You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@hbase.apache.org by Stack <st...@duboce.net> on 2012/09/14 21:17:57 UTC
DISCUSSION: Component Lieutenants?
At the contributor's pow wow a few days ago [1], during a discussion
about whether or not commits should have more friction applied -- i.e.
have more review before they go in -- it was thought that we might
benefit if we had "lieutenants" over-seeing individual HBase
components. A lieutenant would be someone who has an interest and an
understanding of how a particular component works (or should work). A
lieutenant does not need to be a committer. Before committing a patch
that touched on a particular component, the patch would have to have
been +1'd by the component lieutenant before it could go in (or if the
lieutenant is MIA, it was suggested by the Mighty Jon Hsieh that two
+1s by other contributors/committers would do instead; this latter
rule would probably also apply when a patch spanned components).
We already have a few folks signed up, knowingly or otherwise, as
component owners [1].
What do folks think?
Should we go ahead w/ this project? If so, any volunteers (I signed
up a few of the obvious component leads)? I can add you as component
lieutenant into JIRA. We can add more components if you don't see
your interest listed.
St.Ack
1. http://www.meetup.com/hbaseusergroup/events/80621872/
RE: DISCUSSION: Component Lieutenants?
Posted by "Ramkrishna.S.Vasudevan" <ra...@huawei.com>.
I can volunteer for Assignments( though the trunk code I need some more
hands on),
Split regions, HLog replay.
Regards
Ram
> -----Original Message-----
> From: Ted Yu [mailto:yuzhihong@gmail.com]
> Sent: Tuesday, September 18, 2012 9:36 AM
> To: dev@hbase.apache.org; lars hofhansl
> Subject: Re: DISCUSSION: Component Lieutenants?
>
> I volunteer for snapshots and WAL components.
>
> Thanks
>
> On Mon, Sep 17, 2012 at 4:13 PM, lars hofhansl <lh...@yahoo.com>
> wrote:
>
> > Maybe just make it an informal list of (self declared :) )
> "specialists".
> > For example if I see changes in the Assignment code that I do not
> > understand I usually defer to Ram. If there's some HFile stuff, I
> defer to
> > Mikhail...
> >
> > If we had a list of specialists, it would be easier to defer to them,
> or
> > to pull them into a review. I think that would be better than strict
> > guidelines.
> >
> >
> > I'd volunteer for: Transactions/MVCC, Scanners/Scanning/QueryMatcher,
> > Client, Deletion, Performance.
> >
> >
> >
> > ________________________________
> > From: Andrew Purtell <an...@gmail.com>
> > To: "dev@hbase.apache.org" <de...@hbase.apache.org>
> > Cc: "dev@hbase.apache.org" <de...@hbase.apache.org>; lars hofhansl <
> > lhofhansl@yahoo.com>
> > Sent: Monday, September 17, 2012 3:08 PM
> > Subject: Re: DISCUSSION: Component Lieutenants?
> >
> > Why doesn't every committer or contributor with interest volunteer?
> Some
> > overlap there would be good. Beyond that we can list the remaining
> areas
> > without good coverage and nominate for them?
> >
> > I volunteer for Coprocessors, REST, security, filters, and client.
> >
> > On Sep 17, 2012, at 2:12 PM, Todd Lipcon <to...@cloudera.com> wrote:
> >
> > > On Fri, Sep 14, 2012 at 9:15 PM, lars hofhansl
> <lh...@yahoo.com>
> > wrote:
> > >> I like that idea.
> > >>
> > >> Should all PMC members or committers be at top level of the source
> > tree? Or will that just take us back to the status-quo?
> > >>
> > >
> > > I feel like that would take us back to the status quo.
> > >
> > > The downside of this proposal is that we should probably have some
> > > well-principled way of determining who gets "ownership" (whether
> > > co-ownership or alone) of each part of the heirarchy. I fear it
> could
> > > become political or discourage people from contributing or
> reviewing
> > > code outside their area of expertise. So, if people have good ideas
> on
> > > how to go about doing this, please shout them out!
> > >
> > >>
> > >> I certainly like that a typical patch then will involve multiple
> > reviewer, and it will be more defined who should look at what patch.
> > >>
> > >> -- Lars
> > >>
> > >>
> > >> ----- Original Message -----
> > >> From: Todd Lipcon <to...@cloudera.com>
> > >> To: dev@hbase.apache.org
> > >> Cc:
> > >> Sent: Friday, September 14, 2012 1:15 PM
> > >> Subject: Re: DISCUSSION: Component Lieutenants?
> > >>
> > >> I like the idea of lieutenants, but another option would be a
> > >> "multi-lieutenant" model.
> > >>
> > >> The model used at google is that each directory has a file called
> > >> "OWNERS" which lists several usernames, one per line.
> > >>
> > >> For any given patch, you are expected to get a review such that,
> for
> > >> each modified file, one of the OWNERS listed in that directory (or
> any
> > >> parent thereof) has +1ed.
> > >>
> > >> So, for example, imagine that hbase/OWNERS has only Stack, and
> > >> hbase/foo/component1/OWNERS has "jxiang,larsh". If I make a patch
> > >> which touches something in foo/component1/bar/, I'd need a review
> from
> > >> at least one of Jimmy, Lars, or Stack.
> > >>
> > >> The assumption is that you try to get review from the most
> specific
> > >> owner, but if those people are MIA, you get review from someone
> higher
> > >> up the stack. The multi-person-per-dir model also ensures that, if
> > >> someone's on vacation or otherwise busy, we don't get blocked. And
> it
> > >> formalizes in the actual source tree who you should probably email
> if
> > >> you have questions about an area.
> > >>
> > >> It also means that wide-ranging patches that touch multiple
> components
> > >> need a lot of reviewers (or someone higher up the chain of command
> who
> > >> has "permission" on the whole tree). So if I had a mondo patch
> that
> > >> touched the region server, the master, and the IPC layer, I'd
> probably
> > >> need at least three separate people to sign off.
> > >>
> > >> Whatever we do, rather than making it a strict policy, let's start
> out
> > >> with a soft touch. Perhaps declare the component maintainers and
> try
> > >> to pick reviewers based on the criteria. But if people are busy
> and
> > >> work needs to get done, we don't need to be anal about it :)
> > >>
> > >> -Todd
> > >>
> > >> On Fri, Sep 14, 2012 at 12:17 PM, Stack <st...@duboce.net> wrote:
> > >>> At the contributor's pow wow a few days ago [1], during a
> discussion
> > >>> about whether or not commits should have more friction applied --
> i.e.
> > >>> have more review before they go in -- it was thought that we
> might
> > >>> benefit if we had "lieutenants" over-seeing individual HBase
> > >>> components. A lieutenant would be someone who has an interest
> and an
> > >>> understanding of how a particular component works (or should
> work). A
> > >>> lieutenant does not need to be a committer. Before committing a
> patch
> > >>> that touched on a particular component, the patch would have to
> have
> > >>> been +1'd by the component lieutenant before it could go in (or
> if the
> > >>> lieutenant is MIA, it was suggested by the Mighty Jon Hsieh that
> two
> > >>> +1s by other contributors/committers would do instead; this
> latter
> > >>> rule would probably also apply when a patch spanned components).
> > >>>
> > >>> We already have a few folks signed up, knowingly or otherwise, as
> > >>> component owners [1].
> > >>>
> > >>> What do folks think?
> > >>>
> > >>> Should we go ahead w/ this project? If so, any volunteers (I
> signed
> > >>> up a few of the obvious component leads)? I can add you as
> component
> > >>> lieutenant into JIRA. We can add more components if you don't
> see
> > >>> your interest listed.
> > >>>
> > >>> St.Ack
> > >>>
> > >>> 1. http://www.meetup.com/hbaseusergroup/events/80621872/
> > >>
> > >>
> > >>
> > >> --
> > >> Todd Lipcon
> > >> Software Engineer, Cloudera
> > >>
> > >
> > >
> > >
> > > --
> > > Todd Lipcon
> > > Software Engineer, Cloudera
> >
Re: DISCUSSION: Component Lieutenants?
Posted by Stack <st...@duboce.net>.
On Tue, Sep 18, 2012 at 2:40 AM, Anoop Sam John <an...@huawei.com> wrote:
>
> It will be my pleasure to be there for co processor, filters, scan flows and put& delete flows... [If you guys allow me :)]
>
Excellent. Keep volunteering and then I'll write up a list and post
it here for quick review and then I'll cast it into JIRA into the
component description section as Jon suggested.
As I see it, we're talking about a reviewers system that is flatter
than the google system Todd proposed suggested. I'm thinking that a
flatter structure is probably OK given the hbase source is probably
0.00000000001% of the google repository.
Regards who should review or guide at the x-component level, we could
have a fella w/ a chestful of medals and big billed cap or we could
set up a committee of cranky veterans -- Andrew Purtell, Myself, and
say Lars or Jon or... -- but my thinking is that we punt on this
'role' for now and see how this bit of policy that we've instituted
here works a while?
Good stuff,
St.Ack
RE: DISCUSSION: Component Lieutenants?
Posted by Anoop Sam John <an...@huawei.com>.
It will be my pleasure to be there for co processor, filters, scan flows and put& delete flows... [If you guys allow me :)]
-Anoop-
________________________________________
From: Ted Yu [yuzhihong@gmail.com]
Sent: Tuesday, September 18, 2012 9:35 AM
To: dev@hbase.apache.org; lars hofhansl
Subject: Re: DISCUSSION: Component Lieutenants?
I volunteer for snapshots and WAL components.
Thanks
On Mon, Sep 17, 2012 at 4:13 PM, lars hofhansl <lh...@yahoo.com> wrote:
> Maybe just make it an informal list of (self declared :) ) "specialists".
> For example if I see changes in the Assignment code that I do not
> understand I usually defer to Ram. If there's some HFile stuff, I defer to
> Mikhail...
>
> If we had a list of specialists, it would be easier to defer to them, or
> to pull them into a review. I think that would be better than strict
> guidelines.
>
>
> I'd volunteer for: Transactions/MVCC, Scanners/Scanning/QueryMatcher,
> Client, Deletion, Performance.
>
>
>
> ________________________________
> From: Andrew Purtell <an...@gmail.com>
> To: "dev@hbase.apache.org" <de...@hbase.apache.org>
> Cc: "dev@hbase.apache.org" <de...@hbase.apache.org>; lars hofhansl <
> lhofhansl@yahoo.com>
> Sent: Monday, September 17, 2012 3:08 PM
> Subject: Re: DISCUSSION: Component Lieutenants?
>
> Why doesn't every committer or contributor with interest volunteer? Some
> overlap there would be good. Beyond that we can list the remaining areas
> without good coverage and nominate for them?
>
> I volunteer for Coprocessors, REST, security, filters, and client.
>
> On Sep 17, 2012, at 2:12 PM, Todd Lipcon <to...@cloudera.com> wrote:
>
> > On Fri, Sep 14, 2012 at 9:15 PM, lars hofhansl <lh...@yahoo.com>
> wrote:
> >> I like that idea.
> >>
> >> Should all PMC members or committers be at top level of the source
> tree? Or will that just take us back to the status-quo?
> >>
> >
> > I feel like that would take us back to the status quo.
> >
> > The downside of this proposal is that we should probably have some
> > well-principled way of determining who gets "ownership" (whether
> > co-ownership or alone) of each part of the heirarchy. I fear it could
> > become political or discourage people from contributing or reviewing
> > code outside their area of expertise. So, if people have good ideas on
> > how to go about doing this, please shout them out!
> >
> >>
> >> I certainly like that a typical patch then will involve multiple
> reviewer, and it will be more defined who should look at what patch.
> >>
> >> -- Lars
> >>
> >>
> >> ----- Original Message -----
> >> From: Todd Lipcon <to...@cloudera.com>
> >> To: dev@hbase.apache.org
> >> Cc:
> >> Sent: Friday, September 14, 2012 1:15 PM
> >> Subject: Re: DISCUSSION: Component Lieutenants?
> >>
> >> I like the idea of lieutenants, but another option would be a
> >> "multi-lieutenant" model.
> >>
> >> The model used at google is that each directory has a file called
> >> "OWNERS" which lists several usernames, one per line.
> >>
> >> For any given patch, you are expected to get a review such that, for
> >> each modified file, one of the OWNERS listed in that directory (or any
> >> parent thereof) has +1ed.
> >>
> >> So, for example, imagine that hbase/OWNERS has only Stack, and
> >> hbase/foo/component1/OWNERS has "jxiang,larsh". If I make a patch
> >> which touches something in foo/component1/bar/, I'd need a review from
> >> at least one of Jimmy, Lars, or Stack.
> >>
> >> The assumption is that you try to get review from the most specific
> >> owner, but if those people are MIA, you get review from someone higher
> >> up the stack. The multi-person-per-dir model also ensures that, if
> >> someone's on vacation or otherwise busy, we don't get blocked. And it
> >> formalizes in the actual source tree who you should probably email if
> >> you have questions about an area.
> >>
> >> It also means that wide-ranging patches that touch multiple components
> >> need a lot of reviewers (or someone higher up the chain of command who
> >> has "permission" on the whole tree). So if I had a mondo patch that
> >> touched the region server, the master, and the IPC layer, I'd probably
> >> need at least three separate people to sign off.
> >>
> >> Whatever we do, rather than making it a strict policy, let's start out
> >> with a soft touch. Perhaps declare the component maintainers and try
> >> to pick reviewers based on the criteria. But if people are busy and
> >> work needs to get done, we don't need to be anal about it :)
> >>
> >> -Todd
> >>
> >> On Fri, Sep 14, 2012 at 12:17 PM, Stack <st...@duboce.net> wrote:
> >>> At the contributor's pow wow a few days ago [1], during a discussion
> >>> about whether or not commits should have more friction applied -- i.e.
> >>> have more review before they go in -- it was thought that we might
> >>> benefit if we had "lieutenants" over-seeing individual HBase
> >>> components. A lieutenant would be someone who has an interest and an
> >>> understanding of how a particular component works (or should work). A
> >>> lieutenant does not need to be a committer. Before committing a patch
> >>> that touched on a particular component, the patch would have to have
> >>> been +1'd by the component lieutenant before it could go in (or if the
> >>> lieutenant is MIA, it was suggested by the Mighty Jon Hsieh that two
> >>> +1s by other contributors/committers would do instead; this latter
> >>> rule would probably also apply when a patch spanned components).
> >>>
> >>> We already have a few folks signed up, knowingly or otherwise, as
> >>> component owners [1].
> >>>
> >>> What do folks think?
> >>>
> >>> Should we go ahead w/ this project? If so, any volunteers (I signed
> >>> up a few of the obvious component leads)? I can add you as component
> >>> lieutenant into JIRA. We can add more components if you don't see
> >>> your interest listed.
> >>>
> >>> St.Ack
> >>>
> >>> 1. http://www.meetup.com/hbaseusergroup/events/80621872/
> >>
> >>
> >>
> >> --
> >> Todd Lipcon
> >> Software Engineer, Cloudera
> >>
> >
> >
> >
> > --
> > Todd Lipcon
> > Software Engineer, Cloudera
>
RE: DISCUSSION: Component Lieutenants?
Posted by "Ramkrishna.S.Vasudevan" <ra...@huawei.com>.
I can add the HalfStoreFile related things that happens after split. (forgot
to add last time)
Regards
Ram
> -----Original Message-----
> From: Jonathan Hsieh [mailto:jon@cloudera.com]
> Sent: Tuesday, September 18, 2012 12:45 PM
> To: dev@hbase.apache.org
> Subject: Re: DISCUSSION: Component Lieutenants?
>
> For me, I'd say hbck, copytable, snapshots, and likely assignment soon.
>
> Jon.
>
> On Mon, Sep 17, 2012 at 9:22 PM, Ramkrishna.S.Vasudevan <
> ramkrishna.vasudevan@huawei.com> wrote:
>
> > I can volunteer for Assignments( though the trunk code I need some
> more
> > hands on),
> > Split regions, HLog replay.
> >
> > Regards
> > Ram
> >
> > > -----Original Message-----
> > > From: Ted Yu [mailto:yuzhihong@gmail.com]
> > > Sent: Tuesday, September 18, 2012 9:36 AM
> > > To: dev@hbase.apache.org; lars hofhansl
> > > Subject: Re: DISCUSSION: Component Lieutenants?
> > >
> > > I volunteer for snapshots and WAL components.
> > >
> > > Thanks
> > >
> > > On Mon, Sep 17, 2012 at 4:13 PM, lars hofhansl
> <lh...@yahoo.com>
> > > wrote:
> > >
> > > > Maybe just make it an informal list of (self declared :) )
> > > "specialists".
> > > > For example if I see changes in the Assignment code that I do not
> > > > understand I usually defer to Ram. If there's some HFile stuff, I
> > > defer to
> > > > Mikhail...
> > > >
> > > > If we had a list of specialists, it would be easier to defer to
> them,
> > > or
> > > > to pull them into a review. I think that would be better than
> strict
> > > > guidelines.
> > > >
> > > >
> > > > I'd volunteer for: Transactions/MVCC,
> Scanners/Scanning/QueryMatcher,
> > > > Client, Deletion, Performance.
> > > >
> > > >
> > > >
> > > > ________________________________
> > > > From: Andrew Purtell <an...@gmail.com>
> > > > To: "dev@hbase.apache.org" <de...@hbase.apache.org>
> > > > Cc: "dev@hbase.apache.org" <de...@hbase.apache.org>; lars hofhansl
> <
> > > > lhofhansl@yahoo.com>
> > > > Sent: Monday, September 17, 2012 3:08 PM
> > > > Subject: Re: DISCUSSION: Component Lieutenants?
> > > >
> > > > Why doesn't every committer or contributor with interest
> volunteer?
> > > Some
> > > > overlap there would be good. Beyond that we can list the
> remaining
> > > areas
> > > > without good coverage and nominate for them?
> > > >
> > > > I volunteer for Coprocessors, REST, security, filters, and
> client.
> > > >
> > > > On Sep 17, 2012, at 2:12 PM, Todd Lipcon <to...@cloudera.com>
> wrote:
> > > >
> > > > > On Fri, Sep 14, 2012 at 9:15 PM, lars hofhansl
> > > <lh...@yahoo.com>
> > > > wrote:
> > > > >> I like that idea.
> > > > >>
> > > > >> Should all PMC members or committers be at top level of the
> source
> > > > tree? Or will that just take us back to the status-quo?
> > > > >>
> > > > >
> > > > > I feel like that would take us back to the status quo.
> > > > >
> > > > > The downside of this proposal is that we should probably have
> some
> > > > > well-principled way of determining who gets "ownership"
> (whether
> > > > > co-ownership or alone) of each part of the heirarchy. I fear it
> > > could
> > > > > become political or discourage people from contributing or
> > > reviewing
> > > > > code outside their area of expertise. So, if people have good
> ideas
> > > on
> > > > > how to go about doing this, please shout them out!
