VC: 7 Update 3w
TCP: 5
The patch update fails due to a vmafdd RPC bind failure preventing the appliance management framework and vmware-statsmonitor service from retrieving machine certificates from the VMware Endpoint Certificate Store (VECS), which results in a fatal unhandled C++ exception during the SSO domain query phase of the upgrade.
Workaround (To restore services outside of a maintenance window):
Exit the appliance shell by running the appliancesh command.
Unstage the software packages by executing software-packages unstage.
Manually restart all vCenter services by executing: service-control --stop --all && service-control --start --all.
Note: If these steps do not restore the services, rollback the appliance to the latest backup
Resolution (To resolve the root cause and complete the patch update):
Log in to the vCenter Server Appliance via SSH.
Validate certificate health across all VECS stores by executing the following command: for store in $(/usr/lib/vmware-vmafd/bin/vecs-cli store list); do echo "[*] Store :" $store; /usr/lib/vmware-vmafd/bin/vecs-cli entry list --store $store --text | grep -ie "Alias" -ie "Not After"; done
Identify any expired or invalid certificates, specifically focusing on the Machine SSL or STS certificates.
If expired certificates are identified, utilize the vSphere Certificate Manager (/usr/lib/vmware-vmca/bin/certificate-manager) to renew or replace them.
Restart the affected services and verify that vmafd and vmware-statsmonitor start cleanly without logging further C++ or RPC bind exceptions.
Re-initiate the patch sequence from the Appliance Management Interface (VAMI).