v2.22.0 dockerimages moves to debian bullseye
mrhawkin opened this issue · 13 comments
ref:
https://khassel.gitlab.io/magicmirror/
⚠️ Beginning with MagicMirror² version v2.22.0 all images are based on debian bullseye, before this release the linux/arm images were based on debian buster. So if you are still running debian buster on your raspberry pi the karsten13/magicmirror:v2.22.0 and karsten13/magicmirror:latest images won’t work anymore, you have to switch to the compatible image karsten13/magicmirror:buster.
Workaround; modify run/docker-compose.yml to point to image: karsten13/magicmirror:buster instead of magicmirror.latest
Long term solution; make a new release of MagicMirrorOS built on debian bullseye
thanks for bringing this up here.
Solutions:
- for running instances (as written above): modify
run/docker-compose.yml
to point to imagekarsten13/magicmirror:buster
instead ofkarsten13/magicmirror:latest
- for getting rid of above manual action a new (buster) iso should be built containing #55
- long term solution (as written above): iso should be built on bullseye
Hi team !
I encountered the update error with the v2.22 MM-docker-image...
So I'm thinking to make a fresh installation of my rPI with MM-OS.
Question : Do you plan to make a new release soon with an up-to-date RaspianOS 'bullseye' ?
Thank you in advance for your reply ;)
Sent an image to build :)
We can also move it to run on github actions if it helps you.
Hey, there is a build you can try now. Its available on pi-imager or on the nightly server:
https://unofficialpi.org/Distros/MagicMirrorOS/nightly/2023-07-24_2023-05-03-magicmirroros-bullseye-armhf-lite-0.3.0.zip
does not work on my side.
I get a login screen at startup, if I manually start $HOME/scripts/start_gui
the mm scripts are executed but the container fails because there is no x-server started.
Same Issue.
Is there a Workaround or a Fix for it?
Hello, do we have any news on the subject ?
It would be awesome to have a up-to-date Bookworm image of MagicMirror OS 🙏
I am fully available for testing 😉
Hi @guysoft, do it... I don't know if I would have time to test but it could interest other people 😉
Also available on the unofficial rpi imager:
https://github.com/guysoft/pi-imager