none
How to move a Virtual Machine from one host to another in Hyper V

    Question

  • HI,

    We have two Infra server (infra1, infra2), infra1 and infra2 are clustered. When we shutdown infra1, a VM from infra1 automatically moved to infra2, now my question is when we start the infra1 how can we shift that VM from infra2 to infra1 again.


    Swaprakash..

    Thursday, August 29, 2013 8:39 AM

Answers

  • Within the cluster, you can select any resource and check the properties.  One setting is to auto-failback after a failure.  This means that if the resource is running on node1 and it fails, the resource will fail over to node2.  When node1 comes back online, the resource is automatically moved back.  By default, this is turned off.  I generally leave it off in all my clusters because I don't want a situation where a recurring, transient error causes the resource to bounce back and forth.  Node1 goes down causing the resource to fail to node 2.  Node 1 automatically reboots and the resource goes back to node 1.  Node 1 fails again sending the resource back to 2.  Node 1 reboots ... and so on.

    To manually move, within the cluster manager console right click the VM and select move and use Live Migration.


    .:|:.:|:. tim

    Friday, August 30, 2013 12:21 AM

All replies

  • if it's a true failover cluster it may fail back to the original when the original server comes back up, there are ways to disable failback, and ways to force failure to specific owners, when all hosts are back up you can live migrate to the desired host too.
    Thursday, August 29, 2013 12:18 PM
  • Hi,

    Thanks for sharing your views, but my infra1 server is now up and that particular VM from infra1 is still showing in infra2. I guess I need to manually shift that VM, could you please share the steps. I am using WS 2008R2.


    Swaprakash..

    Thursday, August 29, 2013 2:49 PM
  • Within the cluster, you can select any resource and check the properties.  One setting is to auto-failback after a failure.  This means that if the resource is running on node1 and it fails, the resource will fail over to node2.  When node1 comes back online, the resource is automatically moved back.  By default, this is turned off.  I generally leave it off in all my clusters because I don't want a situation where a recurring, transient error causes the resource to bounce back and forth.  Node1 goes down causing the resource to fail to node 2.  Node 1 automatically reboots and the resource goes back to node 1.  Node 1 fails again sending the resource back to 2.  Node 1 reboots ... and so on.

    To manually move, within the cluster manager console right click the VM and select move and use Live Migration.


    .:|:.:|:. tim

    Friday, August 30, 2013 12:21 AM
  • Hi tim,

    It worked.. thanks... I went to fail over cluster manager, and then I select that particular VM, and live migrate to another node... thank u.. cheers....


    Swaprakash..

    Friday, August 30, 2013 3:39 PM
  • Oh, I had the same questions about VM migration..

    Now it's clear for me and it works! Thank you, Tim

    Wednesday, September 18, 2013 10:50 AM