Duplicating a large database.

From: Dan Hrabarchuk <dan(at)kwasar(dot)biz>
To: "pgsql-admin(at)postgresql(dot)org" <pgsql-admin(at)postgresql(dot)org>
Subject: Duplicating a large database.
Date: 2004-10-06 00:26:16
Message-ID: 1097022376.3710.55.camel@dan.geotivity.net
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-admin

I need to move a large database to new machine. I'd also like to upgrade
my postgres server from 7.3 to 7.4 (at least).

My plan is to set up a query log on my current postgres server. Then I
will make a copy of the database using pg_dump. When the dump is done I
will to shut down the original server, then run the query log on the new
server. This should bring up to date, and reduce the space used.
(Vacuuming is to way too slow).

I've been trying to find the "query logger" portion of the plan. I've
found mini-projects like taillog
(http://gborg.postgresql.org/project/tablelog/projdisplay.php) but these
over complicate what I want. I simply want to reply all the inserts,
updates and deletes that happen on the orignal server while the new
server is dumping/restoring.

What can I use? A text file that I can pipe into psql would be AWESOME!
Any ideas? Any half finished projects that I can complete (or at least
get working enough :)?

Thanks for the help

Dan Hrabarchuk

Responses

Browse pgsql-admin by date

  From Date Subject
Next Message John McBride 2004-10-06 03:12:22 Re: fedora core 2 postgresql regression tests fail
Previous Message Simon Riggs 2004-10-05 20:55:31 Re: Enabling --log-update