[Duplicity-talk] Duplicity as general sync/copy tool between backends WAS:Re: Duplicity filenames

[edgar . soldin]

On 21.09.2015 01:49, Ameir A. wrote:
> Hello,
> 
> I currently use Duplicity in a few projects, and it's been great.  There are 
> some scenarios, though, where I resort to aws-cli and swiftclient for S3 and 
> Openstack, respectively.  This is mainly due to needing filenames to be 
> retained as-is (and to be able to download them directly via API or another 
> client), while Duplicity (AFAIK) expects that all uploads/downloads happen 
> through it.

actually no. you are free to backup eg. to a file:// target and sync with the 
software of your choice w/ your backend.

after that you can operate on the backend (use eg. swift://) as if you'd 
uploaded through duplicity.
 
> Is there a way to have Duplicity be a simple rsync frontend to the multiple 
> storage backends, ultimately being the one hammer I need to sync files?
> 

not as sophisticated as rsync i am afraid. traditionally duplicity treats 
backends as dumb and did implement only the most basic operations that can be 
expected of a file based backend - namely put/get/delete/list . with only these 
duplicity will have a hard time of "proper" syncing let alone rolling checksum 
patching file on the backend.

having written all that there is still this (unimplemented so far) idea 
lingering around
 https://lists.nongnu.org/archive/html/duplicity-talk/2012-07/msg00019.html

..ede/duply.net