public inbox for openembedded-core@lists.openembedded.org
 help / color / mirror / Atom feed
From: "Mathieu Dubois-Briand" <mathieu.dubois-briand@bootlin.com>
To: <adam.duskett@amarulasolutions.com>,
	<openembedded-core@lists.openembedded.org>
Subject: Re: [OE-core] [PATCH v7 5/6] dnf: Upgrade to 5.4.0.0
Date: Wed, 01 Apr 2026 15:56:49 +0200	[thread overview]
Message-ID: <DHHVCNZC2CKY.3GCMUUBL3FHN2@bootlin.com> (raw)
In-Reply-To: <DHHVAMTM0ULQ.3QO0VGCFBKNL4@bootlin.com>

On Wed Apr 1, 2026 at 3:54 PM CEST, Mathieu Dubois-Briand wrote:
> On Mon Mar 30, 2026 at 11:58 AM CEST, Adam Duskett via lists.openembedded.org wrote:
>> Completely rewritten in C++, there are several changes needed
>> to facilitate moving from DNF4 to DNF5. Below are some significant
>> items to note:
>>
>>     - The "-v/--verbose" cli option is removed. As such, remove the argument
>>     everywhere it was called.[1]
>>
>>     - the rpmverbosity cli option is replaced with --setopt=rpmverbosity
>>
>>     - Change gpgcheck to pkg_gpgcheck as this is what the documentation
>>     recommends.[2]
>>
>>     - --nogpgcheck is now --no-gpgchecks
>>
>>     - --setopt=protect_running_kernel=False has to be set or else every dnf
>>     transaction complains that /boot/vmlinuz-${kernel_version} does not
>>     exist in /var/log/dnf5.log. There is an open issue on github[3] to enable
>>     the currently not-yet implemented `autocheck_running_kernel option`, but
>>     it has been open for quite some time and the protect_running_kernel
>>     option supresses the message.
>>
>>     - Move --skip-broken to after the install argument as the global
>>       --skip-broken argument is removed.
>>
>>     - /var/cache/dnf is replaced with /var/cache/libdnf5
>>
>>     - Add `features += 'PACKAGECONFIG:append:pn-rpm = " sequoia"\n'`
>>           `features += 'PACKAGECONFIG:append:pn-librepo = " sequoia"\n'`
>>       to test_testimage_dnf as librepo currently does not work
>>       with gpgme.
>>
>> While DNF5 has many config options, unfortunately, the majority
>> of them depends on sdbus-c++, which is not included in OE-core.
>> However, the package is included in meta-openembedded, so it's
>> best to include the PACKAGECONFIG options in the recipe and
>> leave them off by default.
>>
>> 1: https://github.com/rpm-software-management/dnf5/issues/464
>> 2: https://dnf5.readthedocs.io/en/latest/dnf5.conf.5.html
>> 3: https://github.com/rpm-software-management/dnf5/issues/1255
>> Signed-off-by: Adam Duskett <adam.duskett@amarulasolutions.com>
>> ---
>
> Hi Adam,
>
> Thanks for the new version.
>
> I'm sorry, but we have some selftest failures now. :(
>
> The test_incremental_image_generation one was bisected to this commit,
> but I believe other failures have a similar source.
>
> 2026-04-01 09:45:42,563 - oe-selftest - INFO - imagefeatures.ImageFeatures.test_mandb (subunit.RemotedTestCase)
> 2026-04-01 09:45:42,569 - oe-selftest - INFO -  ... FAIL
> ...
> 2026-04-01 09:46:07,826 - oe-selftest - INFO - buildoptions.ImageOptionsTests.test_incremental_image_generation (subunit.RemotedTestCase)
> 2026-04-01 09:46:07,827 - oe-selftest - INFO -  ... FAIL
> ...
> 2026-04-01 10:00:20,138 - oe-selftest - INFO - oelib.utils.TestDefaultFeatures.test_filter_default_features (subunit.RemotedTestCase)
> 2026-04-01 10:00:20,139 - oe-selftest - INFO -  ... FAIL
> ...
> 2026-04-01 10:04:17,834 - oe-selftest - INFO - runtime_test.Postinst.test_failing_postinst (subunit.RemotedTestCase)
> 2026-04-01 10:04:17,834 - oe-selftest - INFO -  ... FAIL
> ...
> 2026-04-01 10:30:17,066 - oe-selftest - INFO - runtime_test.TestImage.test_testimage_install (subunit.RemotedTestCase)
> 2026-04-01 10:30:17,066 - oe-selftest - INFO -  ... FAIL
>
> https://autobuilder.yoctoproject.org/valkyrie/#/builders/48/builds/3456
> Some other builds are still in progress but already show the same
> errors:
> https://autobuilder.yoctoproject.org/valkyrie/#/builders/35/builds/3567
> https://autobuilder.yoctoproject.org/valkyrie/#/builders/23/builds/3666
>
> Can you have a look at the issue?
>
> Thanks,
> Mathieu

Sorry, this comment was for the v8.

-- 
Mathieu Dubois-Briand, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com



  reply	other threads:[~2026-04-01 13:57 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-03-30  9:58 [PATCH v7 1/6] signing-keys.bb: Fix DISTRO_CODENAME truncation Adam Duskett
2026-03-30  9:58 ` [PATCH v7 2/6] libtoml11: new recipe Adam Duskett
2026-03-30  9:58 ` [PATCH v7 3/6] libsolv: explicitly enable comps for rpm packageconfig Adam Duskett
2026-03-30  9:58 ` [PATCH v7 4/6] librepo: add PACKAGECONFIG[sequoia] Adam Duskett
2026-03-30  9:58 ` [PATCH v7 5/6] dnf: Upgrade to 5.4.0.0 Adam Duskett
2026-04-01 13:54   ` [OE-core] " Mathieu Dubois-Briand
2026-04-01 13:56     ` Mathieu Dubois-Briand [this message]
2026-03-30  9:58 ` [PATCH v7 6/6] libdnf: remove recipe Adam Duskett
2026-03-31  7:56 ` [OE-core] [PATCH v7 1/6] signing-keys.bb: Fix DISTRO_CODENAME truncation Mathieu Dubois-Briand

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=DHHVCNZC2CKY.3GCMUUBL3FHN2@bootlin.com \
    --to=mathieu.dubois-briand@bootlin.com \
    --cc=adam.duskett@amarulasolutions.com \
    --cc=openembedded-core@lists.openembedded.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