Re: Strategy for moving a large DB to another machine with least possible down-time - Mailing list pgsql-general
På søndag 21. september 2014 kl. 13:51:00, skrev Bill Moran <wmoran@potentialtech.com>:
On Sun, 21 Sep 2014 13:36:18 +0200 (CEST)
Andreas Joseph Krogh <andreas@visena.com> wrote:
> Hi all. PG-version: 9.3.5 I have a DB large enough for it to be impractical
> to pg_dump/restore it (would require too much down-time for customer). Note
> that I'm noe able to move the whole cluster, only *one* DB in that cluster.
> What is the best way to perform such a move, can i use PITR, rsync +
> webl-replay magic, what else? Can Barman help with this, maybe? Thanks. --
I've used Slony to do this kind of thing with great success in the past.
The biggest advantage of slony is that you can install it without stopping the
DB server, wait patiently while it takes however long is needed to synch up
the two servers without having much impact (if any) on operations, then switch
over when you're ready. The disadvantage to Slony is that the setup/config is
a bit involved.
I see this limitation in Slyny:
http://slony.info/documentation/2.2/limitations.html
Slony-I does not automatically replicate
Changes to large objects (BLOBS)
Changes made by DDL commands
Changes to users and roles
Not being able to replicate BLOBS is a show-stopper for me as we have lots of them.
Seems PITR is my only option?
-- 
Andreas Joseph Krogh
CTO / Partner - Visena AS
Mobile: +47 909 56 963
Attachment
pgsql-general by date: