Building Trixie Alpha 1 images

[Cyril Brulebois]

Hello images team,

This just happened:

Copying installer version 20241227 from unstable to testing

Will copy installer version 20241227 from suite unstable to
testing.
Architectures to copy: amd64, ppc64el, s390x, armhf, arm64, mips64el, 
riscv64
Architectures to skip: 
Installer has been copied successfully.
Done

As planned we're removing support for armel and i386.

Looking at riscv64, it seems to have become an official release arch a
little while back[1]. Its images tarball features netboot stuff (hence
this question regarding d-i-n-i[2]), but also cdrom things:

./installer-riscv64/20241227/images/cdrom/
./installer-riscv64/20241227/images/cdrom/debian-cd_info.tar.gz
./installer-riscv64/20241227/images/cdrom/initrd.gz
./installer-riscv64/20241227/images/cdrom/vmlinux

so I suppose debian-cd should do something with it. :)

(I've cc'd Aurélien again, just in case…)

  1. 
https://salsa.debian.org/release-team/britney2/-/commit/593e01d703966ba8602eeff7594b0553900ea84f
  2. https://lists.debian.org/debian-boot/2024/12/msg00218.html

I'll look into the website next, checking whether it got prepared for
trixie yet and tweaking things as needed. Then see if the massive
changelog from bookworm can be built as usual. Finally check whether
it can be massaged into something decent, or if it's better to stick
to a couple of important points followed by “lots of other stuff
changed”.


Cheers,
-- 
Cyril Brulebois (k...@debian.org)
D-I release manager -- Release team member -- Freelance Consultant


signature.asc
Description: PGP signature

Re: Bug#1076723: grub-common: duplicate GNU/Linux in menuentry and submenu lines of /boot/grub/grub.cfg

[Pascal Hambourg]

Control: tags -1 patch

On 26/12/2024 at 20:37, Cyril Brulebois wrote:

Vincent Lefevre  (2024-07-22):


After a recent grub2 upgrade, there are now duplicate "GNU/Linux"
in menuentry and submenu lines of /boot/grub/grub.cfg, e.g.

menuentry 'Debian GNU/Linux GNU/Linux' --class debian --class gnu-linux --class 
gnu --class os $menuentry_id_option 
'gnulinux-simple-f3a51d21-c3a1-405d-b0fb-b0b25662b752' {

submenu 'Advanced options for Debian GNU/Linux GNU/Linux' $menuentry_id_option 
'gnulinux-advanced-f3a51d21-c3a1-405d-b0fb-b0b25662b752' {


This also happens with fresh Trixie installations, and that looks
utterly silly… :(

There were a number of follow-ups by Vincent, it'd be nice if someone
could look into getting that fixed.


It seems that 
 aims to 
address the issue in an alternative way.

Re: Building Trixie Alpha 1 images

[Aurelien Jarno]

Hi,

On 2024-12-28 11:12, Cyril Brulebois wrote:
> Hello images team,
> 
> This just happened:
> 
> Copying installer version 20241227 from unstable to testing
> 
> Will copy installer version 20241227 from suite unstable to
> testing.
> Architectures to copy: amd64, ppc64el, s390x, armhf, arm64, mips64el, 
> riscv64
> Architectures to skip: 
> Installer has been copied successfully.
> Done
> 
> As planned we're removing support for armel and i386.
> 
> Looking at riscv64, it seems to have become an official release arch a
> little while back[1]. Its images tarball features netboot stuff (hence
> this question regarding d-i-n-i[2]), but also cdrom things:
> 
> ./installer-riscv64/20241227/images/cdrom/
> ./installer-riscv64/20241227/images/cdrom/debian-cd_info.tar.gz
> ./installer-riscv64/20241227/images/cdrom/initrd.gz
> ./installer-riscv64/20241227/images/cdrom/vmlinux
> 
> so I suppose debian-cd should do something with it. :)
> 
> (I've cc'd Aurélien again, just in case…)

riscv64 CD images are generated daily for some time, so I guess that
should just work.

