This page is what confused me. Seems like installing on existing ‘linux’ would be ssh’ing to the system and pulling HA down with apt-get or wget and then going through install, these instructions say " Flash the downloaded image to an SD card using balenaEtcher." That sounds like you are taking a bootable img file and booting your device to run Hass OS.
Flashing the Raspbian image, not HassOS. Read my guide, lots have used it and had success with it.
I can’t contact you directly, but when you connect the Pi to the network in step 6 and then say to ssh to it in step 7, I assume the Pi will use DHCP and I will have to look in the router for a ‘new’ mac address to get the ip address ?
If you can, give your RPi a static IP in your router. If not, you can set a static IP from the RPi itself. It’s very important the machine that is running HA, your MQTT broker etc, has a static IP.
I understand this completely. Unless the Pi has the mac printed on it somewhere, you’d have to use the technique I mentioned. Find the IP it was given in the router by noticing the mac prefixes of the Pi foundation B8:27:EB and DC-A6-32, then static-fy it. Actually alluding to this in step 7 might be a good edit.
“Step 7: …Check your router for the IP address of your Pi and use Putty to connect via SSH, on port 22 by default.”
Check your router for the IP address of your Pi (you can identify the Pi by the Pi foundation Mac address prefixes of B8:27:EB and DC-A6-32) and use Putty to connect via SSH, on port 22 by default.”
Thanks for the feedback, feel free to create your own guide.
My router showed the new device as … wait for it … raspberry pi. Thanks for making the guide. I was able to install a lighttpd webserver and other things. Now I have to learn how to use HA. Thanks
I’m not sure if this guide that you guys mention is still in the google drive folder. Would like to check it! Is it possible to reshare it?
Thank you for your fast response! I will be trying this inside an LXC container. Initially I was looking to install a HAOS VM in my RPI4 but there isn’t much information about this, and so far I got stuck.
It is mentioned in HA docs, that Supervised vs OS is only the Managed OS missing. What is this Managed OS? Is it about other software packages or libraries needed to be installed in the OS to properly run the HA project? If so how stable is a Supervised installation?
Because it a far from a recommended way to install. I would go as far to say it’s completely pointless.
HA OS has a base OS built on Buildroot.
Very.
try this guide which I believe is suitable with you.
https://krdesigns.com/articles/installation-home-assistant-with-supervisor-on-debian-10
I bumped into an issue running the installer. When the script creates /etc/docker/daemon.json file and restarts the service, docker does not start again. I need to remove that file and restart the container in order to get docker back up.
I’m using LXC to create a container:
lxc launch images:debian/10 v2 -p container -c security.nesting=true
Had no problems testing out docker hello-world on the container: hello-world (docker.com)
Not sure why it is so pointless. For me would make lot of sense as I would like to run HAOS VM on pi and other containers running other services
You can do that with Supervised. There is no need to stress the limited resources of a Pi running a VM.
So, I registered because I was irritated about the non understanding answers.
Did you look at the referenced thread, meanwhile non supported OS like ubuntu etc. are blocked for supervised install, which is absolutely great…
Why don’t you see the use case?
I want a supervised/managed Home assistant but also want to use other services/container like seafile or whatever…
How do others manage this on a PI?
So what is the only option when you look at the fancy installation matrix of HA? Yes, it seems like HaOS and then in a VM.
You can actually force install it on non supported OS: Block Unsupported OS by ikifar2012 · Pull Request #262 · home-assistant/supervised-installer · GitHub