> > > > >
> > > > >>
> > > > >> I certainly like that a typical patch then will involve
> multiple
> > > > reviewer, and it will be more defined who should look at what
> patch.
> > > > >>
> > > > >> -- Lars
> > > > >>
> > > > >>
> > > > >> ----- Original Message -----
> > > > >> From: Todd Lipcon <to...@cloudera.com>
> > > > >> To: dev@hbase.apache.org
> > > > >> Cc:
> > > > >> Sent: Friday, September 14, 2012 1:15 PM
> > > > >> Subject: Re: DISCUSSION: Component Lieutenants?
> > > > >>
> > > > >> I like the idea of lieutenants, but another option would be a
> > > > >> "multi-lieutenant" model.
> > > > >>
> > > > >> The model used at google is that each directory has a file
> called
> > > > >> "OWNERS" which lists several usernames, one per line.
> > > > >>
> > > > >> For any given patch, you are expected to get a review such
> that,
> > > for
> > > > >> each modified file, one of the OWNERS listed in that directory
> (or
> > > any
> > > > >> parent thereof) has +1ed.
> > > > >>
> > > > >> So, for example, imagine that hbase/OWNERS has only Stack, and
> > > > >> hbase/foo/component1/OWNERS has "jxiang,larsh". If I make a
> patch
> > > > >> which touches something in foo/component1/bar/, I'd need a
> review
> > > from
> > > > >> at least one of Jimmy, Lars, or Stack.
> > > > >>
> > > > >> The assumption is that you try to get review from the most
> > > specific
> > > > >> owner, but if those people are MIA, you get review from
> someone
> > > higher
> > > > >> up the stack. The multi-person-per-dir model also ensures
> that, if
> > > > >> someone's on vacation or otherwise busy, we don't get blocked.
> And
> > > it
> > > > >> formalizes in the actual source tree who you should probably
> email
> > > if
> > > > >> you have questions about an area.
> > > > >>
> > > > >> It also means that wide-ranging patches that touch multiple
> > > components
> > > > >> need a lot of reviewers (or someone higher up the chain of
> command
> > > who
> > > > >> has "permission" on the whole tree). So if I had a mondo patch
> > > that
> > > > >> touched the region server, the master, and the IPC layer, I'd
> > > probably
> > > > >> need at least three separate people to sign off.
> > > > >>
> > > > >> Whatever we do, rather than making it a strict policy, let's
> start
> > > out
> > > > >> with a soft touch. Perhaps declare the component maintainers
> and
> > > try
> > > > >> to pick reviewers based on the criteria. But if people are
> busy
> > > and
> > > > >> work needs to get done, we don't need to be anal about it :)
> > > > >>
> > > > >> -Todd
> > > > >>
> > > > >> On Fri, Sep 14, 2012 at 12:17 PM, Stack <st...@duboce.net>
> wrote:
> > > > >>> At the contributor's pow wow a few days ago [1], during a
> > > discussion
> > > > >>> about whether or not commits should have more friction
> applied --
> > > i.e.
> > > > >>> have more review before they go in -- it was thought that we
> > > might
> > > > >>> benefit if we had "lieutenants" over-seeing individual HBase
> > > > >>> components. A lieutenant would be someone who has an
> interest
> > > and an
> > > > >>> understanding of how a particular component works (or should
> > > work). A
> > > > >>> lieutenant does not need to be a committer. Before
> committing a
> > > patch
> > > > >>> that touched on a particular component, the patch would have
> to
> > > have
> > > > >>> been +1'd by the component lieutenant before it could go in
> (or
> > > if the
> > > > >>> lieutenant is MIA, it was suggested by the Mighty Jon Hsieh
> that
> > > two
> > > > >>> +1s by other contributors/committers would do instead; this
> > > latter
> > > > >>> rule would probably also apply when a patch spanned
> components).
> > > > >>>
> > > > >>> We already have a few folks signed up, knowingly or
> otherwise, as
> > > > >>> component owners [1].
> > > > >>>
> > > > >>> What do folks think?
> > > > >>>
> > > > >>> Should we go ahead w/ this project? If so, any volunteers (I
> > > signed
> > > > >>> up a few of the obvious component leads)? I can add you as
> > > component
> > > > >>> lieutenant into JIRA. We can add more components if you
> don't
> > > see
> > > > >>> your interest listed.
> > > > >>>
> > > > >>> St.Ack
> > > > >>>
> > > > >>> 1. http://www.meetup.com/hbaseusergroup/events/80621872/
> > > > >>
> > > > >>
> > > > >>
> > > > >> --
> > > > >> Todd Lipcon
> > > > >> Software Engineer, Cloudera
> > > > >>
> > > > >
> > > > >
> > > > >
> > > > > --
> > > > > Todd Lipcon
> > > > > Software Engineer, Cloudera
> > > >
> >
> >
>
>
> --
> // Jonathan Hsieh (shay)
> // Software Engineer, Cloudera
> // jon@cloudera.com
Re: DISCUSSION: Component Lieutenants?
Posted by Jonathan Hsieh <jo...@cloudera.com>.
For me, I'd say hbck, copytable, snapshots, and likely assignment soon.
Jon.
On Mon, Sep 17, 2012 at 9:22 PM, Ramkrishna.S.Vasudevan <
ramkrishna.vasudevan@huawei.com> wrote:
> I can volunteer for Assignments( though the trunk code I need some more
> hands on),
> Split regions, HLog replay.
>
> Regards
> Ram
>
> > -----Original Message-----
> > From: Ted Yu [mailto:yuzhihong@gmail.com]
> > Sent: Tuesday, September 18, 2012 9:36 AM
> > To: dev@hbase.apache.org; lars hofhansl
> > Subject: Re: DISCUSSION: Component Lieutenants?
> >
> > I volunteer for snapshots and WAL components.
> >
> > Thanks
> >
> > On Mon, Sep 17, 2012 at 4:13 PM, lars hofhansl <lh...@yahoo.com>
> > wrote:
> >
> > > Maybe just make it an informal list of (self declared :) )
> > "specialists".
> > > For example if I see changes in the Assignment code that I do not
> > > understand I usually defer to Ram. If there's some HFile stuff, I
> > defer to
> > > Mikhail...
> > >
> > > If we had a list of specialists, it would be easier to defer to them,
> > or
> > > to pull them into a review. I think that would be better than strict
> > > guidelines.
> > >
> > >
> > > I'd volunteer for: Transactions/MVCC, Scanners/Scanning/QueryMatcher,
> > > Client, Deletion, Performance.
> > >
> > >
> > >
> > > ________________________________
> > > From: Andrew Purtell <an...@gmail.com>
> > > To: "dev@hbase.apache.org" <de...@hbase.apache.org>
> > > Cc: "dev@hbase.apache.org" <de...@hbase.apache.org>; lars hofhansl <
> > > lhofhansl@yahoo.com>
> > > Sent: Monday, September 17, 2012 3:08 PM
> > > Subject: Re: DISCUSSION: Component Lieutenants?
> > >
> > > Why doesn't every committer or contributor with interest volunteer?
> > Some
> > > overlap there would be good. Beyond that we can list the remaining
> > areas
> > > without good coverage and nominate for them?
> > >
> > > I volunteer for Coprocessors, REST, security, filters, and client.
> > >
> > > On Sep 17, 2012, at 2:12 PM, Todd Lipcon <to...@cloudera.com> wrote:
> > >
> > > > On Fri, Sep 14, 2012 at 9:15 PM, lars hofhansl
> > <lh...@yahoo.com>
> > > wrote:
> > > >> I like that idea.
> > > >>
> > > >> Should all PMC members or committers be at top level of the source
> > > tree? Or will that just take us back to the status-quo?
> > > >>
> > > >
> > > > I feel like that would take us back to the status quo.
> > > >
> > > > The downside of this proposal is that we should probably have some
> > > > well-principled way of determining who gets "ownership" (whether
> > > > co-ownership or alone) of each part of the heirarchy. I fear it
> > could
> > > > become political or discourage people from contributing or
> > reviewing
> > > > code outside their area of expertise. So, if people have good ideas
> > on
> > > > how to go about doing this, please shout them out!
> > > >
> > > >>
> > > >> I certainly like that a typical patch then will involve multiple
> > > reviewer, and it will be more defined who should look at what patch.
> > > >>
> > > >> -- Lars
> > > >>
> > > >>
> > > >> ----- Original Message -----
> > > >> From: Todd Lipcon <to...@cloudera.com>
> > > >> To: dev@hbase.apache.org
> > > >> Cc:
> > > >> Sent: Friday, September 14, 2012 1:15 PM
> > > >> Subject: Re: DISCUSSION: Component Lieutenants?
> > > >>
> > > >> I like the idea of lieutenants, but another option would be a
> > > >> "multi-lieutenant" model.
> > > >>
> > > >> The model used at google is that each directory has a file called
> > > >> "OWNERS" which lists several usernames, one per line.
> > > >>
> > > >> For any given patch, you are expected to get a review such that,
> > for
> > > >> each modified file, one of the OWNERS listed in that directory (or
> > any
> > > >> parent thereof) has +1ed.
> > > >>
> > > >> So, for example, imagine that hbase/OWNERS has only Stack, and
> > > >> hbase/foo/component1/OWNERS has "jxiang,larsh". If I make a patch
> > > >> which touches something in foo/component1/bar/, I'd need a review
> > from
> > > >> at least one of Jimmy, Lars, or Stack.
> > > >>
> > > >> The assumption is that you try to get review from the most
> > specific
> > > >> owner, but if those people are MIA, you get review from someone
> > higher
> > > >> up the stack. The multi-person-per-dir model also ensures that, if
> > > >> someone's on vacation or otherwise busy, we don't get blocked. And
> > it
> > > >> formalizes in the actual source tree who you should probably email
> > if
> > > >> you have questions about an area.
> > > >>
> > > >> It also means that wide-ranging patches that touch multiple
> > components
> > > >> need a lot of reviewers (or someone higher up the chain of command
> > who
> > > >> has "permission" on the whole tree). So if I had a mondo patch
> > that
> > > >> touched the region server, the master, and the IPC layer, I'd
> > probably
> > > >> need at least three separate people to sign off.
> > > >>
> > > >> Whatever we do, rather than making it a strict policy, let's start
> > out
> > > >> with a soft touch. Perhaps declare the component maintainers and
> > try
> > > >> to pick reviewers based on the criteria. But if people are busy
> > and
> > > >> work needs to get done, we don't need to be anal about it :)
> > > >>
> > > >> -Todd
> > > >>
> > > >> On Fri, Sep 14, 2012 at 12:17 PM, Stack <st...@duboce.net> wrote:
> > > >>> At the contributor's pow wow a few days ago [1], during a
> > discussion
> > > >>> about whether or not commits should have more friction applied --
> > i.e.
> > > >>> have more review before they go in -- it was thought that we
> > might
> > > >>> benefit if we had "lieutenants" over-seeing individual HBase
> > > >>> components. A lieutenant would be someone who has an interest
> > and an
> > > >>> understanding of how a particular component works (or should
> > work). A
> > > >>> lieutenant does not need to be a committer. Before committing a
> > patch
> > > >>> that touched on a particular component, the patch would have to
> > have
> > > >>> been +1'd by the component lieutenant before it could go in (or
> > if the
> > > >>> lieutenant is MIA, it was suggested by the Mighty Jon Hsieh that
> > two
> > > >>> +1s by other contributors/committers would do instead; this
> > latter
> > > >>> rule would probably also apply when a patch spanned components).
> > > >>>
> > > >>> We already have a few folks signed up, knowingly or otherwise, as
> > > >>> component owners [1].
> > > >>>
> > > >>> What do folks think?
> > > >>>
> > > >>> Should we go ahead w/ this project? If so, any volunteers (I
> > signed
> > > >>> up a few of the obvious component leads)? I can add you as
> > component
> > > >>> lieutenant into JIRA. We can add more components if you don't
> > see
> > > >>> your interest listed.
> > > >>>
> > > >>> St.Ack
> > > >>>
> > > >>> 1. http://www.meetup.com/hbaseusergroup/events/80621872/
> > > >>
> > > >>
> > > >>
> > > >> --
> > > >> Todd Lipcon
> > > >> Software Engineer, Cloudera
> > > >>
> > > >
> > > >
> > > >
> > > > --
> > > > Todd Lipcon
> > > > Software Engineer, Cloudera
> > >
>
>
--
// Jonathan Hsieh (shay)
// Software Engineer, Cloudera
// jon@cloudera.com
Re: DISCUSSION: Component Lieutenants?
Posted by Ted Yu <yu...@gmail.com>.
I volunteer for snapshots and WAL components.
Thanks
On Mon, Sep 17, 2012 at 4:13 PM, lars hofhansl <lh...@yahoo.com> wrote:
> Maybe just make it an informal list of (self declared :) ) "specialists".
> For example if I see changes in the Assignment code that I do not
> understand I usually defer to Ram. If there's some HFile stuff, I defer to
> Mikhail...
>
> If we had a list of specialists, it would be easier to defer to them, or
> to pull them into a review. I think that would be better than strict
> guidelines.
>
>
> I'd volunteer for: Transactions/MVCC, Scanners/Scanning/QueryMatcher,
> Client, Deletion, Performance.
>
>
>
> ________________________________
> From: Andrew Purtell <an...@gmail.com>
> To: "dev@hbase.apache.org" <de...@hbase.apache.org>
> Cc: "dev@hbase.apache.org" <de...@hbase.apache.org>; lars hofhansl <
> lhofhansl@yahoo.com>
> Sent: Monday, September 17, 2012 3:08 PM
> Subject: Re: DISCUSSION: Component Lieutenants?
>
> Why doesn't every committer or contributor with interest volunteer? Some
> overlap there would be good. Beyond that we can list the remaining areas
> without good coverage and nominate for them?
>
> I volunteer for Coprocessors, REST, security, filters, and client.
>
> On Sep 17, 2012, at 2:12 PM, Todd Lipcon <to...@cloudera.com> wrote:
>
> > On Fri, Sep 14, 2012 at 9:15 PM, lars hofhansl <lh...@yahoo.com>
> wrote:
> >> I like that idea.
> >>
> >> Should all PMC members or committers be at top level of the source
> tree? Or will that just take us back to the status-quo?
> >>
> >
> > I feel like that would take us back to the status quo.
> >
> > The downside of this proposal is that we should probably have some
> > well-principled way of determining who gets "ownership" (whether
> > co-ownership or alone) of each part of the heirarchy. I fear it could
> > become political or discourage people from contributing or reviewing
> > code outside their area of expertise. So, if people have good ideas on
> > how to go about doing this, please shout them out!
> >
> >>
> >> I certainly like that a typical patch then will involve multiple
> reviewer, and it will be more defined who should look at what patch.
> >>
> >> -- Lars
> >>
> >>
> >> ----- Original Message -----
> >> From: Todd Lipcon <to...@cloudera.com>
> >> To: dev@hbase.apache.org
> >> Cc:
> >> Sent: Friday, September 14, 2012 1:15 PM
> >> Subject: Re: DISCUSSION: Component Lieutenants?
> >>
> >> I like the idea of lieutenants, but another option would be a
> >> "multi-lieutenant" model.
> >>
> >> The model used at google is that each directory has a file called
> >> "OWNERS" which lists several usernames, one per line.
> >>
> >> For any given patch, you are expected to get a review such that, for
> >> each modified file, one of the OWNERS listed in that directory (or any
> >> parent thereof) has +1ed.
> >>
> >> So, for example, imagine that hbase/OWNERS has only Stack, and
> >> hbase/foo/component1/OWNERS has "jxiang,larsh". If I make a patch
> >> which touches something in foo/component1/bar/, I'd need a review from
> >> at least one of Jimmy, Lars, or Stack.
> >>
> >> The assumption is that you try to get review from the most specific
> >> owner, but if those people are MIA, you get review from someone higher
> >> up the stack. The multi-person-per-dir model also ensures that, if
> >> someone's on vacation or otherwise busy, we don't get blocked. And it
> >> formalizes in the actual source tree who you should probably email if
> >> you have questions about an area.
> >>
> >> It also means that wide-ranging patches that touch multiple components
> >> need a lot of reviewers (or someone higher up the chain of command who
> >> has "permission" on the whole tree). So if I had a mondo patch that
> >> touched the region server, the master, and the IPC layer, I'd probably
> >> need at least three separate people to sign off.
> >>
> >> Whatever we do, rather than making it a strict policy, let's start out
> >> with a soft touch. Perhaps declare the component maintainers and try
> >> to pick reviewers based on the criteria. But if people are busy and
> >> work needs to get done, we don't need to be anal about it :)
> >>
> >> -Todd
> >>
> >> On Fri, Sep 14, 2012 at 12:17 PM, Stack <st...@duboce.net> wrote:
> >>> At the contributor's pow wow a few days ago [1], during a discussion
> >>> about whether or not commits should have more friction applied -- i.e.
> >>> have more review before they go in -- it was thought that we might
> >>> benefit if we had "lieutenants" over-seeing individual HBase
> >>> components. A lieutenant would be someone who has an interest and an
> >>> understanding of how a particular component works (or should work). A
> >>> lieutenant does not need to be a committer. Before committing a patch
> >>> that touched on a particular component, the patch would have to have
> >>> been +1'd by the component lieutenant before it could go in (or if the
> >>> lieutenant is MIA, it was suggested by the Mighty Jon Hsieh that two
> >>> +1s by other contributors/committers would do instead; this latter
> >>> rule would probably also apply when a patch spanned components).
> >>>
> >>> We already have a few folks signed up, knowingly or otherwise, as
> >>> component owners [1].
> >>>
> >>> What do folks think?
> >>>
> >>> Should we go ahead w/ this project? If so, any volunteers (I signed
> >>> up a few of the obvious component leads)? I can add you as component
> >>> lieutenant into JIRA. We can add more components if you don't see
> >>> your interest listed.
> >>>
> >>> St.Ack
> >>>
> >>> 1. http://www.meetup.com/hbaseusergroup/events/80621872/
> >>
> >>
> >>
> >> --
> >> Todd Lipcon
> >> Software Engineer, Cloudera
> >>
> >
> >
> >
> > --
> > Todd Lipcon
> > Software Engineer, Cloudera
>
Re: DISCUSSION: Component Lieutenants?
Posted by Jesse Yates <je...@gmail.com>.
I'll be up for maven/build (apparently I'm a masochist), snapshots, client and coprocessors.
- Jesse Yates
Sent from my iPhone.
On Sep 18, 2012, at 11:58 AM, Amandeep Khurana <am...@gmail.com> wrote:
> I'd like to volunteer for client, tools (copytable, export/import, etc and
> others that will come up in the future).
>
> On Tue, Sep 18, 2012 at 2:47 PM, lars hofhansl <lh...@yahoo.com> wrote:
>
>> I'd add WAL/HLog, Mutations (Put/Delete), Memstore, and Coprocessors to
>> what I'd volunteer for since I've been in that code a lot.
>>
>>
>>
>> ________________________________
>> From: lars hofhansl <lh...@yahoo.com>
>> To: "dev@hbase.apache.org" <de...@hbase.apache.org>
>> Sent: Monday, September 17, 2012 4:13 PM
>> Subject: Re: DISCUSSION: Component Lieutenants?
>>
>> Maybe just make it an informal list of (self declared :) ) "specialists".
>> For example if I see changes in the Assignment code that I do not
>> understand I usually defer to Ram. If there's some HFile stuff, I defer to
>> Mikhail...
>>
>> If we had a list of specialists, it would be easier to defer to them, or
>> to pull them into a review. I think that would be better than strict
>> guidelines.
>>
>>
>> I'd volunteer for: Transactions/MVCC, Scanners/Scanning/QueryMatcher,
>> Client, Deletion, Performance.
>>
>>
>>
>> ________________________________
>> From: Andrew Purtell <an...@gmail.com>
>> To: "dev@hbase.apache.org" <de...@hbase.apache.org>
>> Cc: "dev@hbase.apache.org" <de...@hbase.apache.org>; lars hofhansl <
>> lhofhansl@yahoo.com>
>> Sent: Monday, September 17, 2012 3:08 PM
>> Subject: Re: DISCUSSION: Component Lieutenants?
>>
>> Why doesn't every committer or contributor with interest volunteer? Some
>> overlap there would be good. Beyond that we can list the remaining areas
>> without good coverage and nominate for them?
>>
>> I volunteer for Coprocessors, REST, security, filters, and client.
>>
>> On Sep 17, 2012, at 2:12 PM, Todd Lipcon <to...@cloudera.com> wrote:
>>
>>> On Fri, Sep 14, 2012 at 9:15 PM, lars hofhansl <lh...@yahoo.com>
>> wrote:
>>>> I like that idea.
>>>>
>>>> Should all PMC members or committers be at top level of the source
>> tree? Or will that just take us back to the status-quo?
>>>>
>>>
>>> I feel like that would take us back to the status quo.
>>>
>>> The downside of this proposal is that we should probably have some
>>> well-principled way of determining who gets "ownership" (whether
>>> co-ownership or alone) of each part of the heirarchy. I fear it could
>>> become political or discourage people from contributing or reviewing
>>> code outside their area of expertise. So, if people have good ideas on
>>> how to go about doing this, please shout them out!
>>>
>>>>
>>>> I certainly like that a typical patch then will involve multiple
>> reviewer, and it will be more defined who should look at what patch.
>>>>
>>>> -- Lars
>>>>
>>>>
>>>> ----- Original Message -----
>>>> From: Todd Lipcon <to...@cloudera.com>
>>>> To: dev@hbase.apache.org
>>>> Cc:
>>>> Sent: Friday, September 14, 2012 1:15 PM
>>>> Subject: Re: DISCUSSION: Component Lieutenants?
>>>>
>>>> I like the idea of lieutenants, but another option would be a
>>>> "multi-lieutenant" model.
>>>>
>>>> The model used at google is that each directory has a file called
>>>> "OWNERS" which lists several usernames, one per line.
>>>>
>>>> For any given patch, you are expected to get a review such that, for
>>>> each modified file, one of the OWNERS listed in that directory (or any
>>>> parent thereof) has +1ed.
>>>>
>>>> So, for example, imagine that hbase/OWNERS has only Stack, and
>>>> hbase/foo/component1/OWNERS has "jxiang,larsh". If I make a patch
>>>> which touches something in foo/component1/bar/, I'd need a review from
>>>> at least one of Jimmy, Lars, or Stack.
>>>>
>>>> The assumption is that you try to get review from the most specific
>>>> owner, but if those people are MIA, you get review from someone higher
>>>> up the stack. The multi-person-per-dir model also ensures that, if
>>>> someone's on vacation or otherwise busy, we don't get blocked. And it
>>>> formalizes in the actual source tree who you should probably email if
>>>> you have questions about an area.
>>>>
>>>> It also means that wide-ranging patches that touch multiple components
>>>> need a lot of reviewers (or someone higher up the chain of command who
>>>> has "permission" on the whole tree). So if I had a mondo patch that
>>>> touched the region server, the master, and the IPC layer, I'd probably
>>>> need at least three separate people to sign off.
>>>>
>>>> Whatever we do, rather than making it a strict policy, let's start out
>>>> with a soft touch. Perhaps declare the component maintainers and try
>>>> to pick reviewers based on the criteria. But if people are busy and
>>>> work needs to get done, we don't need to be anal about it :)
>>>>
>>>> -Todd
>>>>
>>>> On Fri, Sep 14, 2012 at 12:17 PM, Stack <st...@duboce.net> wrote:
>>>>> At the contributor's pow wow a few days ago [1], during a discussion
>>>>> about whether or not commits should have more friction applied -- i.e.
>>>>> have more review before they go in -- it was thought that we might
>>>>> benefit if we had "lieutenants" over-seeing individual HBase
>>>>> components. A lieutenant would be someone who has an interest and an
>>>>> understanding of how a particular component works (or should work). A
>>>>> lieutenant does not need to be a committer. Before committing a patch
>>>>> that touched on a particular component, the patch would have to have
>>>>> been +1'd by the component lieutenant before it could go in (or if the
>>>>> lieutenant is MIA, it was suggested by the Mighty Jon Hsieh that two
>>>>> +1s by other contributors/committers would do instead; this latter
>>>>> rule would probably also apply when a patch spanned components).
>>>>>
>>>>> We already have a few folks signed up, knowingly or otherwise, as
>>>>> component owners [1].
>>>>>
>>>>> What do folks think?
>>>>>
>>>>> Should we go ahead w/ this project? If so, any volunteers (I signed
>>>>> up a few of the obvious component leads)? I can add you as component
>>>>> lieutenant into JIRA. We can add more components if you don't see
>>>>> your interest listed.
>>>>>
>>>>> St.Ack
>>>>>
>>>>> 1. http://www.meetup.com/hbaseusergroup/events/80621872/
>>>>
>>>>
>>>>
>>>> --
>>>> Todd Lipcon
>>>> Software Engineer, Cloudera
>>>>
>>>
>>>
>>>
>>> --
>>> Todd Lipcon
>>> Software Engineer, Cloudera
>>
Re: DISCUSSION: Component Lieutenants?
Posted by Himanshu Vashishtha <hv...@cs.ualberta.ca>.
+1 for trying this out.
I'd like to volunteer for io, replication, and coprocessors (ACL, security).
Himanshu
On Tue, Sep 18, 2012 at 12:58 PM, Amandeep Khurana <am...@gmail.com> wrote:
> I'd like to volunteer for client, tools (copytable, export/import, etc and
> others that will come up in the future).
>
> On Tue, Sep 18, 2012 at 2:47 PM, lars hofhansl <lh...@yahoo.com> wrote:
>
>> I'd add WAL/HLog, Mutations (Put/Delete), Memstore, and Coprocessors to
>> what I'd volunteer for since I've been in that code a lot.
>>
>>
>>
>> ________________________________
>> From: lars hofhansl <lh...@yahoo.com>
>> To: "dev@hbase.apache.org" <de...@hbase.apache.org>
>> Sent: Monday, September 17, 2012 4:13 PM
>> Subject: Re: DISCUSSION: Component Lieutenants?
>>
>> Maybe just make it an informal list of (self declared :) ) "specialists".
>> For example if I see changes in the Assignment code that I do not
>> understand I usually defer to Ram. If there's some HFile stuff, I defer to
>> Mikhail...
>>
>> If we had a list of specialists, it would be easier to defer to them, or
>> to pull them into a review. I think that would be better than strict
>> guidelines.
>>
>>
>> I'd volunteer for: Transactions/MVCC, Scanners/Scanning/QueryMatcher,
>> Client, Deletion, Performance.
>>
>>
>>
>> ________________________________
>> From: Andrew Purtell <an...@gmail.com>
>> To: "dev@hbase.apache.org" <de...@hbase.apache.org>
>> Cc: "dev@hbase.apache.org" <de...@hbase.apache.org>; lars hofhansl <
>> lhofhansl@yahoo.com>
>> Sent: Monday, September 17, 2012 3:08 PM
>> Subject: Re: DISCUSSION: Component Lieutenants?
>>
>> Why doesn't every committer or contributor with interest volunteer? Some
>> overlap there would be good. Beyond that we can list the remaining areas
>> without good coverage and nominate for them?
>>
>> I volunteer for Coprocessors, REST, security, filters, and client.
>>
>> On Sep 17, 2012, at 2:12 PM, Todd Lipcon <to...@cloudera.com> wrote:
>>
>> > On Fri, Sep 14, 2012 at 9:15 PM, lars hofhansl <lh...@yahoo.com>
>> wrote:
>> >> I like that idea.
>> >>
>> >> Should all PMC members or committers be at top level of the source
>> tree? Or will that just take us back to the status-quo?
>> >>
>> >
>> > I feel like that would take us back to the status quo.
>> >
>> > The downside of this proposal is that we should probably have some
>> > well-principled way of determining who gets "ownership" (whether
>> > co-ownership or alone) of each part of the heirarchy. I fear it could
>> > become political or discourage people from contributing or reviewing
>> > code outside their area of expertise. So, if people have good ideas on
>> > how to go about doing this, please shout them out!
>> >
>> >>
>> >> I certainly like that a typical patch then will involve multiple
>> reviewer, and it will be more defined who should look at what patch.
>> >>
>> >> -- Lars
>> >>
>> >>
>> >> ----- Original Message -----
>> >> From: Todd Lipcon <to...@cloudera.com>
>> >> To: dev@hbase.apache.org
>> >> Cc:
>> >> Sent: Friday, September 14, 2012 1:15 PM
>> >> Subject: Re: DISCUSSION: Component Lieutenants?
>> >>
>> >> I like the idea of lieutenants, but another option would be a
>> >> "multi-lieutenant" model.
>> >>
>> >> The model used at google is that each directory has a file called
>> >> "OWNERS" which lists several usernames, one per line.
>> >>
>> >> For any given patch, you are expected to get a review such that, for
>> >> each modified file, one of the OWNERS listed in that directory (or any
>> >> parent thereof) has +1ed.
>> >>
>> >> So, for example, imagine that hbase/OWNERS has only Stack, and
>> >> hbase/foo/component1/OWNERS has "jxiang,larsh". If I make a patch
>> >> which touches something in foo/component1/bar/, I'd need a review from
>> >> at least one of Jimmy, Lars, or Stack.
>> >>
>> >> The assumption is that you try to get review from the most specific
>> >> owner, but if those people are MIA, you get review from someone higher
>> >> up the stack. The multi-person-per-dir model also ensures that, if
>> >> someone's on vacation or otherwise busy, we don't get blocked. And it
>> >> formalizes in the actual source tree who you should probably email if
>> >> you have questions about an area.
>> >>
>> >> It also means that wide-ranging patches that touch multiple components
>> >> need a lot of reviewers (or someone higher up the chain of command who
>> >> has "permission" on the whole tree). So if I had a mondo patch that
>> >> touched the region server, the master, and the IPC layer, I'd probably
>> >> need at least three separate people to sign off.
>> >>
>> >> Whatever we do, rather than making it a strict policy, let's start out
>> >> with a soft touch. Perhaps declare the component maintainers and try
>> >> to pick reviewers based on the criteria. But if people are busy and
>> >> work needs to get done, we don't need to be anal about it :)
>> >>
>> >> -Todd
>> >>
>> >> On Fri, Sep 14, 2012 at 12:17 PM, Stack <st...@duboce.net> wrote:
>> >>> At the contributor's pow wow a few days ago [1], during a discussion
>> >>> about whether or not commits should have more friction applied -- i.e.
>> >>> have more review before they go in -- it was thought that we might
>> >>> benefit if we had "lieutenants" over-seeing individual HBase
>> >>> components. A lieutenant would be someone who has an interest and an
>> >>> understanding of how a particular component works (or should work). A
>> >>> lieutenant does not need to be a committer. Before committing a patch
>> >>> that touched on a particular component, the patch would have to have
>> >>> been +1'd by the component lieutenant before it could go in (or if the
>> >>> lieutenant is MIA, it was suggested by the Mighty Jon Hsieh that two
>> >>> +1s by other contributors/committers would do instead; this latter
>> >>> rule would probably also apply when a patch spanned components).
>> >>>
>> >>> We already have a few folks signed up, knowingly or otherwise, as
>> >>> component owners [1].
>> >>>
>> >>> What do folks think?
>> >>>
>> >>> Should we go ahead w/ this project? If so, any volunteers (I signed
>> >>> up a few of the obvious component leads)? I can add you as component
>> >>> lieutenant into JIRA. We can add more components if you don't see
>> >>> your interest listed.
>> >>>
>> >>> St.Ack
>> >>>
>> >>> 1. http://www.meetup.com/hbaseusergroup/events/80621872/
>> >>
>> >>
>> >>
>> >> --
>> >> Todd Lipcon
>> >> Software Engineer, Cloudera
>> >>
>> >
>> >
>> >
>> > --
>> > Todd Lipcon
>> > Software Engineer, Cloudera
>>
Re: DISCUSSION: Component Lieutenants?
Posted by Andrew Purtell <ap...@apache.org>.
+1
Vetos are vetos. We're not changing how votes are done, as far as I know,
just guiding contributors to helpful and knowledgeable folks, and
streamlining RTC a bit.
On Tuesday, September 18, 2012, Todd Lipcon wrote:
> -1s should always act as vetos
>
> -Todd
>
> On Tue, Sep 18, 2012 at 1:24 PM, Ted Yu <yu...@gmail.com> wrote:
> > What if we have a +1 from someone on the list and at the same time a -1
> > from someone off the list ?
> >
> > Cheers
> >
> > On Tue, Sep 18, 2012 at 12:47 PM, Andrew Purtell <apurtell@apache.org
> >wrote:
> >
> >> I agree. I think those listed on the component should be those to ping
> for
> >> a review but it's just that... With the expectation that such volunteers
> >> will do reviews as needed. And a +1 from any on the list will do, or two
> >> +1s from any committer.
> >>
> >> If we have too many volunteers in one area and not enough in another,
> let's
> >> allow Stack to spread some effort around.
> >>
> >> On Tuesday, September 18, 2012, Gregory Chanan wrote:
> >>
> >> > How are we deciding what counts as a component? Based on what people
> say
> >> > here? Some of these seem vastly different in scope (e.g. Client vs
> >> > HalfStoreFile).
> >> >
> >> > Also, will it be obvious, from the JIRA, who I need to get reviews
> from
> >> and
> >> > how many? From Stack's e-mail it sounds like clicking on the
> component
> >> > will give you a list of names; perhaps we should make it explicit in
> that
> >> > link that one +1 is enough for anyone on this list, otherwise two +1s
> are
> >> > necessary. We need to make it clear what the process is for new
> >> > contributors (more process is okay, but it needs to be fair and
> >> explicit).
> >> >
> >> > What about patches that touch more than one component?
> >> >
> >> > Greg
> >> >
> >> > On Tue, Sep 18, 2012 at 11:58 AM, Amandeep Khurana <am...@gmail.com>
> >> > wrote:
> >> >
> >> > > I'd like to volunteer for client, tools (copytable, export/import,
> etc
> >> > and
> >> > > others that will come up in the future).
> >> > >
> >> > > On Tue, Sep 18, 2012 at 2:47 PM, lars hofhansl <lhofhansl@yahoo.com
> >
> >> > > wrote:
> >> > >
> >> > > > I'd add WAL/HLog, Mutations (Put/Delete), Memstore, and
> Coprocessors
> >> to
> >> > > > what I'd volunteer for since I've been in that code a lot.
> >> > > >
> >> > > >
> >> > > >
> >> > > > ________________________________
> >> > > > From: lars hofhansl <lh...@yahoo.com>
> >> > > > To: "dev@hbase.apache.org" <de...@hbase.apache.org>
> >> > > > Sent: Monday, September 17, 2012 4:13 PM
> >> > > > Subject: Re: DISCUSSION: Component Lieutenants?
> >> > > >
> >> > > > Maybe just make it an informal list of (self declared :) )
> >> > "specialists".
> >> > > > For example if I see changes in the Assignment code that I do not
> >> > > > understand I usually defer to Ram. If there's some HFile stuff, I
> >> defer
> >> > > to
> >> > > > Mikhail...
> >> > > >
> >> > > > If we had a list of specialists, it would be easier to defer to
> them,
> >> > or
> >> > > > to pull them into a review. I think that would be better than
> strict
> >> > > > guidelines.
> >> > > >
> >> > > >
> >> > > > I'd volunteer for: Transactions/MVCC,
> Scanners/Scanning/QueryMatcher,
> >> > > > Client, Deletion, Performance.
> >> > > >
> >> > > >
> >> > > >
> >> > > > ________________________________
> >> > > > From: Andrew Purtell <an...@gmail.com>
> >> > > > To: "dev@hbase.apache.org" <--
> Todd Lipcon
> Software Engineer, Cloudera
>
--
Best regards,
- Andy
Problems worthy of attack prove their worth by hitting back. - Piet Hein
(via Tom White)
Re: DISCUSSION: Component Lieutenants?
Posted by Todd Lipcon <to...@cloudera.com>.
-1s should always act as vetos
-Todd
On Tue, Sep 18, 2012 at 1:24 PM, Ted Yu <yu...@gmail.com> wrote:
> What if we have a +1 from someone on the list and at the same time a -1
> from someone off the list ?
>
> Cheers
>
> On Tue, Sep 18, 2012 at 12:47 PM, Andrew Purtell <ap...@apache.org>wrote:
>
>> I agree. I think those listed on the component should be those to ping for
>> a review but it's just that... With the expectation that such volunteers
>> will do reviews as needed. And a +1 from any on the list will do, or two
>> +1s from any committer.
>>
>> If we have too many volunteers in one area and not enough in another, let's
>> allow Stack to spread some effort around.
>>
>> On Tuesday, September 18, 2012, Gregory Chanan wrote:
>>
>> > How are we deciding what counts as a component? Based on what people say
>> > here? Some of these seem vastly different in scope (e.g. Client vs
>> > HalfStoreFile).
>> >
>> > Also, will it be obvious, from the JIRA, who I need to get reviews from
>> and
>> > how many? From Stack's e-mail it sounds like clicking on the component
>> > will give you a list of names; perhaps we should make it explicit in that
>> > link that one +1 is enough for anyone on this list, otherwise two +1s are
>> > necessary. We need to make it clear what the process is for new
>> > contributors (more process is okay, but it needs to be fair and
>> explicit).
