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

Re: Updates in stable releases

Shouldn't this be on debian-release instead/as well?

On Sun, 2013-12-29 at 22:04 +0100, Kurt Roeckx wrote:
> Hi,
> I think in general we are either too strict in what we allow as
> updates to stable or people think it's not going to be allowed and
> so don't even try to get updates to stable.
> The last time I asked about this, I got this as reply:
> https://lists.debian.org/debian-devel/2013/09/msg00466.html
> I want to start by giving some examples of things that got updated
> in stable point releases that I know about:
> - linux was 3.2.41-2 in 7.0, 3.2.51-1 in 7.3, 3.2.53-2 in
>   proposed-updates
> - iceweasel was 10.0.12esr-1 in 7.0, is now 17.0.10esr-1~deb7u1
> - postgresql-9.1 was 9.1.9-1, now 9.1.11-0wheezy1
> Clearly new upstream releases are acceptable under some
> conditions.  But it's not clear to me what those conditions are.
> The rules seem to suggest that we need a priority important bug
> in the Debian BTS.  Does that mean that if upstream makes a bugfix
> release we need to file bugs in the Debian BTS for each fix that
> we consider important and backports just those bugfixes, or would
> uploading such bugfix releases be allowed?

I think it depends on how well upstream's criteria for such releases
match ours for stable updates.

For Linux 2.6.32 in squeeze, I was asked to open a bug for each upstream
stable update, briefly explaining the importance of the changes in it.
I have not been doing the same for 3.2, however.

> How about more than just bugfixes?  For instance would new
> features be allowed, and in what case?  It seems that at least for
> the linux kernel support gets added for new hardware.

Missing hardware support is considered to be an important bug.

> One thing I had in mind for an update to apache is to have the
> version in stable support ECDHE which the version in stable
> currently doesn't do.  And I think the general feeling from people
> is that this is going to be rejected and so don't even try and
> ask.

The lack of PFS might be considered an important bug and it is worth
having that discussion.


Ben Hutchings
Klipstein's 4th Law of Prototyping and Production:
                                    A fail-safe circuit will destroy others.

Attachment: signature.asc
Description: This is a digitally signed message part

Reply to: