I’m running HA in docker. I’ve shut down all other add-on containers but still can’t get in. I can’t get into access other addons because HA won’t start.
I have access to the default config folder but can’t work out where my addon’s are installed there or even if they’re installed there. Any ideas what to do?
Thanks. Just removed init from the other container, and it worked.
The way i rolled back , was i duplicated the container, and then changed the image from homeassistant/homeassistant:stable to id id sha from the image like sha256:22526c9da9deb1d3f0c054ba33d1d464039ca8302d92165920c8a98da8ac59f6
Then started the container. This way i both have the old and the new version that i can change back and forth easily. Just stop one, and start the other This doesn’t work with major breaking things. Like when they upgrade databases and such
While I try to understand how to make the removal of /init stick in Portainer, here’s how I managed to revert to the old version, hoping it helps other Portainer users:
stop the Home Assistant container
go to the Images tab and locate the latest image with tags homeassistant/home-assistant:none
open the image and copy the image ID (line starting with sha256: )
go back to containers tab, open your Home Assistant container and click Duplicate/Edit
under Image Configuration, select “Advanced mode” (the one that gives you one line instead of two) and paste the Image ID you copied
Take a look at your docker command for starting up a new instance of Home Assistant. Do you see an –init parameter? If so, remove it and save your changes and re-run.
Anyone know if this would also affect a supervised installation? In the end that also gives you a bunch of docker containers and I can see an /init here in Portainer. In case the upgrade breaks my installation would this be the setting to change to ‘default’?
I’m always so nervous to Update HA, but I just redeployed my entire OS. Almost always something breaks :(. It takes long enough to get it setup the way I want, but I can’t help but loose the will to set it up repeatedly after breaking ‘updates’.
Thanks all for this discussion. I just updated from something in April to 2022.11.2 and ran into this. I thought I’d share my findings so that anyone else who encounters this issue has a bit more information on how to identify and resolve it.
My setup consists of
Raspberry Pi 3B+
Boot and run from an SSD
WiFi networking
Fully up to date R-Pi OS, 64 bits
Docker CE (according to dpkg) 5:20.10.21~3-0~debian-bullseye
Update strategy is to
pull the new container docker pull homeassistant/home-assistant:stable
run HASS backup
stop container
run my custom backup (which now includes the HASS backup) and copy off to another host
rename old container `docker rename homeassistant homeassistant.2022.04
restart container with the command previously used.
Next step was to run the container interactively to see if anything came up.
hbarta@polana:~ $ docker start -i homeassistant
hass s6-overlay-suexec: fatal: can only run as pid 1
hbarta@polana:~ $
( I clobbered the exact text in my notes but I do know it included the error message above.) Searching for that led me to this post and the solution. I restarted my container w/out the --init argument and all seems well.