>> >
>> > What about patches that touch more than one component?
>> >
>> > Greg
>> >
>> > On Tue, Sep 18, 2012 at 11:58 AM, Amandeep Khurana <am...@gmail.com>
>> > wrote:
>> >
>> > > I'd like to volunteer for client, tools (copytable, export/import, etc
>> > and
>> > > others that will come up in the future).
>> > >
>> > > On Tue, Sep 18, 2012 at 2:47 PM, lars hofhansl <lh...@yahoo.com>
>> > > wrote:
>> > >
>> > > > I'd add WAL/HLog, Mutations (Put/Delete), Memstore, and Coprocessors
>> to
>> > > > what I'd volunteer for since I've been in that code a lot.
>> > > >
>> > > >
>> > > >
>> > > > ________________________________
>> > > > From: lars hofhansl <lh...@yahoo.com>
>> > > > To: "dev@hbase.apache.org" <de...@hbase.apache.org>
>> > > > Sent: Monday, September 17, 2012 4:13 PM
>> > > > Subject: Re: DISCUSSION: Component Lieutenants?
>> > > >
>> > > > Maybe just make it an informal list of (self declared :) )
>> > "specialists".
>> > > > For example if I see changes in the Assignment code that I do not
>> > > > understand I usually defer to Ram. If there's some HFile stuff, I
>> defer
>> > > to
>> > > > Mikhail...
>> > > >
>> > > > If we had a list of specialists, it would be easier to defer to them,
>> > or
>> > > > to pull them into a review. I think that would be better than strict
>> > > > guidelines.
>> > > >
>> > > >
>> > > > I'd volunteer for: Transactions/MVCC, Scanners/Scanning/QueryMatcher,
>> > > > Client, Deletion, Performance.
>> > > >
>> > > >
>> > > >
>> > > > ________________________________
>> > > > From: Andrew Purtell <an...@gmail.com>
>> > > > To: "dev@hbase.apache.org" <de...@hbase.apache.org>
>> > > > Cc: "dev@hbase.apache.org" <de...@hbase.apache.org>; lars hofhansl <
>> > > > lhofhansl@yahoo.com>
>> > > > Sent: Monday, September 17, 2012 3:08 PM
>> > > > Subject: Re: DISCUSSION: Component Lieutenants?
>> > > >
>> > > > Why doesn't every committer or contributor with interest volunteer?
>> > Some
>> > > > overlap there would be good. Beyond that we can list the remaining
>> > areas
>> > > > without good coverage and nominate for them?
>> > > >
>> > > > I volunteer for Coprocessors, REST, security, filters, and client.
>> > > >
>> > > > On Sep 17, 2012, at 2:12 PM, Todd Lipcon <to...@cloudera.com> wrote:
>> > > >
>> > > > > On Fri, Sep 14, 2012 at 9:15 PM, lars hofhansl <
>> lhofhansl@yahoo.com>
>> > > > wrote:
>> > > > >> I like that idea.
>> > > > >>
>> > > > >> Should all PMC members or committers be at top level of the source
>> > > > tree? Or will that just take us back to the status-quo?
>> > > > >>
>> > > > >
>> > > > > I feel like that would take us back to the status quo.
>> > > > >
>> > > > > The downside of this proposal is that we should probably have some
>> > > > > well-principled way of determining who gets "ownership" (whether
>> > > > > co-ownership or alone) of each part of the heirarchy. I fear it
>> could
>> > > > > become political or discourage people from contributing or
>> reviewing
>> > > > > code outside their area of expertise. So, if people have good ideas
>> > on
>> > > > > how to go about doing this, please shout them out!
>> > > > >
>> > > > >>
>> > > > >> I certainly like that a typical patch then will involve multiple
>> > > > reviewer, and it will be more defined who should look at what patch.
>> > > > >>
>> > > > >> -- Lars
>> > > > >>
>> > > > >>
>> > > > >> ----- Original Message -----
>> > > > >> From: Todd Lipcon <
>>
>>
>>
>> --
>> Best regards,
>>
>> - Andy
>>
>> Problems worthy of attack prove their worth by hitting back. - Piet Hein
>> (via Tom White)
>>
--
Todd Lipcon
Software Engineer, Cloudera
Re: DISCUSSION: Component Lieutenants?
Posted by Ted Yu <yu...@gmail.com>.
What if we have a +1 from someone on the list and at the same time a -1
from someone off the list ?
Cheers
On Tue, Sep 18, 2012 at 12:47 PM, Andrew Purtell <ap...@apache.org>wrote:
> I agree. I think those listed on the component should be those to ping for
> a review but it's just that... With the expectation that such volunteers
> will do reviews as needed. And a +1 from any on the list will do, or two
> +1s from any committer.
>
> If we have too many volunteers in one area and not enough in another, let's
> allow Stack to spread some effort around.
>
> On Tuesday, September 18, 2012, Gregory Chanan wrote:
>
> > How are we deciding what counts as a component? Based on what people say
> > here? Some of these seem vastly different in scope (e.g. Client vs
> > HalfStoreFile).
> >
> > Also, will it be obvious, from the JIRA, who I need to get reviews from
> and
> > how many? From Stack's e-mail it sounds like clicking on the component
> > will give you a list of names; perhaps we should make it explicit in that
> > link that one +1 is enough for anyone on this list, otherwise two +1s are
> > necessary. We need to make it clear what the process is for new
> > contributors (more process is okay, but it needs to be fair and
> explicit).
> >
> > What about patches that touch more than one component?
> >
> > Greg
> >
> > On Tue, Sep 18, 2012 at 11:58 AM, Amandeep Khurana <am...@gmail.com>
> > wrote:
> >
> > > I'd like to volunteer for client, tools (copytable, export/import, etc
> > and
> > > others that will come up in the future).
> > >
> > > On Tue, Sep 18, 2012 at 2:47 PM, lars hofhansl <lh...@yahoo.com>
> > > wrote:
> > >
> > > > I'd add WAL/HLog, Mutations (Put/Delete), Memstore, and Coprocessors
> to
> > > > what I'd volunteer for since I've been in that code a lot.
> > > >
> > > >
> > > >
> > > > ________________________________
> > > > From: lars hofhansl <lh...@yahoo.com>
> > > > To: "dev@hbase.apache.org" <de...@hbase.apache.org>
> > > > Sent: Monday, September 17, 2012 4:13 PM
> > > > Subject: Re: DISCUSSION: Component Lieutenants?
> > > >
> > > > Maybe just make it an informal list of (self declared :) )
> > "specialists".
> > > > For example if I see changes in the Assignment code that I do not
> > > > understand I usually defer to Ram. If there's some HFile stuff, I
> defer
> > > to
> > > > Mikhail...
> > > >
> > > > If we had a list of specialists, it would be easier to defer to them,
> > or
> > > > to pull them into a review. I think that would be better than strict
> > > > guidelines.
> > > >
> > > >
> > > > I'd volunteer for: Transactions/MVCC, Scanners/Scanning/QueryMatcher,
> > > > Client, Deletion, Performance.
> > > >
> > > >
> > > >
> > > > ________________________________
> > > > From: Andrew Purtell <an...@gmail.com>
> > > > To: "dev@hbase.apache.org" <de...@hbase.apache.org>
> > > > Cc: "dev@hbase.apache.org" <de...@hbase.apache.org>; lars hofhansl <
> > > > lhofhansl@yahoo.com>
> > > > Sent: Monday, September 17, 2012 3:08 PM
> > > > Subject: Re: DISCUSSION: Component Lieutenants?
> > > >
> > > > Why doesn't every committer or contributor with interest volunteer?
> > Some
> > > > overlap there would be good. Beyond that we can list the remaining
> > areas
> > > > without good coverage and nominate for them?
> > > >
> > > > I volunteer for Coprocessors, REST, security, filters, and client.
> > > >
> > > > On Sep 17, 2012, at 2:12 PM, Todd Lipcon <to...@cloudera.com> wrote:
> > > >
> > > > > On Fri, Sep 14, 2012 at 9:15 PM, lars hofhansl <
> lhofhansl@yahoo.com>
> > > > wrote:
> > > > >> I like that idea.
> > > > >>
> > > > >> Should all PMC members or committers be at top level of the source
> > > > tree? Or will that just take us back to the status-quo?
> > > > >>
> > > > >
> > > > > I feel like that would take us back to the status quo.
> > > > >
> > > > > The downside of this proposal is that we should probably have some
> > > > > well-principled way of determining who gets "ownership" (whether
> > > > > co-ownership or alone) of each part of the heirarchy. I fear it
> could
> > > > > become political or discourage people from contributing or
> reviewing
> > > > > code outside their area of expertise. So, if people have good ideas
> > on
> > > > > how to go about doing this, please shout them out!
> > > > >
> > > > >>
> > > > >> I certainly like that a typical patch then will involve multiple
> > > > reviewer, and it will be more defined who should look at what patch.
> > > > >>
> > > > >> -- Lars
> > > > >>
> > > > >>
> > > > >> ----- Original Message -----
> > > > >> From: Todd Lipcon <
>
>
>
> --
> Best regards,
>
> - Andy
>
> Problems worthy of attack prove their worth by hitting back. - Piet Hein
> (via Tom White)
>
Re: DISCUSSION: Component Lieutenants?
Posted by Andrew Purtell <ap...@apache.org>.
I agree. I think those listed on the component should be those to ping for
a review but it's just that... With the expectation that such volunteers
will do reviews as needed. And a +1 from any on the list will do, or two
+1s from any committer.
If we have too many volunteers in one area and not enough in another, let's
allow Stack to spread some effort around.
On Tuesday, September 18, 2012, Gregory Chanan wrote:
> How are we deciding what counts as a component? Based on what people say
> here? Some of these seem vastly different in scope (e.g. Client vs
> HalfStoreFile).
>
> Also, will it be obvious, from the JIRA, who I need to get reviews from and
> how many? From Stack's e-mail it sounds like clicking on the component
> will give you a list of names; perhaps we should make it explicit in that
> link that one +1 is enough for anyone on this list, otherwise two +1s are
> necessary. We need to make it clear what the process is for new
> contributors (more process is okay, but it needs to be fair and explicit).
>
> What about patches that touch more than one component?
>
> Greg
>
> On Tue, Sep 18, 2012 at 11:58 AM, Amandeep Khurana <am...@gmail.com>
> wrote:
>
> > I'd like to volunteer for client, tools (copytable, export/import, etc
> and
> > others that will come up in the future).
> >
> > On Tue, Sep 18, 2012 at 2:47 PM, lars hofhansl <lh...@yahoo.com>
> > wrote:
> >
> > > I'd add WAL/HLog, Mutations (Put/Delete), Memstore, and Coprocessors to
> > > what I'd volunteer for since I've been in that code a lot.
> > >
> > >
> > >
> > > ________________________________
> > > From: lars hofhansl <lh...@yahoo.com>
> > > To: "dev@hbase.apache.org" <de...@hbase.apache.org>
> > > Sent: Monday, September 17, 2012 4:13 PM
> > > Subject: Re: DISCUSSION: Component Lieutenants?
> > >
> > > Maybe just make it an informal list of (self declared :) )
> "specialists".
> > > For example if I see changes in the Assignment code that I do not
> > > understand I usually defer to Ram. If there's some HFile stuff, I defer
> > to
> > > Mikhail...
> > >
> > > If we had a list of specialists, it would be easier to defer to them,
> or
> > > to pull them into a review. I think that would be better than strict
> > > guidelines.
> > >
> > >
> > > I'd volunteer for: Transactions/MVCC, Scanners/Scanning/QueryMatcher,
> > > Client, Deletion, Performance.
> > >
> > >
> > >
> > > ________________________________
> > > From: Andrew Purtell <an...@gmail.com>
> > > To: "dev@hbase.apache.org" <de...@hbase.apache.org>
> > > Cc: "dev@hbase.apache.org" <de...@hbase.apache.org>; lars hofhansl <
> > > lhofhansl@yahoo.com>
> > > Sent: Monday, September 17, 2012 3:08 PM
> > > Subject: Re: DISCUSSION: Component Lieutenants?
> > >
> > > Why doesn't every committer or contributor with interest volunteer?
> Some
> > > overlap there would be good. Beyond that we can list the remaining
> areas
> > > without good coverage and nominate for them?
> > >
> > > I volunteer for Coprocessors, REST, security, filters, and client.
> > >
> > > On Sep 17, 2012, at 2:12 PM, Todd Lipcon <to...@cloudera.com> wrote:
> > >
> > > > On Fri, Sep 14, 2012 at 9:15 PM, lars hofhansl <lh...@yahoo.com>
> > > wrote:
> > > >> I like that idea.
> > > >>
> > > >> Should all PMC members or committers be at top level of the source
> > > tree? Or will that just take us back to the status-quo?
> > > >>
> > > >
> > > > I feel like that would take us back to the status quo.
> > > >
> > > > The downside of this proposal is that we should probably have some
> > > > well-principled way of determining who gets "ownership" (whether
> > > > co-ownership or alone) of each part of the heirarchy. I fear it could
> > > > become political or discourage people from contributing or reviewing
> > > > code outside their area of expertise. So, if people have good ideas
> on
> > > > how to go about doing this, please shout them out!
> > > >
> > > >>
> > > >> I certainly like that a typical patch then will involve multiple
> > > reviewer, and it will be more defined who should look at what patch.
> > > >>
> > > >> -- Lars
> > > >>
> > > >>
> > > >> ----- Original Message -----
> > > >> From: Todd Lipcon <
--
Best regards,
- Andy
Problems worthy of attack prove their worth by hitting back. - Piet Hein
(via Tom White)
Re: DISCUSSION: Component Lieutenants?
Posted by Stack <st...@duboce.net>.
How do we think this 'component owner' project (or 'component guide' or
'component shepherd' or 'captain' *) [1] is going? I see the definite
emergence of recognized domain experts consulted whenever a significant
patch touches their area (for example, Jimmy being consulted whenever
Assignment is touched, etc.). This is good. Lets have more of this
(smile).
We do seem to be accumulating a good backlog of patches in need of review.
What do we think we can do about that? If issues were assigned to a
component, would it be fair that component owners at least register the
submission perhaps prioritizing or even doing first a first pass on the
issue**?
St.Ack
P.S. Anyone can volunteer to take on a component -- just write me off list
and I'll update JIRA accordingly (as per [1]).
* I made note of this 'component owner' project of ours when I sent in our
last hbase report. Two folks objected that 'owner' is exclusionary.
Suggestions for alternate names welcome. I'll update the doc,
http://hbase.apache.org/book.html#OWNER, accordingly.
** I think we need to do a better job assigning priorities to issues and
working on higher priority issues first. Priority seems to mean little
going by what folks are working on, on cursory survey, so its tough getting
a sense of what is important just by looking at JIRA.
1. http://hbase.apache.org/book.html#OWNER
On Thu, Sep 20, 2012 at 3:42 PM, Stack <st...@duboce.net> wrote:
> I made a first pass on assign component owners:
>
> https://issues.apache.org/jira/plugins/servlet/project-config/HBASE/components
>
> I signed you all up for whatever you volunteered for. Let me know if
> you'd have me edit it otherwise.
>
> I removed a few components (mapred -- we have mapreduce) and added a
> few new ones; e.g. protobuf.
>
> St.Ack
>
Re: 答复: DISCUSSION: Component Lieutenants?
Posted by Stack <st...@duboce.net>.
On Sat, Sep 22, 2012 at 1:35 AM, ryan rawson <ry...@gmail.com> wrote:
> I'd like to contribute, we are learning some very interesting things here and I'd like to feed back as much as possible but I just can't guarantee a response time.
>
No worries if you don't have time; just contrib back whatever you are finding.
Good stuff,
St.Ack
Re: 答复: DISCUSSION: Component Lieutenants?
Posted by ryan rawson <ry...@gmail.com>.
I'd like to contribute, we are learning some very interesting things here and I'd like to feed back as much as possible but I just can't guarantee a response time.
Sent from your iPhone
On Sep 21, 2012, at 10:14 AM, Andrew Purtell <ap...@apache.org> wrote:
> On Fri, Sep 21, 2012 at 1:49 AM, Ryan Rawson <ry...@gmail.com> wrote:
>> This is a cool idea, I'd like to contribute, but I'll need coverage
>> since I cannot guarantee my time (since it doesnt belong to me
>> anyways).
>
> What do you need Ryan?
>
> --
> Best regards,
>
> - Andy
>
> Problems worthy of attack prove their worth by hitting back. - Piet
> Hein (via Tom White)
Re: 答复: DISCUSSION: Component Lieutenants?
Posted by Gregory Chanan <gc...@cloudera.com>.
Looks good.
I'll volunteer for protobufs and master.
Greg
On Fri, Sep 21, 2012 at 10:14 AM, Andrew Purtell <ap...@apache.org>wrote:
> On Fri, Sep 21, 2012 at 1:49 AM, Ryan Rawson <ry...@gmail.com> wrote:
> > This is a cool idea, I'd like to contribute, but I'll need coverage
> > since I cannot guarantee my time (since it doesnt belong to me
> > anyways).
>
> What do you need Ryan?
>
> --
> Best regards,
>
> - Andy
>
> Problems worthy of attack prove their worth by hitting back. - Piet
> Hein (via Tom White)
>
Re: 答复: DISCUSSION: Component Lieutenants?
Posted by Andrew Purtell <ap...@apache.org>.
On Fri, Sep 21, 2012 at 1:49 AM, Ryan Rawson <ry...@gmail.com> wrote:
> This is a cool idea, I'd like to contribute, but I'll need coverage
> since I cannot guarantee my time (since it doesnt belong to me
> anyways).
What do you need Ryan?
--
Best regards,
- Andy
Problems worthy of attack prove their worth by hitting back. - Piet
Hein (via Tom White)
Re: 答复: DISCUSSION: Component Lieutenants?
Posted by Ryan Rawson <ry...@gmail.com>.
This is a cool idea, I'd like to contribute, but I'll need coverage
since I cannot guarantee my time (since it doesnt belong to me
anyways).
Re: 答复: DISCUSSION: Component Lieutenants?
Posted by Stack <st...@duboce.net>.
On Thu, Sep 20, 2012 at 6:45 PM, 谢良 <xi...@xiaomi.com> wrote:
> I'd like to volunteer for thrift/zookeeper/performance components, if you guys trust me:)
>
I added you Liang. Thanks for volunteering.
St.Ack
答复: DISCUSSION: Component Lieutenants?
Posted by 谢良 <xi...@xiaomi.com>.
I'd like to volunteer for thrift/zookeeper/performance components, if you guys trust me:)
Thanks,
Liang
________________________________________
发件人: saint.ack@gmail.com [saint.ack@gmail.com] 代表 Stack [stack@duboce.net]
发送时间: 2012年9月21日 7:02
到: dev@hbase.apache.org
主题: Re: DISCUSSION: Component Lieutenants?
On Thu, Sep 20, 2012 at 3:59 PM, Elliott Clark <ec...@stumbleupon.com> wrote:
> If you want to put in a ui or loadbalancer components I would volunteer for
> those.
>
Added. Thanks Elliott.
St.Ack
Re: DISCUSSION: Component Lieutenants?
Posted by Stack <st...@duboce.net>.
On Thu, Sep 20, 2012 at 3:59 PM, Elliott Clark <ec...@stumbleupon.com> wrote:
> If you want to put in a ui or loadbalancer components I would volunteer for
> those.
>
Added. Thanks Elliott.
St.Ack
Re: DISCUSSION: Component Lieutenants?
Posted by Elliott Clark <ec...@stumbleupon.com>.
If you want to put in a ui or loadbalancer components I would volunteer for
those.
On Thu, Sep 20, 2012 at 3:45 PM, Andrew Purtell <ap...@apache.org> wrote:
> I took ownership of "hadoop2" as well. Bring them on!
>
> On Thu, Sep 20, 2012 at 3:42 PM, Stack <st...@duboce.net> wrote:
> > I made a first pass on assign component owners:
> >
> https://issues.apache.org/jira/plugins/servlet/project-config/HBASE/components
> >
> > I signed you all up for whatever you volunteered for. Let me know if
> > you'd have me edit it otherwise.
> >
> > I removed a few components (mapred -- we have mapreduce) and added a
> > few new ones; e.g. protobuf.
> >
> > St.Ack
>
>
>
> --
> Best regards,
>
> - Andy
>
> Problems worthy of attack prove their worth by hitting back. - Piet
> Hein (via Tom White)
>
Re: DISCUSSION: Component Lieutenants?
Posted by Andrew Purtell <ap...@apache.org>.
I took ownership of "hadoop2" as well. Bring them on!
On Thu, Sep 20, 2012 at 3:42 PM, Stack <st...@duboce.net> wrote:
> I made a first pass on assign component owners:
> https://issues.apache.org/jira/plugins/servlet/project-config/HBASE/components
>
> I signed you all up for whatever you volunteered for. Let me know if
> you'd have me edit it otherwise.
>
> I removed a few components (mapred -- we have mapreduce) and added a
> few new ones; e.g. protobuf.
>
> St.Ack
--
Best regards,
- Andy
Problems worthy of attack prove their worth by hitting back. - Piet
Hein (via Tom White)
Re: DISCUSSION: Component Lieutenants?
Posted by Elliott Clark <ec...@stumbleupon.com>.
https://issues.apache.org/jira/browse/HBASE#selectedTab=com.atlassian.jira.plugin.system.project%3Acomponents-panel
On Thu, Sep 20, 2012 at 4:28 PM, Amandeep Khurana <am...@gmail.com> wrote:
> Can't access the link. It says "You cannot edit the configuration of this
> project."
>
>
> http://www.evernote.com/shard/s5/sh/2c3ac4ed-5fd7-4fb2-b331-8138d670a820/8591216e8be37dcc86bbdede6b5b4cf2
>
> I need to be special?
>
> On Thu, Sep 20, 2012 at 3:42 PM, Stack <st...@duboce.net> wrote:
>
> > I made a first pass on assign component owners:
> >
> >
> https://issues.apache.org/jira/plugins/servlet/project-config/HBASE/components
> >
> > I signed you all up for whatever you volunteered for. Let me know if
> > you'd have me edit it otherwise.
> >
> > I removed a few components (mapred -- we have mapreduce) and added a
> > few new ones; e.g. protobuf.
> >
> > St.Ack
> >
>
Re: DISCUSSION: Component Lieutenants?
Posted by Amandeep Khurana <am...@gmail.com>.
Can't access the link. It says "You cannot edit the configuration of this
project."
http://www.evernote.com/shard/s5/sh/2c3ac4ed-5fd7-4fb2-b331-8138d670a820/8591216e8be37dcc86bbdede6b5b4cf2
I need to be special?
On Thu, Sep 20, 2012 at 3:42 PM, Stack <st...@duboce.net> wrote:
> I made a first pass on assign component owners:
>
> https://issues.apache.org/jira/plugins/servlet/project-config/HBASE/components
>
> I signed you all up for whatever you volunteered for. Let me know if
> you'd have me edit it otherwise.
>
> I removed a few components (mapred -- we have mapreduce) and added a
> few new ones; e.g. protobuf.
>
> St.Ack
>
Re: DISCUSSION: Component Lieutenants?
Posted by Stack <st...@duboce.net>.
I made a first pass on assign component owners:
https://issues.apache.org/jira/plugins/servlet/project-config/HBASE/components
I signed you all up for whatever you volunteered for. Let me know if
you'd have me edit it otherwise.
I removed a few components (mapred -- we have mapreduce) and added a
few new ones; e.g. protobuf.
St.Ack
Re: DISCUSSION: Component Lieutenants?
Posted by Todd Lipcon <to...@cloudera.com>.
Looks good to me.
Though I'm somewhat absent in the code base these days, I'll volunteer
for the nebulous areas of: "filesystem integration" and performance
-Todd
On Thu, Sep 20, 2012 at 1:39 PM, Stack <st...@duboce.net> wrote:
> Here is the paragraph I will add to the community chapter of our book
> on what I think we worked out by way of policy over the course of this
> discussion.
>
> In the below, I dropped the term 'lieutenant' and instead use 'owner'.
> Owner is from Todds' recounting of how google does it above. Owner,
> unlike lieutenant, undoes any implication of hierarchy.
>
>
> PATCH +1'ing POLICY[1]
>
> The below policy is something we put in place 09/2012. It is a
> suggested policy rather than a hard requirement. We want to try it
> first to see if it works before we cast it in stone.
>
> HBase is made of components[2]. Components have one or more
> owners[3]. See the 'Description' field on [1] for who the current
> owners are by component.
>
> Patches that fit within the scope of a single HBase component require,
> at least, a +1 by one of the component's owners before commit. If
> owners are absent -- busy or otherwise -- two +1s by non-owners will
> suffice.
>
> Patches that span components need at least two +1s before they can be
> committed, preferably +1s by owners of components touched by the
> x-component patch (TODO: This needs tightening up but I think fine for
> first pass).
>
> Any -1 on a patch by anyone vetos a patch; it cannot be committed
> until the justification for the -1 is addressed.
>
>
> 1. http://search-hadoop.com/m/jtoYcKNU7l1 and
> 2. https://issues.apache.org/jira/plugins/servlet/project-config/HBASE/components.
> 3. Link to Component Owners section (follows)
>
>
> COMPONENT OWNERS
>
> Component owners are listed in the description field on this page [1].
> They are listed in the description field rather than in the 'Component
> Lead' field because the latter only allows us list one individual
> whereas components are encouraged to have multiple owners.
>
> Owners are volunteers who are (usually, but not necessarily) expert in
> their component domain and may have an agenda on how they think their
> HBase component should evolve.
>
> Duties:
>
> + Owners will try and review patches that land within their component's scope.
> + If applicable, if an owner has an agenda, they will publish their
> goals or the design toward which they are driving their component
>
> If you would like to be volunteer as a component owner, just write the
> dev list and we'll sign you up. Owners do not need to be committers.
>
>
> Let me know if you'd add or edit above. Otherwise I'll add it to the
> refguide in next day or so.
>
> St.Ack
> P.S. I will edit the current list of components and add in those who
> have volunteered so far on this thread. Thanks.
--
Todd Lipcon
Software Engineer, Cloudera
Re: DISCUSSION: Component Lieutenants?
Posted by Stack <st...@duboce.net>.
Here is the paragraph I will add to the community chapter of our book
on what I think we worked out by way of policy over the course of this
discussion.
In the below, I dropped the term 'lieutenant' and instead use 'owner'.
Owner is from Todds' recounting of how google does it above. Owner,
unlike lieutenant, undoes any implication of hierarchy.
PATCH +1'ing POLICY[1]
The below policy is something we put in place 09/2012. It is a
suggested policy rather than a hard requirement. We want to try it
first to see if it works before we cast it in stone.
HBase is made of components[2]. Components have one or more
owners[3]. See the 'Description' field on [1] for who the current
owners are by component.
Patches that fit within the scope of a single HBase component require,
at least, a +1 by one of the component's owners before commit. If
owners are absent -- busy or otherwise -- two +1s by non-owners will
suffice.
Patches that span components need at least two +1s before they can be
committed, preferably +1s by owners of components touched by the
x-component patch (TODO: This needs tightening up but I think fine for
first pass).
Any -1 on a patch by anyone vetos a patch; it cannot be committed
until the justification for the -1 is addressed.
1. http://search-hadoop.com/m/jtoYcKNU7l1 and
2. https://issues.apache.org/jira/plugins/servlet/project-config/HBASE/components.
3. Link to Component Owners section (follows)
COMPONENT OWNERS
Component owners are listed in the description field on this page [1].
They are listed in the description field rather than in the 'Component
Lead' field because the latter only allows us list one individual
whereas components are encouraged to have multiple owners.
Owners are volunteers who are (usually, but not necessarily) expert in
their component domain and may have an agenda on how they think their
HBase component should evolve.
Duties:
+ Owners will try and review patches that land within their component's scope.
+ If applicable, if an owner has an agenda, they will publish their
goals or the design toward which they are driving their component
If you would like to be volunteer as a component owner, just write the
dev list and we'll sign you up. Owners do not need to be committers.
Let me know if you'd add or edit above. Otherwise I'll add it to the
refguide in next day or so.
St.Ack
P.S. I will edit the current list of components and add in those who
have volunteered so far on this thread. Thanks.
Re: DISCUSSION: Component Lieutenants?
Posted by 宾莉金 <bi...@gmail.com>.
I would like to volunteer for client, IPC
Thank you:)
--
*Best Regards,*
lijin bin
Re: DISCUSSION: Component Lieutenants?
Posted by Devaraj Das <dd...@hortonworks.com>.
I'd like to volunteer for IPC, PB, Security, and, Replication stuff.
On Tue, Sep 18, 2012 at 5:51 PM, Bijieshan <bi...@huawei.com> wrote:
> I would like to volunteer for client, Scan, Filter, Region Splitting, Create Table(Table related operations)...
> Thank you:)
>
> Jieshan
> -----Original Message-----
> From: saint.ack@gmail.com [mailto:saint.ack@gmail.com] On Behalf Of Stack
> Sent: Wednesday, September 19, 2012 6:15 AM
> To: dev@hbase.apache.org
> Subject: Re: DISCUSSION: Component Lieutenants?
>
> On Tue, Sep 18, 2012 at 12:37 PM, Gregory Chanan <gc...@cloudera.com> wrote:
>> How are we deciding what counts as a component? Based on what people say
>> here?
>
> Yes.
>
> I think that is ok. If the thing can be described and someone has
> volunteered to look after it, then it 'exists'.
>
>> Some of these seem vastly different in scope (e.g. Client vs
>> HalfStoreFile).
>>
>
> Agreed.
>
>> Also, will it be obvious, from the JIRA, who I need to get reviews from and
>> how many? From Stack's e-mail it sounds like clicking on the component
>> will give you a list of names; perhaps we should make it explicit in that
>> link that one +1 is enough for anyone on this list, otherwise two +1s are
>> necessary.
>
> Good suggestion. I'll do this.
>
>
>> We need to make it clear what the process is for new
>> contributors (more process is okay, but it needs to be fair and explicit).
>>
>
> Agreed.
>
> I'll add a section to our community chapter in refguide w/ the few
> rules we came up w/ here.
>
>> What about patches that touch more than one component?
>>
>
> Both component owners? Or a component owner and a +1 from someone
> other than author (That'd align w/ our two +1s from any contributor
> equating to a component owners' +1... sort of).
>
> St.Ack
RE: DISCUSSION: Component Lieutenants?
Posted by Bijieshan <bi...@huawei.com>.
I would like to volunteer for client, Scan, Filter, Region Splitting, Create Table(Table related operations)...
Thank you:)
Jieshan
-----Original Message-----
From: saint.ack@gmail.com [mailto:saint.ack@gmail.com] On Behalf Of Stack
Sent: Wednesday, September 19, 2012 6:15 AM
To: dev@hbase.apache.org
Subject: Re: DISCUSSION: Component Lieutenants?
On Tue, Sep 18, 2012 at 12:37 PM, Gregory Chanan <gc...@cloudera.com> wrote:
> How are we deciding what counts as a component? Based on what people say
> here?
Yes.
I think that is ok. If the thing can be described and someone has
volunteered to look after it, then it 'exists'.
> Some of these seem vastly different in scope (e.g. Client vs
> HalfStoreFile).
>
Agreed.
> Also, will it be obvious, from the JIRA, who I need to get reviews from and
> how many? From Stack's e-mail it sounds like clicking on the component
> will give you a list of names; perhaps we should make it explicit in that
> link that one +1 is enough for anyone on this list, otherwise two +1s are
> necessary.
Good suggestion. I'll do this.
> We need to make it clear what the process is for new
> contributors (more process is okay, but it needs to be fair and explicit).
>
Agreed.
I'll add a section to our community chapter in refguide w/ the few
rules we came up w/ here.
> What about patches that touch more than one component?
>
Both component owners? Or a component owner and a +1 from someone
other than author (That'd align w/ our two +1s from any contributor
equating to a component owners' +1... sort of).
St.Ack
Re: DISCUSSION: Component Lieutenants?
Posted by Stack <st...@duboce.net>.
On Tue, Sep 18, 2012 at 12:37 PM, Gregory Chanan <gc...@cloudera.com> wrote:
> How are we deciding what counts as a component? Based on what people say
> here?
Yes.
I think that is ok. If the thing can be described and someone has
volunteered to look after it, then it 'exists'.
> Some of these seem vastly different in scope (e.g. Client vs
> HalfStoreFile).
>
Agreed.
> Also, will it be obvious, from the JIRA, who I need to get reviews from and
> how many? From Stack's e-mail it sounds like clicking on the component
> will give you a list of names; perhaps we should make it explicit in that
> link that one +1 is enough for anyone on this list, otherwise two +1s are
> necessary.
Good suggestion. I'll do this.
> We need to make it clear what the process is for new
> contributors (more process is okay, but it needs to be fair and explicit).
>
Agreed.
I'll add a section to our community chapter in refguide w/ the few
rules we came up w/ here.
> What about patches that touch more than one component?
>
Both component owners? Or a component owner and a +1 from someone
other than author (That'd align w/ our two +1s from any contributor
equating to a component owners' +1... sort of).
St.Ack
Re: DISCUSSION: Component Lieutenants?
Posted by Gregory Chanan <gc...@cloudera.com>.
How are we deciding what counts as a component? Based on what people say
here? Some of these seem vastly different in scope (e.g. Client vs
HalfStoreFile).
Also, will it be obvious, from the JIRA, who I need to get reviews from and
how many? From Stack's e-mail it sounds like clicking on the component
will give you a list of names; perhaps we should make it explicit in that
link that one +1 is enough for anyone on this list, otherwise two +1s are
necessary. We need to make it clear what the process is for new
contributors (more process is okay, but it needs to be fair and explicit).
What about patches that touch more than one component?
Greg
On Tue, Sep 18, 2012 at 11:58 AM, Amandeep Khurana <am...@gmail.com> wrote:
> I'd like to volunteer for client, tools (copytable, export/import, etc and
> others that will come up in the future).
>
> On Tue, Sep 18, 2012 at 2:47 PM, lars hofhansl <lh...@yahoo.com>
> wrote:
>
> > I'd add WAL/HLog, Mutations (Put/Delete), Memstore, and Coprocessors to
> > what I'd volunteer for since I've been in that code a lot.
> >
> >
> >
> > ________________________________
> > From: lars hofhansl <lh...@yahoo.com>
> > To: "dev@hbase.apache.org" <de...@hbase.apache.org>
> > Sent: Monday, September 17, 2012 4:13 PM
> > Subject: Re: DISCUSSION: Component Lieutenants?
> >
> > Maybe just make it an informal list of (self declared :) ) "specialists".
> > For example if I see changes in the Assignment code that I do not
> > understand I usually defer to Ram. If there's some HFile stuff, I defer
> to
> > Mikhail...
> >
> > If we had a list of specialists, it would be easier to defer to them, or
> > to pull them into a review. I think that would be better than strict
> > guidelines.
> >
> >
> > I'd volunteer for: Transactions/MVCC, Scanners/Scanning/QueryMatcher,
> > Client, Deletion, Performance.
> >
> >
> >
> > ________________________________
> > From: Andrew Purtell <an...@gmail.com>
> > To: "dev@hbase.apache.org" <de...@hbase.apache.org>
> > Cc: "dev@hbase.apache.org" <de...@hbase.apache.org>; lars hofhansl <
> > lhofhansl@yahoo.com>
> > Sent: Monday, September 17, 2012 3:08 PM
> > Subject: Re: DISCUSSION: Component Lieutenants?
> >
> > Why doesn't every committer or contributor with interest volunteer? Some
> > overlap there would be good. Beyond that we can list the remaining areas
> > without good coverage and nominate for them?
> >
> > I volunteer for Coprocessors, REST, security, filters, and client.
> >
> > On Sep 17, 2012, at 2:12 PM, Todd Lipcon <to...@cloudera.com> wrote:
> >
> > > On Fri, Sep 14, 2012 at 9:15 PM, lars hofhansl <lh...@yahoo.com>
> > wrote:
> > >> I like that idea.
> > >>
> > >> Should all PMC members or committers be at top level of the source
> > tree? Or will that just take us back to the status-quo?
> > >>
> > >
> > > I feel like that would take us back to the status quo.
> > >
> > > The downside of this proposal is that we should probably have some
> > > well-principled way of determining who gets "ownership" (whether
> > > co-ownership or alone) of each part of the heirarchy. I fear it could
> > > become political or discourage people from contributing or reviewing
> > > code outside their area of expertise. So, if people have good ideas on
> > > how to go about doing this, please shout them out!
> > >
> > >>
> > >> I certainly like that a typical patch then will involve multiple
> > reviewer, and it will be more defined who should look at what patch.
> > >>
> > >> -- Lars
> > >>
> > >>
> > >> ----- Original Message -----
> > >> From: Todd Lipcon <to...@cloudera.com>
> > >> To: dev@hbase.apache.org
> > >> Cc:
> > >> Sent: Friday, September 14, 2012 1:15 PM
> > >> Subject: Re: DISCUSSION: Component Lieutenants?
> > >>
> > >> I like the idea of lieutenants, but another option would be a
> > >> "multi-lieutenant" model.
