Is there anywhere in the Duplicacy documentation where the required Wasabi/S3 permissions are documented for each operation (e.g. backup, stage-1 prune, stage-2 prune, etc.)? If so, please point me to it (I didn’t find it).
If someone has looked into this please let me know if this makes sense. Goal: Limit damage that can be done if the machine getting backed-up get compromised.
Here’s my proposed plan:
Backup a headless VPS
-Wasabi credentials for a user that can only write and not delete from a bucket are used for Duplicacy config that is only accessible by root to backup the VPS. If the VPS gets compromised by a hacker and they get access to the Wasabi bucket credentials through the Duplicacy config, the backups already in the Wasabi bucket cannot be edited/deleted by the hacker.
Pruning
-On a different machine (e.g. a server I trust with no incoming ports open), I periodically run the pruning operation using a different set of credentials that have read/write access to the same storage bucket on Wasabi that was used to backup the VPS.