Cheers
Aurelien

-- 
Aurelien Jarno  GPG: 4096R/1DDD8C9B
aurel...@aurel32.net http://aurel32.net


signature.asc
Description: PGP signature

Bug#1091617: mdadm: broken debian-installer support

[Cyril Brulebois]

Package: mdadm
Version: 4.2-5
Severity: serious
Tags: d-i
Justification: breaks d-i
X-Debbugs-Cc: debian-b...@lists.debian.org, debian-cd@lists.debian.org

[ X-D-Cc: -boot@ and -cd@ for awareness. ]

Hi,

Spotted by Steve while testing tentative Trixie Alpha 1 images: mdadm is
broken, with a systemd-related problem.

This part of the .config script doesn't seem to work:

# if system has root on md
if findmnt / 2>/dev/null | grep -qs \/dev\/md && \
# and system is not a container
! grep -qs ':cpuset:/$' /proc/1/cgroup && \
# and system is not a chroot
[ "$(stat -c %d/%i /)" != "$(stat -Lc %d/%i /proc/1/root 2>/dev/null)" ] && 
\
# and systemd is not running
[ -z "$(pidof systemd)" ]
then
db_input high mdadm/init-system-not-systemd || true
db_go
fi

The chroot test seems inverted (compared to the intent/comment):
 - On my laptop I'm getting this: 64770/2 == 64770/2
 - In my sid devel schroot I'm getting: 64774/1196038 != 64770/2

I'm awaiting some feedback from the images team to see if my plan would
work for them, but my initial reaction would be to confirm the trivial
bugfix (if it's easy enough to test, within the installer's environment)
then upload + fast-track mdadm into testing, before respinning images.


Cheers,
-- 
Cyril Brulebois (k...@debian.org)
D-I release manager -- Release team member -- Freelance Consultant

Re: Bug#1091617: mdadm: broken debian-installer support

[Cyril Brulebois]

Control: notfound -1 4.2-5
Control: found -1 4.2+20230508-3

Sorry, missed that edit before submitting.


Cheers,
-- 
Cyril Brulebois (k...@debian.org)
D-I release manager -- Release team member -- Freelance Consultant


signature.asc
Description: PGP signature

Re: Trixie

[Cyril Brulebois]

Cyril Brulebois  (2024-12-26):
> Seeing how (at least these past few days, weeks, months, or so) udebs
> have piled up in the archive, it's probably more a matter of making sure
> to include the right set of udebs onto the image, instead of picking
> whatever is the latest.
> 
> From a cursory look at the context of the recent bugfix (which is
> deeff0673833ae6ae97e3f7f96419254763f1d0f in debian-cd), debian-cd picks
> whatever is the highest version, regardless of what's actually used by
> d-i. In other words that traded a constant problem (not having enough
> space to include linux-image debs) with a transient one (the wrong set
> of udebs might get picked).
> 
> Provided there's no immediate decruft in the archive (the current number
> of udebs for past kernels suggests that's not happening), it should just
> be a matter of:
>   1. picking the right set (beware of archs with several flavours) based
>  on info collected from the d-i build;
>   2. bailing out if that version can't be found.
> 
> (1) would fix the issue, and only fail to work if and when the decruft
> becomes super aggressive; (2) would avoid building an image that cannot
> work if and when that happens.

Some heavy decrufting happened earlier, see [1] for the full list.
That's why the dose report for i386[2] got much worse all of a sudden
(it was surprisingly good, similar to say amd64 until today).

=
[Date: Sat, 28 Dec 2024 13:10:09 -] [ftpmaster: Ansgar]
Removed the following packages from unstable:
[…]
--- Reason ---
[auto-cruft] NBS (no longer built by linux)
--
=

  1. https://ftp-master.debian.org/removals.txt
  2. https://d-i.debian.org/dose/graph-unstable-i386.png

I'm not sure whether that was a one-off clean-up or whether some cruft
detection got changed to make those show up (which in turn would
probably determine whether udebs for older kernels are going to stick
around for a while in the future, or just go away immediately).

And I'm not going to try and learn more about this right now, got some
mdadm fun (and maybe other things) to figure out still.


