Last month I had a problem with a customer who was unsuccessful in creating a new Instant Clone Golden Image in the HorizonAdmin portal.
The creating a new Golden Image task failed with the error message “Failed to primeResources for VM “%VM-ID%”; Convert the image back to the virtual machine – Success”
We looked in the Horizon DaaS log, Sysprep log, and Tenant appliance log files and noticed there were no errors we could work with.
It was not until we examined the Resource Manager appliance log files we noticed this error: “Problem connecting to VirtualCenter at https://%IP-Address% – Hostname in certificate didn’t match”
This was the reason why the Instant Clone Golden Image process had failed. In Service Center, the customer has added the Virtual Center (Desktop Manager) with an IP address instead of FQDN.
Unfortunately, there is no way to change the IP address to FQDN within Service Center. In order to do this, we must SSH to the primary Service Provider IP and edit the FDB database. But before you do this, it’s a good practice to perform a backup up front.
# psql -Uadmin fdb
> Supply fdb password
# UPDATE hypervisor_manager SET url=’https://FQDN VC/sdk’;
# UPDATE hypervisor_manager SET address=’FQDN VC’;
> Verify changes to specified tables (url, address)
# select * from hypervisor_manager;
After you did this, you go back to Service Center, go to the Service Grid menu, choose Resources, select Compute Resources, select vCenter and click the Update Compute Resources button (select the “Update Server Certificate” checkbox before clicking)
Turned out, it was a typical case of RTFM…Or in this case, RTF Release Notes:
In short, providing an FQDN instead of IP Address is always a Best Practice!