All is well and all the docker containers are created, by i constantly get notifications of unsupported system, bad docker config, etc. Now recently I can no longer update supervisor as it has used some form of “dev” version also flagged as latest.
Is there a way to remove these warnings? I tried adding to jobs.json but it seems depreciated/ignored?
I looked at the “supported” method on install page, but this is not a supervised install so is missing all the good stuff!
And in the same doc…it shows a different (!) install … not supervised.
I am aware of many attempts and it may work but it will not (!) be supported, it also makes not sense to run a supervised (container) setup inside a container.
Maybe back to Q1: do you want supervised and why?
I only used supervised as I thought that was only way to get addons (node red, ssh/terminal and file manager) as well as HACS (for custom integrations, I use about 5 of them).
Are you saying with vanilla HA I will get ability to install HACS and then get everything that way? I have a feeling that even though it will not be supervised I will still get unsupported OS errors etc?
BTW, the supervisor, dns etc are all running in separate containers to homeassistant, not all within the same one.
Lots of wording and if you are not (yet) HA tech savvy it is a bit confusing at times.
Supervised allows to have a more automated way of installing applications/plug-ins, they are called add-ons and in reality they are also containers. The big + of supervised is that you are free of the hassle of knowing how to deal with containers and much is taken care off also wrt upgrade
I myself like being free and less controlled and no issues with techy things as containers.
If you are not into docker and no interest in getting there, install HAOS via the Syno virtual machine manager…see above recommendation on SSD
For me the downside of HAOS/super is to not briefly try a new version (docker/image-version) and go back when I donot like it. …this however hardly happens
bushbrother: if you do not have time in your hands to mess around with supervised, do yourself a favour and go with HAOS. It is the only long term solution that will give you the least problems. HAOS is like an appliance, with lots of options that all work; and in the unlikely scenario they do not work there will be a fix for it, faster than any other HA flavour.
If your focus is smart control and automation, HAOS is the way to go. If you like to fiddle with Debian and Docker, HAOS is not your thing.
If you want the most robust and most flexible environment, you should consider a type-1 hypervisor. It all depends on budget, your objectives and the time you have to devote to HA.
I run it in VM and it doesn’t suck at all…even more: it runs perfectly. But, as i said: ssd part is a must. You didn’t say which syno model do you have…but if it has ssd cache option, install it.
But, if i know correctly running HA in docker is possible only UNsupervised. VM is only option for “complete” HA sytem.
…but the point is that it “kinda” doesn’t work fine, right?
Investing in SSD cache on synology (if you have that option) is not only good for HA, but for all you have on syno…all works quite noticeably faster.
It’s working fine except the notifications, no point spending money on SSD. I already fixed the update and supervisor upgrade using docker. I will just put up with the notifications. I understand their purpose but there should be an option to accept them forever with an understanding there is no official support
My modified code below, simply adjust the release version it each time there is a supervisor upgrade (i.e. amd64-hassio-supervisor:2023.10.0) . Stop the container, delete it, re-run the script, works fine.
If you are on DSM7.2 you can achieve the same thing in the “projects” part of container manager using the yaml below, again just stop the project, edit the yaml and the version of the supervisor release, then re-build the project.