I’m trying to set up the Elk M1 Integration and I’m noticing weird behavior. I’m on the latest version of core 2021.2.1 and I’m connect via a serial connection. This is the behavior I observe:
Setting up the integration graphically results in a “success” message but no devices appear
Setting up the integration via config (auto-configuration: true) results in an integration appearing but no devices appear
If I delete the integration and then immediately try to add it, it fails to connect (reboot required)
The only way I can get devices to appear is by adding the integration graphically (no configuration entry), and then adding the configuration entry followed by a reboot
Auto configuration works reasonably well, except all 8 areas appear regardless of only having/defining 1 area
If I turn auto-configure off just for getting rid of the extra Areas, exclusion of devices don’t make the entities disappear, but they have a “restored” icon in the entity view. I assume I can delete these manually
Am I setting things up correctly? My main concern is that I would like to keep the Elk on-board configuration and HA configuration in sync, and I do this by deleting and re-adding an integration as necessary but this seems like it would require several steps each time. BTW, iif you know a better way to do this, please share, I’m still new at this.
I’m a little unsure of your setup here. Are you trying to connect via a different machine than the one where HA is hosted? If so you’re probably going to need tools beyond what HA can give out of the box (ser2net comes to mind).
I have a USB to serial device on my RPi4 where HA is hosted. It’s just a local device which I reference in the config like /dev/ttyUSB0 or something like that.
I’m assuming you’re running HAOS on bare metal, and not on top of Windows (like a VM or similar). Otherwise you’d have to make sure your port mapping is set correctly.
Either way, you may be missing a forward slash and have an extra hyphen. My config is: