Taking a Fresh Look at Hyper-V Clusters (Part 3)

If you would like to read the other parts in this article series please go to:

Introduction

In the previous two articles, I talked about some of the new failover clustering features that are included in Windows Server 2012 R2, especially as they relate to Hyper-V. In this article, I want to continue the discussion by walking you through the process of setting up a Hyper-V cluster.

Before I Begin

Before I get started, I want to point out that the Failover Clustering service is very picky with regard to the Hyper-V virtual switch. Any fault tolerant virtual machine that is running within a Hyper-V cluster could potentially fail over to any other node in the cluster. That being the case, all of the cluster nodes must be comparably equipped. This is especially true for Hyper-V virtual switches and for virtual Fibre Channel adapters.

Imagine what would happen if a virtual machine were to fail over to a cluster node that did not have a copy of the virtual switch that the virtual machine had been connected to on the previous node. The virtual machine would be unable to establish network connectivity on the new node. This lack of connectivity would completely negate the benefits of being able to fail over.

That being the case, every node in a Hyper-V cluster must have identical Hyper-V virtual switches. In fact, you won’t even be able to validate a Hyper-V cluster unless identical switches exist on each cluster node. If you happen to be using a virtual SAN, then every cluster nodes must also have matching virtual Fibre Channel adapters.

Building a Cluster

For the purposes of this article, I am going to build a Hyper-V cluster consisting of three Hyper-V hosts, each running Windows Server 2012 R2. In this particular case, each of the servers has its own internal storage array, and the servers all exist on a common subnet. That being the case, this particular cluster will not make use of a cluster shared volume.

Installing the Failover Clustering Feature

The first step in building a failover cluster is to install the Failover Clustering feature. For the purposes of this article, I am going to assume that the Hyper-V role is already installed.

Begin the process by opening the Server Manager and choosing the Add Roles and Features command from the Manage menu. When the Add Roles and Features Wizard appears, click Next to bypass the Welcome screen. You should now see the Installation Type screen. Select the Role Based or Feature Based Installation option and click Next.

Now, you will be taken to the Server Selection screen. Choose the local server and then click Next. You should now be taken to the Server Roles screen. We don’t need to do anything here (assuming that the Hyper-V role is already installed), so just click Next.

Now you will be taken to the Features screen. Select the Failover Clustering feature. Click Next and you will be taken to the Confirmation screen. Take a moment to verify that the information shown on the Confirmation screen is correct, and then click the Install button. When the installation process completes, you can close the wizard.

Before continuing on, you must repeat this process and install the Failover Clustering feature on each Hyper-V host that will act as a failover cluster node.

Creating a Cluster

Now that the Failover Clustering feature is installed on all of your cluster nodes, it is time to actually build a failover cluster. This is done through a native Windows Server tool called the Failover Cluster Manager. You can access the Failover Cluster Manager by choosing the Failover Cluster Manager command from the Server Manager’s Tools menu. You can see what the Failover Cluster Manager looks like in Figure A.

Image
Figure A: You can use the Failover Cluster Manager to create the Hyper-V cluster.

To create a failover cluster, click on the Create Cluster link, found in the Actions pane. When you do, Windows will open the Create Cluster Wizard. Click Next to bypass the wizard’s Welcome screen.

The next screen that you will see asks you to enter the names of all of the servers that you want to include in the cluster. The easiest way to do this is to click on the browse button, enter the server’s computer name, and click OK. You must add each of the cluster nodes to the list, as shown in Figure B. It is worth mentioning however, that it is possible to add additional nodes to the cluster later on.

Image
Figure B: 
You must specify the names of the servers that you want to add to the cluster.

After specifying the host server names, click Next and you will be prompted to specify a name for the cluster. The cluster name must be unique. You can think of the cluster name as being similar to a computer name, except that the name refers to the cluster as a whole rather than referring to an individual computer.

This screen also requires you to enter an IP address for the cluster to use. Just as the cluster name refers to the cluster as a whole, the IP address is assigned to the cluster rather than being assigned to an individual cluster node. You can see what this screen looks like in Figure C.

Image
Figure C: You must assign a name and an IP address to the failover cluster.

Click Next and you will be taken to a confirmation screen. Make sure that all of the information shown on the screen appears to be correct. In most cases, you will also want to make sure that the Add all Eligible Storage to the Cluster check box is selected.

Click Next and the cluster will be created. When the process completes, click Finish.

Validating the Cluster

Even though the cluster has been created at this point, you will need to validate the cluster before you can start using it. To validate the cluster, select the listing for the cluster within the Failover Cluster Manager and then click on the Validate Cluster link. This will cause Windows to launch the Validate Configuration Wizard.

Click Next and you will see the Testing Options screen. Choose the option to run all tests and then click Next. If you are prompted to select any storage, it is a good idea to go ahead and test it as well. When you reach the confirmation screen, simply click Next to begin the testing process.

The validation tests take a few minutes to complete, but the process is not overly time-consuming. It is normal to receive some warnings within the test results, but there must not be any errors. Otherwise you will have to correct the errors prior to putting your cluster into service.

Conclusion

In this article, I have walked you through the process of building a failover cluster for the purpose of hosting Hyper-V virtual machines. In the next article in this series, we will begin making virtual machines fault-tolerant, and I will show you some options for interacting with the virtual machines at the cluster level.

If you would like to read the other parts in this article series please go to:

Leave a Comment

Your email address will not be published.

Scroll to Top