I wrote a summary, when I reinstalled my home cluster (AFAIK I changed from CentOS 7 to 8 in the meanwhile)I tried several ways to upgrade a node and the hosted-engine and it didn´t work. The upgrade of a standard installation of Rocky 8.x to 9 works fine, but when you add the Ovirt/CentOS repositories, dnf starts to throw a bunch of conflicts and it doesn´t work anymore, and you have just bricked your host. Reinstall a node is easy. The problem is reinstall a node with GlusterFS hyper converged. For the hosted-engine, the only option would be reinstall it and restore a backup, right?
Erdősi Péter
Informatikus,
IKT Fejlesztési Főosztály
Kormányzati
Informatikai Fejlesztési Ügynökség
cím: 1134 Budapest, Váci út 35.
tel: +36 1 450 3080
e-mail: erdosi.peter@kifu.gov.hu