Hosts moved out of a cluster incorrectly appear in the Standalone host list in vSphere Lifecycle Manager 8.0u3 after moving back into a cluster

Products

• VMware vCenter Server

Issue/Introduction

In vSphere 8.0u3, you may see hosts that are currently members of a cluster incorrectly listed under the "Standalone" host inventory summary in vSphere Lifecycle Manager (vLCM). This typically occurs when viewing the inventory at the top level of the vCenter inventory.

 

• The Standalone host list is inaccurate or contains hosts that are part of a cluster and not standalone.

• Lifecycle Manager overview reporting shows incorrect data regarding standalone vs. cluster-managed hosts.

• The issue is specific to hosts that were managed as standalone vLCM hosts prior to being moved into a cluster or were moved out of a cluster and back in.

 

Environment

vCenter Server 8.0u3

Cause

A code defect in the inventory filtering logic causes the system to skip certain hosts during the removal process when querying for standalone entities. If a host was previously managed as standalone, a residual database record exists that causes it to be incorrectly included in the standalone summary view.

Resolution

This issue is fixed in vSphere 9.0 and higher. If you cannot upgrade, you can resolve the incorrect UI listing by removing the stale records from the vCenter Database.

Note: Always take a fresh backup of the vCenter Server and its database before performing manual SQL deletions.

1. Identify the Managed Object ID (Moid) for the hosts that are incorrectly showing as standalone (e.g., host-##).
   • The above can be done via the vCenter MOB or by locating the host id in the browser url when the host is selected from vCenter inventory.
2. Log in to the vCenter Server Appliance (VCSA) via SSH and access the database by using the shell command.
3. Execute the following command to access the SQL DB:

   /opt/vmware/vpostgres/current/bin/psql -d VCDB -U postgres

4. Verify the existence of the stale record for the affected host (example):

   SELECT * FROM vumdbc.pm_software_desired_states WHERE entity_id = 'host-##';

5. Delete the stale record:

   DELETE FROM vumdbc.pm_software_desired_states WHERE entity_id = 'host-##';

6. Repeat for any other affected hosts. The UI should reflect the correct inventory immediately or after a browser refresh.

Note: Make sure this deletion happens ONLY for those hosts that were earlier set as standalone and now have been moved under a cluster.