From: Daniel Gomez <da.gomez@kernel.org>
To: Chuck Lever <chuck.lever@oracle.com>,
Daniel Gomez <da.gomez@samsung.com>
Cc: kdevops@lists.linux.dev, Luis Chamberlain <mcgrof@kernel.org>
Subject: Re: [PATCH 0/2] Locales fixes
Date: Sat, 18 Oct 2025 00:21:49 +0200 [thread overview]
Message-ID: <7e469eeb-1c96-47e2-8a7a-ccf1aa1245fa@kernel.org> (raw)
In-Reply-To: <410fd616-811d-423c-b2ed-25f86d851b51@oracle.com>
On 17/10/2025 23.59, Chuck Lever wrote:
> On 10/16/25 2:56 PM, Chuck Lever wrote:
>> On 10/15/25 5:17 PM, Chuck Lever wrote:
>>>> Note, I can't fully test this or the patch, as it appears to be broken (at least
>>>> for me) when using a Debian host and a Fedora guest with kdevops/libguestfs:
>>>>
>>>> virt-builder fedora-41 --arch x86_64 -o /xfs1/libvirt/images/kdevops/guestfs/base_images/fedora-41.raw --size 20G --format raw --commands-from-file /tmp/ansible.8r5zc3so
>>>> [ 3.7] Downloading: http://builder.libguestfs.org/fedora-41.xz
>>>> ###################################################################################################################################################################################################### 100.0%
>>>> [ 140.6] Planning how to build this image
>>>> [ 140.6] Uncompressing
>>>> [ 141.7] Resizing (using virt-resize) to expand the disk to 20.0G
>>>> [ 171.5] Opening the new disk
>>>> [ 181.7] Setting a random seed
>>>> [ 181.7] Installing packages: sudo qemu-guest-agent python3 bash policycoreutils-python-utils
>>>> [ 194.7] Uninstalling packages: dracut-config-rescue
>>>> [ 195.0] Deleting: /boot/*rescue*
>>>> [ 195.0] Deleting: /boot/loader/entries/*rescue*
>>>> [ 195.0] Running: useradd -u 1012 -s /bin/bash -m kdevops
>>>> [ 195.1] Appending line to /etc/sudoers.d/kdevops
>>>> [ 195.1] Editing: /etc/default/grub
>>>> [ 195.2] Running: /usr/sbin/grub2-mkconfig -o /boot/grub2/grub.cfg
>>>> [ 195.9] Installing packages: isc-dhcp-client ifupdown
>>>> Updating and loading repositories:
>>>> Fedora 41 - x86_64 100% | 221.4 KiB/s | 30.3 KiB | 00m00s
>>>> Fedora 41 - x86_64 - Updates 100% | 289.6 KiB/s | 29.0 KiB | 00m00s
>>>> Fedora 41 openh264 (From Cisco) - x86_ 100% | 3.2 KiB/s | 986.0 B | 00m00s
>>>> Fedora 41 - x86_64 100% | 241.7 KiB/s | 480.5 KiB | 00m02s
>>>> Fedora 41 - x86_64 - Updates 100% | 4.4 MiB/s | 11.5 MiB | 00m03s
>>>> Fedora 41 openh264 (From Cisco) - x86_ 100% | 7.3 KiB/s | 5.8 KiB | 00m01s
>>>> Repositories loaded.
>>>> Package "sudo-1.9.15-5.p5.fc41.x86_64" is already installed.
>>>> ...
>>>> Complete!
>>>> Generating grub configuration file ...
>>>> Adding boot menu entry for UEFI Firmware Settings ...
>>>> done
>>>> Updating and loading repositories:
>>>> Repositories loaded.
>>>> Failed to resolve the transaction:
>>>> No match for argument: isc-dhcp-client
>>>> No match for argument: ifupdown
>>>> You can try to add to command line:
>>>> --skip-unavailable to skip unavailable packages
>>>> virt-builder: error: dnf -y install 'isc-dhcp-client' 'ifupdown': command
>>>> exited with an error
>>>>
>>>> If reporting bugs, run virt-builder with debugging enabled and include the
>>>> complete output:
>>>>
>>>> virt-builder -v -x [...]
>>> This appears to come from the new base_image virt_builder template,
>>> where we have:
>>>
>>> {% if distro_debian_based is defined and distro_debian_based %}
>>> {# Ugh, debian has to be told to bring up the network and regenerate ssh
>>> keys #}
>>> {# Hope we get that interface name right! #}
>>> install isc-dhcp-client,ifupdown
>>> mkdir /etc/network/interfaces.d/
>>>
>>> Which I would expect to apply only when the /guest/ is Debian, not
>>> the host. It could be that the distro detection logic works only when
>>> both the host and guest are the same?
>>
>> I think the problem arises because the new base_image role uses these
>> distro checks instead of what bringup_guestfs.sh did, which was select
>> these commands based on {{ virtbuilder_os_version }}
>>
>> virt-builder.j2 needs to use virtbuilder_os_version for these checks.
> virtbuild_os_version doesn't give us quite everything that is needed.
> Looks like the template needs to add some more cruft if Trixie is in
> use. So I've put together a possible fix, pushed to the cel-fixes branch
> in linux-kdevops. Please have a look.
>
> Passes some (very) light testing with a Fedora 41 host and all of a
> Debian 12, Debian 13, or Fedora 42 guest. Bringup gets past the
> virt-builder stage now for all three.
>
I was just preparing the final patch for the fix. I'm adding your patch on top
and test it.
Thanks!
next prev parent reply other threads:[~2025-10-17 22:21 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-09-29 19:04 [PATCH 0/2] Locales fixes Daniel Gomez
2025-09-29 19:04 ` [PATCH 1/2] devconfig: drop locales-all leftovers Daniel Gomez
2025-10-09 21:33 ` Daniel Gomez
2025-09-29 19:04 ` [PATCH 2/2] base_image: ensure base image permissions are set Daniel Gomez
2025-10-14 18:02 ` [PATCH 0/2] Locales fixes Daniel Gomez
2025-10-14 21:35 ` Chuck Lever
2025-10-15 18:21 ` Daniel Gomez
2025-10-15 21:17 ` Chuck Lever
2025-10-16 18:56 ` Chuck Lever
2025-10-17 21:59 ` Chuck Lever
2025-10-17 22:21 ` Daniel Gomez [this message]
2025-10-17 22:53 ` Daniel Gomez
2025-10-17 23:06 ` Daniel Gomez
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=7e469eeb-1c96-47e2-8a7a-ccf1aa1245fa@kernel.org \
--to=da.gomez@kernel.org \
--cc=chuck.lever@oracle.com \
--cc=da.gomez@samsung.com \
--cc=kdevops@lists.linux.dev \
--cc=mcgrof@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox