Migration of Highly Available Virtual Machines Using the Migrate a Cluster Wizard
Updated: November 6, 2013
Applies To: Windows Server 2012 R2, Windows Server 2012
This topic provides a process overview and step-by-step instructions for migrating a Hyper-V highly available virtual machine (HAVM) from a failover cluster running Windows Server 2008, Windows Server 2008 R2, or Windows Server 2012 to a failover cluster running Windows Server 2012 by using the Migrate a Cluster Wizard in Failover Cluster Manager. This is accomplished by migrating the clustered Virtual Machine role from the old cluster to the new cluster. The migrated HAVMs use the same storage that they used in the old cluster. The wizard cannot migrate virtual machines to new storage.
You can use either of the two migration scenarios to migrate an HAVM: migrate between two multi-node clusters or perform an in-place migration.
You can also use this method to migrate HAVMs to a failover cluster running Windows Server 2012 R2 from a failover cluster running Windows Server 2008 R2 with Service Pack 1 (SP1), Windows Server 2012, or Windows Server 2012 R2. In Windows Server 2012 R2, the name of the Migrate a Cluster Wizard was changed to Copy Cluster Roles, and the wizard is opened by using the Copy Roles action. For consistency with labeling in Windows Server 2012 R2, the items being migrated are referred to as clustered roles instead of clustered services and applications. However, the steps for performing the wizard-based migration are the same.
The Migrate a Cluster Wizard in Windows Server 2012 can migrate highly available virtual machines running on any of the following Windows Server operating system versions to Windows Server 2012:
Windows Server 2008 with Server Pack 2 (SP2)
Windows Server 2008 R2 with Service Pack 1 (SP1)
Windows Server 2012
Before you migrate any clustered role or service, you should install the latest operating system updates on all nodes in the old and new clusters.
To migrate a highly available virtual machine from one failover cluster to another, you use the Migrate a Cluster Wizard in Failover Cluster Manager to migrate the clustered Virtual Machine role. After you select the Virtual Machine role, you select the role instances (virtual machines) that you want to migrate.
Be aware that if you migrate one virtual machine that resides on a Cluster Shared Volume (CSV) volume, the wizard migrates all virtual machines on that volume. The wizard allows you to update virtual network settings on the virtual machines to the network settings on the new cluster. You cannot use the wizard to migrate virtual machines to new storage. This restriction does not apply if you are migrating a Scale-out File Server cluster. A Scale-out File Server cluster does not use CSV volumes, so you can migrate one virtual machine at a time.
To prepare the virtual machines for the migration, you must merge or discard all shadow copies on the volumes that contain the virtual machines. To prepare the new cluster for the migration, you must add the Hyper-V role to the cluster nodes and configure storage and virtual networks on the cluster.
After the migration, you will need to take the virtual machines offline on the old cluster, follow your plans to mask the volumes that contain the virtual machines to the old cluster and unmask the volumes to the new cluster, and then bring the virtual machines online on the new cluster. After you bring the virtual machines on the new cluster online, you must also install the latest integration services on the virtual machines.
To schedule local backups of the virtual machines, you will need to configure Volume Shadow Copy Service (VSS) tasks, which the wizard does not migrate. If you migrated from a Windows Server 2012 failover cluster, you will also need to configure Hyper-V Replica Manager settings and Cluster-Aware Updates (CAU) if you were using them; those settings also do not migrate.
For a step-by-step walk-through, with screenshots, of migrating a Hyper-V host cluster from Windows Server 2008 R2 to Windows Server 2012 by using the Migrate a Cluster Wizard, see the blog entry How to Move Highly Available (Clustered) VMs to Windows Server 2012 with the Cluster Migration Wizard on MSDN.
There will be a brief service interruption during the migration. To minimize the effects on users, schedule the migration during a maintenance window. We also recommend that you pretest and verify the migration before you migrate the virtual machines in your production environment.
You cannot use live migration to migrate a highly available virtual machine to a new failover cluster.
To migrate a clustered service or application by using the Migrate a Cluster Wizard, you must be a local administrator on the destination failover cluster and on the cluster or cluster node from which you are migrating.
While you prepare to migrate the virtual machines, the virtual machines can remain online and continue providing service.
To prepare virtual machine storage for migration
Before you begin working with shadow copies, you should back up all volumes that are attached to the virtual machine(s).
Merge or discard all shadow copies for the volumes that store the virtual machines.
Ensure that no virtual machines that you do not want to migrate share a CSV volume with virtual machines that you plan to migrate. If you migrate one virtual machine on a CSV volume, the Migrate a Cluster Wizard migrates all virtual machines on that volume.
To prepare the old failover cluster for the migration
Install the latest operating system updates on each cluster node. A Windows Server 2008 failover cluster must be running Windows Server 2008 SP2 or later. A Windows Server 2008 R2 failover cluster must be running Windows Server 2008 R2 SP1 or later.
To prepare the new failover cluster for the migration
To create the new failover cluster in Windows Server 2012, use Failover Cluster Manager, or use the New-Cluster cmdlet in Windows PowerShell (for information, see New-Cluster). For a detailed description of the steps for preparing a new failover cluster, see Migration Between Two Multi-Node Clusters or In-Place Migration for a Two-Node Cluster.
