Support for --flatten? -- Another business case


The feature discussed earlier already is exactly what I would need for my backup scenario.
Here is the business case.

HCL Domino creates files for each attachment ( *.NLO ) which is moved outside the database for dedup and also to make the database file (NSF) smaller.

The files are stored in many directories and the name is the sha1 of the object + the size in hex as an unique hash.

The idea is to store unique NLO files in a single directory on a target to deduplicate from multiple servers.
In this case duplicates should be ignored and only new files should be copied.

This would be a good use-case for --flatten.

Today the only way I found is to specify the directory on the source side. But we can have many directories because by default only 40000 files are stored per folder.

And yes this only makes sense with a S3 target with support for large number of files in one directory.

Today on Linux I will just use a script to find the directory and sync per directory.

Thanks for your great application!


You can follow this issue:

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.