On Fri, May 29, 2015 at 05:23:41PM +0100, Patrick O'Callaghan wrote:
On Fri, 2015-05-29 at 18:08 +0200, Suvayu Ali wrote:
> > This may be a far more controllable and flexible way to give rsync
> access,
> > because you can use the .conf file at the server end to decide what
> may be
> > backed up, and the rsync command in all its glory at the client end
> to
> > decide what to ask for in a given run.
>
> For me the server does the pulling. I could try to explore the
> option
> where the backup is initiated by the client. I'll look at these
> options, thank you!
Note that incremental rsync-based backup systems such as rsnapshot are
designed for the server to pull from the client rather than vice versa. They will work
the other way, but you'll take a big performance hit
(about 10:1 in my experience).
Ah! This is useful information. I think with Gordon's script, I should
be able to setup a fairly automatic system :).
Cheers,
--
Suvayu
Open source is the future. It sets us free.