Re: Replication status & point-in-time recovery

From: Andrew Sullivan <andrew(at)libertyrms(dot)info>
To: pgsql-hackers(at)postgresql(dot)org, pgreplication-general(at)gborg(dot)postgresql(dot)org
Subject: Re: Replication status & point-in-time recovery
Date: 2002-05-28 14:50:25
Message-ID: 20020528105025.D4231@mail.libertyrms.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Tue, May 28, 2002 at 09:53:53AM +0200, Karel Zak wrote:
> On Mon, May 27, 2002 at 11:40:20AM -0400, Michael Meskes wrote:
>
> > could anyone please enlighten me about the status of replication? I do
> > expect lots of questions about this, and I'm not really sure if I can
> > promise it for 7.3. :-)

> I add the other quesion: how is current status of on-line backup log
> based on WAL?

I've cc:d the replication list, in order to cast the net wider.

It's funny that this should come up just now. I've been fishing
around for some answers to these questions as well.

Darren Johnson tells me that 2 developers, working on the replication
code full time, and sufficiently familiar with the PostgreSQL
internals, could probably finish that project (really finished,
clustering and all) in 18-24 months. I'm pretty happy with the
eRserver code we're using, but it's not a competitor to ORAC, and I'm
getting some flack about that these days. For me, then, the
master-slave replication problem is solved; but I need something
which will go beyond it and do the sort of cluster that will make
marketing guys go "ooh-ahh". Sorry as I am to say it, that's the
truth.

Also, I'm getting some flack about the lack of point-in-time
recovery, which _might_ be possible to get by playing with WAL.

Those are the two big technical reasons I can get beat up with the
Oracle stick. Happily for me, Oracle costs a mint, and producing
cost line-items with fewer zeros after the one is something that
carries a lot of weight.

So I'm thinking of making a sales pitch, internally, to get some cash
to sponsor some development. I have no authorisation to spend
anything -- I'm just trying to get the info I need to propose to
spend money. Think of this post as pre-speculation. (Gee, I hope
the corpoate guys don't read -hackers. Well, not likely, is it?)

I need to know the following in order to make my pitch:

1. Who's interested (in which project) and how soon available.
2. Degree of comfort with the internals (yes, in some cases, I
guess I'll know perfectly well).
3. Rates.

Please understand that I am (as one of the corp guys called me) just
one of the official geeks around here (which is depressing,
considering how many meetings I have to go to). So I am Not Allowed
to Touch the Money. But I might be able to get them to open the
purse strings a little, especially when they contemplate the cost of
moving to Oracle with ORAC.

If people respond in confidence to me, I'll post something here to
the effect of additional interest I've heard expressed (assuming such
posts don't also go here). I can't guarantee any actual results, of
course (have I said that enough?). Also, if they decide to spend
money, I won't be allowed to announce it -- the marketing guys will
make some splashy thing (possibly which annoys everyone at the same
time), I suppose. But I'll try to tell everyone as much as I'm
allowed.

Thanks.

A

--
----
Andrew Sullivan 87 Mowat Avenue
Liberty RMS Toronto, Ontario Canada
<andrew(at)libertyrms(dot)info> M6K 3E3
+1 416 646 3304 x110

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Tom Lane 2002-05-28 14:50:48 Re: revised sample SRF C function; proposed SRF API
Previous Message Kovacs Zoltan 2002-05-28 14:26:09 Re: cache lookup failed: hack pg_* tables?