We need this for upgrade tests now I fixed the upgrade test
scheduling so Rawhide upgrade tests run from Rawhide-1 and
Rawhide-2, not currrel and prevrel...
Signed-off-by: Adam Williamson <awilliam@redhat.com>
Some of the version config directives make it possible to wind
up with dupes - for instance, our config for the 'minimal' image
can result in multiple instances of VirtInstallImage for F31
appearing in the list, meaning we'll build the same image two
or more times in the same run. That's silly! Let's not do that.
Using a dict keyed on the release number and arch we wind up with
after interpreting the config file should avoid it.
Signed-off-by: Adam Williamson <awilliam@redhat.com>
It's only used for upgrade tests, and we don't test upgrading
from Branched, only from current stables.
Signed-off-by: Adam Williamson <awilliam@redhat.com>
This is related to enabling update tests for KDE. While testing
that it transpired that the KDE disk image is missing various
stuff that is on the live image and systems installed from the
live image; this is because the kde-desktop-environment env
group doesn't by default include several groups and individual
packages that the live kickstart *does* include. So our current
disk image closely matches what you get if you do a network
install and pick the KDE env group and don't pick any option
groups, but that's not a very common thing to do, and it causes
problems for the tests (e.g. okular and firefox not being
there). I think it's reasonable to make the image resemble a
live install more closely; this is more convenient but I think
it's *also* more useful, as there are probably far more KDE
installs out there *with* these things than without them. More
KDE installs are probably done with the live than via network
install, and even KDE installs done via netinst may well have
these bits added subsequently.
Obviously, we bump the image version with this change, so we'll
need to update the scheduler and templates also.
Signed-off-by: Adam Williamson <awilliam@redhat.com>
its correct type.
Previously, it was not possible to set the type of the partition via
the specific GUID. This commit adds support for adding the GUID into
the gtp_type of the partition description in hdds.json and this field
will be utilized in to code.
This commit adds support for boot options, that can be passed
from `hdds.json` to control the creation of the virtual
machines, such as enabling of EFI based machines, boot order
control, etc. It also adds EFI based machine to `hdds.json`
and adds a kickstart file for such machine.
We previously did a filter like the one for 'i686 on F31+' for
'ppc64 on F29+', only in a different place which was actually a
better place. That filter is now unneeded as F28 went EOL, so
drop it...but move the i686 F31+ filter into the same place, as
it's better done there than in create().
Signed-off-by: Adam Williamson <awilliam@redhat.com>
This just isn't possible any more, we no longer produce all
the necessary bits to generate images for i686.
Signed-off-by: Adam Williamson <awilliam@redhat.com>
This reverts commit 3f9cbd27d4.
I verified that related bug#1571860 is not present anymore on f29.
Signed-off-by: Michel Normand <normand@linux.vnet.ibm.com>
to avoid error like:
===
$~/createhdds/createhdds.py support
...
Traceback (most recent call last):
File "/root/createhdds/createhdds.py", line 821, in <module>
main()
File "/root/createhdds/createhdds.py", line 811, in main
args.func(args, hdds)
File "/root/createhdds/createhdds.py", line 701, in cli_image
img.create(args.textinst)
File "/root/createhdds/createhdds.py", line 291, in create
loctmp.format(fedoradir, str(self.release), variant, arch), "--name", "createhdds",
NameError: name 'variant' is not defined
===
Signed-off-by: Michel Normand <normand@linux.vnet.ibm.com>
Archive now has up to F27, we're not testing ppc64 any more, and
the old workaround is no longer needed.
Signed-off-by: Adam Williamson <awilliam@redhat.com>
We need minimal to be 18GB now, because we're booting from it for
the Workstation live image build test. Not doing a version bump
as they turn out to be a bit of a pain to deal with, I should
enhance that somehow.
Signed-off-by: Adam Williamson <awilliam@redhat.com>
We need it bigger for the new 'update installer image' test,
and when changing it we must bump its version.
Signed-off-by: Adam Williamson <awilliam@redhat.com>
ppc64 (as opposed to ppc64le) was retired in Fedora 29. Tweaking
hdds.json to reflect this now is a bit awkward, so instead let's
leave it as-is, but add code to createhdds to not really include
a ppc64 image in the expected list if it's for F29 or later.
Once F28 goes EOL we can just drop all ppc64 from hdds.json and
lose this code.
Signed-off-by: Adam Williamson <awilliam@redhat.com>
Rawhide tests are failing as there isn't enough disk space in
this image to copy the whole Server DVD...
Signed-off-by: Adam Williamson <awilliam@redhat.com>
README was badly outdated, so give it a coat of paint. A person
trying to get up to speed on openQA was misled by the current
message into thinking something was wrong.
Signed-off-by: Adam Williamson <awilliam@redhat.com>
This ensures that gnome-initial-setup will run in user mode on
first login as 'test' user for F26+, as it did for <=F25 and as
the tests expected (we can now remove the workaround I added
because this stopped happening with F26). Bump the image version
so the images get rebuilt.
...But also be careful not to treat them as 'unknown' either.
This avoids the ansible plays regenerating outdated images (we
try to avoid that and just have the cron job do it).
Various things were being set in odd orders and not used right
later on, this is a general clean-up that solves that and makes
things a bit easier to read (I hope).
Whoops, that last commit just wasn't very good. We need to
re-open the libvirt connection and re-get the domain object,
and also we should catch errors while destroying and undefining
the domain...
Signed-off-by: Adam Williamson <awilliam@redhat.com>
