Understanding vSphere Replication Enhanced Replication Mapping Tests and Intermittent Connection Errors

Products

• VMware Live Recovery

Issue/Introduction

This article explains the mechanics of the Enhanced Replication Mapping Test in vSphere Replication and addresses why users may see intermittent connection errors in the Site Recovery UI.

Environment

VMware Live Recovery 9.0.x

Resolution

What is the Enhanced Replication Mapping Test?

The Enhanced Replication Mapping Test is a health-check mechanism designed to support the "Direct Path" architecture of Enhanced Replication.

• Optimization:
  • Enhanced Replication optimizes data transfer by allowing direct host-to-host (ESXi to ESXi) traffic. This bypasses the vSphere Replication Server eliminating it as a potential data-plane bottleneck.
• Path Mapping:
  • Because traffic moves directly between hosts, vSphere Replication automatically maps and tests every possible network path between all source ESXi hosts and target ESXi hosts within a replication cluster.
• Redundancy:
  • This continuous mapping ensures that replication traffic is distributed efficiently and can be maintained even if specific hosts or network paths become unavailable.

 

Why does it show errors Intermittently?

Intermittent errors in the mapping matrix are typically not random but are tied to automated background processes:

• Hourly Background Polling:
  • vSphere Replication automatically runs a quick health check every hour. This background task tests every single host-pair combination in the connection matrix.
• Environmental Variables:
  • The background test may intermittently encounter specific network bottlenecks, strict firewall rules (specifically on port 32032), or slight NTP time drifts (clock skew) that only exist on specific host-to-host paths.
• The IP vs. FQDN Certificate Issue (9.0.x):
  • A known issue in version 9.0.x causes the background test to attempt to reach the Host-Based Replication (HBR) REST endpoint using an IP address instead of a Fully Qualified Domain Name (FQDN). Refer 373696 

 

Is it safe to modify or delete these mappings?

Yes. Managing these entries is safe and often required to resolve UI discrepancies:

• Safe Deletion:
  • Deleting incorrect or redundant mappings from the Enhanced Replication Mappings page does not delete replicated data, break existing replications, or impact the VM’s protection status.
• Resolving Stale Data:
  • If the UI displays incorrect resources (e.g., a VM is showing a mapping to the wrong target cluster), it is usually due to stale cached data in the database. The recommended fix is to remove the incorrect mapping and allow the system to rediscover or manually re-assign the correct resources.
• Future Proofing:
  •  Ensuring correct mappings are in place after removing stale entries prevents configuration errors during future VM migrations or reconfigurations.