I am getting this error for the 2nd time in 24 hours, after which it is impossible to restart hass.io. I have to resort to the following (now I know the value of snapshots!):
Restore a snapshot from ssh shell:
To get a list of snapshots: hassio snapshots list
To restore one: hassio snapshots restore --name xyz
hass.io seem to have become very sensitive to config errors lately. Also, quite often a hassio ha restart seems to take forever. I am running 0.65.5.
The strange thing is that after restoring a snapshot and after that copy back the config files that (maybe) caused the issue, hass.io restarts without problems.
The “Panic:…” issue thereby doesn’t seem to be related to a faulty config.
I’ve just had this. Restarts, supervisor reloads and powering off & on all made no difference. I’ve just attempted a snapshots restore (it asked for “slug”), which appeared to be happening but then all my terminal sessions got disconnected, and I get “connection refused” when trying to reconnect to the box!
It eventually was accessible again, but Home Assistant no longer running, and any attempt to start / restart it gives the panic. Further restore attempts result in the same disconnection and having to wait quite a few minutes before being able to reconnect, after which the situation is no better.
Do I have any options beyond a complete re-install?
I’m running 0.65.3 (thought I was later than this but that’s what is reported by hassio ha info) on a Pi model 3 original.
I’ve got the same thing running on 0.69
Edit: Back up and running again. Deleting the home-assistant_v2.db and doing a full reboot did the trick. The restart did take a long time…
I eventually sussed out mine was simply happening because of a syntax error in my configuration file. I spotted a missing quote mark and all was well. Strange it’s not happened to me before…
this is still here on 0.70, not able to reboot the host anymore, hassio ha restart does work mostly, only causing panic attacks when timeout errors are at hand :
having same issue on .70 . I was messing with config files a bit, but ended up reverting everything and doing a reboot. Now, I cannot get HASS to load up the UI at all. When I try to do a reboot through the command line I get this error:
I’ve been getting this error as well when trying to update Hass.IO.
Otherwise everything else is working but I’m unable to update.
Has anyone been able to solve this ?
ERROR
panic: interface conversion: interface is nil, not string
I believe the cause of my issue was switching from using an external db_url for recorder back to using the sqlite version. (I’m moving to using influxdb for long-term storage). I had meant to delete the existing sqlite db, but neglected to.
After I removed the db file, and rebooted the entire host, things went back to normal.
Same problem here, I literally spent an entire day rebuilding HA from scratch, adding little pieces of my configuration until it stopped working. In the end it seems like the culprit somehow was that I was running AVAHI on my PFSense router in order for discovery to work across networks, once I turned AVAHI off on the firewall HA would work normally. This is in addition to using the new cast configuration. So for example if I brought up HA with all of my configuration (minus cast / google devices defined as media players) then everything was fine. The second I allowed HA to add the cast configuration my HA would never start again (as long as AVAHI was running on my firewall). I’m sorry I don’t have any more detailed technical information, it took me literally all day to figure this out so now I will begin working through why having AVAHI running disrupts HA. Hopefully this information will help someone…
UPDATE: If i enable AVAHI on PFSense AFTER I start HA, everything seems to work fine and HA finds my google devices on the other networks and they work fine.