vSphere UI displays "No healthy upstream" error when accessing vSAN Data Protection

Products

• VMware vSAN

Issue/Introduction

After a successful deployment of the vSAN Snapshot Service appliance, the vSphere Client UI fails to load the Data Protection section.

The vSphere Client displays a "No healthy upstream" or "503 Service Unavailable" error message when navigating to Cluster > Configure > vSAN > Data Protection.

Environment

Snapshot Service Appliance 8.x

Cause

This issue occurs when the vSAN Snapshot Service appliance is configured with a short hostname (NetBIOS name) instead of a Fully Qualified Domain Name (FQDN).

During registration, the appliance provides its manifest URL to the vCenter Server. If only a short name is used, the vCenter Server or the user's browser may be unable to resolve the hostname via DNS, leading to a failure in the communication path between the vSphere Client and the appliance's UI service.

Resolution

To resolve this issue, the appliance must be updated to use its FQDN.

 

Option 1: Update the Hostname via VAMI

• Log in to the vSAN Snapshot Service appliance Virtual Appliance Management Interface (VAMI) at https://<appliance-ip>:5480.
• Navigate to the Network settings.
• Update the Hostname field to the proper FQDN (e.g., appliance.domain.com).
• Save the settings and restart the appliance services or reboot the appliance.
• Allow a few minutes for the vCenter extension to auto-update during its periodic poll.

 

Option 2: Redeploy the Appliance If the hostname update via VAMI does not resolve the registration error, redeploy the appliance:

• Delete the existing vSAN Snapshot Service appliance VM.
• Deploy a new appliance using the OVF wizard.
• In the Customize Template section, ensure the Hostname field is explicitly set to the full FQDN.
• Complete the deployment and verify the plugin registration in the vSphere Client.

Verification: 

• SSH into the vCenter Server Appliance (VCSA).

Run the following command to check the registered remote plugins:

python3 -m json.tool /etc/vmware-rhttpproxy/endpoints.conf.d/remote-plugins.json | grep -A5 "remote-plugin-cluster"

 

• Confirm that the sni and socket_address.address fields reflect the FQDN of the appliance.
• Refresh the vSphere Client and verify the Data Protection view populates correctly.