Re: Best high availability solution ?

> I guess the users would start over anyway. So easiest if you
> provide a copy of the app with that other connection and
> signal them if the first server dies to just close the first
> and start the backup-application.

This requires that all the users do the right thing. Problem is, what if
there is a network problem for some users only, and they incorrectly
conclude that the server is down?

I prefer the approach of keeping the backup server up to date, whether using
PITR or Slony or your own home-grown synching, and then changing IPs. My
process involves someone making the decision that server is indeed down,
then UNPLUGGING it from the network, then changing the IP of the backup.
Actually bringing the backup online involves a bit more than changing the
IP, because PG is not the only service running on it, but there's a simple
script for the user to run.

Of course this requires that the backup be on the same subnet as the server,
a restriction which is not shared by the alternate app/DSN nor the DNS
techniques...

--
Scott Ribe
scott_ribe(at)killerbytes(dot)com
http://www.killerbytes.com/
(303) 722-0567 voice