I’ve never needed to employ an HA config in my lab environment when I was building it out, mostly because the hardware I was using at the time was scarce and I had planned as many other services to be offered by it over vCenter HA. That said, I have always wanted to learn a bit more about vCenter Server HA – What it is, what it does and when you would (and would not) use it.
So I got reading up on it, and learnt a lot from the various blogs, whitepapers etc the community has written on it!. When I find some downtime, I’ll be spending half an afternoon setting it up on my own ESXi server.
What is it?
vCenter High Availability protects the VMware vCenter Server Appliance against host and hardware failures.
When would you use it?
This becomes extremely useful more so when you are applying updates\patches to VCSA itself – traditionally this would take all management functions down until the VCSA came back up but with the active-passive architecture of the vCenter HA solution, you can now flip over to the second VCSA (configured in HA) and reduce downtime.
One of the key components here is the witness node service which is more or less there to detect the outage in one of the nodes. The other consideration is that only one VCSA can be the master node at any one time – The idea here is that the slave node is there to mimic the master node and be ready to become active should the master node fall over.