David Glasser wrote on Mon, 28 Apr 2008 at 20:42 -0700:
> On Mon, Apr 28, 2008 at 8:30 PM, Branko Čibej <brane_at_xbc.nu> wrote:
> >
> > Daniel Shahaf wrote:
> >
> > > David Glasser wrote on Mon, 28 Apr 2008 at 11:56 -0700:
> > >
> > >
> > > > On Mon, Apr 28, 2008 at 10:36 AM, Daniel Shahaf
> > <d.s_at_daniel.shahaf.co.il> wrote:
> > > >
> > > >
> > > > > Possibly, but I don't think a check that prevents running
> > > > > 'svnsync init $R $R' (where $R is an empty repository) is
> > > > > very useful. Most likely $R can be thrown away and
> > > > > recreated.
> > > > OK, in that case then I don't think any such check at all is useful :-)
> > > :)
> > >
> > > Thanks David.
> >
> > Perhaps that was meant as a joke, but I'd take it quite seriously. I can't
> > think of any 100% certain way of detecting a "same instance", short of
> > actually inventing another hidden repository UUID just for svnsync. I also
> > don't think it's worth the trouble; presumably admins that do svnsync init
> > sort of know what they're doing. And if they don't, what real harm can it
> > do?
>
Agreed. Someday it might be nice to tell mirrors apart, but I can't see
any other use for it now.
> That is in fact what I meant.
>
> Perhaps in the future I will make sure to include :-( after all of my
> serious statements.
>
> --dave
>
>
> --
> David Glasser | glasser@davidglasser.net | http://www.davidglasser.net/
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe_at_subversion.tigris.org
For additional commands, e-mail: dev-help_at_subversion.tigris.org
Received on 2008-04-29 07:28:24 CEST