Avoiding backup of duplicate files in a backup set

[johnbatty]

I am backing up a single machine that has multiple parallel directory trees that are largely identical (archived source code trees). So each tree is multi-MB, and contains 99% files identical to the other trees.

 

I was hoping that Retrospect would spot the duplicate files and avoid backing them up (even though they are in a different location). However, it does not appear to do this (judging by the increase in backup size when I add an new tree). I gather that if multiple different clients backup to the same Backup Set, then duplicates across clients WILL be detected. However, it looks as though maybe the check for duplicates only occurs at the start of the backup operation, comparing against what is currently in the backup.

 

Is there any way to achieve what I want to do with the current Retrospect? If not, will you consider adding this feature in a future release?

 

Perhaps I can backup a single tree. Then run another backup that has all the other trees. However, this would be a bit of a pain to manage - would be simpler if it "just worked"!

 

Thanks!