Two different dep issues started breaking tests today -
a libiscsi soname bump broke deps in qemu and libvirt, and a
gnome-shell-extension-background-logo update had an invalid dep.
This adds fixes for all three as workarounds so I can get the
tests of each to pass, and other tests to pass while these are
going through gating.
Signed-off-by: Adam Williamson <awilliam@redhat.com>
This enables tests on the KDE aarch64 disk image. The test set
should be the equivalent of what we run on the Workstation disk
image - most of the same tests we run on x86_64, a few left out.
I might do a follow-up to enable the 'missing' tests on both
arches later, at least ones which don't seem to be left out for
any good reason beyond resources.
Signed-off-by: Adam Williamson <awilliam@redhat.com>
@kparal pointed out that it's a bit inconvenient to quickly check
the journal and any SELinux AVCs for a failed test. All the info
is in the /var/log tarball, but you have to download it, extract
it, and know how to run journalctl on it. For convenience, this
adds a text dump of the journal and one of any AVCs found with
ausearch.
Signed-off-by: Adam Williamson <awilliam@redhat.com>
This PR adds a start-stop test for Kleopatra that will replace
the KGpg application. It does not, however, remove the KGpg
start-stop test from the suite, as the KGpg is not gone yet.
We will remove it in a separate commit when time has come.
Fixes: https://pagure.io/fedora-qa/os-autoinst-distri-fedora/issue/405
Change the start command to kleopatra
Move needle to generic space
Remove checking the selection box
The test failed because one of the applications ceased to exist
on the installation media. The fix brings a different set of
applications and more to it - the applications get installed
during the test so we probably won't have to deal with similar
issue in the future. Or at least not that often.
Fixes: https://pagure.io/fedora-qa/os-autoinst-distri-fedora/issue/412
For some reason, on aarch64, we get the 'person' icon, not the
first letter of the user's name. No idea why.
Signed-off-by: Adam Williamson <awilliam@redhat.com>
timedatectl now refuses to set the date to before the date when
systemd was built, so let's just use date instead.
Signed-off-by: Adam Williamson <awilliam@redhat.com>
This makes us create a generic cloud disk image and then run
tests on it, just like we do for Workstation and KDE lives,
everything netinst, Silverblue, and a container image. This is
intended to help avoid anything like
https://bugzilla.redhat.com/show_bug.cgi?id=2390898 happening
again - if we'd had these tests set up and gating enabled, the
update that caused that bug would've been gated.
This requires some changes that affect other things: bind mounting
the whole of /dev in our mock config when using simple isolation,
and forcing START_AFTER_TEST to "" for kiwi_build (this is needed
to override the setting of it in the updates-cloud flavor). bind
mounting is what koji does for 'real' kiwi build jobs so it's fine
(in fact, better); previously the mock config was shared with
livemedia-creator, but we don't use that any more so it's not a
problem.
It's a bit icky when we get to the point of having *multiple*
+START_AFTER_TEST settings, but I can't see another way to get
everything right here and still share the test suite definition
with compose tests.
Moving the definition of KIWI_PROFILE for container tests is not
strictly related, but I noticed it was weird while working on
this and couldn't leave it alone.
Signed-off-by: Adam Williamson <awilliam@redhat.com>
As reported by @yselkowitz , ftp is no longer in ELN, so this
test fails on ELN composes. autofs is in ELN, has minimal
dependencies and doesn't change often, so it looks like a good
option.
Signed-off-by: Adam Williamson <awilliam@redhat.com>
Test "TCP/IPv4 larger transfer, tap" is flaking currently, so
let's disable it. Paul has reproduced the issue and is working on
the fix so no further purpose is served by failing on it.
Signed-off-by: Adam Williamson <awilliam@redhat.com>
The security fix has been removed for now; we're removing this
workaround so we'll notice if it gets broken again.
Signed-off-by: Adam Williamson <awilliam@redhat.com>
I must have done this back at branching but not committed them
then somehow wiped them without committing them last night?
Signed-off-by: Adam Williamson <awilliam@redhat.com>
Probably mostly the GNOME 49 Alpha. These built up while I was on
PTO and Kamil didn't have time to do them.
Signed-off-by: Adam Williamson <awilliam@redhat.com>
we really don't need to, and the new design that's currently in
Workstation but not yet Silverblue makes it awkward. Just check
one page.
Signed-off-by: Adam Williamson <awilliam@redhat.com>
This is intended for things like system monitor where there's
constant motion on the app screen so we can't wait_still_screen.
Signed-off-by: Adam Williamson <awilliam@redhat.com>
If the test settings indicate not to create a local user account,
let's uncheck the relevant box. This should fix the KDE
install_no_user test.
Signed-off-by: Adam Williamson <awilliam@redhat.com>
We used to use +QEMUCPU Haswell for BaseOS to override the old
machine default of Nehalem, because ELN has a higher CPU baseline
and doesn't run on Nehalem. We changed the machine default to
host, which is fine for ELN, so we should have dropped the
override. But there's a wrinkle - we set base_services_start to
run with MINCPU to fix mcelog.service failing with host, but if
we leave it that way the test will fail to run on ELN. So let's
override MINCPU to Haswell for ELN-on-x86_64 products. That ought
to be the right thing.
Signed-off-by: Adam Williamson <awilliam@redhat.com>
KDE needs a kde-settings update, but everything else should be
showing the new background now. Add a needle for it.
Signed-off-by: Adam Williamson <awilliam@redhat.com>
