> > > > > > > > > > Andrew
> > > > > > > > > >
> > > > > > > > > > Words like orphans lost among the crosstalk, meaning torn
> > > from
> > > > > > > truth's
> > > > > > > > > > decrepit hands
> > > > > > > > > >    - A23, Crosstalk
> > > > > > > > > >
> > > > > > > > >
> > > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > --
> > > > > > > Best regards,
> > > > > > > Andrew
> > > > > > >
> > > > > > > Words like orphans lost among the crosstalk, meaning torn from
> > > > truth's
> > > > > > > decrepit hands
> > > > > > >    - A23, Crosstalk
> > > > > > >
> > > > > >
> > > > >
> > > >
> > > >
> > > > --
> > > > Best regards,
> > > > Andrew
> > > >
> > > > Words like orphans lost among the crosstalk, meaning torn from
> truth's
> > > > decrepit hands
> > > >    - A23, Crosstalk
> > > >
> > >
> >
> >
> > --
> > Best regards,
> > Andrew
> >
> > Words like orphans lost among the crosstalk, meaning torn from truth's
> > decrepit hands
> >    - A23, Crosstalk
> >
>


-- 
Best regards,
Andrew

Words like orphans lost among the crosstalk, meaning torn from truth's
decrepit hands
   - A23, Crosstalk

Re: [DISCUSS] A plan for 2.4

Posted by Huaxiang Sun <hu...@gmail.com>.

Hi Andrew,

    For meta replica work (HBASE-18070), we made some good progress. The
current estimate is that it needs one more week to commit patches.
    Hope the schedule works for your 2.4 release schedule. If not, it will
be pushed to 2.5 release.

    Thanks,
    Huaxiang

On Fri, Oct 16, 2020 at 11:32 AM Andrew Purtell <ap...@apache.org> wrote:

> Thank you, Critical seems more appropriate. As RM (assuming that's me now
> :-) ) I think it's fine to wait the estimated week for commit, then we can
> proceed.
>
> On Thu, Oct 15, 2020 at 2:51 PM Stack <st...@duboce.net> wrote:
>
> > On Thu, Oct 15, 2020 at 12:00 PM Andrew Purtell <ap...@apache.org>
> > wrote:
> >
> > > Is it a blocker? Maybe mark it Critical? Either way, a week is well
> > within
> > > the timeframe discussed earlier. Will come back here to check progress
> at
> > > the end of next week. In the meantime it's fine to wait in my opinion.
> > >
> > >
> > Sorry. Critical yes, blocker no. My bad. Let me change priority.
> > S
> >
> >
> > >
> > > On Wed, Oct 14, 2020 at 8:55 PM Stack <st...@duboce.net> wrote:
> > >
> > > > On Wed, Oct 14, 2020 at 6:58 PM 张铎(Duo Zhang) <palomino219@gmail.com
> >
> > > > wrote:
> > > >
> > > > > I do not think HBASE-18070 is a blocker for 2.4 as it is still on a
> > > > feature
> > > > > branch.
> > > > > And in the title of several jiras we have something like
> 'root/meta',
> > > > which
> > > > > seems to me that the intention is to also solve something related
> to
> > > > > splittable meta, then I do not think the work can be done in the
> near
> > > > > future...
> > > > >
> > > > >
> > > > A few of us want it in 2.4. I've just made it a blocker (Hope that ok
> > > > Andrew). Should be done in the next week or so. Will shout if we run
> > into
> > > > obstacles (and we can punt on it then). We'd like to deploy the
> feature
> > > to
> > > > address hbase:meta issues we are seeing in production.
> > > >
> > > > What we're building should work for root if needed but will worry
> about
> > > > that when root shows up -- if it shows up.
> > > >
> > > > Thanks,
> > > > S
> > > >
> > > >
> > > >
> > > > > Andrew Purtell <ap...@apache.org> 于2020年10月15日周四 上午1:53写道：
> > > > >
> > > > > > To meet Nick's objective of a 2.4 release in November, we could
> > wait
> > > > two
> > > > > or
> > > > > > three weeks for the meta replica work to completely land. Will
> that
> > > be
> > > > > > sufficient?
> > > > > >
> > > > > > On Wed, Oct 14, 2020 at 10:03 AM Huaxiang Sun <
> > huaxiangsun@gmail.com
> > > >
> > > > > > wrote:
> > > > > >
> > > > > > > For meta replica work, the task jira is HBASE-18070
> > > > > > > <https://issues.apache.org/jira/browse/HBASE-18070>. The
> source
> > > side
> > > > > > > change
> > > > > > > has been committed, Endpoint change is under review.
> > > > > > > Meta Replica load balance support at client side is in
> progress.
> > > > > > >
> > > > > > > Thanks,
> > > > > > > Huaxiang
> > > > > > >
> > > > > > > On Wed, Oct 14, 2020 at 9:42 AM Nick Dimiduk <
> > ndimiduk@apache.org>
> > > > > > wrote:
> > > > > > >
> > > > > > > > Thank you Andrew. I was of a similar mind -- if I'm doing
> 2.3,
> > > 2.4
> > > > > > would
> > > > > > > be
> > > > > > > > only of minimal incremental overhead. It would probably
> > encourage
> > > > the
> > > > > > > early
> > > > > > > > retirement of 2.3 :)
> > > > > > > >
> > > > > > > > Does anyone else have outstanding work they're hoping to see
> in
> > > > 2.4?
> > > > > > I'd
> > > > > > > > love to see an initial RC posted in the first half of
> November.
> > > > > > > >
> > > > > > > > Thanks,
> > > > > > > > Nick
> > > > > > > >
> > > > > > > > On Fri, Oct 9, 2020 at 2:54 PM Andrew Purtell <
> > > apurtell@apache.org
> > > > >
> > > > > > > wrote:
> > > > > > > >
> > > > > > > > > If nobody else wants to take this, I will.
> > > > > > > > >
> > > > > > > > > Reid is co-RMing the pending 1.7 release with me. We've
> been
> > > > going
> > > > > > slow
> > > > > > > > > with it waiting for backports but can move forward with it
> > now
> > > > that
> > > > > > the
> > > > > > > > > desired work has landed. I am not sensing any particular
> > > urgency
> > > > > with
> > > > > > > 1.x
> > > > > > > > > releasing. Anyway, doing 2.4 as well would be only
> > incremental
> > > > > > effort.
> > > > > > > > >
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > On Tue, Oct 6, 2020 at 3:38 PM Nick Dimiduk <
> > > ndimiduk@apache.org
> > > > >
> > > > > > > wrote:
> > > > > > > > >
> > > > > > > > > > Bump.
> > > > > > > > > >
> > > > > > > > > > We still need a volunteer for a release manager for 2.4.
> If
> > > we
> > > > > > delay
> > > > > > > > much
> > > > > > > > > > longer, we'll need another discussion, which is whether
> we
> > > > still
> > > > > > want
> > > > > > > > to
> > > > > > > > > > target the 6-month cadence on minor releases.
> > > > > > > > > >
> > > > > > > > > > In terms of outstanding work I'm hoping to see in 2.4,
> I'd
> > > like
> > > > > to
> > > > > > > get
> > > > > > > > a
> > > > > > > > > > number of these normalizer improvements in [0], or
> > determine
> > > > that
> > > > > > > > they're
> > > > > > > > > > not a problem. I think there's work on meta region
> replicas
> > > > > that's
> > > > > > > also
> > > > > > > > > > targeting this release, but I don't have a JIRA handy.
> Are
> > > > there
> > > > > > > > > > other features folks are actively working on with 2.4 in
> > > mind?
> > > > > > Please
> > > > > > > > > keep
> > > > > > > > > > in mind that there's really only about 4 weeks worth of
> dev
> > > > time
> > > > > > left
> > > > > > > > > > before we'll need a feature freeze for stabilizing and
> the
> > > > first
> > > > > > RC.
> > > > > > > > > >
> > > > > > > > > > Thanks,
> > > > > > > > > > Nick
> > > > > > > > > >
> > > > > > > > > > [0]:
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > >
> > > > > > > >
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20HBASE%20AND%20resolution%20%3D%20Unresolved%20AND%20component%20%3D%20Normalizer%20ORDER%20BY%20issuetype%20DESC
> > > > > > > > > >
> > > > > > > > > > On Tue, Aug 25, 2020 at 10:36 AM Nick Dimiduk <
> > > > > ndimiduk@apache.org
> > > > > > >
> > > > > > > > > wrote:
> > > > > > > > > >
> > > > > > > > > > > Heya!
> > > > > > > > > > >
> > > > > > > > > > > Now that 2.3.x is in maintenance mode, I'd like to
> start
> > a
> > > > > > > discussion
> > > > > > > > > on
> > > > > > > > > > > 2.4. We've long agreed that a more frequent cadence of
> > > minor
> > > > > > > releases
> > > > > > > > > is
> > > > > > > > > > > superior to a long gap between them. I think we should
> > > > target a
> > > > > > > 2.4.0
> > > > > > > > > > > release sometime before the end of the calendar year,
> > > perhaps
> > > > > the
> > > > > > > > first
> > > > > > > > > > > weeks of December.
> > > > > > > > > > >
> > > > > > > > > > > Hopefully the workload for the next release manager
> isn't
> > > as
> > > > > high
> > > > > > > as
> > > > > > > > it
> > > > > > > > > > > was for 2.3, given that much of the test stabilization
> > > effort
> > > > > > > > happened
> > > > > > > > > on
> > > > > > > > > > > branch-2. However, we are running in a new jenkins
> > > > environment,
> > > > > > and
> > > > > > > > > > entropy
> > > > > > > > > > > is nothing if not persistent, which puts that hard-won
> > > > > stability
> > > > > > > back
> > > > > > > > > > into
> > > > > > > > > > > doubt. Thus, I think it would be good if we can start
> > > sooner
> > > > > than
> > > > > > > > > later.
> > > > > > > > > > >
> > > > > > > > > > > Would anyone like to volunteer to run the 2.4 release
> > line?
> > > > The
> > > > > > > level
> > > > > > > > > of
> > > > > > > > > > > effort is admittedly non-negligible: (1) stabilize
> > branch-2
> > > > (2)
> > > > > > > test
> > > > > > > > > > > new-for-branch-2 features (3) run initial release (4)
> cut
> > > > > monthly
> > > > > > > > patch
> > > > > > > > > > > releases until 2.5 is out. I have a strong interest in
> > > seeing
> > > > > > this
> > > > > > > > > > release
> > > > > > > > > > > through, so I volunteer to assist with these
> activities.
> > > > > > > > > > >
> > > > > > > > > > > Thanks,
> > > > > > > > > > > Nick
> > > > > > > > > > >
> > > > > > > > > >
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > --
> > > > > > > > > Best regards,
> > > > > > > > > Andrew
> > > > > > > > >
> > > > > > > > > Words like orphans lost among the crosstalk, meaning torn
> > from
> > > > > > truth's
> > > > > > > > > decrepit hands
> > > > > > > > >    - A23, Crosstalk
> > > > > > > > >
> > > > > > > >
> > > > > > >
> > > > > >
> > > > > >
> > > > > > --
> > > > > > Best regards,
> > > > > > Andrew
> > > > > >
> > > > > > Words like orphans lost among the crosstalk, meaning torn from
> > > truth's
> > > > > > decrepit hands
> > > > > >    - A23, Crosstalk
> > > > > >
> > > > >
> > > >
> > >
> > >
> > > --
> > > Best regards,
> > > Andrew
> > >
> > > Words like orphans lost among the crosstalk, meaning torn from truth's
> > > decrepit hands
> > >    - A23, Crosstalk
> > >
> >
>
>
> --
> Best regards,
> Andrew
>
> Words like orphans lost among the crosstalk, meaning torn from truth's
> decrepit hands
>    - A23, Crosstalk
>

Re: [DISCUSS] A plan for 2.4

Posted by Andrew Purtell <ap...@apache.org>.

Thank you, Critical seems more appropriate. As RM (assuming that's me now
:-) ) I think it's fine to wait the estimated week for commit, then we can
proceed.

On Thu, Oct 15, 2020 at 2:51 PM Stack <st...@duboce.net> wrote:

> On Thu, Oct 15, 2020 at 12:00 PM Andrew Purtell <ap...@apache.org>
> wrote:
>
> > Is it a blocker? Maybe mark it Critical? Either way, a week is well
> within
> > the timeframe discussed earlier. Will come back here to check progress at
> > the end of next week. In the meantime it's fine to wait in my opinion.
> >
> >
> Sorry. Critical yes, blocker no. My bad. Let me change priority.
> S
>
>
> >
> > On Wed, Oct 14, 2020 at 8:55 PM Stack <st...@duboce.net> wrote:
> >
> > > On Wed, Oct 14, 2020 at 6:58 PM 张铎(Duo Zhang) <pa...@gmail.com>
> > > wrote:
> > >
> > > > I do not think HBASE-18070 is a blocker for 2.4 as it is still on a
> > > feature
> > > > branch.
> > > > And in the title of several jiras we have something like 'root/meta',
> > > which
> > > > seems to me that the intention is to also solve something related to
> > > > splittable meta, then I do not think the work can be done in the near
> > > > future...
> > > >
> > > >
> > > A few of us want it in 2.4. I've just made it a blocker (Hope that ok
> > > Andrew). Should be done in the next week or so. Will shout if we run
> into
> > > obstacles (and we can punt on it then). We'd like to deploy the feature
> > to
> > > address hbase:meta issues we are seeing in production.
> > >
> > > What we're building should work for root if needed but will worry about
> > > that when root shows up -- if it shows up.
> > >
> > > Thanks,
> > > S
> > >
> > >
> > >
> > > > Andrew Purtell <ap...@apache.org> 于2020年10月15日周四 上午1:53写道：
> > > >
> > > > > To meet Nick's objective of a 2.4 release in November, we could
> wait
> > > two
> > > > or
> > > > > three weeks for the meta replica work to completely land. Will that
> > be
> > > > > sufficient?
> > > > >
> > > > > On Wed, Oct 14, 2020 at 10:03 AM Huaxiang Sun <
> huaxiangsun@gmail.com
> > >
> > > > > wrote:
> > > > >
> > > > > > For meta replica work, the task jira is HBASE-18070
> > > > > > <https://issues.apache.org/jira/browse/HBASE-18070>. The source
> > side
> > > > > > change
> > > > > > has been committed, Endpoint change is under review.
> > > > > > Meta Replica load balance support at client side is in progress.
> > > > > >
> > > > > > Thanks,
> > > > > > Huaxiang
> > > > > >
> > > > > > On Wed, Oct 14, 2020 at 9:42 AM Nick Dimiduk <
> ndimiduk@apache.org>
> > > > > wrote:
> > > > > >
> > > > > > > Thank you Andrew. I was of a similar mind -- if I'm doing 2.3,
> > 2.4
> > > > > would
> > > > > > be
> > > > > > > only of minimal incremental overhead. It would probably
> encourage
> > > the
> > > > > > early
> > > > > > > retirement of 2.3 :)
> > > > > > >
> > > > > > > Does anyone else have outstanding work they're hoping to see in
> > > 2.4?
> > > > > I'd
> > > > > > > love to see an initial RC posted in the first half of November.
> > > > > > >
> > > > > > > Thanks,
> > > > > > > Nick
> > > > > > >
> > > > > > > On Fri, Oct 9, 2020 at 2:54 PM Andrew Purtell <
> > apurtell@apache.org
> > > >
> > > > > > wrote:
> > > > > > >
> > > > > > > > If nobody else wants to take this, I will.
> > > > > > > >
> > > > > > > > Reid is co-RMing the pending 1.7 release with me. We've been
> > > going
> > > > > slow
> > > > > > > > with it waiting for backports but can move forward with it
> now
> > > that
> > > > > the
> > > > > > > > desired work has landed. I am not sensing any particular
> > urgency
> > > > with
> > > > > > 1.x
> > > > > > > > releasing. Anyway, doing 2.4 as well would be only
> incremental
> > > > > effort.
> > > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > > > On Tue, Oct 6, 2020 at 3:38 PM Nick Dimiduk <
> > ndimiduk@apache.org
> > > >
> > > > > > wrote:
> > > > > > > >
> > > > > > > > > Bump.
> > > > > > > > >
> > > > > > > > > We still need a volunteer for a release manager for 2.4. If
> > we
> > > > > delay
> > > > > > > much
> > > > > > > > > longer, we'll need another discussion, which is whether we
> > > still
> > > > > want
> > > > > > > to
> > > > > > > > > target the 6-month cadence on minor releases.
> > > > > > > > >
> > > > > > > > > In terms of outstanding work I'm hoping to see in 2.4, I'd
> > like
> > > > to
> > > > > > get
> > > > > > > a
> > > > > > > > > number of these normalizer improvements in [0], or
> determine
> > > that
> > > > > > > they're
> > > > > > > > > not a problem. I think there's work on meta region replicas
> > > > that's
> > > > > > also
> > > > > > > > > targeting this release, but I don't have a JIRA handy. Are
> > > there
> > > > > > > > > other features folks are actively working on with 2.4 in
> > mind?
> > > > > Please
> > > > > > > > keep
> > > > > > > > > in mind that there's really only about 4 weeks worth of dev
> > > time
> > > > > left
> > > > > > > > > before we'll need a feature freeze for stabilizing and the
> > > first
> > > > > RC.
> > > > > > > > >
> > > > > > > > > Thanks,
> > > > > > > > > Nick
> > > > > > > > >
> > > > > > > > > [0]:
> > > > > > > > >
> > > > > > > > >
> > > > > > > >
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20HBASE%20AND%20resolution%20%3D%20Unresolved%20AND%20component%20%3D%20Normalizer%20ORDER%20BY%20issuetype%20DESC
> > > > > > > > >
> > > > > > > > > On Tue, Aug 25, 2020 at 10:36 AM Nick Dimiduk <
> > > > ndimiduk@apache.org
> > > > > >
> > > > > > > > wrote:
> > > > > > > > >
> > > > > > > > > > Heya!
> > > > > > > > > >
> > > > > > > > > > Now that 2.3.x is in maintenance mode, I'd like to start
> a
> > > > > > discussion
> > > > > > > > on
> > > > > > > > > > 2.4. We've long agreed that a more frequent cadence of
> > minor
> > > > > > releases
> > > > > > > > is
> > > > > > > > > > superior to a long gap between them. I think we should
> > > target a
> > > > > > 2.4.0
> > > > > > > > > > release sometime before the end of the calendar year,
> > perhaps
> > > > the
> > > > > > > first
> > > > > > > > > > weeks of December.
> > > > > > > > > >
> > > > > > > > > > Hopefully the workload for the next release manager isn't
> > as
> > > > high
> > > > > > as
> > > > > > > it
> > > > > > > > > > was for 2.3, given that much of the test stabilization
> > effort
> > > > > > > happened
> > > > > > > > on
> > > > > > > > > > branch-2. However, we are running in a new jenkins
> > > environment,
> > > > > and
> > > > > > > > > entropy
> > > > > > > > > > is nothing if not persistent, which puts that hard-won
> > > > stability
> > > > > > back
> > > > > > > > > into
> > > > > > > > > > doubt. Thus, I think it would be good if we can start
> > sooner
> > > > than
> > > > > > > > later.
> > > > > > > > > >
> > > > > > > > > > Would anyone like to volunteer to run the 2.4 release
> line?
> > > The
> > > > > > level
> > > > > > > > of
> > > > > > > > > > effort is admittedly non-negligible: (1) stabilize
> branch-2
> > > (2)
> > > > > > test
> > > > > > > > > > new-for-branch-2 features (3) run initial release (4) cut
> > > > monthly
> > > > > > > patch
> > > > > > > > > > releases until 2.5 is out. I have a strong interest in
> > seeing
> > > > > this
> > > > > > > > > release
> > > > > > > > > > through, so I volunteer to assist with these activities.
> > > > > > > > > >
> > > > > > > > > > Thanks,
> > > > > > > > > > Nick
> > > > > > > > > >
> > > > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > > > --
> > > > > > > > Best regards,
> > > > > > > > Andrew
> > > > > > > >
> > > > > > > > Words like orphans lost among the crosstalk, meaning torn
> from
> > > > > truth's
> > > > > > > > decrepit hands
> > > > > > > >    - A23, Crosstalk
> > > > > > > >
> > > > > > >
> > > > > >
> > > > >
> > > > >
> > > > > --
> > > > > Best regards,
> > > > > Andrew
> > > > >
> > > > > Words like orphans lost among the crosstalk, meaning torn from
> > truth's
> > > > > decrepit hands
> > > > >    - A23, Crosstalk
> > > > >
> > > >
> > >
> >
> >
> > --
> > Best regards,
> > Andrew
> >
> > Words like orphans lost among the crosstalk, meaning torn from truth's
> > decrepit hands
> >    - A23, Crosstalk
> >
>


-- 
Best regards,
Andrew

Words like orphans lost among the crosstalk, meaning torn from truth's
decrepit hands
   - A23, Crosstalk

Re: [DISCUSS] A plan for 2.4

Posted by Stack <st...@duboce.net>.

On Thu, Oct 15, 2020 at 12:00 PM Andrew Purtell <ap...@apache.org> wrote:

> Is it a blocker? Maybe mark it Critical? Either way, a week is well within
> the timeframe discussed earlier. Will come back here to check progress at
> the end of next week. In the meantime it's fine to wait in my opinion.
>
>
Sorry. Critical yes, blocker no. My bad. Let me change priority.
S


>
> On Wed, Oct 14, 2020 at 8:55 PM Stack <st...@duboce.net> wrote:
>
> > On Wed, Oct 14, 2020 at 6:58 PM 张铎(Duo Zhang) <pa...@gmail.com>
> > wrote:
> >
> > > I do not think HBASE-18070 is a blocker for 2.4 as it is still on a
> > feature
> > > branch.
> > > And in the title of several jiras we have something like 'root/meta',
> > which
> > > seems to me that the intention is to also solve something related to
> > > splittable meta, then I do not think the work can be done in the near
> > > future...
> > >
> > >
> > A few of us want it in 2.4. I've just made it a blocker (Hope that ok
> > Andrew). Should be done in the next week or so. Will shout if we run into
> > obstacles (and we can punt on it then). We'd like to deploy the feature
> to
> > address hbase:meta issues we are seeing in production.
> >
> > What we're building should work for root if needed but will worry about
> > that when root shows up -- if it shows up.
> >
> > Thanks,
> > S
> >
> >
> >
> > > Andrew Purtell <ap...@apache.org> 于2020年10月15日周四 上午1:53写道：
> > >
> > > > To meet Nick's objective of a 2.4 release in November, we could wait
> > two
> > > or
> > > > three weeks for the meta replica work to completely land. Will that
> be
> > > > sufficient?
> > > >
> > > > On Wed, Oct 14, 2020 at 10:03 AM Huaxiang Sun <huaxiangsun@gmail.com
> >
> > > > wrote:
> > > >
> > > > > For meta replica work, the task jira is HBASE-18070
> > > > > <https://issues.apache.org/jira/browse/HBASE-18070>. The source
> side
> > > > > change
> > > > > has been committed, Endpoint change is under review.
> > > > > Meta Replica load balance support at client side is in progress.
> > > > >
> > > > > Thanks,
> > > > > Huaxiang
> > > > >
> > > > > On Wed, Oct 14, 2020 at 9:42 AM Nick Dimiduk <nd...@apache.org>
> > > > wrote:
> > > > >
> > > > > > Thank you Andrew. I was of a similar mind -- if I'm doing 2.3,
> 2.4
> > > > would
> > > > > be
> > > > > > only of minimal incremental overhead. It would probably encourage
> > the
> > > > > early
> > > > > > retirement of 2.3 :)
> > > > > >
> > > > > > Does anyone else have outstanding work they're hoping to see in
> > 2.4?
> > > > I'd
> > > > > > love to see an initial RC posted in the first half of November.
> > > > > >
> > > > > > Thanks,
> > > > > > Nick
> > > > > >
> > > > > > On Fri, Oct 9, 2020 at 2:54 PM Andrew Purtell <
> apurtell@apache.org
> > >
> > > > > wrote:
> > > > > >
> > > > > > > If nobody else wants to take this, I will.
> > > > > > >
> > > > > > > Reid is co-RMing the pending 1.7 release with me. We've been
> > going
> > > > slow
> > > > > > > with it waiting for backports but can move forward with it now
> > that
> > > > the
> > > > > > > desired work has landed. I am not sensing any particular
> urgency
> > > with
> > > > > 1.x
> > > > > > > releasing. Anyway, doing 2.4 as well would be only incremental
> > > > effort.
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > On Tue, Oct 6, 2020 at 3:38 PM Nick Dimiduk <
> ndimiduk@apache.org
> > >
> > > > > wrote:
> > > > > > >
> > > > > > > > Bump.
> > > > > > > >
> > > > > > > > We still need a volunteer for a release manager for 2.4. If
> we
> > > > delay
> > > > > > much
> > > > > > > > longer, we'll need another discussion, which is whether we
> > still
> > > > want
> > > > > > to
> > > > > > > > target the 6-month cadence on minor releases.
> > > > > > > >
> > > > > > > > In terms of outstanding work I'm hoping to see in 2.4, I'd
> like
> > > to
> > > > > get
> > > > > > a
> > > > > > > > number of these normalizer improvements in [0], or determine
> > that
> > > > > > they're
> > > > > > > > not a problem. I think there's work on meta region replicas
> > > that's
> > > > > also
> > > > > > > > targeting this release, but I don't have a JIRA handy. Are
> > there
> > > > > > > > other features folks are actively working on with 2.4 in
> mind?
> > > > Please
> > > > > > > keep
> > > > > > > > in mind that there's really only about 4 weeks worth of dev
> > time
> > > > left
> > > > > > > > before we'll need a feature freeze for stabilizing and the
> > first
> > > > RC.
> > > > > > > >
> > > > > > > > Thanks,
> > > > > > > > Nick
> > > > > > > >
> > > > > > > > [0]:
> > > > > > > >
> > > > > > > >
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20HBASE%20AND%20resolution%20%3D%20Unresolved%20AND%20component%20%3D%20Normalizer%20ORDER%20BY%20issuetype%20DESC
> > > > > > > >
> > > > > > > > On Tue, Aug 25, 2020 at 10:36 AM Nick Dimiduk <
> > > ndimiduk@apache.org
> > > > >
> > > > > > > wrote:
> > > > > > > >
> > > > > > > > > Heya!
> > > > > > > > >
> > > > > > > > > Now that 2.3.x is in maintenance mode, I'd like to start a
> > > > > discussion
> > > > > > > on
> > > > > > > > > 2.4. We've long agreed that a more frequent cadence of
> minor
> > > > > releases
> > > > > > > is
> > > > > > > > > superior to a long gap between them. I think we should
> > target a
> > > > > 2.4.0
> > > > > > > > > release sometime before the end of the calendar year,
> perhaps
> > > the
> > > > > > first
> > > > > > > > > weeks of December.
> > > > > > > > >
> > > > > > > > > Hopefully the workload for the next release manager isn't
> as
> > > high
> > > > > as
> > > > > > it
> > > > > > > > > was for 2.3, given that much of the test stabilization
> effort
> > > > > > happened
> > > > > > > on
> > > > > > > > > branch-2. However, we are running in a new jenkins
> > environment,
> > > > and
> > > > > > > > entropy
> > > > > > > > > is nothing if not persistent, which puts that hard-won
> > > stability
> > > > > back
> > > > > > > > into
> > > > > > > > > doubt. Thus, I think it would be good if we can start
> sooner
> > > than
> > > > > > > later.
> > > > > > > > >
> > > > > > > > > Would anyone like to volunteer to run the 2.4 release line?
> > The
> > > > > level
> > > > > > > of
> > > > > > > > > effort is admittedly non-negligible: (1) stabilize branch-2
> > (2)
> > > > > test
> > > > > > > > > new-for-branch-2 features (3) run initial release (4) cut
> > > monthly
> > > > > > patch
> > > > > > > > > releases until 2.5 is out. I have a strong interest in
> seeing
> > > > this
> > > > > > > > release
> > > > > > > > > through, so I volunteer to assist with these activities.
> > > > > > > > >
> > > > > > > > > Thanks,
> > > > > > > > > Nick
> > > > > > > > >
> > > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > --
> > > > > > > Best regards,
> > > > > > > Andrew
> > > > > > >
> > > > > > > Words like orphans lost among the crosstalk, meaning torn from
> > > > truth's
> > > > > > > decrepit hands
> > > > > > >    - A23, Crosstalk
> > > > > > >
> > > > > >
> > > > >
> > > >
> > > >
> > > > --
> > > > Best regards,
> > > > Andrew
> > > >
> > > > Words like orphans lost among the crosstalk, meaning torn from
> truth's
> > > > decrepit hands
> > > >    - A23, Crosstalk
> > > >
> > >
> >
>
>
> --
> Best regards,
> Andrew
>
> Words like orphans lost among the crosstalk, meaning torn from truth's
> decrepit hands
>    - A23, Crosstalk
>

Re: [DISCUSS] A plan for 2.4

Posted by Andrew Purtell <ap...@apache.org>.

Is it a blocker? Maybe mark it Critical? Either way, a week is well within
the timeframe discussed earlier. Will come back here to check progress at
the end of next week. In the meantime it's fine to wait in my opinion.


On Wed, Oct 14, 2020 at 8:55 PM Stack <st...@duboce.net> wrote:

> On Wed, Oct 14, 2020 at 6:58 PM 张铎(Duo Zhang) <pa...@gmail.com>
> wrote:
>
> > I do not think HBASE-18070 is a blocker for 2.4 as it is still on a
> feature
> > branch.
> > And in the title of several jiras we have something like 'root/meta',
> which
> > seems to me that the intention is to also solve something related to
> > splittable meta, then I do not think the work can be done in the near
> > future...
> >
> >
> A few of us want it in 2.4. I've just made it a blocker (Hope that ok
> Andrew). Should be done in the next week or so. Will shout if we run into
> obstacles (and we can punt on it then). We'd like to deploy the feature to
> address hbase:meta issues we are seeing in production.
>
> What we're building should work for root if needed but will worry about
> that when root shows up -- if it shows up.
>
> Thanks,
> S
>
>
>
> > Andrew Purtell <ap...@apache.org> 于2020年10月15日周四 上午1:53写道：
> >
> > > To meet Nick's objective of a 2.4 release in November, we could wait
> two
> > or
> > > three weeks for the meta replica work to completely land. Will that be
> > > sufficient?
> > >
> > > On Wed, Oct 14, 2020 at 10:03 AM Huaxiang Sun <hu...@gmail.com>
> > > wrote:
> > >
> > > > For meta replica work, the task jira is HBASE-18070
> > > > <https://issues.apache.org/jira/browse/HBASE-18070>. The source side
> > > > change
> > > > has been committed, Endpoint change is under review.
> > > > Meta Replica load balance support at client side is in progress.
> > > >
> > > > Thanks,
> > > > Huaxiang
> > > >
> > > > On Wed, Oct 14, 2020 at 9:42 AM Nick Dimiduk <nd...@apache.org>
> > > wrote:
> > > >
> > > > > Thank you Andrew. I was of a similar mind -- if I'm doing 2.3, 2.4
> > > would
> > > > be
> > > > > only of minimal incremental overhead. It would probably encourage
> the
> > > > early
> > > > > retirement of 2.3 :)
> > > > >
> > > > > Does anyone else have outstanding work they're hoping to see in
> 2.4?
> > > I'd
> > > > > love to see an initial RC posted in the first half of November.
> > > > >
> > > > > Thanks,
> > > > > Nick
> > > > >
> > > > > On Fri, Oct 9, 2020 at 2:54 PM Andrew Purtell <apurtell@apache.org
> >
> > > > wrote:
> > > > >
> > > > > > If nobody else wants to take this, I will.
> > > > > >
> > > > > > Reid is co-RMing the pending 1.7 release with me. We've been
> going
> > > slow
> > > > > > with it waiting for backports but can move forward with it now
> that
> > > the
> > > > > > desired work has landed. I am not sensing any particular urgency
> > with
> > > > 1.x
> > > > > > releasing. Anyway, doing 2.4 as well would be only incremental
> > > effort.
> > > > > >
> > > > > >
> > > > > >
> > > > > > On Tue, Oct 6, 2020 at 3:38 PM Nick Dimiduk <ndimiduk@apache.org
> >
> > > > wrote:
> > > > > >
> > > > > > > Bump.
> > > > > > >
> > > > > > > We still need a volunteer for a release manager for 2.4. If we
> > > delay
> > > > > much
> > > > > > > longer, we'll need another discussion, which is whether we
> still
> > > want
> > > > > to
> > > > > > > target the 6-month cadence on minor releases.
> > > > > > >
> > > > > > > In terms of outstanding work I'm hoping to see in 2.4, I'd like
> > to
> > > > get
> > > > > a
> > > > > > > number of these normalizer improvements in [0], or determine
> that
> > > > > they're
> > > > > > > not a problem. I think there's work on meta region replicas
> > that's
> > > > also
> > > > > > > targeting this release, but I don't have a JIRA handy. Are
> there
> > > > > > > other features folks are actively working on with 2.4 in mind?
> > > Please
> > > > > > keep
> > > > > > > in mind that there's really only about 4 weeks worth of dev
> time
> > > left
> > > > > > > before we'll need a feature freeze for stabilizing and the
> first
> > > RC.
> > > > > > >
> > > > > > > Thanks,
> > > > > > > Nick
> > > > > > >
> > > > > > > [0]:
> > > > > > >
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20HBASE%20AND%20resolution%20%3D%20Unresolved%20AND%20component%20%3D%20Normalizer%20ORDER%20BY%20issuetype%20DESC
> > > > > > >
> > > > > > > On Tue, Aug 25, 2020 at 10:36 AM Nick Dimiduk <
> > ndimiduk@apache.org
> > > >
> > > > > > wrote:
> > > > > > >
> > > > > > > > Heya!
> > > > > > > >
> > > > > > > > Now that 2.3.x is in maintenance mode, I'd like to start a
> > > > discussion
> > > > > > on
> > > > > > > > 2.4. We've long agreed that a more frequent cadence of minor
> > > > releases
> > > > > > is
> > > > > > > > superior to a long gap between them. I think we should
> target a
> > > > 2.4.0
> > > > > > > > release sometime before the end of the calendar year, perhaps
> > the
> > > > > first
> > > > > > > > weeks of December.
> > > > > > > >
> > > > > > > > Hopefully the workload for the next release manager isn't as
> > high
> > > > as
> > > > > it
> > > > > > > > was for 2.3, given that much of the test stabilization effort
> > > > > happened
> > > > > > on
> > > > > > > > branch-2. However, we are running in a new jenkins
> environment,
> > > and
> > > > > > > entropy
> > > > > > > > is nothing if not persistent, which puts that hard-won
> > stability
> > > > back
> > > > > > > into
> > > > > > > > doubt. Thus, I think it would be good if we can start sooner
> > than
> > > > > > later.
> > > > > > > >
> > > > > > > > Would anyone like to volunteer to run the 2.4 release line?
> The
> > > > level
> > > > > > of
> > > > > > > > effort is admittedly non-negligible: (1) stabilize branch-2
> (2)
> > > > test
> > > > > > > > new-for-branch-2 features (3) run initial release (4) cut
> > monthly
> > > > > patch
> > > > > > > > releases until 2.5 is out. I have a strong interest in seeing
> > > this
> > > > > > > release
> > > > > > > > through, so I volunteer to assist with these activities.
> > > > > > > >
> > > > > > > > Thanks,
> > > > > > > > Nick
> > > > > > > >
> > > > > > >
> > > > > >
> > > > > >
> > > > > > --
> > > > > > Best regards,
> > > > > > Andrew
> > > > > >
> > > > > > Words like orphans lost among the crosstalk, meaning torn from
> > > truth's
> > > > > > decrepit hands
> > > > > >    - A23, Crosstalk
> > > > > >
> > > > >
> > > >
> > >
> > >
> > > --
> > > Best regards,
> > > Andrew
> > >
> > > Words like orphans lost among the crosstalk, meaning torn from truth's
> > > decrepit hands
> > >    - A23, Crosstalk
> > >
> >
>


-- 
Best regards,
Andrew

Words like orphans lost among the crosstalk, meaning torn from truth's
decrepit hands
   - A23, Crosstalk

Re: [DISCUSS] A plan for 2.4

Posted by Stack <st...@duboce.net>.

On Wed, Oct 14, 2020 at 6:58 PM 张铎(Duo Zhang) <pa...@gmail.com> wrote:

> I do not think HBASE-18070 is a blocker for 2.4 as it is still on a feature
> branch.
> And in the title of several jiras we have something like 'root/meta', which
> seems to me that the intention is to also solve something related to
> splittable meta, then I do not think the work can be done in the near
> future...
>
>
A few of us want it in 2.4. I've just made it a blocker (Hope that ok
Andrew). Should be done in the next week or so. Will shout if we run into
obstacles (and we can punt on it then). We'd like to deploy the feature to
address hbase:meta issues we are seeing in production.

What we're building should work for root if needed but will worry about
that when root shows up -- if it shows up.

Thanks,
S



> Andrew Purtell <ap...@apache.org> 于2020年10月15日周四 上午1:53写道：
>
> > To meet Nick's objective of a 2.4 release in November, we could wait two
> or
> > three weeks for the meta replica work to completely land. Will that be
> > sufficient?
> >
> > On Wed, Oct 14, 2020 at 10:03 AM Huaxiang Sun <hu...@gmail.com>
> > wrote:
> >
> > > For meta replica work, the task jira is HBASE-18070
> > > <https://issues.apache.org/jira/browse/HBASE-18070>. The source side
> > > change
> > > has been committed, Endpoint change is under review.
> > > Meta Replica load balance support at client side is in progress.
> > >
> > > Thanks,
> > > Huaxiang
> > >
> > > On Wed, Oct 14, 2020 at 9:42 AM Nick Dimiduk <nd...@apache.org>
> > wrote:
> > >
> > > > Thank you Andrew. I was of a similar mind -- if I'm doing 2.3, 2.4
> > would
> > > be
> > > > only of minimal incremental overhead. It would probably encourage the
> > > early
> > > > retirement of 2.3 :)
> > > >
> > > > Does anyone else have outstanding work they're hoping to see in 2.4?
> > I'd
> > > > love to see an initial RC posted in the first half of November.
> > > >
> > > > Thanks,
> > > > Nick
> > > >
> > > > On Fri, Oct 9, 2020 at 2:54 PM Andrew Purtell <ap...@apache.org>
> > > wrote:
> > > >
> > > > > If nobody else wants to take this, I will.
> > > > >
> > > > > Reid is co-RMing the pending 1.7 release with me. We've been going
> > slow
> > > > > with it waiting for backports but can move forward with it now that
> > the
> > > > > desired work has landed. I am not sensing any particular urgency
> with
> > > 1.x
> > > > > releasing. Anyway, doing 2.4 as well would be only incremental
> > effort.
> > > > >
> > > > >
> > > > >
> > > > > On Tue, Oct 6, 2020 at 3:38 PM Nick Dimiduk <nd...@apache.org>
> > > wrote:
> > > > >
> > > > > > Bump.
> > > > > >
> > > > > > We still need a volunteer for a release manager for 2.4. If we
> > delay
> > > > much
> > > > > > longer, we'll need another discussion, which is whether we still
> > want
> > > > to
> > > > > > target the 6-month cadence on minor releases.
> > > > > >
> > > > > > In terms of outstanding work I'm hoping to see in 2.4, I'd like
> to
> > > get
> > > > a
> > > > > > number of these normalizer improvements in [0], or determine that
> > > > they're
> > > > > > not a problem. I think there's work on meta region replicas
> that's
> > > also
> > > > > > targeting this release, but I don't have a JIRA handy. Are there
> > > > > > other features folks are actively working on with 2.4 in mind?
> > Please
> > > > > keep
> > > > > > in mind that there's really only about 4 weeks worth of dev time
> > left
> > > > > > before we'll need a feature freeze for stabilizing and the first
> > RC.
> > > > > >
> > > > > > Thanks,
> > > > > > Nick
> > > > > >
> > > > > > [0]:
> > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20HBASE%20AND%20resolution%20%3D%20Unresolved%20AND%20component%20%3D%20Normalizer%20ORDER%20BY%20issuetype%20DESC
> > > > > >
> > > > > > On Tue, Aug 25, 2020 at 10:36 AM Nick Dimiduk <
> ndimiduk@apache.org
> > >
> > > > > wrote:
> > > > > >
> > > > > > > Heya!
> > > > > > >
> > > > > > > Now that 2.3.x is in maintenance mode, I'd like to start a
> > > discussion
> > > > > on
> > > > > > > 2.4. We've long agreed that a more frequent cadence of minor
> > > releases
> > > > > is
> > > > > > > superior to a long gap between them. I think we should target a
> > > 2.4.0
> > > > > > > release sometime before the end of the calendar year, perhaps
> the
> > > > first
> > > > > > > weeks of December.
> > > > > > >
> > > > > > > Hopefully the workload for the next release manager isn't as
> high
> > > as
> > > > it
> > > > > > > was for 2.3, given that much of the test stabilization effort
> > > > happened
> > > > > on
> > > > > > > branch-2. However, we are running in a new jenkins environment,
> > and
> > > > > > entropy
> > > > > > > is nothing if not persistent, which puts that hard-won
> stability
> > > back
> > > > > > into
> > > > > > > doubt. Thus, I think it would be good if we can start sooner
> than
> > > > > later.
> > > > > > >
> > > > > > > Would anyone like to volunteer to run the 2.4 release line? The
> > > level
> > > > > of
> > > > > > > effort is admittedly non-negligible: (1) stabilize branch-2 (2)
> > > test
> > > > > > > new-for-branch-2 features (3) run initial release (4) cut
> monthly
> > > > patch
> > > > > > > releases until 2.5 is out. I have a strong interest in seeing
> > this
> > > > > > release
> > > > > > > through, so I volunteer to assist with these activities.
> > > > > > >
> > > > > > > Thanks,
> > > > > > > Nick
> > > > > > >
> > > > > >
> > > > >
> > > > >
> > > > > --
> > > > > Best regards,
> > > > > Andrew
> > > > >
> > > > > Words like orphans lost among the crosstalk, meaning torn from
> > truth's
> > > > > decrepit hands
> > > > >    - A23, Crosstalk
> > > > >
> > > >
> > >
> >
> >
> > --
> > Best regards,
> > Andrew
> >
> > Words like orphans lost among the crosstalk, meaning torn from truth's
> > decrepit hands
> >    - A23, Crosstalk
> >
>

