VM High Availability Modes

Default: This does not require any configuration and is included by default when an Acropolis Hypervisor-based Nutanix cluster is installed. When an AHV host becomes unavailable, the failed VMs that were running on the failed AHV host restart on the remaining hosts, depending on the available resources. Not all of the failed VMs will restart if the remaining hosts do not have sufficient resources.

Guarantee: This non-default configuration reserves space to guarantee that all failed VMs will restart on other hosts in the AHV cluster during a host failure.

• Acropolis Master responsible for restarting VMs
• Master monitors host health by monitoring all connections to libvirt

• If Master is Isolated/Partitions/Failed, new Master elected.
• If cluster becomes partitioned, side with quorum will remain up and VMs restarted on those hosts.

Restart Policy

• By default, AHV will do its best to restart VMs
• Best effort = ability to restart dependent on AHV resources
• With 5.0 and later we now only support a segment based reservation which is automatically implemented when the Guarantee HA mode is selected.

Reserve Segments Calculation

• Automatically calculates reserved segments per host reservation
• Fixed size segments are used
• Size corresponds to largest VM in system
• Unlike VMware, AHV can pack multiple smaller VMs in single fixed segment
• Reduces fragmentation