In a two-instance scenario, each instance has access only to its own files.
The instances can run on either physical node. If one node fails, the
instance on which it was running fails over to the other node. Now, you are
running both instances on the same node. The "failed" instance is now up
and running and has access to its own files. An instance runs on only one
node at a time.
The BOL have a lot to say on clustering.
--
Tom
---------------------------------------------------------------
Thomas A. Moreau, BSc, PhD, MCSE, MCDBA
SQL Server MVP
Columnist, SQL Server Professional
Toronto, ON Canada
www.pinnaclepublishing.com [quoted text, click to view] "PeterD" <PeterD@discussions.microsoft.com> wrote in message
news:6B6314D2-CA66-42E3-BC94-5605ABEE16B0@microsoft.com...
I have 2 DELL Windows 2003 Enterprise server machines with SQL 2000
Enterprise Edition and 1 PowerVault 220S SCSI Array set in Cluster mode
hooked up to both of these machines.
I want to know is it possible to use this configuration with SQL in a ACTIVE
/ ACTIVE state.
By Active/Active, I am looking to have both physical nodes actively serving
updates to the same database. A change via node1 is instantly accessible by
node2. If one node fails, the remaining node picks up the load (doubling
load
roughly).
I was assuming that the actual MDF of the database file was shared in this
case (a partition on the shared SCSI Array that hosts this file that can be
viewed and updates from either SQL Node). I am not sure if this is the case.
If it is, what happens with the LDF file?
Are there any examples or documents about how to setup such a senario or the
alternative that would suit this case?