> > >>
> > >> The model used at google is that each directory has a file called
> > >> "OWNERS" which lists several usernames, one per line.
> > >>
> > >> For any given patch, you are expected to get a review such that, for
> > >> each modified file, one of the OWNERS listed in that directory (or any
> > >> parent thereof) has +1ed.
> > >>
> > >> So, for example, imagine that hbase/OWNERS has only Stack, and
> > >> hbase/foo/component1/OWNERS has "jxiang,larsh". If I make a patch
> > >> which touches something in foo/component1/bar/, I'd need a review from
> > >> at least one of Jimmy, Lars, or Stack.
> > >>
> > >> The assumption is that you try to get review from the most specific
> > >> owner, but if those people are MIA, you get review from someone higher
> > >> up the stack. The multi-person-per-dir model also ensures that, if
> > >> someone's on vacation or otherwise busy, we don't get blocked. And it
> > >> formalizes in the actual source tree who you should probably email if
> > >> you have questions about an area.
> > >>
> > >> It also means that wide-ranging patches that touch multiple components
> > >> need a lot of reviewers (or someone higher up the chain of command who
> > >> has "permission" on the whole tree). So if I had a mondo patch that
> > >> touched the region server, the master, and the IPC layer, I'd probably
> > >> need at least three separate people to sign off.
> > >>
> > >> Whatever we do, rather than making it a strict policy, let's start out
> > >> with a soft touch. Perhaps declare the component maintainers and try
> > >> to pick reviewers based on the criteria. But if people are busy and
> > >> work needs to get done, we don't need to be anal about it :)
> > >>
> > >> -Todd
> > >>
> > >> On Fri, Sep 14, 2012 at 12:17 PM, Stack <st...@duboce.net> wrote:
> > >>> At the contributor's pow wow a few days ago [1], during a discussion
> > >>> about whether or not commits should have more friction applied --
> i.e.
> > >>> have more review before they go in -- it was thought that we might
> > >>> benefit if we had "lieutenants" over-seeing individual HBase
> > >>> components. A lieutenant would be someone who has an interest and an
> > >>> understanding of how a particular component works (or should work).
> A
> > >>> lieutenant does not need to be a committer. Before committing a
> patch
> > >>> that touched on a particular component, the patch would have to have
> > >>> been +1'd by the component lieutenant before it could go in (or if
> the
> > >>> lieutenant is MIA, it was suggested by the Mighty Jon Hsieh that two
> > >>> +1s by other contributors/committers would do instead; this latter
> > >>> rule would probably also apply when a patch spanned components).
> > >>>
> > >>> We already have a few folks signed up, knowingly or otherwise, as
> > >>> component owners [1].
> > >>>
> > >>> What do folks think?
> > >>>
> > >>> Should we go ahead w/ this project? If so, any volunteers (I signed
> > >>> up a few of the obvious component leads)? I can add you as component
> > >>> lieutenant into JIRA. We can add more components if you don't see
> > >>> your interest listed.
> > >>>
> > >>> St.Ack
> > >>>
> > >>> 1. http://www.meetup.com/hbaseusergroup/events/80621872/
> > >>
> > >>
> > >>
> > >> --
> > >> Todd Lipcon
> > >> Software Engineer, Cloudera
> > >>
> > >
> > >
> > >
> > > --
> > > Todd Lipcon
> > > Software Engineer, Cloudera
> >
>
Re: DISCUSSION: Component Lieutenants?
Posted by Jimmy Xiang <jx...@cloudera.com>.
If it is ok, I'd like to volunteer for hbck, trunk version assignment,
distributed log splitting,
and pb related.
Thanks,
Jimmy
On Tue, Sep 18, 2012 at 11:58 AM, Amandeep Khurana <am...@gmail.com> wrote:
> I'd like to volunteer for client, tools (copytable, export/import, etc and
> others that will come up in the future).
>
> On Tue, Sep 18, 2012 at 2:47 PM, lars hofhansl <lh...@yahoo.com> wrote:
>
>> I'd add WAL/HLog, Mutations (Put/Delete), Memstore, and Coprocessors to
>> what I'd volunteer for since I've been in that code a lot.
>>
>>
>>
>> ________________________________
>> From: lars hofhansl <lh...@yahoo.com>
>> To: "dev@hbase.apache.org" <de...@hbase.apache.org>
>> Sent: Monday, September 17, 2012 4:13 PM
>> Subject: Re: DISCUSSION: Component Lieutenants?
>>
>> Maybe just make it an informal list of (self declared :) ) "specialists".
>> For example if I see changes in the Assignment code that I do not
>> understand I usually defer to Ram. If there's some HFile stuff, I defer to
>> Mikhail...
>>
>> If we had a list of specialists, it would be easier to defer to them, or
>> to pull them into a review. I think that would be better than strict
>> guidelines.
>>
>>
>> I'd volunteer for: Transactions/MVCC, Scanners/Scanning/QueryMatcher,
>> Client, Deletion, Performance.
>>
>>
>>
>> ________________________________
>> From: Andrew Purtell <an...@gmail.com>
>> To: "dev@hbase.apache.org" <de...@hbase.apache.org>
>> Cc: "dev@hbase.apache.org" <de...@hbase.apache.org>; lars hofhansl <
>> lhofhansl@yahoo.com>
>> Sent: Monday, September 17, 2012 3:08 PM
>> Subject: Re: DISCUSSION: Component Lieutenants?
>>
>> Why doesn't every committer or contributor with interest volunteer? Some
>> overlap there would be good. Beyond that we can list the remaining areas
>> without good coverage and nominate for them?
>>
>> I volunteer for Coprocessors, REST, security, filters, and client.
>>
>> On Sep 17, 2012, at 2:12 PM, Todd Lipcon <to...@cloudera.com> wrote:
>>
>> > On Fri, Sep 14, 2012 at 9:15 PM, lars hofhansl <lh...@yahoo.com>
>> wrote:
>> >> I like that idea.
>> >>
>> >> Should all PMC members or committers be at top level of the source
>> tree? Or will that just take us back to the status-quo?
>> >>
>> >
>> > I feel like that would take us back to the status quo.
>> >
>> > The downside of this proposal is that we should probably have some
>> > well-principled way of determining who gets "ownership" (whether
>> > co-ownership or alone) of each part of the heirarchy. I fear it could
>> > become political or discourage people from contributing or reviewing
>> > code outside their area of expertise. So, if people have good ideas on
>> > how to go about doing this, please shout them out!
>> >
>> >>
>> >> I certainly like that a typical patch then will involve multiple
>> reviewer, and it will be more defined who should look at what patch.
>> >>
>> >> -- Lars
>> >>
>> >>
>> >> ----- Original Message -----
>> >> From: Todd Lipcon <to...@cloudera.com>
>> >> To: dev@hbase.apache.org
>> >> Cc:
>> >> Sent: Friday, September 14, 2012 1:15 PM
>> >> Subject: Re: DISCUSSION: Component Lieutenants?
>> >>
>> >> I like the idea of lieutenants, but another option would be a
>> >> "multi-lieutenant" model.
>> >>
>> >> The model used at google is that each directory has a file called
>> >> "OWNERS" which lists several usernames, one per line.
>> >>
>> >> For any given patch, you are expected to get a review such that, for
>> >> each modified file, one of the OWNERS listed in that directory (or any
>> >> parent thereof) has +1ed.
>> >>
>> >> So, for example, imagine that hbase/OWNERS has only Stack, and
>> >> hbase/foo/component1/OWNERS has "jxiang,larsh". If I make a patch
>> >> which touches something in foo/component1/bar/, I'd need a review from
>> >> at least one of Jimmy, Lars, or Stack.
>> >>
>> >> The assumption is that you try to get review from the most specific
>> >> owner, but if those people are MIA, you get review from someone higher
>> >> up the stack. The multi-person-per-dir model also ensures that, if
>> >> someone's on vacation or otherwise busy, we don't get blocked. And it
>> >> formalizes in the actual source tree who you should probably email if
>> >> you have questions about an area.
>> >>
>> >> It also means that wide-ranging patches that touch multiple components
>> >> need a lot of reviewers (or someone higher up the chain of command who
>> >> has "permission" on the whole tree). So if I had a mondo patch that
>> >> touched the region server, the master, and the IPC layer, I'd probably
>> >> need at least three separate people to sign off.
>> >>
>> >> Whatever we do, rather than making it a strict policy, let's start out
>> >> with a soft touch. Perhaps declare the component maintainers and try
>> >> to pick reviewers based on the criteria. But if people are busy and
>> >> work needs to get done, we don't need to be anal about it :)
>> >>
>> >> -Todd
>> >>
>> >> On Fri, Sep 14, 2012 at 12:17 PM, Stack <st...@duboce.net> wrote:
>> >>> At the contributor's pow wow a few days ago [1], during a discussion
>> >>> about whether or not commits should have more friction applied -- i.e.
>> >>> have more review before they go in -- it was thought that we might
>> >>> benefit if we had "lieutenants" over-seeing individual HBase
>> >>> components. A lieutenant would be someone who has an interest and an
>> >>> understanding of how a particular component works (or should work). A
>> >>> lieutenant does not need to be a committer. Before committing a patch
>> >>> that touched on a particular component, the patch would have to have
>> >>> been +1'd by the component lieutenant before it could go in (or if the
>> >>> lieutenant is MIA, it was suggested by the Mighty Jon Hsieh that two
>> >>> +1s by other contributors/committers would do instead; this latter
>> >>> rule would probably also apply when a patch spanned components).
>> >>>
>> >>> We already have a few folks signed up, knowingly or otherwise, as
>> >>> component owners [1].
>> >>>
>> >>> What do folks think?
>> >>>
>> >>> Should we go ahead w/ this project? If so, any volunteers (I signed
>> >>> up a few of the obvious component leads)? I can add you as component
>> >>> lieutenant into JIRA. We can add more components if you don't see
>> >>> your interest listed.
>> >>>
>> >>> St.Ack
>> >>>
>> >>> 1. http://www.meetup.com/hbaseusergroup/events/80621872/
>> >>
>> >>
>> >>
>> >> --
>> >> Todd Lipcon
>> >> Software Engineer, Cloudera
>> >>
>> >
>> >
>> >
>> > --
>> > Todd Lipcon
>> > Software Engineer, Cloudera
>>
Re: DISCUSSION: Component Lieutenants?
Posted by Amandeep Khurana <am...@gmail.com>.
I'd like to volunteer for client, tools (copytable, export/import, etc and
others that will come up in the future).
On Tue, Sep 18, 2012 at 2:47 PM, lars hofhansl <lh...@yahoo.com> wrote:
> I'd add WAL/HLog, Mutations (Put/Delete), Memstore, and Coprocessors to
> what I'd volunteer for since I've been in that code a lot.
>
>
>
> ________________________________
> From: lars hofhansl <lh...@yahoo.com>
> To: "dev@hbase.apache.org" <de...@hbase.apache.org>
> Sent: Monday, September 17, 2012 4:13 PM
> Subject: Re: DISCUSSION: Component Lieutenants?
>
> Maybe just make it an informal list of (self declared :) ) "specialists".
> For example if I see changes in the Assignment code that I do not
> understand I usually defer to Ram. If there's some HFile stuff, I defer to
> Mikhail...
>
> If we had a list of specialists, it would be easier to defer to them, or
> to pull them into a review. I think that would be better than strict
> guidelines.
>
>
> I'd volunteer for: Transactions/MVCC, Scanners/Scanning/QueryMatcher,
> Client, Deletion, Performance.
>
>
>
> ________________________________
> From: Andrew Purtell <an...@gmail.com>
> To: "dev@hbase.apache.org" <de...@hbase.apache.org>
> Cc: "dev@hbase.apache.org" <de...@hbase.apache.org>; lars hofhansl <
> lhofhansl@yahoo.com>
> Sent: Monday, September 17, 2012 3:08 PM
> Subject: Re: DISCUSSION: Component Lieutenants?
>
> Why doesn't every committer or contributor with interest volunteer? Some
> overlap there would be good. Beyond that we can list the remaining areas
> without good coverage and nominate for them?
>
> I volunteer for Coprocessors, REST, security, filters, and client.
>
> On Sep 17, 2012, at 2:12 PM, Todd Lipcon <to...@cloudera.com> wrote:
>
> > On Fri, Sep 14, 2012 at 9:15 PM, lars hofhansl <lh...@yahoo.com>
> wrote:
> >> I like that idea.
> >>
> >> Should all PMC members or committers be at top level of the source
> tree? Or will that just take us back to the status-quo?
> >>
> >
> > I feel like that would take us back to the status quo.
> >
> > The downside of this proposal is that we should probably have some
> > well-principled way of determining who gets "ownership" (whether
> > co-ownership or alone) of each part of the heirarchy. I fear it could
> > become political or discourage people from contributing or reviewing
> > code outside their area of expertise. So, if people have good ideas on
> > how to go about doing this, please shout them out!
> >
> >>
> >> I certainly like that a typical patch then will involve multiple
> reviewer, and it will be more defined who should look at what patch.
> >>
> >> -- Lars
> >>
> >>
> >> ----- Original Message -----
> >> From: Todd Lipcon <to...@cloudera.com>
> >> To: dev@hbase.apache.org
> >> Cc:
> >> Sent: Friday, September 14, 2012 1:15 PM
> >> Subject: Re: DISCUSSION: Component Lieutenants?
> >>
> >> I like the idea of lieutenants, but another option would be a
> >> "multi-lieutenant" model.
> >>
> >> The model used at google is that each directory has a file called
> >> "OWNERS" which lists several usernames, one per line.
> >>
> >> For any given patch, you are expected to get a review such that, for
> >> each modified file, one of the OWNERS listed in that directory (or any
> >> parent thereof) has +1ed.
> >>
> >> So, for example, imagine that hbase/OWNERS has only Stack, and
> >> hbase/foo/component1/OWNERS has "jxiang,larsh". If I make a patch
> >> which touches something in foo/component1/bar/, I'd need a review from
> >> at least one of Jimmy, Lars, or Stack.
> >>
> >> The assumption is that you try to get review from the most specific
> >> owner, but if those people are MIA, you get review from someone higher
> >> up the stack. The multi-person-per-dir model also ensures that, if
> >> someone's on vacation or otherwise busy, we don't get blocked. And it
> >> formalizes in the actual source tree who you should probably email if
> >> you have questions about an area.
> >>
> >> It also means that wide-ranging patches that touch multiple components
> >> need a lot of reviewers (or someone higher up the chain of command who
> >> has "permission" on the whole tree). So if I had a mondo patch that
> >> touched the region server, the master, and the IPC layer, I'd probably
> >> need at least three separate people to sign off.
> >>
> >> Whatever we do, rather than making it a strict policy, let's start out
> >> with a soft touch. Perhaps declare the component maintainers and try
> >> to pick reviewers based on the criteria. But if people are busy and
> >> work needs to get done, we don't need to be anal about it :)
> >>
> >> -Todd
> >>
> >> On Fri, Sep 14, 2012 at 12:17 PM, Stack <st...@duboce.net> wrote:
> >>> At the contributor's pow wow a few days ago [1], during a discussion
> >>> about whether or not commits should have more friction applied -- i.e.
> >>> have more review before they go in -- it was thought that we might
> >>> benefit if we had "lieutenants" over-seeing individual HBase
> >>> components. A lieutenant would be someone who has an interest and an
> >>> understanding of how a particular component works (or should work). A
> >>> lieutenant does not need to be a committer. Before committing a patch
> >>> that touched on a particular component, the patch would have to have
> >>> been +1'd by the component lieutenant before it could go in (or if the
> >>> lieutenant is MIA, it was suggested by the Mighty Jon Hsieh that two
> >>> +1s by other contributors/committers would do instead; this latter
> >>> rule would probably also apply when a patch spanned components).
> >>>
> >>> We already have a few folks signed up, knowingly or otherwise, as
> >>> component owners [1].
> >>>
> >>> What do folks think?
> >>>
> >>> Should we go ahead w/ this project? If so, any volunteers (I signed
> >>> up a few of the obvious component leads)? I can add you as component
> >>> lieutenant into JIRA. We can add more components if you don't see
> >>> your interest listed.
> >>>
> >>> St.Ack
> >>>
> >>> 1. http://www.meetup.com/hbaseusergroup/events/80621872/
> >>
> >>
> >>
> >> --
> >> Todd Lipcon
> >> Software Engineer, Cloudera
> >>
> >
> >
> >
> > --
> > Todd Lipcon
> > Software Engineer, Cloudera
>
Re: DISCUSSION: Component Lieutenants?
Posted by lars hofhansl <lh...@yahoo.com>.
I'd add WAL/HLog, Mutations (Put/Delete), Memstore, and Coprocessors to what I'd volunteer for since I've been in that code a lot.
________________________________
From: lars hofhansl <lh...@yahoo.com>
To: "dev@hbase.apache.org" <de...@hbase.apache.org>
Sent: Monday, September 17, 2012 4:13 PM
Subject: Re: DISCUSSION: Component Lieutenants?
Maybe just make it an informal list of (self declared :) ) "specialists". For example if I see changes in the Assignment code that I do not understand I usually defer to Ram. If there's some HFile stuff, I defer to Mikhail...
If we had a list of specialists, it would be easier to defer to them, or to pull them into a review. I think that would be better than strict guidelines.
I'd volunteer for: Transactions/MVCC, Scanners/Scanning/QueryMatcher, Client, Deletion, Performance.
________________________________
From: Andrew Purtell <an...@gmail.com>
To: "dev@hbase.apache.org" <de...@hbase.apache.org>
Cc: "dev@hbase.apache.org" <de...@hbase.apache.org>; lars hofhansl <lh...@yahoo.com>
Sent: Monday, September 17, 2012 3:08 PM
Subject: Re: DISCUSSION: Component Lieutenants?
Why doesn't every committer or contributor with interest volunteer? Some overlap there would be good. Beyond that we can list the remaining areas without good coverage and nominate for them?
I volunteer for Coprocessors, REST, security, filters, and client.
On Sep 17, 2012, at 2:12 PM, Todd Lipcon <to...@cloudera.com> wrote:
> On Fri, Sep 14, 2012 at 9:15 PM, lars hofhansl <lh...@yahoo.com> wrote:
>> I like that idea.
>>
>> Should all PMC members or committers be at top level of the source tree? Or will that just take us back to the status-quo?
>>
>
> I feel like that would take us back to the status quo.
>
> The downside of this proposal is that we should probably have some
> well-principled way of determining who gets "ownership" (whether
> co-ownership or alone) of each part of the heirarchy. I fear it could
> become political or discourage people from contributing or reviewing
> code outside their area of expertise. So, if people have good ideas on
> how to go about doing this, please shout them out!
>
>>
>> I certainly like that a typical patch then will involve multiple reviewer, and it will be more defined who should look at what patch.
