I recently onboarded a Proxmox server, and so after watching the 2025.1 Release party issue, I thought I would give it a try at doing a Backup / Restore.
*After it had been installed, I opened in my browser and was presented with an option with ‘RESTORE FROM BACKUP’
Exported from my backup on my RasPi
Moved my existing pi to a new IP. ( Renamed it with -old to make sure it didn’t conflict. )
Began the Restore, but it hung on the screen: Restore in progress
Going to the URL using the name it received form being restored, fails over http. Using https, it loads to https://homeassist.benandtoni.com:8123/lovelace but just shows ‘Loading Data’ and continually going to Retry Now.
The easiest solution I think at this point would be for me to destroy the VM and restart the automation, configure it and then restore what I’m missing. With that said, as this is a new feature, I was wondering if the information regarding what went wrong would be useful.
Things I have found:
Per the screenshot, and after investigating the media, the SSL files are not present.
The setup may also not have like its IP changing, as it was auto assigned an ip via DHCP, and then the restore would have changed its ip to a static address.
Are you trying to restore a backup made in an earlier different version? Elsewhere in these forums there are reports of problems restoring a backup made on 2024.12 into a system running 2025.01…
I ended up building the VM again, running through the same process.
This time, prior to coming online I made sure that the IP of the new machine matched the ip of the machine it was replacing. I then set it up as a new instance, installed the Let’s Encrypt SSL package. I’m not sure that I’m actually using it, but it was part of the steps that I was using before.
In the Let’s encrypt, I went to edit yaml and pasted in the yaml from the old machine.
I then installed the FTP addon, and uploaded my SSL files, renaming the files created by the Let’s Encrypt add-on.
Finally I downloaded the /config/homeassistant.yaml and added the lines:
After that, I went to backup and uploaded my previous backup, and selected to restore from it.
At first it seemed like it was doing the same thing, but as I had FTP enabled, I was able to collect the configuration from before and after the restore and confirm the files configuration files from backup matched.
I ended up rebooting the machine from the console. Upon reboot, I was able to login to it with the correct url and most of the resources were working. ]
I hadn’t yet moved my zWave stick from the old instance, so I manually moved that, and assigned it to the HaOS VM and then did a reboot of the whole ProxMox server and was able to communicate with my zWave devices.
=========================================
A few things I found out:
Even if Home Assistant was off, if I opened my browser to the URL, it would open and show the Loading Data / Retry Now. Based of this, I suspect that I was hitting a locally cached copy instead of the actual server, so I cleared the data for those urls.
I think that even though HomeAssistant was configured for the new IP, my VM was not giving it the ip I requested, so when I tried to access it, I was getting the cached version from above.
I wasn’t able to find the SSL files on the first attempt and there were multiple errors referencing it, hence the steps I took to manually put the files where they were expected.