I am re posting this question in the hope someone can explain / comment / correct me:
With the admission control policy based on % of resources HA will first calculate all resources within the cluster, deduct overhead then deduct reservations from all VMs within the cluster. If reservations do not exist then memory overhead (VM) and 32MHz CPU will be deducted by default, this will yield available resources.
As most of us try not to use per VM reservations then the defaults give the impression that the cluster has a high percentage of available resources when in fact consumed memory (example) could be 70%.
Let’s say that failover capacity is set to 33% (balanced 3 node cluster). Will HA monitor the figures for current cpu / memory failover capacity and only prevent VMs being powered on when these figures drop below 33%? If so then this is not a true reflection???
HA does not use the metrics you use. HA is about powering on virtual machines, not guaranteeing resources... that is what DRS does. indeed this doesn't always, or probably does not reflect the real world resource usage. It is not recommended to use HA admission control as a resource management / indication tool.
Thanks for the response Duncan.
So I should ignore these metrics with respect to HA being in a position to failover a VM. In my example of 33% then my cluster would cope with the failure of one host (not taking into account fragmentation).