Hi,
after a sequence of unfunny events, I had to do a full restore from a 1 year old backup.
This backup is now running for almost a week (on a RPI4) and I still have the onboarding assistant that tells me, that the backup is in process.
On the other hand I was able to log-in to the web frontend and can see that my devices and my addons are there.
Is there a way to check the state of the restore to be save to reboot the device and put it back into the rack for production?
regards
Dom
I am in a similar boat as you right now. Maybe we can help each other. How do you log into the Web frontend?
I did a restore of a pretty big instance. A couple of months ago, restore was optimized a lot, so it should definitely not take this long.
Did you use the upload function? With that I had a similar experience - it did not upload and looked like it was busy but it was not.
I installed Samba, uploaded the backup and restored it from there. It was done within half an hour if I remember correctly.
If you have access to the cli, here’s how hou monitor progress:
Hi, I did the ha job info and got a quite big YAML back.
I am Still checking, but I am sure that all the entries have a “done: true”:
done: true
errors: []
name: backup_manager_full_restore
progress: 0
reference: 2098f791
stage: check_home_assistant
uuid: ...
Would you consider that the restore is complete in this case?
Hi, I simply accessed the frontend by IP and accepted the invalid SSL certificate. In my case even the FQDN works fine. It simply worked, no magic attached.
This seems like the final stage, where it is checking the results.
I understand, so stage: check_home_assistant is more important dann done: true, right?
I’m no expert, but I think that is what I saw when I restored. A bit later it was back up (pun intended).