>>
>> -- Lars
>>
>>
>> ----- Original Message -----
>> From: Todd Lipcon <to...@cloudera.com>
>> To: dev@hbase.apache.org
>> Cc:
>> Sent: Friday, September 14, 2012 1:15 PM
>> Subject: Re: DISCUSSION: Component Lieutenants?
>>
>> I like the idea of lieutenants, but another option would be a
>> "multi-lieutenant" model.
>>
>> The model used at google is that each directory has a file called
>> "OWNERS" which lists several usernames, one per line.
>>
>> For any given patch, you are expected to get a review such that, for
>> each modified file, one of the OWNERS listed in that directory (or any
>> parent thereof) has +1ed.
>>
>> So, for example, imagine that hbase/OWNERS has only Stack, and
>> hbase/foo/component1/OWNERS has "jxiang,larsh". If I make a patch
>> which touches something in foo/component1/bar/, I'd need a review from
>> at least one of Jimmy, Lars, or Stack.
>>
>> The assumption is that you try to get review from the most specific
>> owner, but if those people are MIA, you get review from someone higher
>> up the stack. The multi-person-per-dir model also ensures that, if
>> someone's on vacation or otherwise busy, we don't get blocked. And it
>> formalizes in the actual source tree who you should probably email if
>> you have questions about an area.
>>
>> It also means that wide-ranging patches that touch multiple components
>> need a lot of reviewers (or someone higher up the chain of command who
>> has "permission" on the whole tree). So if I had a mondo patch that
>> touched the region server, the master, and the IPC layer, I'd probably
>> need at least three separate people to sign off.
>>
>> Whatever we do, rather than making it a strict policy, let's start out
>> with a soft touch. Perhaps declare the component maintainers and try
>> to pick reviewers based on the criteria. But if people are busy and
>> work needs to get done, we don't need to be anal about it :)
>>
>> -Todd
>>
>> On Fri, Sep 14, 2012 at 12:17 PM, Stack <st...@duboce.net> wrote:
>>> At the contributor's pow wow a few days ago [1], during a discussion
>>> about whether or not commits should have more friction applied -- i.e.
>>> have more review before they go in -- it was thought that we might
>>> benefit if we had "lieutenants" over-seeing individual HBase
>>> components. A lieutenant would be someone who has an interest and an
>>> understanding of how a particular component works (or should work). A
>>> lieutenant does not need to be a committer. Before committing a patch
>>> that touched on a particular component, the patch would have to have
>>> been +1'd by the component lieutenant before it could go in (or if the
>>> lieutenant is MIA, it was suggested by the Mighty Jon Hsieh that two
>>> +1s by other contributors/committers would do instead; this latter
>>> rule would probably also apply when a patch spanned components).
>>>
>>> We already have a few folks signed up, knowingly or otherwise, as
>>> component owners [1].
>>>
>>> What do folks think?
>>>
>>> Should we go ahead w/ this project? If so, any volunteers (I signed
>>> up a few of the obvious component leads)? I can add you as component
>>> lieutenant into JIRA. We can add more components if you don't see
>>> your interest listed.
>>>
>>> St.Ack
>>>
>>> 1. http://www.meetup.com/hbaseusergroup/events/80621872/
>>
>>
>>
>> --
>> Todd Lipcon
>> Software Engineer, Cloudera
>>
>
>
>
> --
> Todd Lipcon
> Software Engineer, Cloudera
Re: DISCUSSION: Component Lieutenants?
Posted by lars hofhansl <lh...@yahoo.com>.
Maybe just make it an informal list of (self declared :) ) "specialists". For example if I see changes in the Assignment code that I do not understand I usually defer to Ram. If there's some HFile stuff, I defer to Mikhail...
If we had a list of specialists, it would be easier to defer to them, or to pull them into a review. I think that would be better than strict guidelines.
I'd volunteer for: Transactions/MVCC, Scanners/Scanning/QueryMatcher, Client, Deletion, Performance.
________________________________
From: Andrew Purtell <an...@gmail.com>
To: "dev@hbase.apache.org" <de...@hbase.apache.org>
Cc: "dev@hbase.apache.org" <de...@hbase.apache.org>; lars hofhansl <lh...@yahoo.com>
Sent: Monday, September 17, 2012 3:08 PM
Subject: Re: DISCUSSION: Component Lieutenants?
Why doesn't every committer or contributor with interest volunteer? Some overlap there would be good. Beyond that we can list the remaining areas without good coverage and nominate for them?
I volunteer for Coprocessors, REST, security, filters, and client.
On Sep 17, 2012, at 2:12 PM, Todd Lipcon <to...@cloudera.com> wrote:
> On Fri, Sep 14, 2012 at 9:15 PM, lars hofhansl <lh...@yahoo.com> wrote:
>> I like that idea.
>>
>> Should all PMC members or committers be at top level of the source tree? Or will that just take us back to the status-quo?
>>
>
> I feel like that would take us back to the status quo.
>
> The downside of this proposal is that we should probably have some
> well-principled way of determining who gets "ownership" (whether
> co-ownership or alone) of each part of the heirarchy. I fear it could
> become political or discourage people from contributing or reviewing
> code outside their area of expertise. So, if people have good ideas on
> how to go about doing this, please shout them out!
>
>>
>> I certainly like that a typical patch then will involve multiple reviewer, and it will be more defined who should look at what patch.
>>
>> -- Lars
>>
>>
>> ----- Original Message -----
>> From: Todd Lipcon <to...@cloudera.com>
>> To: dev@hbase.apache.org
>> Cc:
>> Sent: Friday, September 14, 2012 1:15 PM
>> Subject: Re: DISCUSSION: Component Lieutenants?
>>
>> I like the idea of lieutenants, but another option would be a
>> "multi-lieutenant" model.
>>
>> The model used at google is that each directory has a file called
>> "OWNERS" which lists several usernames, one per line.
>>
>> For any given patch, you are expected to get a review such that, for
>> each modified file, one of the OWNERS listed in that directory (or any
>> parent thereof) has +1ed.
>>
>> So, for example, imagine that hbase/OWNERS has only Stack, and
>> hbase/foo/component1/OWNERS has "jxiang,larsh". If I make a patch
>> which touches something in foo/component1/bar/, I'd need a review from
>> at least one of Jimmy, Lars, or Stack.
>>
>> The assumption is that you try to get review from the most specific
>> owner, but if those people are MIA, you get review from someone higher
>> up the stack. The multi-person-per-dir model also ensures that, if
>> someone's on vacation or otherwise busy, we don't get blocked. And it
>> formalizes in the actual source tree who you should probably email if
>> you have questions about an area.
>>
>> It also means that wide-ranging patches that touch multiple components
>> need a lot of reviewers (or someone higher up the chain of command who
>> has "permission" on the whole tree). So if I had a mondo patch that
>> touched the region server, the master, and the IPC layer, I'd probably
>> need at least three separate people to sign off.
>>
>> Whatever we do, rather than making it a strict policy, let's start out
>> with a soft touch. Perhaps declare the component maintainers and try
>> to pick reviewers based on the criteria. But if people are busy and
>> work needs to get done, we don't need to be anal about it :)
>>
>> -Todd
>>
>> On Fri, Sep 14, 2012 at 12:17 PM, Stack <st...@duboce.net> wrote:
>>> At the contributor's pow wow a few days ago [1], during a discussion
>>> about whether or not commits should have more friction applied -- i.e.
>>> have more review before they go in -- it was thought that we might
>>> benefit if we had "lieutenants" over-seeing individual HBase
>>> components. A lieutenant would be someone who has an interest and an
>>> understanding of how a particular component works (or should work). A
>>> lieutenant does not need to be a committer. Before committing a patch
>>> that touched on a particular component, the patch would have to have
>>> been +1'd by the component lieutenant before it could go in (or if the
>>> lieutenant is MIA, it was suggested by the Mighty Jon Hsieh that two
>>> +1s by other contributors/committers would do instead; this latter
>>> rule would probably also apply when a patch spanned components).
>>>
>>> We already have a few folks signed up, knowingly or otherwise, as
>>> component owners [1].
>>>
>>> What do folks think?
>>>
>>> Should we go ahead w/ this project? If so, any volunteers (I signed
>>> up a few of the obvious component leads)? I can add you as component
>>> lieutenant into JIRA. We can add more components if you don't see
>>> your interest listed.
>>>
>>> St.Ack
>>>
>>> 1. http://www.meetup.com/hbaseusergroup/events/80621872/
>>
>>
>>
>> --
>> Todd Lipcon
>> Software Engineer, Cloudera
>>
>
>
>
> --
> Todd Lipcon
> Software Engineer, Cloudera
Re: DISCUSSION: Component Lieutenants?
Posted by Andrew Purtell <an...@gmail.com>.
Why doesn't every committer or contributor with interest volunteer? Some overlap there would be good. Beyond that we can list the remaining areas without good coverage and nominate for them?
I volunteer for Coprocessors, REST, security, filters, and client.
On Sep 17, 2012, at 2:12 PM, Todd Lipcon <to...@cloudera.com> wrote:
> On Fri, Sep 14, 2012 at 9:15 PM, lars hofhansl <lh...@yahoo.com> wrote:
>> I like that idea.
>>
>> Should all PMC members or committers be at top level of the source tree? Or will that just take us back to the status-quo?
>>
>
> I feel like that would take us back to the status quo.
>
> The downside of this proposal is that we should probably have some
> well-principled way of determining who gets "ownership" (whether
> co-ownership or alone) of each part of the heirarchy. I fear it could
> become political or discourage people from contributing or reviewing
> code outside their area of expertise. So, if people have good ideas on
> how to go about doing this, please shout them out!
>
>>
>> I certainly like that a typical patch then will involve multiple reviewer, and it will be more defined who should look at what patch.
>>
>> -- Lars
>>
>>
>> ----- Original Message -----
>> From: Todd Lipcon <to...@cloudera.com>
>> To: dev@hbase.apache.org
>> Cc:
>> Sent: Friday, September 14, 2012 1:15 PM
>> Subject: Re: DISCUSSION: Component Lieutenants?
>>
>> I like the idea of lieutenants, but another option would be a
>> "multi-lieutenant" model.
>>
>> The model used at google is that each directory has a file called
>> "OWNERS" which lists several usernames, one per line.
>>
>> For any given patch, you are expected to get a review such that, for
>> each modified file, one of the OWNERS listed in that directory (or any
>> parent thereof) has +1ed.
>>
>> So, for example, imagine that hbase/OWNERS has only Stack, and
>> hbase/foo/component1/OWNERS has "jxiang,larsh". If I make a patch
>> which touches something in foo/component1/bar/, I'd need a review from
>> at least one of Jimmy, Lars, or Stack.
>>
>> The assumption is that you try to get review from the most specific
>> owner, but if those people are MIA, you get review from someone higher
>> up the stack. The multi-person-per-dir model also ensures that, if
>> someone's on vacation or otherwise busy, we don't get blocked. And it
>> formalizes in the actual source tree who you should probably email if
>> you have questions about an area.
>>
>> It also means that wide-ranging patches that touch multiple components
>> need a lot of reviewers (or someone higher up the chain of command who
>> has "permission" on the whole tree). So if I had a mondo patch that
>> touched the region server, the master, and the IPC layer, I'd probably
>> need at least three separate people to sign off.
>>
>> Whatever we do, rather than making it a strict policy, let's start out
>> with a soft touch. Perhaps declare the component maintainers and try
>> to pick reviewers based on the criteria. But if people are busy and
>> work needs to get done, we don't need to be anal about it :)
>>
>> -Todd
>>
>> On Fri, Sep 14, 2012 at 12:17 PM, Stack <st...@duboce.net> wrote:
>>> At the contributor's pow wow a few days ago [1], during a discussion
>>> about whether or not commits should have more friction applied -- i.e.
>>> have more review before they go in -- it was thought that we might
>>> benefit if we had "lieutenants" over-seeing individual HBase
>>> components. A lieutenant would be someone who has an interest and an
>>> understanding of how a particular component works (or should work). A
>>> lieutenant does not need to be a committer. Before committing a patch
>>> that touched on a particular component, the patch would have to have
>>> been +1'd by the component lieutenant before it could go in (or if the
>>> lieutenant is MIA, it was suggested by the Mighty Jon Hsieh that two
>>> +1s by other contributors/committers would do instead; this latter
>>> rule would probably also apply when a patch spanned components).
>>>
>>> We already have a few folks signed up, knowingly or otherwise, as
>>> component owners [1].
>>>
>>> What do folks think?
>>>
>>> Should we go ahead w/ this project? If so, any volunteers (I signed
>>> up a few of the obvious component leads)? I can add you as component
>>> lieutenant into JIRA. We can add more components if you don't see
>>> your interest listed.
>>>
>>> St.Ack
>>>
>>> 1. http://www.meetup.com/hbaseusergroup/events/80621872/
>>
>>
>>
>> --
>> Todd Lipcon
>> Software Engineer, Cloudera
>>
>
>
>
> --
> Todd Lipcon
> Software Engineer, Cloudera
Re: DISCUSSION: Component Lieutenants?
Posted by Stack <st...@duboce.net>.
On Mon, Sep 17, 2012 at 3:08 PM, Jonathan Hsieh <jo...@cloudera.com> wrote:
> I do like the idea of multiple people "owning" an area to avoid
> politicking.
Yes, multi-owners if possible is the way to go. A +1 by a multi-owner
or two +1s by non-multi-owners before a patch goes into a particular
component.
> For lookup purposes, instead of putting names under a
> component lead, we could just use the description field to list folk's name.
> https://issues.apache.org/jira/browse/HBASE/component/12312140 (what you
> get when you click on the rest component link)
> https://issues.apache.org/jira/browse/HBASE/component/12315702 (what you
> get when you click on the hbck component link)
>
Good one Jon. I actually want to go ahead and do this now and move
over the ownerships we already have and add the ones that you and Lars
just volunteered for.
> To initially dole out components, we could do it the benevolent dictator
> way (stack's velvet glove initially assigns folks), and then folks in the
> component can add others.
Smile. Or better would be folks volunteering (as fellas are doing
here already). I can take care of updating JIRA.
> I think one that that is
> important some sort of design info up so that we know how things are
> supposed to work.
>
I agree.
If it were in place, it'd make it easier rejecting patches that take
us away from the stated goal. It'd also serve as a target for patches
to drive toward.
Who'd do the design or component tenets list? Component owner's
ideally, when they get a moment.
St.Ack
Re: DISCUSSION: Component Lieutenants?
Posted by Andrew Purtell <an...@gmail.com>.
+1 to component partitioning, not source layout. My earlier volunteering was by-component.
On Sep 17, 2012, at 3:08 PM, Jonathan Hsieh <jo...@cloudera.com> wrote:
> I like Todd's idea in principle but I think there are some problems with
> the organization of the code base currently that make this potentially
> painful today (did you look at how much code is in hmaster and
> regionserver?). Also some folk's expertise doesn't lay in one hierarchical
> tree necessarily. Because of this I prefer having it per component. For
> areas where components are a little broad (master, regionserver) we could
> break them down a little more. If we actually start using components
> consistently it also makes it easier for us to setup jira filters for the
> review queues.
>
> I do like the idea of multiple people "owning" an area to avoid
> politicking. For lookup purposes, instead of putting names under a
> component lead, we could just use the description field to list folk's name.
> https://issues.apache.org/jira/browse/HBASE/component/12312140 (what you
> get when you click on the rest component link)
> https://issues.apache.org/jira/browse/HBASE/component/12315702 (what you
> get when you click on the hbck component link)
>
> To initially dole out components, we could do it the benevolent dictator
> way (stack's velvet glove initially assigns folks), and then folks in the
> component can add others. (likely criteria are similar to commit --
> designs docs presented, authorship of patches, support on mailing list).
> If we add components (like snapshots recently), the authors and authors of
> design docs folks get to own it initially. I think one that that is
> important some sort of design info up so that we know how things are
> supposed to work.
>
> Jon.
>
> On Mon, Sep 17, 2012 at 2:12 PM, Todd Lipcon <to...@cloudera.com> wrote:
>
>> On Fri, Sep 14, 2012 at 9:15 PM, lars hofhansl <lh...@yahoo.com>
>> wrote:
>>> I like that idea.
>>>
>>> Should all PMC members or committers be at top level of the source tree?
>> Or will that just take us back to the status-quo?
>>>
>>
>> I feel like that would take us back to the status quo.
>>
>> The downside of this proposal is that we should probably have some
>> well-principled way of determining who gets "ownership" (whether
>> co-ownership or alone) of each part of the heirarchy. I fear it could
>> become political or discourage people from contributing or reviewing
>> code outside their area of expertise. So, if people have good ideas on
>> how to go about doing this, please shout them out!
>>
>>>
>>> I certainly like that a typical patch then will involve multiple
>> reviewer, and it will be more defined who should look at what patch.
>>>
>>> -- Lars
>>>
>>>
>>> ----- Original Message -----
>>> From: Todd Lipcon <to...@cloudera.com>
>>> To: dev@hbase.apache.org
>>> Cc:
>>> Sent: Friday, September 14, 2012 1:15 PM
>>> Subject: Re: DISCUSSION: Component Lieutenants?
>>>
>>> I like the idea of lieutenants, but another option would be a
>>> "multi-lieutenant" model.
>>>
>>> The model used at google is that each directory has a file called
>>> "OWNERS" which lists several usernames, one per line.
>>>
>>> For any given patch, you are expected to get a review such that, for
>>> each modified file, one of the OWNERS listed in that directory (or any
>>> parent thereof) has +1ed.
>>>
>>> So, for example, imagine that hbase/OWNERS has only Stack, and
>>> hbase/foo/component1/OWNERS has "jxiang,larsh". If I make a patch
>>> which touches something in foo/component1/bar/, I'd need a review from
>>> at least one of Jimmy, Lars, or Stack.
>>>
>>> The assumption is that you try to get review from the most specific
>>> owner, but if those people are MIA, you get review from someone higher
>>> up the stack. The multi-person-per-dir model also ensures that, if
>>> someone's on vacation or otherwise busy, we don't get blocked. And it
>>> formalizes in the actual source tree who you should probably email if
>>> you have questions about an area.
>>>
>>> It also means that wide-ranging patches that touch multiple components
>>> need a lot of reviewers (or someone higher up the chain of command who
>>> has "permission" on the whole tree). So if I had a mondo patch that
>>> touched the region server, the master, and the IPC layer, I'd probably
>>> need at least three separate people to sign off.
>>>
>>> Whatever we do, rather than making it a strict policy, let's start out
>>> with a soft touch. Perhaps declare the component maintainers and try
>>> to pick reviewers based on the criteria. But if people are busy and
>>> work needs to get done, we don't need to be anal about it :)
>>>
>>> -Todd
>>>
>>> On Fri, Sep 14, 2012 at 12:17 PM, Stack <st...@duboce.net> wrote:
>>>> At the contributor's pow wow a few days ago [1], during a discussion
>>>> about whether or not commits should have more friction applied -- i.e.
