osbuild-composer can support user specified image sizes, this adds an
optional argument, after start, to specify the size in MiB. eg.
composer-cli compose start --size 2048 example-http-server qcow2
This only works when the backend is not 'lorax-composer', when it is the
user will get a warning that it will be ignored.
Related: rhbz#1844649
At least in our CI, the default place to store VM runtime overlays
(testvm.get_temp_dir()) points to a tmpfs file, so that tests can put VM
overlays into RAM and are not affected by slow I/O. But that doesn't
work for composer, as it tends to produce huge overlays due to real-life
OS composed trees.
Use /var/tmp/ instead, which is meant for large files.
Also simplify the VirtMachine invocation -- if `identity_file` is None,
that's fine -- it's the default value of that argument anyway.
Cherry-picked from master commit 9b8e0e2335
Anaconda creates a user for connecting via ssh during the installation when the sshpw kickstart command is used. The user is created with "/sbin/nologin" shell when /etc/default/useradd is missing, which prevents the user from log into the running installation.
Resolves: rhbz#1838677
This new setting for 'find_multipaths' tries to prevent things like
LVM from going ahead and activating LVM on the individual disks/paths
until there is reasonable certainty (via a timeout) that the device
is not a component of a multipath set.
NOTE: 'smart' is supported by device-mapper-multipath v0.7.7 and later.
Resolves: rhbz#1676777
(cherry picked from commit b8c1e706bb)
Signed-off-by: Brian C. Lane <bcl@redhat.com>
Previously the release package was chosen by picking the first package
that provides 'system-release' that isn't named generic. This can cause
branding issues with repos containing multiple system-release packages.
This patch changes _install_branding so that it will give preference to
a system-release package that ends with lowercase --variant name. If
there isn't one it will fall back to the previous behavior.
Resolves: rhbz#1826479
dracut renamed this module so the files were being removed from the
install.img (BUT not from the initramfs). They are necessary for the
shutdown eject feature to work.
Related: rhbz#1805405
The eject utility moved into util-linux and the package was dropped, but
since the runtime-cleanup template is using `removefrom util-linux
--allbut` it was never added to the boot.iso after the move.
This removes the package request for eject and adds it to the list of
binaries to keep from util-linux.
Related: rhbz#1805405
This also moves the packages to be installed into a file that can be
shared between test methods.
It drops several packages that are unnecessary for testing, and excludes
those paths from pylint checks using pocketlint.
Related: rhbz#1785154
These tests are fragile, order matters in some cases, and a previous
patch was incomplete. The diff test was failing because the blueprint
that it checks must be pushed twice in order to trigger the version bump
and 'Changed Version' output from composer-cli.
Related: rhbz#1785154
Without this, depending on which version of pylint is used, you may see
errors related to the rpm.RPMTAG_* constants. This makes sure that
pylint allows loading the rpm module.
Related: rhbz#1785154
This makes sure that depsolving shim installs the shim-* package, and
that depsolving grub2-efi-*-cdboot installs a specific -cdboot package.
Cherry-picked from 47fd6e85b2
Related: rhbz#1791612
- this is required for arm64 but is present in all latest kernels
so doesn't seem to hurt
- when registering the AMI mark its architecture properly
Cherry-picked from 6c35448feb
Related: rhbz#1789308
When the timeout is too short the http library sends the request again.
We return the last response to the user so even if the cancel works they
get an error about the UUID not being valid.
Resolves: rhbz#1788461
