Re: Logical replication and multimaster

On 13 December 2015 at 11:53, Andres Freund <andres(at)anarazel(dot)de> wrote:

> > Thanks for asking, perhaps our plans weren't public enough. pglogical has
> > already been announced as open source, under the postgres licence and
> that
> > it will be a submission to core PostgreSQL, just as BDR was. pglogical is
> > in development/test right now and will be released when its ready, which
> > hopefully will be "soon", aiming for 9.6.
>
> Well, at the moment not making it public is obviously blocking other
> people, and

What other people are being blocked? What contribution they are making to
PostgreSQL core is being delayed? What is the nature of this delay?

> not doing open design discussions publically seems to make
> it rather unlikely that it'll get accepted close to as-is anyway. It's
> constantly referred to in discussions, and it guided the design of the
> submitted output plugin.

It's a shame you think that, but posting incoherent proposals just wastes
everybody's time.

The focus has been on making the internals more generic, unpicking many of
the parts of BDR that were too specialized. The output plugin has not been
guided by the needs of pglogical at all, its been designed to be way more
open than BDR was.

The UI is a fairly thin layer on top of that and can be recoded without too
much work, but I don't think its a surprising design. It's been quite
difficult to cater for the many complex and sometimes conflicting
requirements and that has only recently come to together into a coherent
form by my hand. Whether the UI makes sense remains to be seen, but I think
building it is an essential part of the evaluation of whether it is
actually a good UI.

Submission to core implies that changes are possible and discussion is
welcome. I expect that to happen. If there were any truly contentious parts
they would have been discussed ahead of time. I see no reason to believe
that pglogical would not or could not be accepted into 9.6.

> > Thanks also for the opportunity to let me ask what your plans are
> regarding
> > contributing to core?
>
> Uh, I am? Stuff like working on upsert, grouping sets et al, was all on
> Citus' time. I've been busy with something else for the last 2-3 months.

Good, thanks; you misread that and I wasn't questioning it. pglogical
developers have a day job too.

> > I couldn't make it to SF recently because of a funding meeting, but I
> > heard something like your company will release something as open
> > source sometime in 2016. Could you clarify what that will be, when it
> > will be, what licence it is under and if it is a contribution to core
> > also? Is that something you're working on also?
>
> I don't see how that belongs to this thread, it's unrelated to
> replication.
>

I assumed your interest in pglogical meant there was some connection.

> Either way, the license is yet to be determined, and it'll be Q1
> 2016. Yes, I've worked on open sourcing it.
>

If its under the Postgres licence and submitted to core, as is BDR, you may
find many people interested in working on it also.

Initial development of major features is IMHO best done by small groups of
dedicated developers. That has got nothing at all to do with what happens
to the code in the longer term.

--
Simon Riggs http://www.2ndQuadrant.com/
<http://www.2ndquadrant.com/>
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services