VM replication fails with a connectivity error despite the vSphere Replication (VR) appliances appearing online in the vSphere Client and Site Pairing appears healthy.
This issue typically occurs when the management plane communication between sites becomes desynchronized, even if the underlying network is functional.
Symptoms:
Replication status shows "Error" or "Not Active".
Exact error message: A replication error occurred at the vSphere Replication Server for replication '<VM_NAME>'. Details: 'No connection to VR Server for virtual machine <VM_NAME> on host <ESXi_HOSTNAME> in cluster <CLUSTER_NAME> in <VCENTER_FQDN>: Unknown'.
2026-03-08T22:28:22.144Z In(166) hbr-agent-bin[18966755]: [0x00000076d6d88700] error: [Proxy [Group: GID-xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxx] -> [xx.xx.xx.xx:32032]] Failed find server request additional error info: Thumbprint and certificate is not allowed to send replication data
vSphere Replication 9.x
The "No connection to VR Server" error in this context is caused by a stale or asymmetric site pairing state. Specifically:
Certificate Mismatch: The certificate thumbprints exchanged during the initial pairing are no longer trusted or have expired/rotated.
HMS Inventory Stale: The vSphere Replication Management Server (HMS) has a stale or missing view of the host-to-VRS assignments, preventing the ESXi hbr-agent from receiving the correct connection parameters for the target site.
To resolve the connectivity state, the site pairing must be refreshed to force a re-handshake between the Management Servers.
Note: Verify port connectivity 31031 first reviewing Port numbers that must be open for vSphere Replication
Log in to the vSphere Client.
Navigate to Site Recovery > Open Site Recovery.
Locate the affected Site Pair and click View Details.
Click the Summary tab.
Click Reconnect
Provide the credentials for the PSC/vCenter of the remote site.
Accept the certificate prompts for the remote site's vCenter and VRMS appliance.
Once the site status returns to Connected (Green), navigate back to the Replications tab.
Select the failed replication and click Sync Now.
If the "Reconnect" button fails, restart the HMS service on the appliance via SSH with command: systemctl restart hms