Hi, If I have two machine in a cluste
- if I install SQL on one will the installation replicate on the other or do I have to install both seperatly
- if I have SQL operating on both nodes and I write SQL to one will it replicate over automatically (does it remove the need for SQL redundancy
- would WebPages do the same

sorry for the newbie questions but was finding it hard to get specific answers, any help would be appreciated
Steve

SQL Cluster installation puts the binaries on whatever host nodes you
specify. You can add or remove nodes later if you need to. The clustered
data store is accessable from both nodes, but is arbitrated so that only one
host at a time owns it owns. The entire data store shifts to another node
during a failover. A failover appears like a SQL stop/start to the outside
world. If you have a SQL instance running on a different host node in the
same cluster, it will have a different set of databases and will appear like
(because it is) a separate SQL server.

I do not understand your question about WebPages.

--
Geoff N. Hiten
Microsoft SQL Server MVP
Senior Database Administrator
Careerbuilder.com

I support the Professional Association for SQL Server
www.sqlpass.org

"Steve" <anonymous (AT) discussions (DOT) microsoft.com> wrote

Setup will install SQL Server 2000 on both nodes. That is, it will install
the binaries and utilities on a local drive on each node and the data and
log files on the shared drive.

There is only one set of shared databses between the nodes. There is no
data redundancy; only high availability. So if you want data redundancy you
will need to use something like replication or log shipping.

Rand
This posting is provided "as is" with no warranties and confers no rights.

Even in a SQL cluster, you have only one set of data (mounted by one node or
the other, but not both).

If you lose the storage solution or corrupt the data, you're toast. That's
where replication comes in.

Look at either NSI product (www.nsisoftware.com).

GeoCluster splits the storage under MSCS, so each node has its own copy -
although only one is active at a time.

DoubleTake replicates data between servers, and also from the active MSCS
node, to another server. This can be built in such a way that you could do
backups or snaps from the redundant server (which gives you some additional
restore and rollback capabilities).

jason


"Steve" <anonymous (AT) discussions (DOT) microsoft.com> wrote