none
DPM 2012 backup of Hyper-V R2/CSV VMs RRS feed

  • Question

  • First, I would like to confirm that what I am seeing is what I am supposed to be seeing. I am using a Dell EqualLogic PS6000 iSCSI SAN as my backend. When DPM begins the VM backup, the CSV moves to the node that is currently running the VM, then states that a backup is in progress, then the CSV goes to redirected access, a snapshot of the volume is taken, then redirected access ends, then the CSV goes to online. All of the snapshots are displayed in the Group Manager console, and are deleted when the backup of the VM is complete.

    So far so good?

    My question is first, is this the expected behavior? I haven't found anything in my limited research that says this is what I am to expect.

    My second question, if this is all as is should be, then why is my cluster reporting all of these redirected access errors? It seems to me that if this is a "by design" procedure, an error should not be generated. For what it's worth, the error is 5121 and states:

    "Cluster Shared Volume 'Volume1' ('Volume 1') is no longer directly accessible from this cluster node. I/O access will be redirected to the storage device over the network through the node that owns the volume. This may result in degraded performance. If redirected access is turned on for this volume, please turn it off. If redirected access is turned off, please troubleshoot this node's connectivity to the storage device and I/O will resume to a healthy state once connectivity to the storage device is reestablished."

    I had one incident when I first set up the protection group on my test cluster that during the backups (at some stage as I stopped watching), the volume must have went offline and several VMs simply stopped. I discovered the volume online, VMs stopped, and little explanation of why except for all of these redirected access errors.


    -Richard

    Tuesday, June 26, 2012 4:19 AM

All replies

  • Hi Richard.

    All your information is exactly what we experience in all our Cluster setups as well. So to make it short: It's by design.
    The redirected mode when backing up, and also the error 5121 is to be expected. So if your VM's works fine and backup runs smoothly, I see nothing wrong.

    Just make sure you have all the latast Cluster, MPIO hotfixes, EQ VSS snapshot drivers etc.  - you should be homefree :-)
    More info on hotfixes are found here: http://social.technet.microsoft.com/wiki/contents/articles/2008.list-of-cluster-hotfixes-for-windows-server-2008-r2.aspx

    Cheers,
    Anders

    Wednesday, June 27, 2012 12:13 PM
  • Thanks. It just seems so wrong for "by design" to generate an error.... :-)

    -Richard

    Wednesday, June 27, 2012 4:18 PM
  • Hi Richard,

    We have had reports of several hardware providers causing the event 5121 errors but seem to be benign and if not causing failures go ahead an ignore them.  However, to be clear, these are NOT normal and not every customer sees them, so what the common denominator is is unknown.


    Please remember to click “Mark as Answer” on the post that helps you, and to click “Unmark as Answer” if a marked post does not actually answer your question. This can be beneficial to other community members reading the thread. Regards, Mike J. [MSFT] This posting is provided "AS IS" with no warranties, and confers no rights.

    Thursday, June 28, 2012 2:01 AM
    Moderator