>>>> have more review before they go in -- it was thought that we might
>>>> benefit if we had "lieutenants" over-seeing individual HBase
>>>> components. A lieutenant would be someone who has an interest and an
>>>> understanding of how a particular component works (or should work). A
>>>> lieutenant does not need to be a committer. Before committing a patch
>>>> that touched on a particular component, the patch would have to have
>>>> been +1'd by the component lieutenant before it could go in (or if the
>>>> lieutenant is MIA, it was suggested by the Mighty Jon Hsieh that two
>>>> +1s by other contributors/committers would do instead; this latter
>>>> rule would probably also apply when a patch spanned components).
>>>>
>>>> We already have a few folks signed up, knowingly or otherwise, as
>>>> component owners [1].
>>>>
>>>> What do folks think?
>>>>
>>>> Should we go ahead w/ this project? If so, any volunteers (I signed
>>>> up a few of the obvious component leads)? I can add you as component
>>>> lieutenant into JIRA. We can add more components if you don't see
>>>> your interest listed.
>>>>
>>>> St.Ack
>>>>
>>>> 1. http://www.meetup.com/hbaseusergroup/events/80621872/
>>>
>>>
>>>
>>> --
>>> Todd Lipcon
>>> Software Engineer, Cloudera
>>>
>>
>>
>>
>> --
>> Todd Lipcon
>> Software Engineer, Cloudera
>>
>
>
>
> --
> // Jonathan Hsieh (shay)
> // Software Engineer, Cloudera
> // jon@cloudera.com
Re: DISCUSSION: Component Lieutenants?
Posted by Jonathan Hsieh <jo...@cloudera.com>.
I like Todd's idea in principle but I think there are some problems with
the organization of the code base currently that make this potentially
painful today (did you look at how much code is in hmaster and
regionserver?). Also some folk's expertise doesn't lay in one hierarchical
tree necessarily. Because of this I prefer having it per component. For
areas where components are a little broad (master, regionserver) we could
break them down a little more. If we actually start using components
consistently it also makes it easier for us to setup jira filters for the
review queues.
I do like the idea of multiple people "owning" an area to avoid
politicking. For lookup purposes, instead of putting names under a
component lead, we could just use the description field to list folk's name.
https://issues.apache.org/jira/browse/HBASE/component/12312140 (what you
get when you click on the rest component link)
https://issues.apache.org/jira/browse/HBASE/component/12315702 (what you
get when you click on the hbck component link)
To initially dole out components, we could do it the benevolent dictator
way (stack's velvet glove initially assigns folks), and then folks in the
component can add others. (likely criteria are similar to commit --
designs docs presented, authorship of patches, support on mailing list).
If we add components (like snapshots recently), the authors and authors of
design docs folks get to own it initially. I think one that that is
important some sort of design info up so that we know how things are
supposed to work.
Jon.
On Mon, Sep 17, 2012 at 2:12 PM, Todd Lipcon <to...@cloudera.com> wrote:
> On Fri, Sep 14, 2012 at 9:15 PM, lars hofhansl <lh...@yahoo.com>
> wrote:
> > I like that idea.
> >
> > Should all PMC members or committers be at top level of the source tree?
> Or will that just take us back to the status-quo?
> >
>
> I feel like that would take us back to the status quo.
>
> The downside of this proposal is that we should probably have some
> well-principled way of determining who gets "ownership" (whether
> co-ownership or alone) of each part of the heirarchy. I fear it could
> become political or discourage people from contributing or reviewing
> code outside their area of expertise. So, if people have good ideas on
> how to go about doing this, please shout them out!
>
> >
> > I certainly like that a typical patch then will involve multiple
> reviewer, and it will be more defined who should look at what patch.
> >
> > -- Lars
> >
> >
> > ----- Original Message -----
> > From: Todd Lipcon <to...@cloudera.com>
> > To: dev@hbase.apache.org
> > Cc:
> > Sent: Friday, September 14, 2012 1:15 PM
> > Subject: Re: DISCUSSION: Component Lieutenants?
> >
> > I like the idea of lieutenants, but another option would be a
> > "multi-lieutenant" model.
> >
> > The model used at google is that each directory has a file called
> > "OWNERS" which lists several usernames, one per line.
> >
> > For any given patch, you are expected to get a review such that, for
> > each modified file, one of the OWNERS listed in that directory (or any
> > parent thereof) has +1ed.
> >
> > So, for example, imagine that hbase/OWNERS has only Stack, and
> > hbase/foo/component1/OWNERS has "jxiang,larsh". If I make a patch
> > which touches something in foo/component1/bar/, I'd need a review from
> > at least one of Jimmy, Lars, or Stack.
> >
> > The assumption is that you try to get review from the most specific
> > owner, but if those people are MIA, you get review from someone higher
> > up the stack. The multi-person-per-dir model also ensures that, if
> > someone's on vacation or otherwise busy, we don't get blocked. And it
> > formalizes in the actual source tree who you should probably email if
> > you have questions about an area.
> >
> > It also means that wide-ranging patches that touch multiple components
> > need a lot of reviewers (or someone higher up the chain of command who
> > has "permission" on the whole tree). So if I had a mondo patch that
> > touched the region server, the master, and the IPC layer, I'd probably
> > need at least three separate people to sign off.
> >
> > Whatever we do, rather than making it a strict policy, let's start out
> > with a soft touch. Perhaps declare the component maintainers and try
> > to pick reviewers based on the criteria. But if people are busy and
> > work needs to get done, we don't need to be anal about it :)
> >
> > -Todd
> >
> > On Fri, Sep 14, 2012 at 12:17 PM, Stack <st...@duboce.net> wrote:
> >> At the contributor's pow wow a few days ago [1], during a discussion
> >> about whether or not commits should have more friction applied -- i.e.
> >> have more review before they go in -- it was thought that we might
> >> benefit if we had "lieutenants" over-seeing individual HBase
> >> components. A lieutenant would be someone who has an interest and an
> >> understanding of how a particular component works (or should work). A
> >> lieutenant does not need to be a committer. Before committing a patch
> >> that touched on a particular component, the patch would have to have
> >> been +1'd by the component lieutenant before it could go in (or if the
> >> lieutenant is MIA, it was suggested by the Mighty Jon Hsieh that two
> >> +1s by other contributors/committers would do instead; this latter
> >> rule would probably also apply when a patch spanned components).
> >>
> >> We already have a few folks signed up, knowingly or otherwise, as
> >> component owners [1].
> >>
> >> What do folks think?
> >>
> >> Should we go ahead w/ this project? If so, any volunteers (I signed
> >> up a few of the obvious component leads)? I can add you as component
> >> lieutenant into JIRA. We can add more components if you don't see
> >> your interest listed.
> >>
> >> St.Ack
> >>
> >> 1. http://www.meetup.com/hbaseusergroup/events/80621872/
> >
> >
> >
> > --
> > Todd Lipcon
> > Software Engineer, Cloudera
> >
>
>
>
> --
> Todd Lipcon
> Software Engineer, Cloudera
>
--
// Jonathan Hsieh (shay)
// Software Engineer, Cloudera
// jon@cloudera.com
Re: DISCUSSION: Component Lieutenants?
Posted by Todd Lipcon <to...@cloudera.com>.
On Fri, Sep 14, 2012 at 9:15 PM, lars hofhansl <lh...@yahoo.com> wrote:
> I like that idea.
>
> Should all PMC members or committers be at top level of the source tree? Or will that just take us back to the status-quo?
>
I feel like that would take us back to the status quo.
The downside of this proposal is that we should probably have some
well-principled way of determining who gets "ownership" (whether
co-ownership or alone) of each part of the heirarchy. I fear it could
become political or discourage people from contributing or reviewing
code outside their area of expertise. So, if people have good ideas on
how to go about doing this, please shout them out!
>
> I certainly like that a typical patch then will involve multiple reviewer, and it will be more defined who should look at what patch.
>
> -- Lars
>
>
> ----- Original Message -----
> From: Todd Lipcon <to...@cloudera.com>
> To: dev@hbase.apache.org
> Cc:
> Sent: Friday, September 14, 2012 1:15 PM
> Subject: Re: DISCUSSION: Component Lieutenants?
>
> I like the idea of lieutenants, but another option would be a
> "multi-lieutenant" model.
>
> The model used at google is that each directory has a file called
> "OWNERS" which lists several usernames, one per line.
>
> For any given patch, you are expected to get a review such that, for
> each modified file, one of the OWNERS listed in that directory (or any
> parent thereof) has +1ed.
>
> So, for example, imagine that hbase/OWNERS has only Stack, and
> hbase/foo/component1/OWNERS has "jxiang,larsh". If I make a patch
> which touches something in foo/component1/bar/, I'd need a review from
> at least one of Jimmy, Lars, or Stack.
>
> The assumption is that you try to get review from the most specific
> owner, but if those people are MIA, you get review from someone higher
> up the stack. The multi-person-per-dir model also ensures that, if
> someone's on vacation or otherwise busy, we don't get blocked. And it
> formalizes in the actual source tree who you should probably email if
> you have questions about an area.
>
> It also means that wide-ranging patches that touch multiple components
> need a lot of reviewers (or someone higher up the chain of command who
> has "permission" on the whole tree). So if I had a mondo patch that
> touched the region server, the master, and the IPC layer, I'd probably
> need at least three separate people to sign off.
>
> Whatever we do, rather than making it a strict policy, let's start out
> with a soft touch. Perhaps declare the component maintainers and try
> to pick reviewers based on the criteria. But if people are busy and
> work needs to get done, we don't need to be anal about it :)
>
> -Todd
>
> On Fri, Sep 14, 2012 at 12:17 PM, Stack <st...@duboce.net> wrote:
>> At the contributor's pow wow a few days ago [1], during a discussion
>> about whether or not commits should have more friction applied -- i.e.
>> have more review before they go in -- it was thought that we might
>> benefit if we had "lieutenants" over-seeing individual HBase
>> components. A lieutenant would be someone who has an interest and an
>> understanding of how a particular component works (or should work). A
>> lieutenant does not need to be a committer. Before committing a patch
>> that touched on a particular component, the patch would have to have
>> been +1'd by the component lieutenant before it could go in (or if the
>> lieutenant is MIA, it was suggested by the Mighty Jon Hsieh that two
>> +1s by other contributors/committers would do instead; this latter
>> rule would probably also apply when a patch spanned components).
>>
>> We already have a few folks signed up, knowingly or otherwise, as
>> component owners [1].
>>
>> What do folks think?
>>
>> Should we go ahead w/ this project? If so, any volunteers (I signed
>> up a few of the obvious component leads)? I can add you as component
>> lieutenant into JIRA. We can add more components if you don't see
>> your interest listed.
>>
>> St.Ack
>>
>> 1. http://www.meetup.com/hbaseusergroup/events/80621872/
>
>
>
> --
> Todd Lipcon
> Software Engineer, Cloudera
>
--
Todd Lipcon
Software Engineer, Cloudera
Re: DISCUSSION: Component Lieutenants?
Posted by lars hofhansl <lh...@yahoo.com>.
I like that idea.
Should all PMC members or committers be at top level of the source tree? Or will that just take us back to the status-quo?
I certainly like that a typical patch then will involve multiple reviewer, and it will be more defined who should look at what patch.
-- Lars
----- Original Message -----
From: Todd Lipcon <to...@cloudera.com>
To: dev@hbase.apache.org
Cc:
Sent: Friday, September 14, 2012 1:15 PM
Subject: Re: DISCUSSION: Component Lieutenants?
I like the idea of lieutenants, but another option would be a
"multi-lieutenant" model.
The model used at google is that each directory has a file called
"OWNERS" which lists several usernames, one per line.
For any given patch, you are expected to get a review such that, for
each modified file, one of the OWNERS listed in that directory (or any
parent thereof) has +1ed.
So, for example, imagine that hbase/OWNERS has only Stack, and
hbase/foo/component1/OWNERS has "jxiang,larsh". If I make a patch
which touches something in foo/component1/bar/, I'd need a review from
at least one of Jimmy, Lars, or Stack.
The assumption is that you try to get review from the most specific
owner, but if those people are MIA, you get review from someone higher
up the stack. The multi-person-per-dir model also ensures that, if
someone's on vacation or otherwise busy, we don't get blocked. And it
formalizes in the actual source tree who you should probably email if
you have questions about an area.
It also means that wide-ranging patches that touch multiple components
need a lot of reviewers (or someone higher up the chain of command who
has "permission" on the whole tree). So if I had a mondo patch that
touched the region server, the master, and the IPC layer, I'd probably
need at least three separate people to sign off.
Whatever we do, rather than making it a strict policy, let's start out
with a soft touch. Perhaps declare the component maintainers and try
to pick reviewers based on the criteria. But if people are busy and
work needs to get done, we don't need to be anal about it :)
-Todd
On Fri, Sep 14, 2012 at 12:17 PM, Stack <st...@duboce.net> wrote:
> At the contributor's pow wow a few days ago [1], during a discussion
> about whether or not commits should have more friction applied -- i.e.
> have more review before they go in -- it was thought that we might
> benefit if we had "lieutenants" over-seeing individual HBase
> components. A lieutenant would be someone who has an interest and an
> understanding of how a particular component works (or should work). A
> lieutenant does not need to be a committer. Before committing a patch
> that touched on a particular component, the patch would have to have
> been +1'd by the component lieutenant before it could go in (or if the
> lieutenant is MIA, it was suggested by the Mighty Jon Hsieh that two
> +1s by other contributors/committers would do instead; this latter
> rule would probably also apply when a patch spanned components).
>
> We already have a few folks signed up, knowingly or otherwise, as
> component owners [1].
>
> What do folks think?
>
> Should we go ahead w/ this project? If so, any volunteers (I signed
> up a few of the obvious component leads)? I can add you as component
> lieutenant into JIRA. We can add more components if you don't see
> your interest listed.
>
> St.Ack
>
> 1. http://www.meetup.com/hbaseusergroup/events/80621872/
--
Todd Lipcon
Software Engineer, Cloudera
Re: DISCUSSION: Component Lieutenants?
Posted by Stack <st...@duboce.net>.
On Fri, Sep 14, 2012 at 1:15 PM, Todd Lipcon <to...@cloudera.com> wrote:
> I like the idea of lieutenants, but another option would be a
> "multi-lieutenant" model.
>
> The model used at google is that each directory has a file called
> "OWNERS" which lists several usernames, one per line.
>
> For any given patch, you are expected to get a review such that, for
> each modified file, one of the OWNERS listed in that directory (or any
> parent thereof) has +1ed.
>
> So, for example, imagine that hbase/OWNERS has only Stack, and
> hbase/foo/component1/OWNERS has "jxiang,larsh". If I make a patch
> which touches something in foo/component1/bar/, I'd need a review from
> at least one of Jimmy, Lars, or Stack.
>
> The assumption is that you try to get review from the most specific
> owner, but if those people are MIA, you get review from someone higher
> up the stack. The multi-person-per-dir model also ensures that, if
> someone's on vacation or otherwise busy, we don't get blocked. And it
> formalizes in the actual source tree who you should probably email if
> you have questions about an area.
>
> It also means that wide-ranging patches that touch multiple components
> need a lot of reviewers (or someone higher up the chain of command who
> has "permission" on the whole tree). So if I had a mondo patch that
> touched the region server, the master, and the IPC layer, I'd probably
> need at least three separate people to sign off.
>
Seems like a nice model to me. I can't use JIRA to implement the
hierarchy. This page,
https://issues.apache.org/jira/plugins/servlet/project-config/HBASE/components,
only lets me have one person per component and it does not let me
specify a supra-layer above components to which I could add a
captains' layer.
We could do the hierarchy in a wiki page if folks like Todds suggestion above.
> Whatever we do, rather than making it a strict policy, let's start out
> with a soft touch. Perhaps declare the component maintainers and try
> to pick reviewers based on the criteria. But if people are busy and
> work needs to get done, we don't need to be anal about it :)
>
Agreed.
St.Ack
Re: DISCUSSION: Component Lieutenants?
Posted by Todd Lipcon <to...@cloudera.com>.
I like the idea of lieutenants, but another option would be a
"multi-lieutenant" model.
The model used at google is that each directory has a file called
"OWNERS" which lists several usernames, one per line.
For any given patch, you are expected to get a review such that, for
each modified file, one of the OWNERS listed in that directory (or any
parent thereof) has +1ed.
So, for example, imagine that hbase/OWNERS has only Stack, and
hbase/foo/component1/OWNERS has "jxiang,larsh". If I make a patch
which touches something in foo/component1/bar/, I'd need a review from
at least one of Jimmy, Lars, or Stack.
The assumption is that you try to get review from the most specific
owner, but if those people are MIA, you get review from someone higher
up the stack. The multi-person-per-dir model also ensures that, if
someone's on vacation or otherwise busy, we don't get blocked. And it
formalizes in the actual source tree who you should probably email if
you have questions about an area.
It also means that wide-ranging patches that touch multiple components
need a lot of reviewers (or someone higher up the chain of command who
has "permission" on the whole tree). So if I had a mondo patch that
touched the region server, the master, and the IPC layer, I'd probably
need at least three separate people to sign off.
Whatever we do, rather than making it a strict policy, let's start out
with a soft touch. Perhaps declare the component maintainers and try
to pick reviewers based on the criteria. But if people are busy and
work needs to get done, we don't need to be anal about it :)
-Todd
On Fri, Sep 14, 2012 at 12:17 PM, Stack <st...@duboce.net> wrote:
> At the contributor's pow wow a few days ago [1], during a discussion
> about whether or not commits should have more friction applied -- i.e.
> have more review before they go in -- it was thought that we might
> benefit if we had "lieutenants" over-seeing individual HBase
> components. A lieutenant would be someone who has an interest and an
> understanding of how a particular component works (or should work). A
> lieutenant does not need to be a committer. Before committing a patch
> that touched on a particular component, the patch would have to have
> been +1'd by the component lieutenant before it could go in (or if the
> lieutenant is MIA, it was suggested by the Mighty Jon Hsieh that two
> +1s by other contributors/committers would do instead; this latter
> rule would probably also apply when a patch spanned components).
>
> We already have a few folks signed up, knowingly or otherwise, as
> component owners [1].
>
> What do folks think?
>
> Should we go ahead w/ this project? If so, any volunteers (I signed
> up a few of the obvious component leads)? I can add you as component
> lieutenant into JIRA. We can add more components if you don't see
> your interest listed.
>
> St.Ack
>
> 1. http://www.meetup.com/hbaseusergroup/events/80621872/
--
Todd Lipcon
Software Engineer, Cloudera