A while back I bought a second enclosure for my EMC san and tried migrating my VM's to a new cluster made up of 15K rpm drives via Migration Wizard in SCVMM 2008 R2. I have two hosts using 5 different failover clusters. This worked on all but one of my VM's, on the one where this failed it put the VM into a failed state on SCVMM, but it was still accessible on the host. However I suspected that it was not highly available in this state. In any case I wanted it manageable from SCVMM, so today I did the following.
Shut down the VM.
Copied VM files to a different cluster
Deleted VM from SCVMM
Deleted remaining files from Cluster
Copied files back to correct Cluster
Tried recreating the VM, got an error that it still existed on the Host. I then ran a SQL script in Microsoft SQL Server Management Studio in order to remove any remaining hidden VMs.
I then successfully recreated the VM in High Availability State – at one point I received an error that there were resources in existence in the cluster for a VM with the same name, so I removed these resources via cluster manager on the host. I did not do this on the other host, though now I feel I should have. After this was done I went through the steps to recreate the VM in SCVMM, and I was successful.
Now here's where my problem is:
After I was done I shut down the VM and I tested migration. Migration via drag and drop doesn’t work. - Error message says: Unable to migrate the virtual machine AVDB2 because the host HV-01 is not a suitable host for the virtual machine. To select from hosts that are best suited for the virtual machine, use the Migrate Virtual Machine Wizard. Do you want to start the Migrate Virtual Machine Wizard now?
At this point I select Yes, and the migration via the Wizard does work.
Does this mean anything for the High Availability of the server? Is there an easy way for me to test the high availability without compromising my VM by disconnecting the server? The page I found in Microsoft Technet for SCVMM errors doesn’t have any suggestions for this error.
Below are links to sites I was using to work through this issue.