What determines a "full" scan?

There is an aspect that I still don’t understand in :d:, even after so long use.

When small changes occur in a repository, backup is insanely fast (seconds), quickly identifying the modified files and updating the storage.

However, when many files are modified / added, :d: seems to redo a full repository scan, obviously skipping 95% of the chunks, but it still takes a long time, as it scans the entire repository. It doesn’t seem to easily identify the modified / added files.

What determines whether a backup will “scan” the entire repository or quickly identify modified / added files? The size of these files? The cache?

Duplicacy checks the size and modified time of every file. If either one is changed since last revision, then the file will be fully read and split into chunks.

Only when the -hash option is specified will Duplicacy run a full repository scan.

1 Like