What’s the objective? If I’ve reactivated Administrative Circumlocution, I must be trying to resurrect my homelab.

That used to be three machines:

File server FreeIPA Virtualization server - OpenVPN - squid - dhcpd - FreeIPA - PXE - GitLab - Userland

Some of those functions (OpenVPN, dhcpd) went to my pfSense instance. Some (PXE, squid) were never really necessary. Those two FreeIPA systems had constant split brain issues, as I didn’t yet know to have three instances for quorum.

Only the file server survives to this day. And I think that’s starting to bug me.

At the same time, I don’t just want to replicate the systems of the past; I want to bring my practices into modern times by running CoreOS and trying once again for this dream of software-defined infrastructure.

I last tried to create a hyperconverged oVirt cluster in 2022, but didn’t get anywhere with it. This year I tried only a single-node OKD system, but it broke for unknown reasons before I could really deploy anything. I think I’m overreaching.

I do want to focus on containers rather than VMs - but I want a system simple enough that I still know how to back it up. And I want it to be scalable: I want to be able to deploy kubernetes - or perhaps something simpler - to one node until I’m familiar with how it works - then add the other two nodes as I reacquire disks for them.

I also want the ability to spin up traditional virtual machines. Sometimes I want a BSD instance to play with, and I want this system to be able to handle that.

I want to be able to manage these systems with Ansible, and I want to learn vagrant. I want to have every VM and configuration change written in a git repository.

And I don’t even care anymore if what I’m doing is industry standard. I’ve been spinning my wheels for three years now, and it’s time to just make something I know will work.

EDIT: I changed some firewall and DNS settings, and now I’m able to connect to my single-node OKD instance. Guess I’ll keep playing with that for the time being.