Commit Graph

348 Commits

Author SHA1 Message Date
Michel Normand
05cf6f03bc Increase from 120 to 180s delay for rescue_select
required at least for install_repository_hd_variation
test that failed systematically for ppc64le since a while
https://openqa.stg.fedoraproject.org/tests/980357#step/preinstall_iso_in_hd/17
https://openqa.stg.fedoraproject.org/tests/980357#next_previous

Signed-off-by: Michel Normand <normand@linux.vnet.ibm.com>
2020-12-16 04:18:51 +01:00
Adam Williamson
940848b628 Use 'test.openqa.fedoraproject.org' not 'domain.local'
Using .local is apparently Bad Form because it's reserved for
mDNS. However there doesn't appear to be any particularly Good
Form for what to call a test domain you never want to exist
outside of a closed system, apparently. Sigh. Let's try this.
Includes a bump to disk_ks version because the kickstarts on
that image also need to have this change applied.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-12-15 14:40:30 -08:00
Adam Williamson
1c692ef6d6 Add a wait to custom_change_fs
Otherwise we can immediately match 'fs is already selected'
for the *previously selected* fs before the UI updates and exit
without actually changing the fs of the intended partition.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-12-04 16:15:33 -08:00
Adam Williamson
c7a1b94c84 Enable aarch64 disk image testing, related fixes
This sets us up to test the release-blocking aarch64 disk images
(Minimal, Server and Workstation). It also allows for testing
armhfp disk images on aarch64 worker hosts (though my testing of
that isn't going too well so far), and fixes the initial-setup
handling for a change upstream ('use password' is now the default
so we don't need to choose it). We rewire disk image deployment
test loading to work through the generic loader code rather than
using ENTRYPOINT, as it allows us to more gracefully handle
graphical (Workstation) vs. console (Server, Minimal), moving
the code for handling console initial-setup to a helper function
just like the code for gnome-initial-setup and having _console_
wait_login call it when appropriate. We also tweak desktop_vt a
bit because now we need to switch from a console running as test
to a desktop, which breaks the assumption that the highest
numbered session of user test is the desktop...

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-10-30 12:39:49 -07:00
Adam Williamson
f51a804357 Handle emergency console entry with no password
This happens on ARM disk images.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-10-28 15:21:03 -07:00
Adam Williamson
cfb226d3fd Simplify anacondatest root_console to always use tty3
Rawhide KDE lives now have the desktop on tty2, and the installer
environment tty3 now has a shell (in Ye Olde Times it didn't, not
sure when that changed but it's the case at least back to F31).
So let's make our lives simple and just always use tty3 here.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-10-26 09:21:53 -07:00
Adam Williamson
e5f396a785 Drop workaround for fixed GNOME submenu bug
https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/2878 was
fixed (or just rendered obsolete by a change to how submenus
work), so we don't need this any more.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-10-15 15:36:09 -07:00
Adam Williamson
1c33d07d38 Drop workaround_ble26, bug was fixed months ago
https://pagure.io/background-logo-extension/issue/26 was fixed
months back, we don't need this any more.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-10-15 15:36:04 -07:00
Adam Williamson
db99e5462c Drop F33 FreeIPA update from workarounds, it went stable
Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-10-10 17:27:33 -07:00
Adam Williamson
9de75af60b Drop f33-backgrounds from workarounds (it went stable)
Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-10-09 16:57:21 -07:00
Adam Williamson
94b47afc53 Tweak setup_tap_static and FreeIPA tests for resolved
This does some of the things suggested by cheimes in
https://bugzilla.redhat.com/show_bug.cgi?id=1880628#c24 . It
seems to make the replica tests work with resolved, still work
with pre-F33 resolving, and not break anything. Also remove the
workaround to disable resolved if it's running, as we can now
work with it.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-10-09 16:54:46 -07:00
Adam Williamson
401ebecc10 Add F33 FreeIPA 4.8.10-5 update as workaround to fix upgrades
FreeIPA upgrade test is failing because of
https://bugzilla.redhat.com/show_bug.cgi?id=1886205. The test
failing every time is not useful as we know what the issue is,
so add the update as a workaround to avoid it.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-10-08 11:28:10 -07:00
Adam Williamson
e89de5fffb Bump a timeout a bit for aarch64
Seems like 10 seconds here isn't always long enough for aarch64,
so bump it.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-10-07 15:21:17 -07:00
Adam Williamson
4fcea15610 Updates testing: add f33-backgrounds workaround
Add FEDORA-2020-27f80050a2 as a workaround because without it the
KDE desktop background test fails due to the bug in -6.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-10-07 12:40:27 -07:00
Adam Williamson
8d9684dd80 Update tests: don't install git any more
We were using it to checkout a git version of python-fedora to
work around a bug, long ago, but we don't do that any more so
we don't need it.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-10-07 10:19:33 -07:00
Adam Williamson
aafe460f85 Update testing: drop all workarounds
They're all stable now.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-10-07 10:09:59 -07:00
Adam Williamson
1fdf7cd567 Handle update build not having any packages for target arch
openQA sometimes winds up testing an update that doesn't have
any packages for x86_64 (or aarch64). The most common case is
s390utils, which is on the critpath but only has packages for
s390x. I would ideally like to skip scheduling entirely if the
update has no packages for the arch we're scheduling on, but
sadly that involves using the Koji API which is XML-RPC and I
don't really want to deal with that again. This deals with it
at the test level instead, by checking the error message if
`koji download-build` fails and carrying on if it's the "no
packages for this arch" error. That means if the update has no
packages at all for our arch we're not really testing anything,
but that's better than a bunch of false failures, I guess.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-10-07 09:49:03 -07:00
Adam Williamson
fb1bc1dad8 Add mock 2.6 to workarounds, drop kernel for F31 and F32
Kernel updates for F31 and F32 went stable so they can come out.
mock 2.6 fixes the bug that occurs when /etc/resolv.conf is a
dangling symlink - this breaks the live_build test.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-09-17 16:40:53 -07:00
Adam Williamson
ee666d5f13 Add kernel 5.8.9 updates to workarounds (fix parted problems)
Kernel 5.8.8 broke the installer by changing return codes for
partition operations, these updates are listed as fixing it.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-09-14 18:48:08 -07:00
Adam Williamson
d86960166d drop FEDORA-2020-2bef864cab as a workaround (it went stable)
Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-09-14 18:47:23 -07:00
Adam Williamson
0350977ba3 Update workarounds: drop stable one, add F33 FreeIPA upgrade one
The F32 FreeIPA update that changed the web UI has gone stable
now, so remove it. A pki-core update has just come out that fixes
F32 -> F33 FreeIPA server upgrade test: add this as a workaround
for F33 so that test stops failing on every update.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-09-08 14:13:29 -07:00
Adam Williamson
0999921fe3 Tweak detection of installed update packages
Previously we were relying on `rpm -q` always outputting the
right package last. We saw some test failures on recent kernel
updates, e.g. https://openqa.fedoraproject.org/tests/658768 ,
which indicate this isn't always the case; there the 'right'
package was second of three for kernel, third of three for
kernel-core and first of three for kernel-modules. So we need to
make it more robust. This uses an additional call:
`rpm -q $pkg --last | head -1` to find the most recent package,
if there are more than one; this should always be the right one,
I hope. Note we cannot just add `--last` to the `rpm -q --qf...`
call because the output when you do that is weird; you get the
output you'd get if you just called `rpm -q --last` first, and
*then* the query-formatted output afterwards (though with the
modified order as expected). There doesn't seem to be any way to
get only the latter.

