Tell me more ×
Database Administrators Stack Exchange is a question and answer site for database professionals who wish to improve their database skills and learn from others in the community. It's 100% free, no registration required.
up vote 0 down vote favorite

I am planning to install two SQL Server 2008 instances on an active/passive cluster. Each instance is supporting the same application, but the DBs will be supporting specific geographic locations. I'd like to name the instances to reflect the name of the application and the site, factoring in that at least one instance will be a named instance.

My understanding is that the SQL Server Virtual Name has to be registered in AD and can only be used once. If this is true, I can not have instances on this cluster named \EAST and \WEST as the can't be registered in AD more than one, or can I?

Can I have two named instances on a cluster with the same base instance name or will the base instance name have to be unique for AD registration purposes?

share|improve this question
1  
I'm confused. Do you have two clusters, or one cluster with two instances? Because the former would be a standard failover cluster (active/passive) while the latter would be multi-instance failover (active/active). Also, is this application COTS or can you influence the architecture? Why not compensate for geography within the model itself? – swasheck Nov 15 '12 at 21:15
It is a COTS product and the original implementation had two uniquely named DBs (with identical schemas) on the same instance. This new approach appears to be their way of scaling out the architecture and tailoring access control by site. – MZDBA Nov 15 '12 at 22:13
Just so you know, clustering is not scaling. Clustering is HA. – swasheck Nov 15 '12 at 22:16

1 Answer

up vote 3 down vote

You are correct in noting that the virtual name needs to be distinct. I'm having a hard time understanding what you're describing so I'm going to try to lay it all out.

SINGLE INSTANCE FAILOVER CLUSTER

If you're truly looking for "Active/Passive" then you're going to need at least four servers - two nodes for each instance. Let's call the instance SQLPROD-EAST.

You could then have:

Node A: SQLPROD-EASTA
Node B: SQLPROD-EASTB

Virtual Name: SQLPROD-EAST

as your nodes.

The virtual name would be SQLPROD-EAST and you'd install the default instance of SQL Server would would take the name SQLPROD-EAST. You'd then replicate this topology with SQLPROD-WEST:

Node A: SQLPROD-WESTA
Node B: SQLPROD-WESTB

Virtual Name: SQLPROD-WEST

MULTI-INSTANCE FAILOVER CLUSTER

This is where your description got confusing because it seems like you began describing Active/Active. This might look like:

Node A: SQLPRODA
Node B: SQLPRODB

Virtual Name: SQLPROD

as your nodes.

The virtual name would be SQLPROD and you'd install two named instances of SQL Server SQLPROD\EAST and SQLPROD\WEST.

share|improve this answer
The multi-instance failover cluster approach is what I'm working with. The part that confuses me is whether you can reuse the virtual name when setting up the second named instance. I was under the impression that you need a new one, but it look like I can reuse it, correct? – MZDBA Nov 15 '12 at 22:08
You won't be re-using it, you'll only be using it once. The clustered nodes will need a virtual server name (SQLPROD, in this case). It is on that server name that you will be installing instances. You'll have to specifically choose to install named instances for each install. – swasheck Nov 15 '12 at 22:14

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.