We originally required there to be only one kernel modules directory
since we typically only have one kernel installed. Change this check
up a bit since some kernel module RPMs aren't delivering a new kernel
but do deliver kmods that were built against a different kernel.
For example v4l2loopback from https://elrepo.org/
```
bash-5.1# rpm -ql kmod-v4l2loopback
/etc/depmod.d/kmod-v4l2loopback.conf
/lib/modules/5.14.0-427.13.1.el9_4.x86_64
/lib/modules/5.14.0-427.13.1.el9_4.x86_64/extra
/lib/modules/5.14.0-427.13.1.el9_4.x86_64/extra/v4l2loopback
/lib/modules/5.14.0-427.13.1.el9_4.x86_64/extra/v4l2loopback/v4l2loopback.ko
/usr/share/doc/kmod-v4l2loopback-0.12.7
/usr/share/doc/kmod-v4l2loopback-0.12.7/GPL-v2.0.txt
/usr/share/doc/kmod-v4l2loopback-0.12.7/greylist.txt
```
This patch now allows an image built from a Containerfile like the
following to work as input to the live artifacts stage.
```
FROM rhcos
RUN <<EOF
set -eux -o pipefail
dnf install -y https://www.elrepo.org/elrepo-release-9.el9.elrepo.noarch.rpm
dnf install -y v4l2loopback
dnf clean all
ostree container commit
EOF
```
Add support for setting environment variables for kernel-install and its
plugins. These can be used to control the behaviour of kernel-install
during the kernel rpm installation. Currently, we only add support for
$BOOT_ROOT, which we need to control the path where the kernel should be
installed.
Normally, kernel-install scripts will detect the correct path based on
the filesystem layout, specifically the ESP path. However, during our
installation the filesystem is not yet set up, so with this option, we
can control the installation path based on the location we know the ESP
will be mounted in the final image.
This is a very similar override to what we do with the
org.osbuild.fix-bls stage.
The stage option is added under a kernel_install_env object so we can
potentially support more of the known env vars in the future.
Printing 'owner' in an fstring triggers the mypy warning 'str-bytes-safe'
Error: If x = b'abc' then f"{x}" or "{}".format(x) produces "b'abc'",
not "abc".
We can't yet use the nicer run() argument, 'text=True', because we still
target the RHEL 7 build root which only has Python 3.6 and 'text' was
introduced in Python 3.7.
Use a temporary directory for the output of grub2-mkimage.
This makes the stage clean up the grub2-core.img from the build root
after its done. It also has the nice side-effect that unit tests that
call the stage are independent. Previously, a bug in the stage *might*
have been missed if a certain configuration of the stage was not
creating the grub2-core.img. One unit test could create an image at the fixed path
(/var/tmp/grub2-core.img) and then another one could call the stage with
the buggy configuration but the `shutil.copyfile()` call at the end of
the stage would succeed because it would find the image from the
previous stage run.
To accommodate for this change, the unit test with the mocked run call
is adjusted to intercept the random tmp output path and use it to create
a fake file for the stage to succeed.
This commit adds a test that mocks subprocess.run() so that we
can regression test the arguments passed to "grub-mkimage".
This is a regression test for
https://github.com/osbuild/osbuild/issues/2037
When writing the grub2 core image, the format should be set based on the
platform. This was accidentally changed to be `i386-pc` for all
platforms except when building ISOs.
Set the format to the platform and only change it to `i386-pc-eltorito`
when building ISOs.
This commit replaces the `/usr/bin/logger` binary in the dracut
chroot with a bind mount to `/usr/bin/true` to silence the spam
that we get from dracut during initramfs generation:
```
logger: socket /dev/log: No such file or directory
```
Unfortunately I could not find a nicer way, it seems it is
not possible to simply pass `sysloglvl=0` via the commandline
or an environment.
The extra complication here is that the dracut stage mounts
`devtmpfs` which will likely include:
```
/dev/log -> /run/systemd/journal/dev-log
```
but of course inside this chroot there is no `/run` which
leads to these messages.
