Renamed files and directories

Jamie Lokier jamie at
Thu Feb 26 12:56:58 GMT 2009

N.J. van der Horn (Nico) wrote:
> >But you need to verify and update the DB contents - which requires
> >stat on all the files mentioned in the DB.  In other words you might
> >have to scan everything :-)
> >  
> This already takes place while Rsync does its job, so it has not to be 
> done separately.

Right, but it has to be done in a separate pass if you're to compare
all files with each other, not just one destination file.  And you
need all the RAM, too.  It's like the worst case of "rsync -H".

> Adding a DB to Rsync would give many more advantages, like:
> - de-duplication (eliminating copies)
> - alternative to "locate"
> - filesystem statistics/analysis
> If the structure is choosen well, it can prove to be very valuable for 
> other purposes also.

I vaguely remember a conscious decision not to expand rsync that much.
Keep the tool simple and stateless.  There are already other tools
which do the things you've described.

> It must be possible to enable/disable checksumming when the
> timestamp and size are unchanged.  That clever trick is pretty
> reliable in normal Rsync usage as well and earns a lot of savings.
> We only do once every while full checksummed Rsyncs to be sure, but
> see seldom transfers then.

You could use that trick, but it's more dangerous when you're looking
through the _whole_ filesystem for a matching timestamp and size, not
just looking at the corresponding paths of a single file at each end.

-- Jamie

More information about the rsync mailing list