This section lists the clustered services or applications (clustered resources) that can be migrated to a cluster running Windows Server 2008 R2.

Important

You cannot use the Migrate a Cluster Wizard to migrate settings for virtualized servers, mail servers, database servers, and print servers, or any other resources that are not listed in the following subsections. Other migration tools exist for some of these applications. For information about migrating mail server applications, see http://go.microsoft.com/fwlink/?LinkId=91732 and http://go.microsoft.com/fwlink/?LinkId=91733.

Resources for which the Migrate a Cluster Wizard performs most or all of the migration steps

After you use the Migrate a Cluster Wizard to migrate the settings of the following resources to a failover cluster running Windows Server 2008 R2, few or no additional steps are needed before the resources can be brought online.

	Caution
	If new storage is used, you must handle the copying or moving of data or folders during a migration from your shared volumes. The wizard for migrating the settings of clustered resource groups does not copy data from one location to another.

• File Server or File Share resources: You can migrate settings for a clustered file server (or from a Windows Server 2003 cluster, a File Share resource group) and for the associated Physical Disk, IP Address, and Network Name resources.
  
  When you migrate from a cluster running Windows Server 2003, the Migrate a Cluster Wizard automatically translates all File Share resource groups to a single clustered file server (with multiple File Share resources within it) in Windows Server 2008 R2. Therefore, some resources might look different after the migration. The following table provides details:
  
  

  Resource as Seen in a Server Cluster Running Windows Server 2003	Migrated Resource as Seen in a Failover Cluster Running Windows Server 2008 R2
  One File Share resource	One File Server resource
  Multiple File Share resources	Multiple File Share resources within a single clustered file server (resource group)
  File Share resource with DFS root	Distributed File System resource and File Share resource (both within a clustered DFS Server)

• Physical Disk: You can migrate settings for Physical Disk resources other than the quorum resource.
  
  You do not need to migrate the quorum resource. When you run the Create a Cluster Wizard, the cluster software automatically chooses the quorum configuration that will provide the highest availability for your new failover cluster. You can change the quorum configuration settings if necessary for your specific environment. For information about changing settings (including quorum configuration settings) for a failover cluster, see Modifying Settings for a Failover Cluster.
  
  
• IP Address: You can migrate IP Address settings other than the cluster IP address. IP addresses are eligible for migration only within the same subnet.
  
  
• Network Name: You can migrate Network Name settings other than the cluster name. If Kerberos authentication is enabled for the Network Name resource, the wizard will prompt you for the password for the Cluster service account that is used by the old cluster.
  
  

Resources for which the Migrate a Cluster Wizard might not perform all of the migration steps

After you use the Migrate a Cluster Wizard to migrate the settings of the following resource groups to a failover cluster running Windows Server 2008 R2, some additional steps might be needed before the resources can be brought online, depending on your original configuration. The migration report indicates what steps, if any, are needed for these resource groups:

• DHCP Service
  
  
• Distributed File System Namespace (DFS-N)
  
  
• Distributed Transaction Coordinator (DTC)
  
  
• Internet Storage Name Service (iSNS)
  
  
• Message Queuing
  
  
• NFS Service
  
  
• WINS Service
  
  
• Generic Application
  
  
• Generic Script
  
  
• Generic Service
  
  

The wizard provides a report that describes the additional steps that are needed. Generally, the steps you must take include:

• Installing server roles or features that are needed in the new cluster (all nodes).
  
  
• Copying or installing any associated applications, services, or scripts on the new cluster (all nodes).
  
  
• Ensuring that any data is copied.
  
  
• Providing static IP addresses if the new cluster is on a different subnet.
  
  
• Updating drive path locations for applications if the new cluster uses a different volume letter.
  
  

The resource settings are migrated, as are the settings for the IP Address and Network Name resources that are in the resource group. If there is a Physical Disk resource in the resource group, the settings for the Physical Disk resource are also migrated.

For this migration scenario, there are three phases:

1. Install two or more new servers, run validation, and create a new cluster. For this phase, while the old cluster continues to run, install Windows Server 2008 R2 and Failover Clustering on at least two servers. Create the networks the servers will use, and connect the storage. Next, run the complete set of cluster validation tests to confirm that the hardware and hardware settings can support a failover cluster. Finally, create the new cluster. At this point, you have two clusters.
   
   Additional information about connecting the storage: If the new cluster is connected to old storage, make at least two logical unit numbers (LUNs) or disks accessible to the servers, and do not make those LUNs or disks accessible to any other servers. (These LUNs or disks are necessary for validation and for the disk witness, which is similar to, although not the same as, the quorum resource in Windows Server 2003.) If the new cluster is connected to new storage, make as many disks or LUNs accessible to it as you think it will need.
   
   The steps for creating a cluster are listed in Checklist: Create a Failover Cluster.
   
   
2. Migrate settings to the new cluster and determine how you will make any existing data available to the new cluster. When the Migrate a Cluster Wizard completes, all migrated resources will be offline. Leave them offline at this stage. The new cluster will remain online and continue serving clients. If the new cluster will reuse old storage, plan how you will make the storage available to it, but leave the old cluster connected to the storage until you are ready to make the transition. If the new cluster will use new storage, copy the appropriate folders and data to the storage.
   
   
3. Make the transition from the old cluster to the new. The first step in the transition is to take clustered services and applications offline on the old cluster. If the new cluster uses old storage, follow your plan for making LUNs or disks inaccessible to the old cluster and accessible to the new cluster. Then, regardless of which storage the new cluster uses, bring clustered services and applications online on the new cluster.
   
   

For this migration scenario, there are four phases:

1. Install a new server and run selected validation tests. For this phase, allow one existing server to keep running Windows Server 2003, Windows Server 2008, or Windows Server 2008 R2 and the Cluster service, while you begin the migration process. Evict the other server from the old cluster, and then install Windows Server 2008 R2 and the Failover Clustering feature on it. On that server, run all tests that the Validate a Configuration Wizard will run. The wizard will recognize that this is a single node without storage and limit the tests that it runs. Tests that require two nodes (for example, tests that compare the nodes or that simulate failover) will not run.
   
   Note that the tests that you run at this stage do not provide complete information about whether the storage will work in a cluster running Windows Server 2008 R2. As described later in this section, you will run the Validate a Configuration Wizard later with all tests included.
   
   
2. Make the new server into a single-node cluster and migrate settings to it. Create a new single-node cluster and use the Migration Wizard to migrate settings to it, but keep the clustered resources offline on the new cluster.
   
   
3. Bring the new cluster online, and make existing data available to it. Take the services and applications in the old cluster offline. If the new cluster will use the old storage, leave the data on the old storage, and make the disks or LUNs accessible to the new cluster. If the new cluster will use new storage, copy the folders and data to appropriate LUNs or disks in the new storage, and make sure that those LUNs or disks are visible to the new cluster (and not visible to any other servers). Confirm that the settings for the migrated services and applications are correct. Bring the services and applications in the new cluster online and make sure that the resources are functioning and can access the storage.
   
   
4. Bring the second node into the new cluster. Destroy the old cluster and on that server, install Windows Server 2008 R2 and the Failover Clustering feature. Connect that server to the networks and storage used by the new cluster. If the appropriate disks or LUNs are not already accessible to both servers, make them accessible. Run the Validate a Configuration Wizard, specifying both servers, and confirm that all tests pass. Finally, add the second server to the new cluster.