Bug#920353: debian-installer configuration on alpha (and others)
Hi JH!
On 1/28/19 9:41 PM, jhcha54008 wrote:
> As you note, including debian-ports-archive-keyring-udeb and
> setting DEBIAN_RELEASE = unstable benefits to all ports architecture.
Yes, I know. I have therefore made your suggested change to all Debian
Ports architectures, not just alpha.
> Including pata-modules-${kernel:Version} is alpha-specific instead
> ('qemu-system-alpha-specific' in fact).
I did not see this particular change, I must have missed it. Was
it part of the patches you sent?
> At runtime, choose-mirror-bin checks the availability of
> the selected mirror and of the required 'suite' on it.
> The suite is taken from the file /etc/default-release
> (if not preseeded in debconf variable 'mirror/suite').
>
> At build time, the content of the variable DEBIAN_RELEASE
> is written into /etc/default-release.
>
> Unfortunately, the cd images from 2019-01-25 have 'buster'
> in /etc/default-release.
Good catch. Thanks for digging this up. I have not (yet) taken
the time to look into how choose-mirror works in detail.
I do hope that at some point we can get a mirror selection
list for Debian Ports like we have for the release architectures
as a lot of users seem to have trouble finding the correct
mirrors for Debian Ports.
> Actually I tested with a (home made) netboot image in qemu
> emulation. I wasn't able to test the cd images with qemu-system-alpha
> precisely because of the missing cdrom driver in initrd.gz :
> it boots but cannot mount the cdrom image afterwards.
Ah, I didn't know. If that's a problem, then I can push the
change to debian-installer to include this package. I will
also build updates installation images with this change.
> Sorry for this late reply, and thank you for your work on the
> ports architectures !
You're welcome. And thanks your help as well!
Adrian
--
.''`. John Paul Adrian Glaubitz
: :' : Debian Developer - glaub...@debian.org
`. `' Freie Universitaet Berlin - glaub...@physik.fu-berlin.de
`-GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913
Bug#920353: debian-installer configuration on alpha (and others)
Le vendredi 25 janvier à 12h 46mn 02s (+0100), John Paul Adrian Glaubitz a
écrit :
> On 1/24/19 4:53 PM, John Paul Adrian Glaubitz wrote:
> > I just never bothered tracking down the problem since I just used the
> > standard
> > installer. If your particular changes makes the installer work on Debian
> > Ports
> > architectures in expert mode, I am happy to commit the change for all ports
> > architectures.
>
> I have committed the changes for all architectures in Debian Ports:
>
> > https://salsa.debian.org/installer-team/debian-installer/commit/c36cade1787ebe8cf4438dce7227cac20c887d32
> > https://salsa.debian.org/installer-team/debian-installer/commit/8c9f81fc412abcef3a8a6fa11d0a7ed2dfd26ff4
>
> Thanks for spotting this!
>
> Adrian
>
> --
> .''`. John Paul Adrian Glaubitz
> : :' : Debian Developer - glaub...@debian.org
> `. `' Freie Universitaet Berlin - glaub...@physik.fu-berlin.de
> `-GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913
Hi Adrian,
Thank you for your quick answer.
As you note, including debian-ports-archive-keyring-udeb and
setting DEBIAN_RELEASE = unstable benefits to all ports architecture.
Including pata-modules-${kernel:Version} is alpha-specific instead
('qemu-system-alpha-specific' in fact).
Here are some details missing in the original bug report :
At runtime, choose-mirror-bin checks the availability of
the selected mirror and of the required 'suite' on it.
The suite is taken from the file /etc/default-release
(if not preseeded in debconf variable 'mirror/suite').
At build time, the content of the variable DEBIAN_RELEASE
is written into /etc/default-release.
Unfortunately, the cd images from 2019-01-25 have 'buster'
in /etc/default-release.
Actually I tested with a (home made) netboot image in qemu
emulation. I wasn't able to test the cd images with qemu-system-alpha
precisely because of the missing cdrom driver in initrd.gz :
it boots but cannot mount the cdrom image afterwards.
Sorry for this late reply, and thank you for your work on the
ports architectures !
Regards,
JH Chatenet
