From: Vincent Lefevre on
On 2010-04-12 12:07:08 +0200, Sven Joachim wrote:
> The conclusion I had drawn from this long ago: don't use testing, it is
> only meant to be a staging area for preparing the next release and not
> as a distribution actually to be used by people, except during freezes
> maybe.

The machine in question was a Debian/unstable machine, which I no
longer use very much, and keeping it in unstable would need time to
maintain it. So, for this machine, I want to progressively upgrade it
to squeeze. That's why I now track the testing distribution; otherwise
I would have to downgrade packages, which is not supported.

Moreover, with the current status of module-init-tools, it seems
that the broken version will get into squeeze anyway (at least,
there doesn't seem to be anything to prevent that) if the problem
on HPPA isn't fixed.

--
Vincent Lef�vre <vincent(a)vinc17.net> - Web: <http://www.vinc17.net/>
100% accessible validated (X)HTML - Blog: <http://www.vinc17.net/blog/>
Work: CR INRIA - computer arithmetic / Ar�naire project (LIP, ENS-Lyon)


--
To UNSUBSCRIBE, email to debian-user-REQUEST(a)lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster(a)lists.debian.org
Archive: http://lists.debian.org/20100412102151.GA17851(a)prunille.vinc17.org
From: Stephen Powell on
On Mon, 12 Apr 2010 06:07:08 -0400 (EDT), Sven Joachim wrote:
> On 2010-04-12 11:19 +0200, Vincent Lefevre wrote:
>> There's something broken in the design of Debian's migration to testing.
>>
>> A grave bug in the testing version of module-init-tools (3.12~pre2-1)
>> was fixed several weeks ago, and the package was uploaded with
>> urgency=high:
>>
>> module-init-tools (3.12~pre2-2) unstable; urgency=high
>>
>> * Fixed an init scripts dependency loop introduced in -1. (Closes: #574535)
>>
>> -- Marco d'Itri <md(a)linux.it> Mon, 22 Mar 2010 12:21:18 +0100
>>
>> but it hasn't migrated to testing yet!
>>
>> The reason is a build failure on HPPA:
>>
>> https://buildd.debian.org/build.php?arch=hppa&pkg=module-init-tools&ver=3.12~pre2-2
>>
>> Isn't it unacceptable that a problem on some platform blocks a fix
>> for a grave bug?
>
> This may not be acceptable for testing users, but the rules are pretty
> clear: all release architectures must be in sync in testing. This often
> holds up important fixes for months.
>
> The conclusion I had drawn from this long ago: don't use testing, it is
> only meant to be a staging area for preparing the next release and not
> as a distribution actually to be used by people, except during freezes
> maybe.
>
> Use stable or unstable instead, or a mixed testing/unstable system where
> you upgrade packages that are broken in testing from unstable.

As a user of one of the non-x86-derivative architectures (s390) I am
grateful for this rule. This rule is often the only thing that motivates
package maintainers to fix the problem on my architecture. Many package
maintainers seem to have the attitude of "Who cares that it doesn't build
on s390? Nobody uses s390 anyway." But some people do, such as yours
truly. And I care. And that rule forces the package maintainer to fix
the build problem on s390. I am also subscribed to the debian-s390 list,
and the package maintainers sometimes post on that list saying, in effect,
"how can I fix this build problem?" And sometimes, I've been able to
help them.

--
.''`. Stephen Powell <zlinuxman(a)wowway.com>
: :' :
`. `'`
`-


--
To UNSUBSCRIBE, email to debian-user-REQUEST(a)lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster(a)lists.debian.org
Archive: http://lists.debian.org/156338502.2645521271087878019.JavaMail.root(a)md01.wow.synacor.com
From: Brad Rogers on
On Mon, 12 Apr 2010 11:19:54 +0200
Vincent Lefevre <vincent(a)vinc17.org> wrote:

Hello Vincent,

> module-init-tools (3.12~pre2-2) unstable; urgency=high

Migrated to testing today.

--
Regards _
/ ) "The blindingly obvious is
/ _)rad never immediately apparent"

A friend of a friend he got beaten
I Predict A Riot - Kaiser Chiefs
From: Andrei Popescu on
On Mon,12.Apr.10, 11:19:54, Vincent Lefevre wrote:
> There's something broken in the design of Debian's migration to testing.
>
> A grave bug in the testing version of module-init-tools (3.12~pre2-1)
> was fixed several weeks ago, and the package was uploaded with
> urgency=high:
>
> module-init-tools (3.12~pre2-2) unstable; urgency=high
>
> * Fixed an init scripts dependency loop introduced in -1. (Closes: #574535)
>
> -- Marco d'Itri <md(a)linux.it> Mon, 22 Mar 2010 12:21:18 +0100
>
> but it hasn't migrated to testing yet!

You can always install the unstable version.

Regards,
Andrei
--
Offtopic discussions among Debian users and developers:
http://lists.alioth.debian.org/mailman/listinfo/d-community-offtopic
From: Vincent Lefevre on
On 2010-04-12 12:55:59 +0200, Sven Joachim wrote:
> On 2010-04-12 12:21 +0200, Vincent Lefevre wrote:
>
> > Moreover, with the current status of module-init-tools, it seems
> > that the broken version will get into squeeze anyway (at least,
> > there doesn't seem to be anything to prevent that) if the problem
> > on HPPA isn't fixed.
>
> There is something to prevent that, the bug is marked as grave. So
> something has to be done about it some day before the release.

The problem has eventually been solved, but note that the bug was
closed on March 22.

Are bugs which are marked as done taken into account before a
release?

--
Vincent Lef�vre <vincent(a)vinc17.net> - Web: <http://www.vinc17.net/>
100% accessible validated (X)HTML - Blog: <http://www.vinc17.net/blog/>
Work: CR INRIA - computer arithmetic / Ar�naire project (LIP, ENS-Lyon)


--
To UNSUBSCRIBE, email to debian-user-REQUEST(a)lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster(a)lists.debian.org
Archive: http://lists.debian.org/20100419104321.GA22570(a)prunille.vinc17.org