You want to choose "Add node to a SQL Server failover cluster". When creating a Failover Cluster Instance (FCI), you will always do one initial cluster installation for the instance, then add nodes to that installation.
Keep in mind that Active/Active and Active/Passive as descriptors is very misleading. By it's very nature, each SQL Instance you install can only be active on a single node at any one time. To leverage multiple physical nodes at the same time, you will need to install multiple named FCIs to your cluster.
For example, let's assume a two node cluster. If we install a single named FCI to this, our layout would look like this:
NODE1: CLSTSQL01\SQLServer01 (Active)
NODE2:
This would be the result of you performing one "New SQL Server Failover Cluster installation" with the additional node additions for each node in your cluster. Now let's add a second instance to it so we can utilize each node. For this, we would install a second FCI to the cluster and end up with something like this:
NODE1: CLSTSQL01\SQLServer01 (Active)
NODE2: CLSTSQL02\SQLServer02 (Active)
Now, keep in mind that in a failover situation, on either node, the SQL instances could end up on the same node. For example, if NODE1 failed:
NODE1: <<DEAD>>
NODE2: CLSTSQL01\SQLServer01, CLSTSQL02\SQLServer02 (Active)
This is the basics of how you would arrange your instances. You will need to monitor your cluster if you want to make sure your instances are properly balanced across your nodes.
As per your comments in your question, what you need to do is two-fold:
- Add SQL Server Agent as a cluster resource type
- Add the SQL Server Agent cluster resource to the failover cluster resource group
For the first step, one way to do this is through PowerShell (utilizing the FailoverClusters module):
Import-Module -Name FailoverClusters
$ClusterName = "YourClusterName"
$FciClusterGroupName = "SQL Server (MSSQLSERVER)" # may need to modify to reflect your environment
# check to see if you have the SQL Server Agent cluster resource type
Get-Cluster -Name $ClusterName |
Get-ClusterResourceType |
Where-Object {$_.Name -eq "SQL Server Agent"}
# if not, add it
Add-ClusterResourceType -Name "SQL Server Agent" -Dll "sqagtres.dll"
Execute each of those blocks of code separately instead of just all at once. You'll be able to confirm whether or not that cluster resource type exists before just blindly attempting to add it.
Then for the second step, here's what you need to do to add the SQL Server Agent as a resource in your failover cluster resource group:
- Right-Click on your FCI resource group within the failover cluster manager
- Select "Add a resource" and select the SQL Server Agent resource type
- Rename the new cluster resource to "SQL Server Agent"
- Go into the new SQL Server Agent resource properties and set the virtual network name and instance name properties
- On the dependency tab, set SQL Server as its dependency resource
You may need to go into the registry at the following location and ensure those keys (one will contain the string "AGENT", another "SQL_ENGINE") are set to 1 (this is where it is recorded on what was installed correctly. 1 denotes correct installation):
HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Microsoft SQL Server\YOURINSTANCE\ConfigurationState
That should get you to a place where you can now install an additional node and have the SQL Server Agent service account appear. Of course, prior to starting ensure that you have a back out here (especially with the registry editing).
Best Answer
It will be a normal backup and restore operation (provided there are no other issues with the actual backup and/or restore).
The fact that you are going from a standalone instance to a failover cluster instance will not negatively affect an otherwise normal backup and restore.
The same rules will apply here, of course, as well as any other backup/restore operation regardless of standalone or FCI.