Re: [DISCUSS] A plan for 2.4

Posted by "张铎 (Duo Zhang)" <pa...@gmail.com>.

I do not think HBASE-18070 is a blocker for 2.4 as it is still on a feature
branch.
And in the title of several jiras we have something like 'root/meta', which
seems to me that the intention is to also solve something related to
splittable meta, then I do not think the work can be done in the near
future...

Andrew Purtell <ap...@apache.org> 于2020年10月15日周四 上午1:53写道：

> To meet Nick's objective of a 2.4 release in November, we could wait two or
> three weeks for the meta replica work to completely land. Will that be
> sufficient?
>
> On Wed, Oct 14, 2020 at 10:03 AM Huaxiang Sun <hu...@gmail.com>
> wrote:
>
> > For meta replica work, the task jira is HBASE-18070
> > <https://issues.apache.org/jira/browse/HBASE-18070>. The source side
> > change
> > has been committed, Endpoint change is under review.
> > Meta Replica load balance support at client side is in progress.
> >
> > Thanks,
> > Huaxiang
> >
> > On Wed, Oct 14, 2020 at 9:42 AM Nick Dimiduk <nd...@apache.org>
> wrote:
> >
> > > Thank you Andrew. I was of a similar mind -- if I'm doing 2.3, 2.4
> would
> > be
> > > only of minimal incremental overhead. It would probably encourage the
> > early
> > > retirement of 2.3 :)
> > >
> > > Does anyone else have outstanding work they're hoping to see in 2.4?
> I'd
> > > love to see an initial RC posted in the first half of November.
> > >
> > > Thanks,
> > > Nick
> > >
> > > On Fri, Oct 9, 2020 at 2:54 PM Andrew Purtell <ap...@apache.org>
> > wrote:
> > >
> > > > If nobody else wants to take this, I will.
> > > >
> > > > Reid is co-RMing the pending 1.7 release with me. We've been going
> slow
> > > > with it waiting for backports but can move forward with it now that
> the
> > > > desired work has landed. I am not sensing any particular urgency with
> > 1.x
> > > > releasing. Anyway, doing 2.4 as well would be only incremental
> effort.
> > > >
> > > >
> > > >
> > > > On Tue, Oct 6, 2020 at 3:38 PM Nick Dimiduk <nd...@apache.org>
> > wrote:
> > > >
> > > > > Bump.
> > > > >
> > > > > We still need a volunteer for a release manager for 2.4. If we
> delay
> > > much
> > > > > longer, we'll need another discussion, which is whether we still
> want
> > > to
> > > > > target the 6-month cadence on minor releases.
> > > > >
> > > > > In terms of outstanding work I'm hoping to see in 2.4, I'd like to
> > get
> > > a
> > > > > number of these normalizer improvements in [0], or determine that
> > > they're
> > > > > not a problem. I think there's work on meta region replicas that's
> > also
> > > > > targeting this release, but I don't have a JIRA handy. Are there
> > > > > other features folks are actively working on with 2.4 in mind?
> Please
> > > > keep
> > > > > in mind that there's really only about 4 weeks worth of dev time
> left
> > > > > before we'll need a feature freeze for stabilizing and the first
> RC.
> > > > >
> > > > > Thanks,
> > > > > Nick
> > > > >
> > > > > [0]:
> > > > >
> > > > >
> > > >
> > >
> >
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20HBASE%20AND%20resolution%20%3D%20Unresolved%20AND%20component%20%3D%20Normalizer%20ORDER%20BY%20issuetype%20DESC
> > > > >
> > > > > On Tue, Aug 25, 2020 at 10:36 AM Nick Dimiduk <ndimiduk@apache.org
> >
> > > > wrote:
> > > > >
> > > > > > Heya!
> > > > > >
> > > > > > Now that 2.3.x is in maintenance mode, I'd like to start a
> > discussion
> > > > on
> > > > > > 2.4. We've long agreed that a more frequent cadence of minor
> > releases
> > > > is
> > > > > > superior to a long gap between them. I think we should target a
> > 2.4.0
> > > > > > release sometime before the end of the calendar year, perhaps the
> > > first
> > > > > > weeks of December.
> > > > > >
> > > > > > Hopefully the workload for the next release manager isn't as high
> > as
> > > it
> > > > > > was for 2.3, given that much of the test stabilization effort
> > > happened
> > > > on
> > > > > > branch-2. However, we are running in a new jenkins environment,
> and
> > > > > entropy
> > > > > > is nothing if not persistent, which puts that hard-won stability
> > back
> > > > > into
> > > > > > doubt. Thus, I think it would be good if we can start sooner than
> > > > later.
> > > > > >
> > > > > > Would anyone like to volunteer to run the 2.4 release line? The
> > level
> > > > of
> > > > > > effort is admittedly non-negligible: (1) stabilize branch-2 (2)
> > test
> > > > > > new-for-branch-2 features (3) run initial release (4) cut monthly
> > > patch
> > > > > > releases until 2.5 is out. I have a strong interest in seeing
> this
> > > > > release
> > > > > > through, so I volunteer to assist with these activities.
> > > > > >
> > > > > > Thanks,
> > > > > > Nick
> > > > > >
> > > > >
> > > >
> > > >
> > > > --
> > > > Best regards,
> > > > Andrew
> > > >
> > > > Words like orphans lost among the crosstalk, meaning torn from
> truth's
> > > > decrepit hands
> > > >    - A23, Crosstalk
> > > >
> > >
> >
>
>
> --
> Best regards,
> Andrew
>
> Words like orphans lost among the crosstalk, meaning torn from truth's
> decrepit hands
>    - A23, Crosstalk
>

Re: [DISCUSS] A plan for 2.4

Posted by Andrew Purtell <ap...@apache.org>.

To meet Nick's objective of a 2.4 release in November, we could wait two or
three weeks for the meta replica work to completely land. Will that be
sufficient?

