I've never done it on NT/2000 - however we do it all the time on IBM RS6000
using HACMP in mutual take-over - works like
a charm...
For NT/2000 you'll need a MS Cluster with Oracle FailSafe (if I'm not
mistaking) - luckely I have one of those in our lab and I'm going
to set up a test environment just as you described ... it should work -
don't know how stable it'll be but I'll soon find out
"sueZ" <sue_zirbes_at_fws.gov> wrote in message
news:3f75e12b.0110310955.10696a6a_at_posting.google.com...
> We have an application for which we would like to cluster a pair of
> servers, using a SAN storage device for the databases. Each of the
> pair would be serving 4 databases, with the idea that if one of the
> servers in the pair went down, it would fail over to the other server,
> which would then serve all 8 databases until the failed server could
> be made operational and fail back could occur.
>
> Does anyone have an opinion on this or an architecture like this
> running in their shop?
>
> -sueZ
Received on Wed Oct 31 2001 - 14:43:49 CST
