If you watched the demonstration blogcast of Virtual Server 2005 R2 I posted up at the weekend, yesterday, I ripped the environment apart to build it again from scratch so you can see how I built it.

In part 3, the new cluster group for the virtual machine was created. In the penultimate part of this mini-series, we place the VHD and VMC for a virtual machine on our shared drive and examine the configuration after adding it to the Virtual Server 2005 R2 admin interface on Node 1 of the cluster. For now, we don't configure the network (that's for part 5). We fail the virtual machine cluster group across to Node 2 and repeat adding the machine through the Virtual Server admin interface. We perform a critical step from the command line to set a private variable for the cluster script to work correctly. This absolutely must be done before the script is brought online in cluster administrator. The command takes the following format:

cluster res "<HA scriptname>" /priv VirtualMachineName=<Name of VM>

With that done, we are ready to bring the script online in the cluster group and watch the Virtual Machine starts up. Once booted, we log on and start working. In cluster administrator, we then proceed to move the cluster group across to the other node, verifying we are able to continue working on the guest now running on the other cluster node. This is an example of "Planned" downtime.

The final part of thi series is where we setup the network for failover, but in an atypical configuration to that described in the whitepaper - in the environment I have, the network cards used by the guest are different between the two hosts. However, it's good to know that this is possible. Stay tuned tomorrow to find out how :)

Click here to view part 4.

Previous parts:

Part 1
Part 2
Part 3