Re: Bits (Nybbles?) from the Vancouver release team meeting
On Mon, Mar 14, 2005 at 04:17:45PM +0100, Matthias Urlichs wrote:
> Hi, Sven Luther wrote:
>
> > What about building the scc (or tier 2 as i would say) arches from testing and
> > not unstable ?
>
> That would negate one of the main points of having Testing, i.e. something
> that's supposed to be in an installable state at all times, for that
> architecture.
Well, yes and no, it would at least elliminate all the problems related to
non-arch problems ? I don't know if an additional step would be worth it,
maybe it could.
> You could treat Testing as that arch's Unstable, though, and set up a
> separate britney for each arch, which migrates Testing=>Running (or
> however you'd like to call it). I think that would mitigate the "keep lots
> of source versions around" bloat problem because (a) testing isn't updated
> as often as Unstable is, and (b) presumably you wouldn't need another
> 10-day waiting period at this point.
Yep, that is even nicer. You would need to fix bugs in two branches for these
packages : 1) in the main-unstable upload, and second in the
per-arch-unstable, but there will be two natural entry point for package
builds. and there will be no need for new code to be developed, since
everything works as usual, you just need to point the autobuilders to the
main-testing source archive and that's all.
But that means that it is waranted for tier-2 arch maintainer to do NMU of
packages in main-unstable (after having tested them on at least one main
arch), if the maintainer doesn't accept the tier-2 arch fixing patch in a
timely manner.
Friendly,
Sven Luther
Reply to: