vSphere Lifecycle Manager failed to prepare some hosts in the cluster for NSX-T Networking.

Problem

In a cluster containing many hosts, vSphere Lifecycle Manager successfully prepared some hosts, whereas vSphere Lifecycle Manager failed to realize NSX-T on one of the host.

Cause

Hosts can take different states when vSphere Lifecycle Manager triggers installation of NSX-T.
  • Cluster goes into Install Failed if vSphere Lifecycle Manager fails to remediate the entire cluster.
  • If one or more individual hosts fail, failed hosts go into Install Failed state. If there are other hosts in the cluster yet to be prepared, those hosts go into Install Skipped state. Both cluster and individual hosts display failure states.

Solution

  1. On the NSX Manager, go to System → Fabric → Nodes → Host Transport Node.
  2. From the Managed by drop-down menu, select the vCenter Server.
  3. Identify the failed cluster to view the error state. Click the error link to open a popup window.
  4. If the cluster is in Install Failed state, click Resolve at the to initiate transport node profile realization on the cluster.
    Important: With the cluster in Install Failed state, first try to resolve the remediation issues at the cluster and then try to remediate individual hosts. If you overlook cluster-level errors and directly try to remediate host-level errors, the UI does not allow you to perform any remediation action at the host-level.
  5. If one or more hosts failed but the cluster remediation status is Success, then navigate to the failed host and click Resolve to remediate hosts.
  6. You can also try to realize the transport node profile on the cluster by executing the following API command, POST /api/v1/transport-node-collections/{tnc id}?action=retry_profile_realization.
    This API command re-triggers the transport node profile on the cluster.