After successfully upgrading Local Managers and/or Global Managers in a federated environment, some clusters may still show the old version number for the manager appliances in the Global Manager UI.
This may also occur during an upgrade pre-check or upgrade attempt and cause the upgrade pre-check to fail.
VMware NSX 4.x
This can be caused by if the Local Managers at a site are restored from backup, or it can be caused by issues with the order of the Local Manager nodes being upgraded.
It can also occur due to a data synchronization issue during an upgrade pre-check or upgrade attempt: see Active Global Manager upgrade fails with 'upgrade Standby first' error despite Standby already upgraded
To resolve this, SSH to a Local Manager or Global Manager of the affected site as root and run the following API:
curl -X POST -ik http://localhost:7999/api/v1/sites?action=refresh -H 'X-NSX-Username:admin;adminPassword'
Replace adminPassword with your NSX admin password
Refresh the NSX Manager user interface and confirm that the displayed software versions are correct.
Refer to the KB article to enable the root access on NSX manager Enable ssh root access for NSX appliances