a question for inhomgenous environments

[Tobias Polzin]

> I have the task to think of a backup system for our local environment
> (lot's of notebooks with windows and linux) and stumbled over your 
> great rdiff-backup and want to give it a try. Because I do not
> want to force everyone to install python2.2, I will use some
> rsync to mirror the filesystem to a backup server, and than make
> a rdiff-backup of this mirror for history backup. This may 
> nearly double the necessary discspace, but brings the installation
> effort to a fraction. Or do you have a better idea?

I thought how I could avoid doubling the discspace. 
a) one could compress the rdiff-backup in a tar file and 
extract it only for the short time it is needed (but this
could take some time...)
b) do not use an additional mirror, but use
  rsync --dry-run 
to find out, which files are new and which are deleted.
Copy the new files to some directory NEWFILES.
But what then? Is there a possibility to update only
a certain list of files (I tried --include, but this also
removes all files that are not in the list). I am looking
for something like the inversion of --restrict, a --allow-from-stdin.
This would be cool.

If that is not possible, I have to copy the current rdiff-backup
to a new file, add the files from NEWFILES, remove those
files that where deleted and than start rdiff-backup.
But this is a waste of cpu-time...

Any idea? Or should I look into the sources to add a patch for
that.

Tobias

By the way: Do you have experience how fast the rdiff-backup-data
directory for the diff-files grows?