Feedback

[angelnu]

Thanks for this project!

I forked it and adjusted it to make it wotk on my env: https://github.com/angelnu/proxmox-openshift-setup/blob/angelnu/NOTES.md

The key parts are:

• I run proxmox with 4 servers so I added support to distribute the OKD nodes properly
• I added the Kea configuration to make DHCP and PXE (iPXE for me) work
• I also re-enabled named so I only need a few DNS records on my primary DNS server

My plan is to transition from my k3s homelab to this as I use OpenShift at work and I was missing some of the OCP features at home,

[stratokumulus]

Nice :) I'm now using PiHole and Unifi in my network, so I don't use dedicated DNS/DHCP servers anymore.

My next update (which is already working, but I still need to clean it up before pushing) : 100% hands-free automated deployment (VM starting/stopping, shutting down bootstrap, ...). With my current Proxmox setup, it takes about 1h05m from nothing to a 6 nodes cluster. The README must absolutely be rewritten :D

[angelnu]

I also have Unifi and vyos so I could move the functionality there but it is quite nice having the OKD specific things in the OKD service node - this way everything can be triggered from one side.

Do you configure DHCP for PXE directly in unify? If so, do you do it via Terraform? I happen to manage my Unifi/vyos via terraform.

I am really looking for your update. I was also thinking on how to integrate the terraform/Ansible parts - so glad to hear you have already figured it out. For now I had made the VM state depend on a terraform variable which takes initial, bootstrapping, running as options.

I am also adding a data disk to the worker VMs so I can deploy DBs on them - I am planning to use https://github.com/topolvm/topolvm to provide local disks to my database clusters.

Are you doing persistent volumes with ceph/ODF or out of cluster? I used the ceph on proxmox before but I am considering longhorn now to keep it in cluster