Best Practices - Microsoft Hyper-V

Configuring Hyper-V Online Backups

The following is a list of requirements that must be met to perform online backups. If these requirements are not met, then only offline backups can run and the guest virtual machine will be unavailable for the duration of the backup.

• Integration services are installed and the backup integration service has not been disabled.
• Integration services on the guest virtual machine must be set to the same level as the host Hyper-V server.
• All disks used by the virtual machine are configured within the guest operating system as NTFS-formatted basic disks. Virtual machines that use storage on which the physical partitions have been formatted as dynamic disks or the FAT32 file system prevent an online backup from being performed. This is not the same as dynamically expanding virtual hard disks, which are fully supported by backup and restore operations.
• Volume Shadow Copy Service must be enabled for all volumes used by the virtual machine. Each volume must also serve as the storage location for shadow copies of the volume.

There are also some forms of storage in Hyper-V guest virtual machines that are not supported for backup through the Virtual Server Agent.

• Physical disks that are directly attached to a virtual machine
• Storage accessed from a virtual machine by using an Internet SCSI (iSCSI) initiator within the guest operating system.

Configuring Cluster Shared Volume

Cluster shared Backup with single volume

Cluster shared volumes allow multiple Hyper-V hosts in a cluster to share the same LUN. The basic concept of CSV can be seen in the graphic.  CSV allows multiple virtual machines that are running on multiple cluster nodes to access their Virtual Hard Disk files at the same time, even if the VHD files are on a single disk (LUN) in the storage. The clustered virtual machines can all fail over independently of one another. CSV enables Live Migration which allows a live virtual machine to migrate from one node to another with zero down time.

CSV in Redirected i/o mode

Disks in a cluster shared volume are placed into redirected I/O mode during management operating-system based backups. This can be seen in the diagram. Once the backup of a virtual machine has been started on Node 1 it has ownership of the volume and all of the I/O traffic from the other nodes with virtual machine’s on this volume will flow through Node 1. This obviously can cause resource contention and requires care when scheduling backups.

Cluster shared Backup with multiple volumes

For parallel backups of all nodes in the cluster, it is recommended to use multiple cluster shared volumes. Distribute the virtual machines on these volumes as shown in the diagram. Backup virtual machines residing on one CSV from one node. Virtual machines residing on other cluster shared volumes can be backed up from another node of the cluster, at the same time. To catch the Virtual machines that migrate from one node to another, it is recommended to leave Auto Discovery ON.

Running Backups using Hardware VSS Provider from Storage Vendor

You can perform backups using the hardware VSS Provider offered by the storage vendor. It is recommended to use a hardware VSS Provider as it improves the performance of backup operations. For Hyper-V clusters, it reduces the time in which the cluster shared volumes are caught in redirected-access mode.

Follow the steps below to create a registry key which will allow you to use the vendor's VSS provider:

1. From the CommCell Browser, navigate to Client Computers.
2. Right-click the <Client>, and the click Properties.
3. Click the Registry Key Settings tab.
4. Click Add.
5. In the Name box, type VSSProvider.
6. In the Location box, type VirtualServer.
7. In the Type box, select REG_SZ.
8. In the Value box, type the hardware VSS Provider ID of the vendor. To find the ID, run the following command on your client computer:

   vssadmin list providers

   An example of a provider ID (for NetApp) is ddd3d232-a96f-4ac5-8f7b-250fd91fd102.

9. Click OK.

Your backup operations will now use the hardware VSS Provider instead of the software VSS Provider from Microsoft.

Location of Job Results Directory

The Job Results directory stores the job results files from backup and restore operations of a client.

The user impersonation account specified for the job results directory will take precedence and will be used to backup and restore data from a virtual machine. This may result into file access related issues during the backup. Therefore, it is recommended to use a local folder on the client computer as the job results directory.

If you have virtual machines with long names or virtual machine files with long paths, it is recommended to create the job results directory at the root of a volume. For example: C:\JobResults\. For more information about, changing the location of job results directory, refer to Changing the Job Results Path of a Client.