On to the final part of the mini-series on clustering, and we round off with the most exciting bit – seeing our creation come to life!
In this video, I perform a number of Live Migrations, for the different VM types (VHD, Pass Through Disk and CSV), including using PowerShell for the migration (Yes, I can use PowerShell, a bit!) and we wrap up with a failover of the workloads, when I cold-shutdown one of the physical servers, which just happened to have all 4 of our VMs on. Needless to say, the technology kicked in, and within a minute or so, the VMs were up and running again. The video weighs in at just over 13 minutes, but well worth sticking with it.
So, that’s it for the clustering mini-series. You can view the video straight in this webpage, or head on over to the VirtualboyTV to see it in higher definition. For the best experience, you can always download it from this page too.
Hopefully you understand how easy it can be to configure and build this type of environment, and in future videos, we’ll be looking at Hyper-V Server 2008 R2, rather than Windows Server 2008 R2 Hyper-V, but we’ll be easily able to take what we’ve learned from this series, over to the GUI-less HVS2008 R2. Sound like a plan? Watch this space!
In this video you have 2 nodes... Let's say you had 6 nodes and 1 node failed, crashed etc. How would the VM's failover in a 6 node setup?
When you are just using pure Hyper-V, and no System Center, the failover is based on logic. You can read all about that here: http://blogs.technet.com/mattmcspirit/archive/2009/02/03/hyper-v-clustering-virtual-machine-failover-behaviour.aspx but if you involve System Center, i.e SCOM and SCVMM, I believe that there is more intelligence involved, but I'd have to look into that one for sure! The link above talks about 2008, and there has been improvements in R2, but I believe it's still logic-based out of the box.
Hope that helps!