Re: glibc plans for the lenny cycle

[Andreas Barth <aba@not.so.argh.org>]

* Aurelien Jarno (aurelien@aurel32.net) [070412 22:59]:
> Basically it looks ok. What about the freeze period for the toolchain? I
> think we usually suffer for a too early freeze of the glibc (it has been
> frozen in July for Etch, even if it has been unblocked a lot of time
> after). In my opinion, it would be better to freeze the upstream version
> at that time and allow minor update until the main freeze.

Hm, we should consider how to do this better then.

Basically, the main reasons for the freeze are:
1. We don't want to have stuff broken (and sometimes, even small and
unrisky looking bugfixes have a devasting potential);
2. Packages related to the installer have their own specialities.

We hope to reduce the impact of 2nd a bit during the release cycle (but
not to none) - for 1st, I think it might be good to have some discussion
on debconf (some ideas float in my head float like "longer default
testing in unstable", "better patch review" or whatever - I think it
might be a good thing to discuss that more in depth, and I hope we come
to an result both the glibc maintainers and we like).


Cheers,
Andi
-- 
  http://home.arcor.de/andreas-barth/