A physical node in a four-node Nutanix AHV cluster has failed and requires hardware replacement. The administrator needs to remove the failed node from the cluster inventory and insert the new node, ensuring that no data is lost and cluster redundancy is maintained throughout the process. Which sequence of actions should the administrator perform first?
Select an answer to reveal the explanation.
Short Explanation
Think of it like changing a tire on the highway: you don't yank the wheel off until the car is stable on the others. In Nutanix, verify data is replicated and cluster health is green before Prism retires the failed node. The trap is treating node replacement like a simple power-cycle; skipping health checks can expose you to data loss.
Full Explanation
The correct answer prioritizes data integrity before physical removal. Nutanix protects data by replicating blocks across nodes, so a failed node must be retired only after the cluster confirms surviving copies and healthy redundancy. Prism Element's node removal workflow handles the controlled decommissioning sequence, updating cluster membership and ensuring the CVM is gracefully removed before the hardware is pulled. This matters because a failed node may not have completed replication, and removing it too soon can leave missing data blocks. The first wrong sequence is unsafe because it removes the node without checking replication or cluster health, assuming hardware replacement is just a shutdown event. The second wrong approach bypasses Prism and the cluster's coordination by manually deleting the CVM, which can corrupt metadata or leave the cluster inconsistent. The third wrong approach treats a single-node failure like a full disaster restore, using backup and re-image when replication and node retirement are the supported path. Exam caveat: Always confirm cluster health and data replication before removing a failed node. Operational check: Review Prism Element health and replication status, then use the node removal workflow before disconnecting the failed node.