An organization has two Nutanix clusters managed by a single Prism Central (PC) instance. They intend to implement Synchronous Replication between the sites. During a failover test, the administrator notices that automatic failover does not occur when the primary site's storage becomes unavailable. What is the most likely cause of this behavior based on an analysis of the requirements?

Metro Availability (synchronous replication) performs an automatic unplanned failover only when a Nutanix Witness VM is deployed and reachable by both clusters. Without the Witness, the system cannot make the quorum decision that the primary storage is down, so replication stops and no automatic failover is triggered; administrators must initiate a manual failover instead.

A. RTT > 5 ms prevents enabling Metro Availability at all; replication would not have been running in the first place.
B. Multiple PC instances are not required; automatic failover logic lives with the clusters and Witness, not Prism Central.
D. Metro Availability reacts to any datastore unavailability (network or storage) once Witness confirms failure.

1. Nutanix Metro Availability Tech Note v5.20, “Automatic Failure Handling with Witness,” p. 6–7.

2. Nutanix Witness VM Deployment Guide v3, “Quorum-based failover requirements,” p. 4–5.

3. Nutanix Data Protection & Disaster Recovery Guide AOS 6.5, Chap. 8 “Auto Unplanned Failover,” para. 1.