Grüß gott! Great stuff this esphome, I am considering migrating my nodes to it and try to help out any way I can.
A silly question perhaps but I am also developing for home assistant and I want be able to start using some esphome nodes on my local mac dev HASS and then “move” them to my production HASS. I can’t seem to find a way to delete a home assistant integration on the node.
What is the proper procedure?
I installed the addon on my RPI hassos and created a new yaml with a new name, flashed it while my hass was stopped on my dev computer. On the hassos RPI instance it doesn’t show up as an integration, even after a restart. I don’t know the protocol used so if someone can help me with how I proceed to add it to my RPI hassos I would be greatful.
Found https://www.home-assistant.io/components/esphome/ which says I need to add them manually. that worked but on my dev (non hassos non docker) dev machine they just showed up beautifully on the integrations page. Is this a known issue with RPI hassos that it requires manual configuration?
This is a general question regarding all yaml configs. If you read my entire post you will find that it seems like auto discovery doesn’t work in hassio but it does on hass on a dev machine.
The question boils down to: is auto discovery supposed to work on hassio or is it needed to add the node manually on the integrations page?
That would be mqtt discovery, so both esphome and ha are connected to same broker.
Esphome advertises thru mDNS, no hub or broker in between however ha can see esphome devices on the network and attempt to connect to them (the way api works ha connects to esphome) esphome accepts all clients who wish to connect.
Also auto discovery does not mean auto add integration. I think I have read somewhere they will stop adding discovered stuff automatically.
If you don’t want to wait you can also manual add it under Configuration -> Integrations -> esphome and just enter your node name (for example fonderbergsupernode.local) and just enter your password (defined in the node’s yaml). Profit!
Cheers @esphomeGuru uru! Once we’ve done that, does that allow us to see the logs, do ota uploads, etc… direct from the new instance. I want to update to a Pi4, but wonder about the overheads involved!