• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 5169
  • Last Modified:

Active Active SQL 2008 cluster

has anyone set one of these up?
Can I get a general outline?
I suspect it is something like this, but would like some confirmation;
setup MSCS Cluster and test working correctly
install failover cluster sql2008 on first node. (I should now have a Active Passive cluster)
(do I need to use a differant shared disk for 2nd active? second IP?)
install sql2008 failover cluster on second node.

any input here would be much appreciated.
  • 3
  • 2
1 Solution
Just wondering why you want A/A when most people see a two-node as A/P ?

In essence you're correct; but when you install SQL2008 for the first time; you need to give it an instance name; not the default.

You would need 2nd shared disk, names, IP's etc.

Jim P.Commented:
Good quick primer:
Getting Started with SQL Server 2008 Failover Clustering

Technically there is no such thing as Active/Active SQL. A named SQL instance can only be active on one node at a time. You can have multiple instances running on the cluster, but each will be holding its own distinct set of databases.

Each distinct instance will need a separate drive, IP, and network name.

You should have a default instance though. some applications can't handle the MyServer\MyInstance naming for SQL.
loftywormAuthor Commented:
I am choosing an A/A because I want to do a bit of load balancing and prep for  patching.  We have 25+ DB's we will put on it, and I do not want to bring them all down for a patch.  So I figure I will have an A/A.

So it should be like this then?
ClusterName------Node1----SQLClusterName1---IP1,disk1,networkname1---Default instance
 ClusterName------Node2----SQLClusterName2---IP2,disk2,networkname2---Second instance
This should allow me to take one of the nodes down for patching/maintenance and not effect the clients, correct?

and clients will connect to ClusterName/Instance, or networkname#/instance?
Nothing ever in the clear!

This technical paper will help you implement VMware’s VM encryption as well as implement Veeam encryption which together will achieve the nothing ever in the clear goal. If a bad guy steals VMs, backups or traffic they get nothing.

loftywormAuthor Commented:
DB load is design so it can be handled by 1 node with little impact on performance.
Jim P.Commented:
>> I am choosing an A/A because I want to do a bit of load
>> balancing and prep for  patching.

That won't help for patching.

When you install a new instance you have to do it from the inactive node. When you install a patch it has to be done from the active node and the other node has to have all instances inactive. You would have to flop all the instances to one node.

The load balancing would probably help -- split the databases between two instances.

>> and clients will connect to ClusterName/Instance, or networkname#/instance?

The default instance will still be just NetworkName. The other would be NetworkName\InstanceName
Jim P.Commented:
Can I ask why a B grade?

Featured Post

Prep for the ITIL® Foundation Certification Exam

December’s Course of the Month is now available! Enroll to learn ITIL® Foundation best practices for delivering IT services effectively and efficiently.

  • 3
  • 2
Tackle projects and never again get stuck behind a technical roadblock.
Join Now