Re: migration of 7.4 to 8.1

From: "Joshua D(dot) Drake" <jd(at)commandprompt(dot)com>
To: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: Geoffrey <lists(at)serioustechnology(dot)com>, pgsql-admin(at)postgresql(dot)org
Subject: Re: migration of 7.4 to 8.1
Date: 2008-03-13 16:28:09
Message-ID: 20080313092809.34f88c71@commandprompt.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-admin

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Thu, 13 Mar 2008 12:04:27 -0400
Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:

> Geoffrey <lists(at)serioustechnology(dot)com> writes:
> > Joshua D. Drake wrote:
> >> I would avoid 8.3 without extensive testing. 8.2.6 is likely a
> >> better shot a minimizing one off incompatibilities.
>
> > We are in a very similar situation in that we are looking to
> > migrate from 7.4 to 8.3. So seeing Scott's and Joshua's responses,
> > I'm not sure if we should go with 8.2 or 8.3.
>
> 7.4 to 8.2 isn't exactly a trivial jump either; there are enough
> incompatibilities that could bite you if you don't test your
> applications. I don't think JD's argument above holds water.

7.4 to 8.2 is not a small jump. You are correct Tom. However, as
someone who is actively in the field doing these migrations, I can tell
you that 7.4 to 8.2 is a smaller jump. If nothing else it will avoid
the casting issues which can be a large burden on a small to medium
shop to go through and fix.

8.3 changed a lot of things. We are still observing (in production) all
the changes that must be made procedurally to make it work correctly
(like the way autovacuum will just quit if it runs into a lock it
doesn't like).

> What does hold water is the observation that 8.3 is still at 8.3.0
> (though not for much longer) and has certainly got more bugs today
> than recent 8.2.x releases. That should equalize out by, say,
> middle of the year. So if you intend to go to production in the
> next month or two then 8.2.x is a safer bet. If your release
> timeframe is a bit longer, then 8.3.x would be a good choice because
> it will perform better and be supported longer. You'd want to be
> sure you were on 8.3.latest before going live of course.

If the have the time and resources to do extensive testing over the
course of the next 3-4 months, then yes I would agree that 8.3 is a
reasonable solution.

Sincerely,

Joshua D. Drake

>
> regards, tom lane
>

- --
The PostgreSQL Company since 1997: http://www.commandprompt.com/
PostgreSQL Community Conference: http://www.postgresqlconference.org/
Donate to the PostgreSQL Project: http://www.postgresql.org/about/donate
PostgreSQL political pundit | Mocker of Dolphins

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFH2VYZATb/zqfZUUQRAjVbAJ9rbhWxTaLCH6gRI7cEkOpWZtzUFwCfX2un
v5qnuULJ4v+W7TVrEjhrM1w=
=7IMO
-----END PGP SIGNATURE-----

In response to

Browse pgsql-admin by date

  From Date Subject
Next Message Gregory Stark 2008-03-13 16:33:19 Re: Encoding problems with migration from 8.0.14 to 8.3.0 on Windows
Previous Message Geoffrey 2008-03-13 16:17:18 Re: migration of 7.4 to 8.1