Closes: https://github.com/osbuild/osbuild/issues/1976
Add an additional option called `kernel_line_size`
to allow setting a maximum cmdline size check
value for custom kernels or other restrictions.
This will override the arch defaults, if not set,
then the size map is checked, and if the current
architecture is not in the map, fallback to
4096, which is the max value allowed for
COMMAND_LINE_SIZE.
Signed-off-by: Albert Esteve <aesteve@redhat.com>
Add check to ensure that the size of
the parameters does not exceed the
maximum kernel cmdline size.
Otherwise, the parameters will
be truncated and the command line
will fail.
The size is arch-dependant. In
order to not to over-complicate
the search of the value in the
kernel files (which will probably
not be installed in most cases),
it uses a map with some values
for common architectures.
If architecture is not found in
the map, defaults to 4096, which
is the maximum posible size for
COMMAND_LINE_SIZE.
Signed-off-by: Albert Esteve <aesteve@redhat.com>
Instead of passing in the path we name the file according to the module
name. Path can be reintroduced later if absolutely necessary.
Signed-off-by: Simon de Vlieger <supakeen@redhat.com>
This helps to clear a risk concern for
automotive.
Add optional flag to org.osbuild.mkfs.ext4 stage
to enable/disable both lazy_itable_init and
lazy_journal_init extended options. Both mke2fs
options are controled by the same flag, namely
`lazy_init`.
Signed-off-by: Albert Esteve <aesteve@redhat.com>
This commit adds support to run the tests on Debian/Ubuntu systems.
Here the binary is called `grub-mkimage` instead of `grub2-mkimage`,
in addition the "grub-pc-bin" package must be installed there.
Don't include the "location" offset, and use just a path for the prefix
section to set the path of the grub2 modules on the iso. eg.
{
"filename": "eltorito.img",
"platform": "i386-pc",
"core": {
"type": "mkimage",
"partlabel": "gpt",
"filesystem": "iso9660"
},
"prefix": {
"path": "/boot/grub2/"
}
}
This creates an iso matching the current method used in Fedora where it
uses grub2 for BIOS and UEFI booting. Pass the path to the grub2 hybrid
mbr to the stage in the 'grub2mbr' field. eg.
"grub2mbr": "/usr/lib/grub/i386-pc/boot_hybrid.img"
Previously, the stage would try to call `os.scandir()` on non-existing
profile directories, which results in an exception. While the
directories that it scans are currently created by the TuneD RPM package
when installed, it it much nicer to gracefully handle their potential
non-existence, than to fail with an exception.
Signed-off-by: Tomáš Hozza <thozza@redhat.com>
The stage schema previously accepted an empty string as a profile name.
This would not work in practice, therefore extend the schema to require
non-empty strings.
Signed-off-by: Tomáš Hozza <thozza@redhat.com>
It's a no-op nowadays, we dropped support for the custom
ostree-container signature verification status.
Signed-off-by: Colin Walters <walters@verbum.org>
The rename to get file names + suffix with < 8 characters
was done in [1] when our initramfs was named initramfs.img.
It was subsequently renamed to initrd.img in [2] and the
rename of the initramfs was dropped but the rename of the
kernel was never dropped. Since vmlinuz is already < 8
characters let's just drop the rename here too.
[1] 60400910bb
[2] 6f533ef55e
This commit adds some unit tests around the coreos live-artifcats
mono stage.
- test/coreos_live_artifacts: add test for align_initrd()
- test/coreos_live_artifacts: add test for extend_initramfs()
- This actually tests the mkinitrd_pipe() function, which
extend_initramfs() calls after opening the file.
- test/coreos_live_artifacts: add test for make_stream_hash()
- test/coreos_live_artifacts: add test for make_efi_bootfile()
Co-authored-by: Achilleas Koutsou <achilleas@koutsou.net>
