Description of problem: The procedure to migrate the engine/Manager database to a separate machine [1] is not working. It should either be fixed or removed. It's not clear we want to keep it, even if it can be fixed. Our testing indicates that usually, running the engine db on the engine machine performs better, overall. The only reason we can think of, for running on a separate machine, is for organizations that know PostgreSQL well, have policies/tooling/etc for managing it, and prefer to put every PG database they can, on their own PG infrastructure. It's probably broken for a long time now. The procedure to migrate DWH (and its DB) to a separate machine, on the other hand, does make sense, from a performance point-of-view, and should definitely be maintained in the future (and is generally up-to-date, although I didn't test it myself recently (and it might require some fixes as well). [1] https://access.redhat.com/documentation/en-us/red_hat_virtualization/4.4-beta/html/administration_guide/migrating-rhv-databases
Lucie, could you please share your thoughts on and experience with this?
I was doing some testing to migrate engine database to separate machine and also create ansible role to do that on 4.3 https://github.com/oVirt/ovirt-ansible-local-to-remote-db. I can see use cases where it can be requested but on large environments it probably will cause performance issues. On the other hand migrating dwh database and service makes sense with Grafana as Didi said and I'm testing it in Bug 1860067.
From @mtessun via gchat in response to Marina: Thanks. @Steve Goodman So let's drop it instead of fixing it. So I will remove this topic from the RHV 4.4 documentation.