NSX-T Version: 3.2.1
This issue occurs due to the following contributing factors:
NSX operations being blocked by vLCM-controlled remediation policies, which conflict with the standard NSX removal workflow.
A stale or orphaned Transport Node Profile (TNP) reference within NSX, preventing the cluster from completing the NSX removal process.
To resolve the issue and avoid low-level interventions such as database surgery, the following recovery workflow can be implemented:
Create a new vSphere cluster that is not managed by vLCM.
Migrate each host from the problematic cluster to the new one by:
Disconnecting the hosts 1 at a time from the original cluster.
Moving the host to the new non-vLCM cluster.
Reconnecting the host in the new cluster and verifying its operational status.
Once all hosts are successfully migrated, the original vLCM-enabled cluster can be deleted from vCenter.
NSX should automatically start removing NSX components from the hosts after they are moved into the new cluster.
If this automatic removal does not initiate, go to the NSX UI, select the new cluster, and manually trigger Remove NSX.
To verify successful VIB removal from each host in the new cluster, run the following command on each ESXi host:
esxcli software vib list | grep -E 'nsx|vsip'
Note: This method ensures that NSX VIBs are removed cleanly and avoids risking further corruption or stuck states in the NSX database.
If the NSX VIBs are still not removed from the hosts after applying the workaround, the removal process may remain stuck at 40% – “Removing NSX bits.” If this status does not progress, proceed with the resolution steps outlined in the following article:
NSX Failed to Remove NSX in a Security-Only Cluster
If neither the automated nor manual workaround is successful, please open a case with Broadcom Support for further assistance.
For instructions, refer to: Creating and Managing Broadcom Support Cases