If you deploy SCVMM to manage your virtual infrastructure, navigating to DPM / Protection Group and expanding the <clusterName> on the “Select Group Members” page will present you with the opportunity to pick from a list of virtual machines prefixed with <SCVMM> for backup.
Assuming that you have a clustered VM resource named Exchange, it will read “SCVMM Exchange Resources” below “<clusterName> (Cluster)”. Drill down to “HyperV” and put a check against “Backup Using Child Partition Snapshot\Exchange”. Complete the remaining steps to finalize your backup requirements (backup media, goals, schedule, etc.).
For the reverse procedure (data restore), click on the Recovery tab and move to the Browse tab. On the left hand pane, expand “Recoverable data” and choose the node representing the Active Directory domain name. Instead of the <clusterName> name, you will find “SCVMM Exchange Resources (<clusterNameFQDN>)” along with other protected servers and resources (SQL, SharePoint, etc.).
In order to recover the entire VM - virtual hard disks (vhd), Virtual Machines folders, XML/bin/vsv files, etc. - it is imperative to first select “All Protected HyperV Data” in the tree view (Protected Data) on the left hand side. Be aware that these files do not show up as individual items for selection.
On the list pane to the right, highlight “Backup Using Child Partition Snapshot\Exchange” and locate the desired recovery point, date, time and backup media using the calendar. Click on Recover where you have a choice on the “Recovery Type” page to perform restore using one of the options:
Recover to original instance
Recover as virtual machine to any host
Copy to a network folder
Copy to tape
For restoration of just the associated vhd files of the VM, double click “Backup Using Child Partition Snapshot\Exchange”. You can only pick a single vhd for recovery one at a time. The recovery target in this case is limited to “Copy to a network folder”. Should you attempt to replace the vhds using this approach, VMM may fail to start the VM and complain with the error:
Error (12711)
VMM cannot complete the WMI operation on server <clusterNodeNameFQDN> because of error: [MSCluster_Resource.Name=”SCVMM Exchange”] The group or resource is not in the correct state to perform the requested operation.
(The group or resource is not in the correct state to perform the requested operation (0×139F))
To learn more about a cluster and its nodes under DPM, simply hover and pause the mouse over the entity to see the corresponding tool tip:
<clusterName> consists of multiple servers, which may have unclustered resources. Unclustered resources are listed separately under server names. Since this list is sorted alphabetically, server names may appear somewhere else in this tree view.
<clusterNodeName> is part of a cluster. Unclustered resources on this server are listed here. Clustered resources belonging to the same cluster are listed under cluster name separately. Since this list is sorted alphabetically, cluster name may appear somewhere else in this tree view.
Environment
Hyper-V R2 in a 2-node Fail-over Cluster with Cluster Shared Volume, CSV*
System Center Virtual Machine Manager 2008 R2, v2.0.4273.0
Data Protection Manager 2010 RTM*
* Windows Server 2008 R2
See also:
Restore of hyper-v VM’s on DPM 2010 in Data Protection Manager
Technorati tags: System Center, Data Protection Manager, Virtual Machine Manager, Windows Server 2008 R2