| From: | "Scott Marlowe" <scott(dot)marlowe(at)gmail(dot)com> |
|---|---|
| To: | rexmabry(at)yahoo(dot)com |
| Cc: | "pgsql-admin(at)postgresql(dot)org" <pgsql-admin(at)postgresql(dot)org> |
| Subject: | Re: replica of database |
| Date: | 2008-09-11 15:43:05 |
| Message-ID: | dcc563d10809110843q24a56ed2i78f1a43ab2473485@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-admin |
On Thu, Sep 11, 2008 at 7:42 AM, Rex Mabry <rexmabry(at)yahoo(dot)com> wrote:
> Version 8.3 has pg_standby. Does anyone know how well that works for
> failover?
pg_standby is part of the Point in time recovery suite, which can be
used to create a cold (recovery on comand) or warm standby (continuous
recovery) server ready to take over with a few commands. It's quite
different from slony, in that there's not hot read-only standby, so
there's no load balancing, however, it usually produces less load on
the master db, so that's a plus.
PITR is quite well tested and used in a lot of production
environments. It has very different design objectives than slony or
pgpool or pgbouncer, which all allow you to have multiple live servers
at once.
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Marc Fromm | 2008-09-11 16:57:51 | starting postgres on red hat |
| Previous Message | Brad Nicholson | 2008-09-11 15:11:09 | Re: replica of database |