Usage meter detected wrong hostname during the upgrade to 9.1

Products

VMware Usage Meter

Issue/Introduction

Upgrading to VCF Usage Meter can result in the below warning message while upgrading to 9.1

Checking if authentication to cloud is configured.
Authentication to cloud is already configured, continue with the upgrade procedure.
Has a snapshot of the VCF Usage Meter appliance VM been created on the vCenter Server environment that it's running on? (y/n): y
Detected wrong hostname. Expected hostname: <HOSTNAME>, but found: <HOSTNAME>.
This May cause issues after upgrade with the existing VCF Usage Meter certificates.
Please change the hostname and then you can either generate a new self-signed certificate or import an internal Certification Authority (CA) - Signed Certificate
Note: The certificate CN and Subject Alternative Name's DNS Name Must Match the hostname of the VCF Usage Meter appliance.
For More information, please refer to the official VCF Usage Meter Certificate Management - https://techdocs.broadcom.com/bin/gethidpage?ux-context-string=uM-003&appid=uM9-1&language=&format=rendered.I understand the issue and want to proceed with upgrade. (y/n):

Running the command below in an SSH session on the Usage Meter appliance displays only the short hostname, rather than the expected Fully Qualified Domain Name (FQDN)

hostname

Environment

Usage Meter 9.x

Cause

The VCF Usage Meter upgrade script validates that the appliance's local hostname matches its DNS-resolvable Fully Qualified Domain Name (FQDN). If only the short hostname is detected, the script triggers a warning message.
Proceeding past this warning without resolving the mismatch will invalidate the appliance's security certificates post-upgrade. Consequently, users will encounter errors when attempting to log in or access the web interface.
Also the upgrade process maps existing certificates against the FQDN, proceeding with a mismatched local hostname will invalidate the appliance's security certificates post-upgrade.

Resolution

Follow the steps below to resolve the issue.

Update the local hostname configuration to match the expected FQDN, regenerate the internal self-signed certificate to maintain cryptographic alignment, and then resume the upgrade process.

Step 1: Update the Appliance Hostname

Ensure a fresh, healthy vSphere snapshot of the Usage Meter VM is taken before proceeding.
Login into SSH session of appliance using the root user and update the hostname to the full FQDN:
```
hostnamectl set-hostname <FQDN>

example: hostnamectl set-hostname <hostname.example.com>
```
Edit the /etc/hosts file to verify and ensure that the local IP address mappings correctly align with the new FQDN.
Reboot the appliance to commit the structural changes:
```
reboot -f
```

Step 2: Regenerate the Self-Signed Certificate

Once the system boots up with the updated FQDN, replace the default appliance self-signed SSL certificate with a new self-signed certificate configured for a VCF Usage Meter appliance with deactivated FIPS mode.
- Replace the Default Appliance Self-Signed SSL Certificate With a New Self-Signed Certificate for a
  VCF Usage Meter
  Appliance with Deactivated FIPS Mode

Step 3: Resume the 9.1.0.0 Upgrade

Remount the upgrade ISO if necessary and run the deployment script again:
```
bash /root/upgrade/upgrade-um.sh
```
Verify that the pre-check passes the validation phase successfully.

Step 4 :Remove snapshot.

Note: Reverting to the snapshot of the source VCF Usage Meter appliance is not recommended, unless an error occurs during the upgrade.

Additional Information

Reference Link: In-Place Upgrade of VCF Usage Meter