HAOS - docker containers no longer starting

[techiem2 (Mark D Montgomery II)]

I’m running HAOS in a VM and it appears that the Supervisor docker image automatically updated last night/this morning.
I was getting some errors this morning and after rebooting the VM now none of the containers other than Observer are starting.
I can successfully manually start some of them, but Supervisor will not stay running - it just immediately exits on start.
I tried reverting the VM to a backup from several days ago and it started to load, but then apparently Supervisor updated itself again and died, leaving me once again with a dead system with only Observer running.

Any suggestions to get back up and running?

Thanks!

[jeffcrum]

I would revert it again, take it off of the internet before starting it back up, and find out where to shut off the auto update before letting it talk to the nets again.

That’ll get you running again. Not sure what to do about the issue after update.

[techiem2 (Mark D Montgomery II)]

Thanks, that helped.
I found my most recent backup that wasn’t borked (morning of 11/21…so apparently the first flaky update was between then and the morning of 11/22…it just didn’t fully fall over until last night) and have the VM blocked from the internet except for the IP that Matter Server needs to pull the cert list and port 123 for NTP since time is kinda important lol.

Now I just need to figure out how to disable the core dockers from auto updating…and remember to make backups before letting them lol.

[techiem2 (Mark D Montgomery II)]

Ok I found this post:

Stopping HA Autoupdate?

Hi guys, I’m running a pretty substantial HA build on top of Arch Linux. Right now I’m fighting with the system as it keeps auto updating on me and I haven’t told it to do so?? It’s getting pretty annoying and frustrating because I have to keep rebuilding the docker instance from scratch which defeats the purpose, almost every day now? . Basically eons ago I installed Hassio, so I have a systemd service called: hassio-supervisor Now currently in docker I am using these containers: …

That gives the command to apparently stop Supervisor from auto updating:
ha supervisor options --auto-update=false

So I did that and I’ll let it have internet back after I finish a little cleanup and see what happens.

[jeffcrum]

Make sure your last backup is saved off somewhere

[techiem2 (Mark D Montgomery II)]

Lol yeah I did a full backup of the VM and did an HA full backup and downloaded it.

[techiem2 (Mark D Montgomery II)]

I ended up doing a fresh install last night. I was getting errors with several addons and integrations that didn’t make sense.
The fresh install wouldn’t load the backup from the previous one, so there was definitely something else going on with that install.
I’m guessing something was conflicting with the newer Supervisor to cause those crashes, but there were obviously other issues.
Probably some random thing I’d installed to test out.
Fortunately I didn’t have all that much setup, so it wasn’t too bad to get the important stuff set back up quickly.

I did turn off auto updates of Supervisor on the fresh install to make sure I remember to run and download a backup before updating it just in case.