Too frequently I hear the expression n+1 as a model for ESX clusters to provide High Availability. If you EVER expect to patch ESX servers without VM downtime then you need at least(†) n+2. When running your clusters to only n+1, you can never safely put one of your hosts in Maintenance Mode; not if High Availability is important to you.
Footnote: If you don’t understand the importance of HA slot sizes, go learn.
I always try to convince the customer of n+1+1. n for the number of hosts needed to run their environment. 1 for a planned server downtime, and 1 for an unplanned server downtime.
Only in very small or money constrained environments, does the customer choose n+1, knowing the consequences.
You’re playing with semantics. What you’re really suggesting is that it’s not possible to keep HA enabled while patching an n+1 environment, which is a lot different than suggesting ESX servers cannot be patched without VM downtime in an n+1 environment.
p.s. HA does NOT prevent VM downtime, even n+100.
(now I’m playing with semantics). You’ll need Fault Tolerance for that.
Hi JC,
I want to create discussion on the subject.
Keeping HA “enabled” is somewhat mute if its useless. During patching and maintenance downtime, if you have an n+1 setup, HA may as well be off because it isn’t going to help at all. Just because it says “I’m enabled – your safe”, means diddly. I won’t work. To say that its enabled is semantics.
The only advantage I can think of is that it saves you having to manually re-enable it once you have more redundancy in the cluster.
But yes, I’m being a bit coarse