The long-lost pg_upgrade (was:Re: 8.2 features status)

From: "Lamar Owen" <lowen(at)pari(dot)edu>
To: pgsql-hackers(at)postgresql(dot)org
Cc: Josh Berkus <josh(at)agliodbs(dot)com>, David Fetter <david(at)fetter(dot)org>, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Subject: The long-lost pg_upgrade (was:Re: 8.2 features status)
Date: 2006-08-13 05:00:49
Message-ID: 200608130100.49772.lowen@pari.edu
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Friday 04 August 2006 02:20, Josh Berkus wrote:
> <grin> Aren't I, the marketing geek, supposed to be the one whining about
> this?
[snip]
> > * In-place upgrades (pg_upgrade)

> BTW, I may get Sun to contribute an engineer for this; will get you posted.

Long time no post. This statement really caught my attention; bravo if true
upgrading can happen, and someone can be put on it and do it right.

As Tom said, a little farther down the thread, we have talked over this many
times. I specifically remember, oh, about a dozen times I personally
have 'gadflied' this issue.

As one who now has a, let's see:
[root(at)pachyderm ~]# du /var/lib/pgsql/data -s
16668528 /var/lib/pgsql/data
[root(at)pachyderm ~]#

Yes, a 16GB inventory database, with in-database large object images. Anyway,
as one who does not look forward to migrating this the old-fashioned way (I
can just imagine how fas^H^H^Hslow a restore of all those large objects is
going to be; backup is slow enough (about 50 minutes on this Xeon 2.4GHz
box)), in place upgrade would cut this considerably; the database is not a
complex one, just a largish one. It's, let's see, only holding a little less
than 5,000 items with associated lo images (due to many factors, this is
handled through ODBC from Microsoft Access; it is a kludge, and a big one,
but it works very smoothly from the users' points of view, where item images
are literally 'dragged and dropped' from the digital camera straight to the
database).

So, anyway, looking forward to seeing some progress in this department... :-)
--
Lamar Owen
Director of Information Technology
Pisgah Astronomical Research Institute
1 PARI Drive
Rosman, NC 28772
(828)862-5554
www.pari.edu

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Sergey E. Koposov 2006-08-13 05:34:34 Re: segfault on rollback
Previous Message Jonah H. Harris 2006-08-13 03:52:24 Re: [PATCHES] Adding fulldisjunctions to the contrib