Looking for opinions on setup.
My current environment is:
I have 3 buildings, House, Detached Garage, Shed, all close enough that I have direct burial Cat5e to support network. Metal siding and insulation is such that I lose nearly all rf between buildings, so I have to use individual rf bridges in each building. (Whether it’s Wifi, Zigbee, ZWave)
My current HA setup is a HA VM running on Freenas + 2x Raspberry Pi’s running Raspbian + Z-Wave JS UI Docker Image + Zooz ZST10-700 USB stick. Pis are just acting as a zwave bridge, nothing else.
Design wise, my goal for the Pis was to have zero commercial 3rd party software layers to trust. With Raspbian pretty well supported, and Zwave JS UI appearing to have a healthy dev path, I felt that was a great long term, DIY, never deal with someones failed business plan(WINK) again path. The zwave JS UI setup is working great, functionality wise. Love that integration, visibility in to zwave network is awesome.
The one sticking point is updates. At this point, from within HA, I can update the TrueNAS box, my Unifi install, HA, and of course all the plugins. But I cannot find a way to keep the Pi + Zwave JS instances updated without SSHing and manually running, or entrusting it all to a bash script that will eventually be an auto-break-things script.
Not a programmer. Not in a position to roll my own component. Network/Sysadmin/Some coding via tinkering & modification
So, after reading about linked HA nodes, I wondered if I should be doing some sort of HA + Zwave JS UI plugin + USB Stick instead of the raspbian. That way the OS layer would be handled, but I really don’t want to have to log into 3 HA boxes to maintain. And somehow that feels more overkill than the raspbian boxes.
Ultimately, I want the Pi + ZWave JS UI + Zooz to work/feel like a zwave bridge appliance, but without a commercial interest eventually pulling the plug.
Thoughts?