I keep having significant problems with Supervisor where it just won’t start - details/screenshots below.
I am managing a few different home assistant instances and several of them had the issue by now, my own has even had it several times. Usually this is fixed by a restart. But since supervisor and addons are unavailable, I have to do this manually.
I’ve found a couple other threads on this as well, but basically all of them either fix it by restarting or by doing something in the terminal. Both of which I can’t do remotely because supervisor is down. And even if SSH was available, I feel quite uncomfortable to make port 22 available on all my friends/family.
So honestly, I’d like this to not happen again
I can’t really make out why this is happening and how to avoid it in the future.
Can you help me out? Please let me know if you need any information from logs - not even sure where to start with this
some screenshots
when trying to access addons settings:
SD cards are not the best for a system like HA that writes a lot.
32Gb is also a bit on the edge, especially if you also store local backups.
The reason I pick on the media is the log output, which looks like it simply stoppe writing there, because there where no more space.
A restart will then delete homeassistant.log.1 and rename homeassistant.log to homeassistant.log.1, which frees up space again to run, but only for a somewhat short period of time.
11Gb is a lot to fill for logs, but if the recorder is not managed, then it can eat up a lot too and maybe it will fail with a corrupt database then too, which means it will do the same as the logs and overwrite an older corrupt database backup and thereby free more space.
I just pulled the logs from a different installation (I know, not the same thing, but it’s all I have right now). Both (homeassistant.log and homeassistant.log.1) combined are 400MB in size. I feel it’s unlikely that this would cause the issue, right?
Backups are done externally, so this shouldn’t add to it. Or at least not too much, only the automated ones from addon updates and so on.
If the installation crashes for some reason (let’s ignore power loss, that shouldn’t be an issue), would that produce a similar log or would that write “complete” logs?
Backups are usually done internally and then copied to the external location, so you still need the space for the backup.
Normally it would write the complete line, because the log text is sent to the logging service and then that service write the line. The crashing of other processes would not interfere, unless it was a critical kernel process.
But these Backups are also only around 250MB max - in most cases more like 150MB.
I find it very confusing because I don’t really see any indication of where this is coming from. It just suddenly fails (and often I don’t even notice for a couple of days because everything else just keeps working more or less).
The complete backups are usually in the Gb size and when the space runs out, then HA just keeps running in memory with no logging and no database writes.
It might also just be areas on the SD card that are defective and when your HA hits them, then your issue arise.
My own installation is running on a 128GB SSD, but I have already had the issue there as well.
But I just checked and it is actually already 93% full (4GB left), so that would also point towards a storage size issue (rather than a storage medium issue, while I do understand that SD is not ideal)
My own installation just ran into the same issue again!
These are the last few log lines:
2025-04-11 00:24:32.904 INFO (MainThread) [supervisor.store.git] Update add-on /data/addons/git/5c53de3b repository from https://github.com/esphome/home-assistant-addon
2025-04-11 00:24:32.905 INFO (MainThread) [supervisor.store.git] Update add-on /data/addons/git/3490a758 repository from https://github.com/ikifar2012/ha-addons
2025-04-11 00:24:32.907 INFO (MainThread) [supervisor.store.git] Update add-on /data/addons/git/cebe7a76 repository from https://github.com/sabeechen/hassio-google-drive-backup
2025-04-11 00:24:32.908 INFO (MainThread) [supervisor.store.git] Update add-on /data/addons/git/d5369777 repository from https://github.com/music-assistant/home-assistant-addon
2025-04-11 00:24:32.909 INFO (MainThread) [supervisor.store.git] Update add-on /data/addons/git/a0d7b954 repository from https://github.com/hassio-addons/repository
2025-04-11 00:24:32.928 INFO (MainThr
Again: stops in the middle of logging.
What would be the ideal next step to troubleshoot this? Restart it and then delete some old backups (this was before I implemented generational backups, so there is a lot of old stuff here) and then see if the issue comes back or not?
Or do you have any other inputs that I should consider now that I have a problematic installation on my hands?
