DPM Was unable to access the cluster resource group SCVMM Server Resources.CVSD for \backup Using Child Partition Snapshot\Server RRS feed

  • Question

  • Hi,

    I've been using DPM 2010 to back up Hyper-V guest machines for some time now but recently have started getting an error message with one of the servers:

    Affected area: \Backup Using Child Partition Snapshot\Server
    Occurred since: 12/07/2013 10:31:02 PM
    Description: Recovery point creation jobs for Microsoft Hyper-V \Backup Using Child Partition Snapshot\Server on SCVMM Server have been failing. The number of failed recovery point creation jobs = 1.
     If the datasource protected is SharePoint, then click on the Error Details to view the list of databases for which recovery point creation failed. (ID 3114)
     DPM was unable to access the cluster resource group SCVMM Server for \Backup Using Child Partition Snapshot\Server of protection group Hyper-V Guests (ID 30148)
     More information
    Recommended action: Verify that the cluster resource group SCVMM Server is accessible by using the Cluster Administration console.

    I can open Failover Cluster Manager and see that the server is online and I am only seeing the error with this one single server.

    If I right-click on the protection group and select Modify Disk Allocation I get an error:

    Operation failed because SCVMM Server Resources.C#MNVSD, is not accessible from the DPM server. Use the cluster management software to check to see if the cluster resource group SCVMM Server is online.

    Please does anyone know how to fix this?

    Monday, July 15, 2013 1:07 AM

All replies

  • Has the VM change location on disk or migrated to another computer in the cluster? Have there been any other changes to the cluster configuration (addition of new node etc?, addition or removal of disk on virtual server?)

    When you modify a protection group and expand out the cluster resource can you see the VM listed under the cluster?

    Are you still able to protect other VM's that are using the same cluster node for compute and storage?

    Assuming you've checked the VSS writer states both inside the VM and on the host managing CSV and compute?

    Have a look at the event log on the host server to see if there are any obvious errors when the protection job is scheduled.

    In some cases I've seen modification to a VM where I need to stop protection (retain protected data) and re-add to a protection group if I've had a configuration change. This may increase the amount of data stored for the replica and recovery point volume though.

    Monday, July 15, 2013 6:19 AM
  • I don't think there have been any changes or that the VM has been migrated to another computer in the cluster.

    If I go to modify protection group and expand the cluster it's on I just get an error saying:

    "DPM is unable to enumerate contents in..." with a list of volumes.

    There are other servers on that cluster that are protected and working fine

    The VSS writers on both server and cluster are fine

    I tried removing the protection but I now can't re-add the server.

    Tuesday, July 16, 2013 11:10 PM
  • Are you able to check the cluster state using the failover cluster manager and see if there are any related errors on the Hyper-V cluster. If you're expanding out the cluster object.

    When you expand out the cluster name (cluster) object if you are not seeing any of the VM's then DPM is having an most likely having an issue communicating with the cluster resource.

    What is the event log showing on the hyper-V host that owns that Virtual Machine and/or storage for the problem VM? Are there any event or system log errors showing up that relate to VSS or to Hyper-V clustering?

    Wednesday, July 17, 2013 12:01 AM