Starting my homelab by writing the handbook before the cluster
"You can rebuild a server in an hour. You can't rebuild the reasons you built it that way"
That's why I started my homelab with a handbook instead of a cluster.
A lab I did but can't point to is a lab I never did. The commands are easy to repeat. The thinking behind them, the mistakes, the "why this and not that", is the part that quietly disappears if you don't write it down.
So Phase 0 gets documented from the first command up:
- 3 mini nodes, one Kubernetes cluster
- Talos Linux on all of them (no SSH, no shell, config is the only front door)
- 1 control plane + 2 workers
Anyone can flash a node. Fewer people can hand you the reasoning that made the node worth flashing.
It all starts here.
What's a decision in your setup you'd struggle to explain today because you never wrote down the why?