Every host in a vSphere HA cluster apart from the elected primary. Each reports its status to the primary, which keeps an eye on it and, should it fail, restarts its VMs on other hosts.
Read more: Broadcom TechDocs
In the Ultra Transcenders books
Each book explains Secondary host in context, with comparison tables and the common traps.
Terms in this definition
- vSphere HA
When a host fails, this cluster capability restarts its VMs elsewhere; through VM Component Protection and VM Monitoring it also reacts to storage and VM failures.
- Virtual machines
Infrastructure-as-a-service compute giving complete control of the operating system, making it a fit for lift-and-shift moves and for software relying on OS-level pieces like COM.
Related terms
- Datastore heartbeating
When a secondary host stops answering over the management network, the HA primary looks at shared heartbeat datastores to tell a dead host apart from one that is merely cut off or partitioned.
- Liveness check
How vSphere HA decides whether a silent secondary host is really down. Once network heartbeats have stopped, the primary looks for datastore heartbeats and tries pinging the host's management addresses.