What is the ultimate difference between installing this on Debian 10 and Debian 9? Besides the Debian version that needs to be installed of course?
Is there any difference in the procedure or scripts after you have some version of Debian installed?
I’m on 9 and somehow I just managed to accidentally delete my HA Supervised HA container trying to update and obviously the old script I used to initially used to install it to try to re-install it won’t work.
All I need is the installation script for HA Supervised itself. Not all the rest of the stuff.
If you sudo -i, then you should just be able to run the script and it will only install what is needed/missing, that’s my understanding anyways. Might be wrong.
Before I stopped in here last night I had tried to run the script as just a normal user using sudo and I kept getting a “please install avahi first” message. I ran the peripheral package install command again in the chance that something changed since I used the old supervised install script and everything installed properly as far as I could tell. But I got the same message again.
then I used your suggested “sudo -i” to get root access and ran your installer script again and then it ran fine. But I still never got the actual HA container to redeploy. I even waited overnight just to be sure (which is why I waited till this morning to respond).
This time all I did was restart the Supervisor container using Portainer and in a couple of minutes the HA container finally showed back up and seems to be working again.
TBH, I’m not really sure what got me into this mess in the first place. For some reason I could never update HA thru the Supervisor before and had to go into Portainer to do it. Which is what I did this time and even tho I thought I had done it exactly the same as I had always done I must have screwed something up.
But the question is why couldn’t I update from the Supervisor before? is it because I was using the old deprecated install script?
Hopefully when the next release comes out I can test things again to see if using your new script fixes that issue.
Luckily this wasn’t on my production system. And this is one of the reasons I’ve been hesitant to move to HA Supervised as my production install. There seems to be too many mysterious “black boxes” interacting between the supervisor service, the supervisor container, the HA container and Docker itself. And with no official support/documentation for this install method and no easy troubleshooting method it kind of sucks when things go wrong.
I am using docker on my RPi3…and I had some problems updating to 114 from the supervisor page in HA. This post has me wondering if I shouldn’t do the update through Home Assistant…and maybe there’s a different procedure when using docker.
Has anyone using the supervised docker install upgraded to 114…and how did you do it
For it was a v114 specific problem. I haven’t ever been able to update HA from within HA ever since I installed the supervised version.
But I also used the old install script to install supervised literally a few days before it was “deprecated” so I never had a chance to know if it ever worked or if it’s because of the old script version I used.
@kanga_who
I noticed after completing your extremely well done guide that there is a message in the Supervisor panel that says
“You are running an unsupported installation. Learn More”
After following the link, I found that Frenck points out a few extra stipulations for a supported system which include using overlayfs2 storage and journald logging in docker.
I was able to make it a ‘supported’ installation by performing the following in a terminal:
# open/create a file with nano
sudo nano /etc/docker/daemon.json
# insert the following content then save (ctrl + s) and exit (ctrl + x)
{
"log-driver": "journald",
"storage-driver": "overlay2"
}
# restart docker
sudo systemctl restart docker
# wait ~10-15 seconds then restart Home Assistant
sudo ha core restart
It would be great if you could add creating the file and restarting docker between the Docker & Home Assistant installation commands; then people will end up with a ‘supported’ configuration.
Me neither. I thought it was strange too but every time I tried to update from within HA it never completed. But if I updated using Portainer it always worked. Until this time…
So how are you doing that with Portainer? Are you sure you are talking about a supervised install?
A few months ago, for some inexplicable reason, one of the images/containers was removed. The audio one or one of those types… had constant errors in the logs about it. Or it might have been there with the wrong tag for the image and supervisor couldn’t start it. I eventually pulled the correct tagged image manually in Portainer, restarted supervisor and it was happy again. So certainly glitches are possible.
I’ve updated the guide with this info after testing as part of a fresh install, I’m getting no warnings or errors in HA. Thank you for the info @mjoshd.
did you read the information on that note? It’s normal…the note just is a warning that you are responsible for some of your own system support…there is another thread on this…many are seeing this note for the first time…but nothing has really changed.
Any idea how that change affects other docker containers?
I assume that it would be fine for everything else as well but I still like knowing what changes are being made systemwide that might affect other things.
Tried this, still get the message “You are running an unsupported installation."
I edit the daemon.json file outside the docker, so on root command line.
That is correct?
If you follow the link and read the document it describes several conditions, all of which must be met, in order for the unsupported message to go away.
From the command line hat does it tell you for the storage driver?
If it doesn’t show the right driver you did it wrong… You would do it on the Host OS where you installed docker.
@finity it hasn’t seemed to affect my other docker stuff.