Add the Hyper-V role to each cluster node.
Configure virtual switches in Hyper-V.
If you are migrating from a Windows Server 2008 or Windows Server 2008 R2 failover cluster, check with your hardware vendor to ensure that the existing storage is supported in Windows Server 2012.
You do not need to configure storage for the virtual machines on the new cluster before you run the wizard. The Migrate a Cluster Wizard will migrate existing storage settings to the new cluster. After the wizard completes, you will mask the storage from the old cluster and then unmask the storage on the new cluster.
Install the latest operating system updates on each cluster node.
You will use the Migrate a Cluster Wizard in Failover Cluster Manager to migrate the virtual machines to the new failover cluster. This is done by migrating the clustered Virtual Machine role. After the virtual machines are migrated, you must make the storage available to the new cluster before you bring the virtual machines online.
Before you migrate the highly available virtual machines
Ensure that the virtual switches are configured on host operating systems in the new cluster.
Prepare for a brief service interruption on the workloads running on the virtual machines. Live migration is not supported during migration of a virtual machine to a new host cluster.
To migrate the virtual machines to a new failover cluster
Log on to any node in the new failover cluster with an Administrator account.
From the Start screen or Server Manager (Tools), open Failover Cluster Manager.
In the console tree, expand Failover Cluster Manager in the console tree, and select the cluster that you want to migrate the virtual machines to.
If the new cluster is not displayed, right-click Failover Cluster Manager, click Connect to Cluster, and then select the cluster to which you want to migrate the virtual machines.
With the destination cluster selected, click Migrate Roles.
The Migrate a Cluster Wizard opens.
Review the instructions on the Before You Begin page, and click Next.
On the Specify Old Cluster page, enter the name or IP address of the source cluster, or use the Browse button to find the cluster, and then click Next.
The wizard connects to the cluster and displays the roles and features that can be migrated. For virtual machines, each virtual machine “role” is listed under the cluster shared volume that stores the virtual machine.
On the Select Services and Applications page, click View Reports, and review the resources that can and cannot be migrated. Note that Available Storage and Cluster Group are never available for migration, and always have a Failed result. For all other resources, review any Warning results to identify any and resolve any issues that might prevent a successful migration. Then close the report.
On the Select Services and Applications page, select each highly available virtual machine that you want to migrate, and then click Next. If a volume stores more than one virtual machine, and you select any virtual machine on that volume, the wizard will migrate all virtual machines on that volume.
On the Customize Virtual Machine Networks page, optionally select a select virtual switch for the virtual machines to use on the destination host cluster. If you do not select a virtual switch, the wizard retains the default switch that is selected automatically the first time the virtual machine starts on its new host.
On the Configuration page, review your settings. Then click Next to start the migration.
After the migration completes, review the Post-Migration Report to verify that the virtual machines were migrated. Then click Finish.
In Failover Cluster Manager, verify the status of the migrated virtual machines and the related resources:
In the console tree, click the name of the new failover cluster, and then click Roles. You should see the migrated virtual machine (roles) in the Roles pane. The virtual machines will be turned off.
Click a virtual machine to display the associated resources at the bottom of the window. For a newly migrated virtual machine, the resources have been registered but are not online.
Before you can start the virtual machines, you must remap the storage to the new cluster and then bring the storage online.
To complete the migration
Prepare for clients to experience downtime, probably briefly.
Shut down the old cluster to ensure that no one will attempt to start the virtual machine during migration and no connections will be made from storage.
At no time should a virtual machine be running on both the old cluster and the new cluster. A virtual machine that runs on both the old cluster and the new cluster at the same time might become corrupted. You can run a virtual machine on the old cluster while you migrate it to a new cluster with no problems; the virtual machine on the new cluster is created in a Stopped state. However, to avoid corruption, it is important that you do not turn on the virtual machine on the new cluster until after you stop the virtual machine on the old cluster.
To complete the transition for the storage:
Make the CSV volume that stores the virtual machines inaccessible to the old cluster, and then make them accessible to the new cluster.
After you move the storage to the new cluster, in Disk Management, bring the CSV volume and Virtual Machine Configuration resource for each virtual machine online.
At this point, you should be able to start the virtual machines. To start the virtual machines in Failover Cluster Manager, display and select the virtual machine role, and then click Start Role.
Install the latest integration services on the new virtual machines. You might need to restart the virtual machine to complete the integration services update.
The Migrate a Cluster Wizard does not migrate Volume Shadow Copy Service (VSS) tasks, Hyper-V Replica Broker settings, Task Scheduler settings, and Cluster Aware Updating (CAU) settings. If you were using any of these features on the old cluster, you will need to configure them on the new cluster.
After you complete the migration, you should bring the virtual machine online, make sure the services that the virtual machine was providing on the old cluster are still available and working as expected, and test failover for the virtual machine on the new cluster. Verify that you can connect to the virtual machines by using Remote Desktop or Virtual Machine Connection. For detailed steps for verifying a successful role migration and testing failover, see Migration Between Two Multi-Node Clusters.