On Mon, 24 Feb 2003, Andrew Sullivan wrote:
> On Tue, Feb 25, 2003 at 01:03:38AM +1030, Justin Clift wrote:
> >
> > > b) is it possible to configure an auto-failover to the secondary
> > > database in case primary dies.
> >
> > Definitely. See the above URLs for further information.
>
> I have to say that we do not use it that way, and I cannot recommend
> such a strategy. (We're the ones using it for .org and .info.)
> Failing over is something of a pain in the neck, and it is not
> 100% reliable in cases where you have flakey hardware (as I learned
> to my chagrin recently).
>
> That said, the replication itself has not failed for us.
>
> PostgreSQL, Inc. sells a version which is somehow linked to a Linux
> system; I don't know very much about it, so it may work perfectly
> well. But that is _not_ the system we are using.
The newest version uses an Lanux (not Linux) layer ... its a set of
monitoring tools that are used to detect failure of the primary server,
and auto-switch over to the secondary ... we're workign right now on
extending that further to include fail-over mid-transaction, so that the
failover itself is more transparent to the end user ...