On Tue, Nov 25, 2008 at 8:48 PM, Jamie Tufnell <diesql@googlemail.com> wrote:
> Hi,
>
> We have been wanting to HA our pgsql server for some time now. We're
> actually reaching the limits of our current server now, so there is a
> pressing need to do something to ease the load too.
>
> There are several individual solutions to both of these problems. I
> was hoping I could get some advice from someone more experienced
> though, about what the simplest most effective way of moving forward
> would be? Hopefully I can squash both HA and LB with the same
> solution.
>
> So far I've narrowed it down to
>
> Slony-I master/slave replication with failover
We use this where I work and it works quite well. We have some custom
scripts that detect failure of the master db and initiates failover to
the slave if necessary. We do this from the applicaiton level.
> pgpool-II with heartbeat for failover. The pgpool-II configuration
> matrix (http://pgpool.projects.postgresql.org/) isn't terribly clear
> (there's no legend) but it leads me to believe failover and load
> balancing are mutually exclusive options. Is that so?
Seems to be. I haven't played with pgpool in a long time so don't
take my word as authoritative.