[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: An alternative analysis of the etch architecture proposal



Frank Lichtenheld wrote:
> On Tue, Mar 15, 2005 at 05:50:23PM +0100, Adrian Bunk wrote:
> [...]
> > - issues with space on ftp.debian.org and on mirrors
> >   (especially hindering amd64)
> > 
> > It might be a better point to start moving non-released architectures 
> > (GNU Hurd and sh) to a different location. Depending on what exactly 
> > hinders amd64 today, this might be sufficient [2].
> 
> No. The plan is definetly to get the main archive down from currently
> over 100 GB to something way more reasonable (as there are currently planned
> two archs, this would mean something between 30 and 40 GB) and to get the
> needed daily pulses down from currently up to 2.5 GB.

Why not improving the mirror network and the mirror software?

I.e. why not requiring only the five tier-1 mirrors to mirror the
entire archive *and* providing Tools for easy mirroring a subset of
DISTR x ARCHS

DISTR = {stable,testing,unstable,stable-proposed-updates,testing-proposed-updates}
ARCHS = {alpha arm hppa i386 ia64 m68k mips mipsel powerpc s390 sparc}
        (plus whatever will be added in the future)

In order to support simple mirroring I could even think about a
slightly modified directory layout, such as:

   .../pool/... contains source + i386 binary packages
   .../ports/... contains all other binary packages

The Package files would then point to the proper locations.  This
would help non-tier-1 mirrors that don't want to mirror all
architectures to do so without more complicated mirror software.  And
it wouldn't drop architectures the way it is planned.

Maybe I'm missing the point.  I'd appreciate an explanation.

Regards,

	Joey

-- 
Ten years and still binary compatible.  -- XFree86

Please always Cc to me when replying to me on the lists.



Reply to: