In 6.5 VMware has changed the name of vSphere High Availability (vSphere HA) to vSphere Availability.
In vSphere 6.0 Version
And in 6.5 Version
New Setting in 6.5 vSphere is Proactive HA which works with DRS and it has two options.
First thing, this is DRS feature, not an HA and basically, it will get data from physical hardware like RAM, CPU like DRS algorithms doing normally and then use DRS to decide whether it needs to move VMs or not.
Proactive HA look at different alerts to see if maybe something is going on with the host in order to make a decision to move workload or not.
Now in order to do this, Different hardware vendor is creating or in process of creating OEM.
If you are using Proactive HA with OEM and there is DIMM or NIC card failure then, in this case, Proactive HA coordinate with DRS and take action on that host. (vMotion)
Now we have two option in this setting:
1. Automation Level is like DRS.
2. Remediation
Quarantine Mode Means: DRS will move all the VMs but no impact on VM performance and affinity rules should not be violated.
when you decided to put the host into quarantine mode, then it will start evacuating the VMs but it will consider the basic facts of and this is DRS affinity and anti affinity rules should not be violated, and also if there is no performance degradation.
After this DRS will not place any VM on that host.
As I said second consideration is if there is no performance degradation on VMs if DRS see any impact on VMs performance then DRS place VM on this Quarantine host.
A) Quarantine mode for all failures
B) Quarantine mode for moderate and Maintenance mode for Severe failure(Mixed).
C) Maintenance mode for all failures
Admission Control: New Change is Autocalculation
Now it can automatically calculate the Percentage of resources which are required for Cluster failover.
Performance degration VMs tolerate:
Here 100% means this setting is disabled and you are fine if there is any degradation in terms of performance and if i put 0% percent then i am not okay or i dont want any performance degradation with VMs.
In the event of failure and VMs restarted and they are not getting sufficient resource and facing performance degradation then it will alert me.
