How much does Duplicacy need to download for incremental backup

I am very new in this “backup & restore” tech so pardon my ignorance if it’s a very basic question. I’ve only been using auto-sync pCloud and Dropbox like apps only recently moving to Backblaze B2. Reading up on what client side program to use, I finally decided to start off with Duplicacy. I have a few questions I’m curious about:

  1. Since i can do client side encryption (duplicacy init blah b2://blah -e), how much and what does Duplicacy need to download in order to do incremental backup work ? With everything (including metadata) encrypted client side, how does Duplicacy know the delta and how much does it need to download in order to find out what to upload ? Asking because B2 charges for downloads so wanted to know this.
  2. Also curious, what does it exactly download to know the above ?
  3. I run backups with “-v” and “-stats”, it show what+how-much it uploaded but not what/how-much it downloaded. Can this be shown too ?

All Duplicacy needs is the metadata of the last backup for the incremental backup to work. These metadata are split into chunks and uploaded to the storage, but Duplicacy keeps a local copy under the .preferences/cache directory. Unless you clean up this cache directory manually, normally Duplicacy is able to reconstruct the required metadata from the local chunks without downloading them from the storage. It just needs to make certain api calls to check these chunks actually exist in the storage.

duplicacy -d backup should show you every chunk that is being downloaded, although there aren’t any summary stats for downloading.

1 Like

Right, thanks for the explanation (and your wonderful work) !