A VM is configured with a 5-minute Nearsync RPO. After several hours, replication transitions back to hourly RPO. Alerts indicate the minute schedule cannot be maintained. What is the most likely root cause?

Nearsync replication in Nutanix Disaster Recovery allows for RPOs as low as one minute. It achieves this by continuously shipping data logs to the recovery site. If the rate of data change (churn) on the protected VMs generates more data than the available network bandwidth can transfer, a backlog occurs. When this backlog becomes too large for the system to catch up, Prism Central automatically transitions the replication schedule to the next available, less aggressive schedule (typically hourly) to ensure consistency, and it generates an alert. This indicates the network is the bottleneck.

A. An RTT greater than 5 ms is a requirement for Metro Availability (synchronous replication), not Nearsync, although very high latency can impact throughput.
B. The number of recovery plans is irrelevant to the technical ability of the system to maintain a specific replication schedule for a given VM.
C. Reverse synchronization is a feature used after a failover to replicate changes back to the original site and does not affect the primary replication performance.

1. Nutanix University. (2023). Nutanix Business Continuity and Disaster Recovery (NCP-BC) 6.5. Section: Nearsync Replication. "If the data change rate exceeds the available bandwidth, the system may not be able to maintain the Nearsync schedule and will revert to a longer RPO."

2. Nutanix. (2023). Prism Central Disaster Recovery Guide v2023.3. Section: Nearsync Replication for Nutanix DR. "If the system cannot maintain the 1-minute RPO schedule (for example, due to network bandwidth constraints or a high rate of change), it switches to the next-best RPO schedule."

3. Nutanix Support Portal. (2022). KB-8878 Troubleshooting Nearsync. "The most common reason for Nearsync schedules failing to be met is insufficient network bandwidth between the primary and recovery sites for the current data change rate."