Cheers,
-- 
Cyril Brulebois (k...@debian.org)
D-I release manager -- Release team member -- Freelance Consultant


signature.asc
Description: PGP signature

Re: Bug#1091617: mdadm: broken debian-installer support

[Cyril Brulebois]

Control: tag -1 patch pending

Cyril Brulebois  (2024-12-28):
> Spotted by Steve while testing tentative Trixie Alpha 1 images: mdadm is
> broken, with a systemd-related problem.
> 
> This part of the .config script doesn't seem to work:
> 
> # if system has root on md
> if findmnt / 2>/dev/null | grep -qs \/dev\/md && \
> # and system is not a container
> ! grep -qs ':cpuset:/$' /proc/1/cgroup && \
> # and system is not a chroot
> [ "$(stat -c %d/%i /)" != "$(stat -Lc %d/%i /proc/1/root 2>/dev/null)" ] 
> && \
> # and systemd is not running
> [ -z "$(pidof systemd)" ]
> then
> db_input high mdadm/init-system-not-systemd || true
> db_go
> fi
> 
> The chroot test seems inverted (compared to the intent/comment):
>  - On my laptop I'm getting this: 64770/2 == 64770/2
>  - In my sid devel schroot I'm getting: 64774/1196038 != 64770/2

I've reproduced the “mdadm shows a note that really doesn't make sense
in that context” problem, and confirmed the proposed attached patch
avoids this problem.

> I'm awaiting some feedback from the images team to see if my plan would
> work for them, but my initial reaction would be to confirm the trivial
> bugfix (if it's easy enough to test, within the installer's environment)
> then upload + fast-track mdadm into testing, before respinning images.

What I've seen doesn't exactly quite match the early report, so I'm
still waiting on some more information before a possible upload. What
I've seen locally was a successful installation of mdadm and tasksel
running as usual, but that was just with a dummy RAID0 and a single
device, so maybe there are other problems lurking in the shadows that
I haven't spotted yet.


Cheers,
-- 
Cyril Brulebois (k...@debian.org)
D-I release manager -- Release team member -- Freelance Consultant
From 10113f6f7b5e66a5e35af6d90df849c104500de4 Mon Sep 17 00:00:00 2001
From: Cyril Brulebois 
Date: Sat, 28 Dec 2024 20:47:46 +0100
Subject: [PATCH] Fix chroot detection (Closes: #1091617).

This avoids displaying the systemd note when mdadm is getting installed
within d-i.
---
 debian/changelog| 8 
 debian/mdadm.config | 2 +-
 2 files changed, 9 insertions(+), 1 deletion(-)

diff --git a/debian/changelog b/debian/changelog
index 965eeb9..358608f 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -1,3 +1,11 @@
+mdadm (4.3+20241202-1.1) sid; urgency=high
+
+  * Non-maintainer upload.
+  * Fix chroot detection to avoid displaying the systemd note when mdadm
+is getting installed within d-i (Closes: #1091617).
+
+ -- Cyril Brulebois   Sat, 28 Dec 2024 20:46:53 +0100
+
 mdadm (4.3+20241202-1) sid; urgency=medium
 
   * Uploading to sid.
diff --git a/debian/mdadm.config b/debian/mdadm.config
index 848e6b6..2bd3489 100644
--- a/debian/mdadm.config
+++ b/debian/mdadm.config
@@ -37,7 +37,7 @@ if findmnt / 2>/dev/null | grep -qs \/dev\/md && \
 # and system is not a container
 ! grep -qs ':cpuset:/$' /proc/1/cgroup && \
 # and system is not a chroot
-[ "$(stat -c %d/%i /)" != "$(stat -Lc %d/%i /proc/1/root 2>/dev/null)" ] && \
+[ "$(stat -c %d/%i /)" = "$(stat -Lc %d/%i /proc/1/root 2>/dev/null)" ] && \
 # and systemd is not running
 [ -z "$(pidof systemd)" ]
 then
-- 
2.39.5



signature.asc
Description: PGP signature