You are not logged in.
Hi!
I'm setting up a small home server for the family with a 5 TB drive. The large storage of the home server will mostly be used for:
File syncing using syncthing or some other software.
Backups, mostly from Linux machines but also a one or two Windows ones.
There are a lot of files which are synced by me and my partner, like a lot of family images and videos. We also have backups which will include these files. That means that a lot of files will be stored 4 times or more in total on the server.
I'm thinking that it would be nice if file backups were stored as plain files instead of in zipped tar archives or similar. Instead I'd utilize Btrfs compression and deduplication (I guess a manual one with a timer trigger) which could deduplicate from both the file syncing solution and the backup solution since the files are plain.
Do you see any problems with my idea? Even though the file contents are identical the metadata like access time etc will be different, will that be a problem? And any solutions to those?
There are vast number of backup software available. Do you know of one that will fit this use case?
I'd be greatful for some feedback. I'm new to both Btrfs and backup solutions.
Offline
according to the btrfs docs https://btrfs.readthedocs.io/en/latest/ … ation.html you can do either file based dedup or block based dedup - there're also two tools linked
btw: 5tb is an odd size - do you plan to use a single drive? as a backup target you should use at least mirrors or a raid6
Offline