Re: [Yaird-devel] Bug#496500: yaird: fails to create initrd when running 2.6.24 etchnhalf kernel

["Maximilian Attems" <maks@sternwelten.at>]

On Mon, Aug 25, 2008 at 12:26 PM, Jonas Smedegaard <dr@jones.dk> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> On Mon, Aug 25, 2008 at 04:29:39PM +0800, James Andrewartha wrote:
>>Package: yaird
>>Version: 0.0.12-18
>>Severity: important
>>
>>This is bug 431534, exposed again because stable has an old version of
>>yaird but a recent kernel.
>
> Acknowledged.
>
> I see some different approaches to this:
>
>  1. Leave this bug open but do nothing about it.
>     * This bug is not a security issue in itself
>     * Security-related kernel updates can switch to initramfs-tools

etch+half linux images are an optional upgrade.
never meant to be the default.


>  2. Release 0.0.12-18+etch1 fixing only this specific issue
>     * Security-updates must be minimal
>  3. Release backport of newest yaird in unstable
>     * Linux changes are large already, so "must be minimal" cannot
>       apply here
>     * Most if not all recent yaird changes are to support the major
>       changes to recent Linux kernels
>  4. Drop yaird from etchnhalf
>     * Yaird has been dropped from testing (see bug#457177)

that would mean dropping yaird out of etch, seems a bit late for that too.

>
>
> If we do 1) when should probably go through and etch-tag all other bugs
> fixed recently.
>
> If we do 3) then a single change must be made compared to current
> package in Sid: LVM workaround must be enabled by default.
>
>
> Cc'ing release team and security team for input.
>
>
> NB! Even if yaird really is "generally too buggy" as judged in
> bug#457177, the current release in unstable is far better than the
> version currently in Etch, (contains no known regressions, and actually
> works out-of-the-box in many cases with recent Linux kernels whereas
> etch release don't).

this diff seems a bit huge for a stable update.
also considering that lenny is about to be released.
as yaird is considered to be working with default etch kernel, that bug
doesn't need any further action.

regards

-- 
maks