I have to say… after just upgrading to v. 2025.3.2 that my opinion of the new “backup” features is a big step backwards.
Regarding the backup features:
You overcomplicated a simple thing that worked well already. I used to simply specify my Google Drive as a Windows target for the backup destination. Now I have to load a new Google Drive API?
I feel that the encryption feature adds no benefit (at least for me) and again overcomplicates what was simple and straightforward.
Google Drive (my former definable location for backups) is already encrypted.
** Where is the Windows folder for “local system” backups (the folder name is not shown anywhere) ???
Just curious. How did the previous Google Drive Backup manage to work without setting up oauth in the cloud panel. I agree, it was complicated although I’m happy to have the backup system integrated.
In addition, I configured all of the OAuth configuration for Google Drive, and added the Google Drive “integration”.
I performed a manual backup using the Google Drive API as a target, yet when I actually go to Google “My Drive” there is no backup file shown in the “Home Assistant” folder.
From within the Home Assistant web app, I CAN see the backup.
*** Why is this so complicated now???
I should have made an “Old school” backup before upgrading, as I can no longer revert back to the usable version of the backup software.
The older code worked very simply and logically.
I just pointed the destination of the backup to a folder on Google My Drive. It worked flawlessly.
How is a backup target to the same Docker container (which might have been completely lost or destroyed) usable anyway?
With the referenced addon, the local backup done stays on the physical drive in the Home Assistant filesystem and is synced to the cloud by the addon. It is not stored in the addon in any way. Then if you want to download it, you grab it from the cloud, or if you have way to get it from the Home Assistant installation, you can get it from there to use if needed.
In the past versions I successfully stored it directly in the cloud via Google Drive “My Drive” using no additional integration, just specifying the target of the backup in the GUI…
*** Why would one want to consume hundreds of megabytes of backup on the very Docker filesystem being “backed up” (even though it is synced via the newly required API)?