Hi,
i just wanted to restore a backup because i corrupted my db file.
HA does not have an internet connection, and now the restore job refuses to work because of
INFO (MainThread) [supervisor.jobs] 'BackupManager.do_restore_full' blocked from execution, no supervisor internet connection
Why the hack does a restore job need an internet connection? 
Thanks in advance
I just ran into the same problem. How can HA claim to be “local” when you have to have an internet connection to restore a local backup file? This is crap.
Experienced the same surprise today. Looking to restore a backup i’ve copied to a machine without network access.
I had the same issue 2 weeks ago. Full internal network available but no internet and restore does not even start.
Also worth to mention that if you want to configure MariaDB you can’t make it without opening the internet because it tries to check the password against the HaveIBeenPwned database. Not very privacy friendly to send your password, even if it’s hashed, to any service. It’s good to block the internet, so you know what shit is happening in the background
has this been fixed ? does anyone know how to restore a backup without internet.
i need to restore to a new server without the old one knowing about it until i switch them over.
Just hit this issue last night after a power outage that took out the internet connection. When HA restarted NodeRed went into a reboot loop and I wanted to downgrade it to the last version, but due to no internet and this enforced restore condition (which suprises me for privacy reasons) I could not restore and thus lost all control of my house as I run nearly everything via MQTT, ZW and NodeRed. I went back to sleep and manually turned the lights and heating off. An option to disable or override this requirement from the command line (a bit like auto patching defeat) would be appreciated.
Would also be interested if it’s still a thing…
Should we file an issue on github?
I will try to make a FR this week, a CLI option is all that is required, in IMHO:
I am going to suggest -n --nointernet
Yup, this is apparently still a thing with the current latest HAOS version including all HA components up-to-date.
Once you temporarily resore internet connection, you’ll need to reboot HAOS I guess due to caching issues.
This whole restriction is defeating the purpose of having a fully local only instance for a smart home, also questions how much tight control do we even have over our own smart home.
I think it needs the time from an NTP server.
Good comment! 
Makes sense for rPi, although I’m using an x64 PC which has RTC. The errmsg still wrong not stating the actual reason needing internet leading to potential misunderstanding.
I’m having the same issue trying to do a test restore in an isolated network.
I have the same problem, is there any solution now?
