Anyone else seeing this when trying to upgrade to Supervisor 240?
Nope. Just updated.
I just tried upgrading again and it appears the first upgrade worked, but the UI didn’t update? Rebooting. Thanks for letting me know its working for you.
Sorry, this may or may not be related to Bad Gateway, but it does seem related. I can post elsewhere if needed.
My system is toast after upgrading to 240. Was on 235 and saw the update today. I think the DNS is shot breaking my mqtt and openzwave. I tried using time shift to restore before the update, but the backup must not be backing up all the needed files since it restarts with the supervisor still at 240 and the rest of my configuration looks like yesterday.
I’m running it in Docker on Linux Mint 19.3.
How do I downgrade my supervisor back to 235 where everything was working? I can’t seem to find anything on this.
Right now my whole home automation is shot and I’ve wasted hours on it today.
My Supervisor log is now saying the update to 240 failed, but yet it is reporting 240 as the version of the supervisor running.
20-09-06 00:55:16 INFO (MainThread) [supervisor.bootstrap] Use the old homeassistant repository for machine extraction
20-09-06 00:55:16 INFO (MainThread) [main] Initialize Supervisor setup
20-09-06 00:55:16 INFO (MainThread) [supervisor.bootstrap] Initialize Supervisor Sentry
20-09-06 00:55:16 INFO (MainThread) [supervisor.bootstrap] Setup coresys for machine: qemux86-64
20-09-06 00:55:16 INFO (SyncWorker_0) [supervisor.docker.supervisor] Attach to Supervisor homeassistant/amd64-hassio-supervisor with version 240
20-09-06 00:55:17 ERROR (MainThread) [supervisor.docker] Docker logging driver json-file is not supported!
20-09-06 00:55:17 ERROR (MainThread) [supervisor.core] Update 235 of Supervisor 240 failed!
Is there a way for me to move to a supported configuration with Docker on Linux? I didn’t realize until today that I’m running a non-supported config.
Thanks,
Mike
You can’t downgrade Supervisor. And if you do (e.g. by restoring an old snapshot), it will automatic upgrade itself to the latest version.
You just needed to refresh your browser. The 502 just means it’s still reloading the supervisor…
You need to set the correct logging driver for docker… the log is telling you that. See the debian 10 install guide for exactly how to do that… it’s discussed at length a few weeks ago there.
Thanks, David. Yes, I had fixed that once and then I restored another time and it was gone at the time I copied the log to this post. It’s fixed again.
My system is showing at Supervisor 240 and I found I had to reinstall all my addons. openzwave daemon is still toast. It’s like it it’s not seeing my /dev/ttyACM0 device. Is it possible that in the move from 235 to 240 I missed where I have to set up something for Docker containers to see outside devices. That’s all I can think of at this point.
Mike
EDIT: OK, well, it turns out sometime through all this mess my USB Z wave stick is now /dev/ttyACM1 and so it seems to be working now. What a day!
When you restart docker it doesn’t restart the addons I have found. There is a new feature in supervisor that restarts them after a crash but not sure if this qualifies as a crash. Sometime after a docker restart (say if you update docker) it is recoverable but often the easiest way is to restart the host then all the addons start again…
Well, I have my system working OK now. Had to rebuild a lot of the Z Wave stuff like when I recently moved to OZW 1.6 (I think that’s right). Anyway, since I’m using an unsupported configuration I’m planning to move this all to a VM in VirtualBox.
It seems like the preferred install of HA is on a Raspberry Pi. I moved away from that a long time ago since it was so slow.
I would have thought that running Home Assistant with the supervisor in Docker would be the preferred installation with the Pi being a nice alternative. Unless I’m misunderstanding this whole mess. It’s been a very frustrating 18 hours.
Mike