On Wed, Oct 14, 2020 at 10:03 AM Huaxiang Sun <hu...@gmail.com> wrote:

> For meta replica work, the task jira is HBASE-18070
> <https://issues.apache.org/jira/browse/HBASE-18070>. The source side
> change
> has been committed, Endpoint change is under review.
> Meta Replica load balance support at client side is in progress.
>
> Thanks,
> Huaxiang
>
> On Wed, Oct 14, 2020 at 9:42 AM Nick Dimiduk <nd...@apache.org> wrote:
>
> > Thank you Andrew. I was of a similar mind -- if I'm doing 2.3, 2.4 would
> be
> > only of minimal incremental overhead. It would probably encourage the
> early
> > retirement of 2.3 :)
> >
> > Does anyone else have outstanding work they're hoping to see in 2.4? I'd
> > love to see an initial RC posted in the first half of November.
> >
> > Thanks,
> > Nick
> >
> > On Fri, Oct 9, 2020 at 2:54 PM Andrew Purtell <ap...@apache.org>
> wrote:
> >
> > > If nobody else wants to take this, I will.
> > >
> > > Reid is co-RMing the pending 1.7 release with me. We've been going slow
> > > with it waiting for backports but can move forward with it now that the
> > > desired work has landed. I am not sensing any particular urgency with
> 1.x
> > > releasing. Anyway, doing 2.4 as well would be only incremental effort.
> > >
> > >
> > >
> > > On Tue, Oct 6, 2020 at 3:38 PM Nick Dimiduk <nd...@apache.org>
> wrote:
> > >
> > > > Bump.
> > > >
> > > > We still need a volunteer for a release manager for 2.4. If we delay
> > much
> > > > longer, we'll need another discussion, which is whether we still want
> > to
> > > > target the 6-month cadence on minor releases.
> > > >
> > > > In terms of outstanding work I'm hoping to see in 2.4, I'd like to
> get
> > a
> > > > number of these normalizer improvements in [0], or determine that
> > they're
> > > > not a problem. I think there's work on meta region replicas that's
> also
> > > > targeting this release, but I don't have a JIRA handy. Are there
> > > > other features folks are actively working on with 2.4 in mind? Please
> > > keep
> > > > in mind that there's really only about 4 weeks worth of dev time left
> > > > before we'll need a feature freeze for stabilizing and the first RC.
> > > >
> > > > Thanks,
> > > > Nick
> > > >
> > > > [0]:
> > > >
> > > >
> > >
> >
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20HBASE%20AND%20resolution%20%3D%20Unresolved%20AND%20component%20%3D%20Normalizer%20ORDER%20BY%20issuetype%20DESC
> > > >
> > > > On Tue, Aug 25, 2020 at 10:36 AM Nick Dimiduk <nd...@apache.org>
> > > wrote:
> > > >
> > > > > Heya!
> > > > >
> > > > > Now that 2.3.x is in maintenance mode, I'd like to start a
> discussion
> > > on
> > > > > 2.4. We've long agreed that a more frequent cadence of minor
> releases
> > > is
> > > > > superior to a long gap between them. I think we should target a
> 2.4.0
> > > > > release sometime before the end of the calendar year, perhaps the
> > first
> > > > > weeks of December.
> > > > >
> > > > > Hopefully the workload for the next release manager isn't as high
> as
> > it
> > > > > was for 2.3, given that much of the test stabilization effort
> > happened
> > > on
> > > > > branch-2. However, we are running in a new jenkins environment, and
> > > > entropy
> > > > > is nothing if not persistent, which puts that hard-won stability
> back
> > > > into
> > > > > doubt. Thus, I think it would be good if we can start sooner than
> > > later.
> > > > >
> > > > > Would anyone like to volunteer to run the 2.4 release line? The
> level
> > > of
> > > > > effort is admittedly non-negligible: (1) stabilize branch-2 (2)
> test
> > > > > new-for-branch-2 features (3) run initial release (4) cut monthly
> > patch
> > > > > releases until 2.5 is out. I have a strong interest in seeing this
> > > > release
> > > > > through, so I volunteer to assist with these activities.
> > > > >
> > > > > Thanks,
> > > > > Nick
> > > > >
> > > >
> > >
> > >
> > > --
> > > Best regards,
> > > Andrew
> > >
> > > Words like orphans lost among the crosstalk, meaning torn from truth's
> > > decrepit hands
> > >    - A23, Crosstalk
> > >
> >
>


-- 
Best regards,
Andrew

Words like orphans lost among the crosstalk, meaning torn from truth's
decrepit hands
   - A23, Crosstalk

Re: [DISCUSS] A plan for 2.4

Posted by Huaxiang Sun <hu...@gmail.com>.

For meta replica work, the task jira is HBASE-18070
<https://issues.apache.org/jira/browse/HBASE-18070>. The source side change
has been committed, Endpoint change is under review.
Meta Replica load balance support at client side is in progress.

Thanks,
Huaxiang

On Wed, Oct 14, 2020 at 9:42 AM Nick Dimiduk <nd...@apache.org> wrote:

> Thank you Andrew. I was of a similar mind -- if I'm doing 2.3, 2.4 would be
> only of minimal incremental overhead. It would probably encourage the early
> retirement of 2.3 :)
>
> Does anyone else have outstanding work they're hoping to see in 2.4? I'd
> love to see an initial RC posted in the first half of November.
>
> Thanks,
> Nick
>
> On Fri, Oct 9, 2020 at 2:54 PM Andrew Purtell <ap...@apache.org> wrote:
>
> > If nobody else wants to take this, I will.
> >
> > Reid is co-RMing the pending 1.7 release with me. We've been going slow
> > with it waiting for backports but can move forward with it now that the
> > desired work has landed. I am not sensing any particular urgency with 1.x
> > releasing. Anyway, doing 2.4 as well would be only incremental effort.
> >
> >
> >
> > On Tue, Oct 6, 2020 at 3:38 PM Nick Dimiduk <nd...@apache.org> wrote:
> >
> > > Bump.
> > >
> > > We still need a volunteer for a release manager for 2.4. If we delay
> much
> > > longer, we'll need another discussion, which is whether we still want
> to
> > > target the 6-month cadence on minor releases.
> > >
> > > In terms of outstanding work I'm hoping to see in 2.4, I'd like to get
> a
> > > number of these normalizer improvements in [0], or determine that
> they're
> > > not a problem. I think there's work on meta region replicas that's also
> > > targeting this release, but I don't have a JIRA handy. Are there
> > > other features folks are actively working on with 2.4 in mind? Please
> > keep
> > > in mind that there's really only about 4 weeks worth of dev time left
> > > before we'll need a feature freeze for stabilizing and the first RC.
> > >
> > > Thanks,
> > > Nick
> > >
> > > [0]:
> > >
> > >
> >
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20HBASE%20AND%20resolution%20%3D%20Unresolved%20AND%20component%20%3D%20Normalizer%20ORDER%20BY%20issuetype%20DESC
> > >
> > > On Tue, Aug 25, 2020 at 10:36 AM Nick Dimiduk <nd...@apache.org>
> > wrote:
> > >
> > > > Heya!
> > > >
> > > > Now that 2.3.x is in maintenance mode, I'd like to start a discussion
> > on
> > > > 2.4. We've long agreed that a more frequent cadence of minor releases
> > is
> > > > superior to a long gap between them. I think we should target a 2.4.0
> > > > release sometime before the end of the calendar year, perhaps the
> first
> > > > weeks of December.
> > > >
> > > > Hopefully the workload for the next release manager isn't as high as
> it
> > > > was for 2.3, given that much of the test stabilization effort
> happened
> > on
> > > > branch-2. However, we are running in a new jenkins environment, and
> > > entropy
> > > > is nothing if not persistent, which puts that hard-won stability back
> > > into
> > > > doubt. Thus, I think it would be good if we can start sooner than
> > later.
> > > >
> > > > Would anyone like to volunteer to run the 2.4 release line? The level
> > of
> > > > effort is admittedly non-negligible: (1) stabilize branch-2 (2) test
> > > > new-for-branch-2 features (3) run initial release (4) cut monthly
> patch
> > > > releases until 2.5 is out. I have a strong interest in seeing this
> > > release
> > > > through, so I volunteer to assist with these activities.
> > > >
> > > > Thanks,
> > > > Nick
> > > >
> > >
> >
> >
> > --
> > Best regards,
> > Andrew
> >
> > Words like orphans lost among the crosstalk, meaning torn from truth's
> > decrepit hands
> >    - A23, Crosstalk
> >
>

Re: [DISCUSS] A plan for 2.4

Posted by Nick Dimiduk <nd...@apache.org>.

Thank you Andrew. I was of a similar mind -- if I'm doing 2.3, 2.4 would be
only of minimal incremental overhead. It would probably encourage the early
retirement of 2.3 :)

Does anyone else have outstanding work they're hoping to see in 2.4? I'd
love to see an initial RC posted in the first half of November.

Thanks,
Nick

On Fri, Oct 9, 2020 at 2:54 PM Andrew Purtell <ap...@apache.org> wrote:

> If nobody else wants to take this, I will.
>
> Reid is co-RMing the pending 1.7 release with me. We've been going slow
> with it waiting for backports but can move forward with it now that the
> desired work has landed. I am not sensing any particular urgency with 1.x
> releasing. Anyway, doing 2.4 as well would be only incremental effort.
>
>
>
> On Tue, Oct 6, 2020 at 3:38 PM Nick Dimiduk <nd...@apache.org> wrote:
>
> > Bump.
> >
> > We still need a volunteer for a release manager for 2.4. If we delay much
> > longer, we'll need another discussion, which is whether we still want to
> > target the 6-month cadence on minor releases.
> >
> > In terms of outstanding work I'm hoping to see in 2.4, I'd like to get a
> > number of these normalizer improvements in [0], or determine that they're
> > not a problem. I think there's work on meta region replicas that's also
> > targeting this release, but I don't have a JIRA handy. Are there
> > other features folks are actively working on with 2.4 in mind? Please
> keep
> > in mind that there's really only about 4 weeks worth of dev time left
> > before we'll need a feature freeze for stabilizing and the first RC.
> >
> > Thanks,
> > Nick
> >
> > [0]:
> >
> >
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20HBASE%20AND%20resolution%20%3D%20Unresolved%20AND%20component%20%3D%20Normalizer%20ORDER%20BY%20issuetype%20DESC
> >
> > On Tue, Aug 25, 2020 at 10:36 AM Nick Dimiduk <nd...@apache.org>
> wrote:
> >
> > > Heya!
> > >
> > > Now that 2.3.x is in maintenance mode, I'd like to start a discussion
> on
> > > 2.4. We've long agreed that a more frequent cadence of minor releases
> is
> > > superior to a long gap between them. I think we should target a 2.4.0
> > > release sometime before the end of the calendar year, perhaps the first
> > > weeks of December.
> > >
> > > Hopefully the workload for the next release manager isn't as high as it
> > > was for 2.3, given that much of the test stabilization effort happened
> on
> > > branch-2. However, we are running in a new jenkins environment, and
> > entropy
> > > is nothing if not persistent, which puts that hard-won stability back
> > into
> > > doubt. Thus, I think it would be good if we can start sooner than
> later.
> > >
> > > Would anyone like to volunteer to run the 2.4 release line? The level
> of
> > > effort is admittedly non-negligible: (1) stabilize branch-2 (2) test
> > > new-for-branch-2 features (3) run initial release (4) cut monthly patch
> > > releases until 2.5 is out. I have a strong interest in seeing this
> > release
> > > through, so I volunteer to assist with these activities.
> > >
> > > Thanks,
> > > Nick
> > >
> >
>
>
> --
> Best regards,
> Andrew
>
> Words like orphans lost among the crosstalk, meaning torn from truth's
> decrepit hands
>    - A23, Crosstalk
>

Re: [DISCUSS] A plan for 2.4

Posted by Andrew Purtell <ap...@apache.org>.

If nobody else wants to take this, I will.

Reid is co-RMing the pending 1.7 release with me. We've been going slow
with it waiting for backports but can move forward with it now that the
desired work has landed. I am not sensing any particular urgency with 1.x
releasing. Anyway, doing 2.4 as well would be only incremental effort.



On Tue, Oct 6, 2020 at 3:38 PM Nick Dimiduk <nd...@apache.org> wrote:

> Bump.
>
> We still need a volunteer for a release manager for 2.4. If we delay much
> longer, we'll need another discussion, which is whether we still want to
> target the 6-month cadence on minor releases.
>
> In terms of outstanding work I'm hoping to see in 2.4, I'd like to get a
> number of these normalizer improvements in [0], or determine that they're
> not a problem. I think there's work on meta region replicas that's also
> targeting this release, but I don't have a JIRA handy. Are there
> other features folks are actively working on with 2.4 in mind? Please keep
> in mind that there's really only about 4 weeks worth of dev time left
> before we'll need a feature freeze for stabilizing and the first RC.
>
> Thanks,
> Nick
>
> [0]:
>
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20HBASE%20AND%20resolution%20%3D%20Unresolved%20AND%20component%20%3D%20Normalizer%20ORDER%20BY%20issuetype%20DESC
>
> On Tue, Aug 25, 2020 at 10:36 AM Nick Dimiduk <nd...@apache.org> wrote:
>
> > Heya!
> >
> > Now that 2.3.x is in maintenance mode, I'd like to start a discussion on
> > 2.4. We've long agreed that a more frequent cadence of minor releases is
> > superior to a long gap between them. I think we should target a 2.4.0
> > release sometime before the end of the calendar year, perhaps the first
> > weeks of December.
> >
> > Hopefully the workload for the next release manager isn't as high as it
> > was for 2.3, given that much of the test stabilization effort happened on
> > branch-2. However, we are running in a new jenkins environment, and
> entropy
> > is nothing if not persistent, which puts that hard-won stability back
> into
> > doubt. Thus, I think it would be good if we can start sooner than later.
> >
> > Would anyone like to volunteer to run the 2.4 release line? The level of
> > effort is admittedly non-negligible: (1) stabilize branch-2 (2) test
> > new-for-branch-2 features (3) run initial release (4) cut monthly patch
> > releases until 2.5 is out. I have a strong interest in seeing this
> release
> > through, so I volunteer to assist with these activities.
> >
> > Thanks,
> > Nick
> >
>


-- 
Best regards,
Andrew

Words like orphans lost among the crosstalk, meaning torn from truth's
decrepit hands
   - A23, Crosstalk

Re: [DISCUSS] A plan for 2.4

Posted by Nick Dimiduk <nd...@apache.org>.

Bump.

We still need a volunteer for a release manager for 2.4. If we delay much
longer, we'll need another discussion, which is whether we still want to
target the 6-month cadence on minor releases.

In terms of outstanding work I'm hoping to see in 2.4, I'd like to get a
number of these normalizer improvements in [0], or determine that they're
not a problem. I think there's work on meta region replicas that's also
targeting this release, but I don't have a JIRA handy. Are there
other features folks are actively working on with 2.4 in mind? Please keep
in mind that there's really only about 4 weeks worth of dev time left
before we'll need a feature freeze for stabilizing and the first RC.

Thanks,
Nick

[0]:
https://issues.apache.org/jira/issues/?jql=project%20%3D%20HBASE%20AND%20resolution%20%3D%20Unresolved%20AND%20component%20%3D%20Normalizer%20ORDER%20BY%20issuetype%20DESC

On Tue, Aug 25, 2020 at 10:36 AM Nick Dimiduk <nd...@apache.org> wrote:

> Heya!
>
> Now that 2.3.x is in maintenance mode, I'd like to start a discussion on
> 2.4. We've long agreed that a more frequent cadence of minor releases is
> superior to a long gap between them. I think we should target a 2.4.0
> release sometime before the end of the calendar year, perhaps the first
> weeks of December.
>
> Hopefully the workload for the next release manager isn't as high as it
> was for 2.3, given that much of the test stabilization effort happened on
> branch-2. However, we are running in a new jenkins environment, and entropy
> is nothing if not persistent, which puts that hard-won stability back into
> doubt. Thus, I think it would be good if we can start sooner than later.
>
> Would anyone like to volunteer to run the 2.4 release line? The level of
> effort is admittedly non-negligible: (1) stabilize branch-2 (2) test
> new-for-branch-2 features (3) run initial release (4) cut monthly patch
> releases until 2.5 is out. I have a strong interest in seeing this release
> through, so I volunteer to assist with these activities.
>
> Thanks,
> Nick
>