We have one project which has Failover Cluster with two nodes and one standalone server in AlwaysON. For example:

Project 1

  • Node1 (ip:
  • Node2 (ip: in Failover Cluster with clustered IP:
  • Node3 (ip: Clustered IP:( and it's AlwaysON IP:

And now we have another project as follows (Cluster IP:

Project 2

  • Node1 (ip:
  • Node2 (ip:

We want to add third Node3 (ip: into Project2 cluster and create AlwaysON. But don't want change Cluster IP. In Project1 we have two WFCI. Can we create with one WFCI and Clustered IP?

  • Please try to clarify the question. It is a bit confusing -- what does project #1 have to do with project #2? How do you have a standalone server in AlwaysOn--that's impossible. – Tony Hinkle Oct 8 at 12:11
  • Project#1 and project#2 are different projects. It's for example. Standalone I mean, that it isn't in FCI. It's in disaster recovery side. – Farik013 Oct 8 at 12:43
  • A node can only be in one cluster at a time. If Node3 is in CLUSTER-1, you have to remove it in order to add it to CLUSTER-2. – Tony Hinkle Oct 8 at 13:28

A node can only be in one cluster at a time. If Node3 is in CLUSTER-1, you have to remove it in order to add it to CLUSTER-2.

If you want to use Node3 with availability groups in both projects, then all five servers will need to be in the same cluster.

There are other options that would allow you to use Node3 for DR purposes without having all 5 nodes in one cluster. For instance, you could perhaps use log shipping to replicate the databases on CLUSTER-2 to Node3. More detail on the requirements would be needed for someone to be able to suggest what would be best.

  • The more nodes in your cluster the more likely you are going to get corruption on your databases. – PPann Oct 8 at 19:45
  • No, you didn't understand me. Maybe I didn't say correct. I need one cluster with three Nodes. Node1 and Node2 are FCI. Now I need to add third Node to this cluster to use AlwaysOn. So I will have Node1 and Node2 with the same storage and Node3 with own storage. Can I do that? – Farik013 Oct 9 at 6:45
  • Yes, it's referred to as FCI+AG. docs.microsoft.com/en-us/sql/database-engine/…. Your question is confusing because it sounds like that's what you have with project #1. If you put the project #2 nodes in the same cluster, you can use Node3 as a replica in both availability groups. – Tony Hinkle Oct 9 at 11:13
  • Yes, I've red this article. But there isn't my question specifically or maybe I don't see. So I ask here. I have Node1 and Node2 with cluster IP like If I add third Node3 to this cluster and create AlwaysOn between Node1/Node2 and Node3, will this HADR work? Do I need second cluster IP or I can change between Primary/Secondary with physical IP? – Farik013 Oct 9 at 12:59
  • If you want to have a listener for the availability group, the listener will create another cluster IP address. The IP address for the FCI is in the cluster group (role) with the disks and SQL Server service objects, which cannot fail over to Node3 because it won't be able to own the disks and services. – Tony Hinkle Oct 9 at 13:27

A better solution is to have several physical ESX/HyperV/VMWare servers that host only your VM with the SQL instance and have the VM host switch, this gives you a better performance than a clustered server as you have hardware commitment and not a distributed transaction and failover is far easyer to manage, also you do not need Enterprise licenses.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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