Good writeup!
At least it wasn't a BTRFS vol filling up, or you'd be waiting for restore ;)
After my dayjob went 95% SaaS I scrapped my HA homelab in favour of a tiny Ubuntu server with Podman containers. Started out ansible controlled, but with little free time I often fix stuff directly, making my config drift hopelessly.
With good backups and wireguard in front I've landed on autoupdates with major version pinning, but as most of us know, there are always new yaks to shave.