From: | "Jaume Sabater" <jsabater(at)gmail(dot)com> |
---|---|
To: | pgsql-admin(at)postgresql(dot)org |
Subject: | Re: suggestions/hints for multimaster solution |
Date: | 2009-01-09 11:13:15 |
Message-ID: | 3786f7bb0901090313l2780aba8ia87d0a41e82acdec@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-admin |
On Fri, Jan 9, 2009 at 11:32 AM, Gerd König <koenig(at)transporeon(dot)com> wrote:
> Perhaps we should have a look at
> * pg_pool(II)
I can only speak of pgpool-II. It's working fine for me in a setup
which is exactly the same one you have (change OpenSuse for Debian,
but the rest is the same). The thing is that pgpool-II itself can
become a single point of failure unless you provide HA for it (I am
doing so with Heartbeat), but IMHO that would then be a failover
solution, not a multimaster (from pgpool-II's point of view, not the
databases behind pgpool-II).
Next thing I want to do is try Slony-I with pgpool-II, but I think
that is not a multimaster solution either.
Of course we may have different points of view on what multimaster mean :)
If you plan on using pgpool-II, take into consideration that
replication is done at a SQL query level, which is cool, but has its
limitations. If you use Slony-I inside pgpool-II I think you can take
over those limitations, but the solution is not so... neat, so to
speak :)
/discuss
--
Jaume Sabater
http://linuxsilo.net/
"Ubi sapientas ibi libertas"
From | Date | Subject | |
---|---|---|---|
Next Message | Jaume Sabater | 2009-01-09 12:46:26 | Re: suggestions/hints for multimaster solution |
Previous Message | Gerd König | 2009-01-09 10:32:48 | suggestions/hints for multimaster solution |