Re: Bug#231538: A possible solution

[GOTO Masanori <gotom@debian.or.jp>]

At Sun, 13 Jun 2004 22:27:16 +0200,
Andreas Barth wrote:
> * Herbert Xu (herbert@gondor.apana.org.au) [040302 08:10]:
> > On Tue, Mar 02, 2004 at 04:22:30PM +0900, GOTO Masanori wrote:
> > > 
> > > > Well newer initrd-tools module-init-tools should be in woody 
> > > > in order to upgrade to sarge smoothly.
> 
> > > Is this ok?  Herbert?  Marco?  If not, I reassign this bug to such
> > > packages.
>  
> > initrd-tools does not depend on glibc so the version in sarge/sid
> > should suffice.
> 
> But initrd-tools from sarge wants cramfsprogs (>= 1.1-4), dash; both
> is not part of woody. I'd propose to use it from backports.org (which
> is changed to cramfsprogs (>= 1.1-3), dash|ash; or is there a problem
> with that)?
> 
> Well, and I'd also propose to take the backport from modutils from the
> same source.
> 
> However, the kernel package could be taken out of the pool / from
> sarge.
> 
> @ftpmasters: How should these packages be upload? Just as "byhand",
> and you sort it in? And how is the packages-file generated? In which
> form do you want to have the README-file for that directory?

Please look at the whole discussion of #231538.  Joey made a good
summary at:

	http://people.debian.org/~joey/pr/3.1/i386.html

#231538 was closed because I introduced a patch for glibc preinst
kernel version check.  But as you pointed out, it's true that we need
to create upgrade-i386-80386 for sarge to support smooth upgrade path
for real i386 machine.

The problem is: (1) we're lazy, no one has prepared upgrade-i386-80386
directory, (2) no one has real i386 80386 machine to test that
directory.

If you have real i386 machine, could you work for this issue?

Regards,
-- gotom