VCF 9.0 Deployment Fails at NSX Transport Node Configuration Due to Host DNS Misconfiguration

Products

• VMware Cloud Foundation

Issue/Introduction

 During VMware Cloud Foundation (VCF) deployment, the NSX transport node configuration may stall or fail at approximately 48%.

The SDDC Manager reports an "Error in configuring VLAN backed transport node collections,"

In NSX, one of the Esxi host failed to be prepared, and affected ESXi hosts remain in a "Waiting for connection to Managers" state .

Environment

VMware Cloud Foundation 9.0.x

Cause

The ESXi host is configured with incorrect or unreachable DNS server settings. This prevents the host's Cluster Agent from resolving the Fully Qualified Domain Name (FQDN) of the vCenter Server Appliance (VCSA), blocking connectivity to management services and depots.
 

a) In Esxi vmkernel.log,  found error message:

 VmkAccess： 149：  ClusterAgent:  running in clusterAgentDom(38):  ipAddr = 127.0.0.1, port=53:  Access denied by vmkernel access control policy 

b) In vSphere Web GUI, check cluster and host compliance, there is an error with the Esxi host:

Failed to connect to depot: "URL:  http://<VCSA_FQDN>:9084/vum/repository/hostupdate/__micro-depot__vendor-vmw__metadata-65index__xml"  Error: '<urlopen error [Errno -2] Name or service not known>'

c) Check DNS settings on this specific Esxi host, DNS server is not configured.

Resolution

    1. Log in to the affected ESXi host   via the Direct Console User Interface (DCUI) or ESXi Shell.
    2. Verify the current DNS configuration and compare it against known working hosts in the same cluster.
    3. Update the DNS server settings to the correct environment-specific DNS IP addresses.
    4. Restart management agents if necessary using the command: `services.sh restart`.
    5. Return to the SDDC Manager dashboard and select **Retry** for the failed NSX deployment task.