Risorse per professionisti IT > Home page del forum > Clustering > Move an unallocated cluster disk between nodes
Formula una domandaFormula una domanda
 

Con rispostaMove an unallocated cluster disk between nodes

  • martedì 30 settembre 2008 1.17JeremyHagan Medaglie utenteMedaglie utenteMedaglie utenteMedaglie utenteMedaglie utente
     
    I have a 2-node WS08 Failover Cluster.  It has about 18 cluster disks ready for use as Hyper-V resources.  I want to move one of the disks that isn't allocated to a clustered resource to the other node, but can't work out how to do it either via the GUI or via the command prompt.  Can someone help me please?

Risposte

Tutte le risposte

  • martedì 30 settembre 2008 2.05VikasKumar Medaglie utenteMedaglie utenteMedaglie utenteMedaglie utenteMedaglie utente
     

    It would be good to know what you are trying to achieve. Usually Un-used Disks reside in Group "Available Storage". All the resources in a group are hosted on a single node. The entire group moves between nodes.

    When you create a Hyper-V Role, the disk would be move to the Hyper-V group and then you can move this group to any node of desire before bringing the Guest Online.

    Another alternative, You would need to create an empty Group and then move the disk to this new group. Now you can move the group to any node you wish.

  • martedì 30 settembre 2008 2.10JeremyHagan Medaglie utenteMedaglie utenteMedaglie utenteMedaglie utenteMedaglie utente
     
    What you have described is exactly what happens.  The disk is grouped under the storage node in the GUI in "Available Storage".  I have a VM I want to move from local disk to the cluster disk that is on the other node and I want to avoid doing it over the network.  There is no way I can see to move the "Available Storage" group to the other node.  I already tried creating an empty service adding the disk to it then moving that across, but when I destroy that service it reverts back to the "Available Storage" group that is on the other node.
  • venerdì 3 ottobre 2008 13.04JohnTonerMVP, ModeratoreMedaglie utenteMedaglie utenteMedaglie utenteMedaglie utenteMedaglie utente
     Con risposta
    You can move the "Available Storage" group via cluster.exe commands. For example,

    cluster group "Available Storage" /move
    Visit my blog about multi-site clustering - http://msmvps.com/blogs/jtoner
  • giovedì 2 luglio 2009 16.45G_Rowley Medaglie utenteMedaglie utenteMedaglie utenteMedaglie utenteMedaglie utente
     
    I am having the exact same problem as Jeremy. Vikas, you mentioned some interesting ideas but could you elaborate on the Hyper-V groups? All of my disks are showing up under "available storage" and I'm having a problem where one of the nodes in the cluster is grabbing all of those disks. This is a problem because I have VM's on those disks that can't load because the node that owns them doesn't have access to the resource.
  • giovedì 2 luglio 2009 20.35JeremyHagan Medaglie utenteMedaglie utenteMedaglie utenteMedaglie utenteMedaglie utente
     
    If you don't intend to use those disks that are in the available storage group for highly available VMs then they should be in the available storage group.  Use the cluster.exe command mentioned above to move the available storage group to the desired node then remove the disk(s) from the group using the Failover Cluster admin tool.  Then they will be on the node you want, but not part of the cluster and therefore wont ever move.

    The alternative is to make the VM(s) highly available then this will allow the VM(s) to come up on any node.  If you have nultiple VMs hosted on the LUN then there is a failover clustering update (that I can't find right now) which will allow this.  If you use SCVMM please be aware that doing this will mean that the VMs are in an "Unsupported cluster configuration" and you will have to manage them with Hyper-V Manager and not SCVMM
  • venerdì 3 luglio 2009 8.39Edwin vMierloMVP, ModeratoreMedaglie utenteMedaglie utenteMedaglie utenteMedaglie utenteMedaglie utente
     
    you have to add the disks to your cluster group where your VM's are. Then make a dependency, which means the disk will come online first prior to the VM(s).
    this way your disk is always online on the node where your VM is

    rgds,
    edwin,.