Shared virtual hard disks

This new and improved feature in Windows Server 2012 R2 allows an administrator to share a virtual hard disk file (the .vhdx file format) between multiple virtual machines. These .vhdx files can be used as shared storage for a failover cluster created between virtual machines (also known as guest clustering). A shared virtual hard disk allows you to create data disks and witness disks using .vhdx files with some advantages:

• Shared disks are ideal for SQL database files and file servers
• Shared disks can be run on generation 1 and generation 2 virtual machines

This new feature allows you to save on storage costs and use the .vhdx files for guest clustering, enabling easier deployment rather than using virtual Fibre Channel or Internet Small Computer System Interface (iSCSI), which are complicated and require storage configuration changes such as zoning and Logic Unit Number (LUN) masking.

In Windows Server 2012 R2, virtual iSCSI disks (both shared and unshared virtual hard disk files) show up as virtual SAS disks when you add an iSCSI hard disk to a virtual machine. Shared virtual hard disks (.vhdx ) files can be placed on Cluster Shared Volumes (CSV) or a Scale-Out File Server cluster

Let's look at the ways you can automate and manage your shared .vhdx guest clustering configuration using PowerShell. In the following example, we will demonstrate how you can create a two-node file server cluster using the shared VHDX feature. After that, let's set up a testing environment within which we can start learning these new features. The steps are as follows:

1. We will start by creating two virtual machines each with 50 GB OS drives, which contains a sysprep image of Windows Server 2012 R2. Each virtual machine will have 4 GB RAM and four virtual CPUs.

   Note

   D:\vhd\base_1.vhdx and D:\vhd\base_2.vhdx are already existing VHDX files with sysprepped image of Windows Server 2012 R2.

   The following code is used to create two virtual machines:

   New-VM –Name "Fileserver_VM1" –MemoryStartupBytes 4GB –NewVHDPath d:\vhd\base_1.vhdx -NewVHDSizeBytes 50GB
   
   New-VM –Name "Fileserver_VM2" –MemoryStartupBytes 4GB –NewVHDPath d:\vhd\base_2.vhdx -NewVHDSizeBytes 50GB
   

2. Next, we will install the file server role and configure a failover cluster on both the virtual machines using PowerShell.

   Note

   You need to enable PowerShell remoting on both the file servers and also have them joined to a domain.

   The following is the code:

   Install-WindowsFeature -computername Fileserver_VM1 File-Services, FS-FileServer, Failover-Clustering
   
   Install-WindowsFeature -computername Fileserver_VM1 RSAT-Clustering –IncludeAllSubFeature
   
   Install-WindowsFeature -computername Fileserver_VM2 File-Services, FS-FileServer, Failover-Clustering
   
   Install-WindowsFeature -computername Fileserver_VM2 RSAT-Clustering -IncludeAllSubFeature
   

3. Once we have the virtual machines created and the file server and failover clustering features installed, we will create the failover cluster as per Microsoft's best practices using the following set of cmdlets:

   New-Cluster -Name Cluster1 -Node FileServer_VM1, FileServer_VM2 -StaticAddress 10.0.0.59 -NoStorage –Verbose
   

   You will need to choose a name and IP address that fits your organization.

4. Next, we will create two vhdx files named sharedvhdx_data.vhdx (which will be used as a data disk) and sharedvhdx_quorum.vhdx (which will be used as the quorum or the witness disk). To do this, the following commands need to be run on the Hyper-V cluster:

   New-VHD -Path c:\ClusterStorage\Volume1\sharedvhdx_data.VHDX -Fixed -SizeBytes 10GB
   
   New-VHD -Path c:\ClusterStorage\Volume1\sharedvhdx_quorum.VHDX -Fixed -SizeBytes 1GB
   

5. Once we have created these virtual hard disk files, we will add them as shared .vhdx files. We will attach these newly created VHDX files to the Fileserver_VM1 and Fileserver_VM2 virtual machines and specify the parameter-shared VHDX files for guest clustering:

   Add-VMHardDiskDrive –VMName Fileserver_VM1 -Path c:\ClusterStorage\Volume1\sharedvhdx_data.VHDX –ShareVirtualDisk
   
   Add-VMHardDiskDrive –VMName Fileserver_VM2 -Path c:\ClusterStorage\Volume1\sharedvhdx_data.VHDX –ShareVirtualDisk
   

6. Finally, we will be making the disks available online and adding them to the failover cluster using the following command:

   Get-ClusterAvailableDisk | Add-ClusterDisk
   

Once we have executed the preceding set of steps, we will have a highly available file server infrastructure using shared VHD files.