I also tweaked the log uploading so we always upload the working
logs even when the test passes; it can't hurt anything and it is
sometimes useful to have them.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-09-08 14:03:23 -07:00
Adam Williamson
478b7eff9e Add initial CoreOS product and test templates
Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-09-02 14:49:29 -07:00
Adam Williamson
170ef0733a Use nmcli for static network stuff, not ifcfg files
This should work even if the ifcfg plugin is not present (hi,
CoreOS) or 'predictable' (har) network names are on.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-09-02 14:49:29 -07:00
Adam Williamson
527cb63152 post_fail_hook: upload PostgreSQL init log if it exists
Why...why...WHY is this not in /var/log. Sheesh.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-08-25 16:18:32 -07:00
Adam Williamson
b9f6ecd72d Conditionalize FreeIPA UI change, add 4.8.9 update to workarounds
The FreeIPA UI change that the previous commit adapted to is in
4.8.9. That's stable for Rawhide and F33 already, but still in
testing for F32, and won't go to F31. So we need to make the
change conditional on release number, and we also add the update
to workarounds for F32 so we don't have to do something awkward
while we wait for it to go stable.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-08-21 14:00:42 -07:00
Adam Williamson
2713ce60dd Add a new not-really-a-Next-button needle for gnome-initial-setup
In g-i-s 3.37.91, the first screen has a 'Start Setup' button
rather than a 'Next' button. Easiest thing for us to do here is
just to add a new needle which has the 'next_button' tag even
though it's clearly not a 'Next' button, because then the code
still works :) So do that, but give the file a suggestive name
and explain the situation in a code comment.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-08-18 12:23:09 -07:00
Adam Williamson
8ee330b5d3 Add krb5 reversions for #1868482 to update workarounds
Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-08-13 15:11:35 -07:00
Adam Williamson
0557a774ac Retry bodhi updates download a few times if it fails
It's failing about one in six tries currently, with Bodhi 5.5 on
the server end: https://github.com/fedora-infra/bodhi/issues/4105
this should work around that.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-08-13 15:03:36 -07:00
Adam Williamson
41ddac0951 Drop all update workarounds
The ones that were in there are stable now, plus downloading them
is hitting a bug in Bodhi and breaking tests.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-08-13 07:55:05 -07:00
Adam Williamson
aa41fe4e4e Automate QA:Testcase_Zezere_Ignition
This is a bit complex to automate, because we cannot really use
the production Zezere server (provision.fedoraproject.org) as
the test case shows, as we'd have to solve authentication and
we also don't really want to constantly keep registering new
hosts to it that are going to disappear and never be seen again.

So, instead we'll do it by setting up our *own* Zezere, and
provisioning our IoT system in that. We run two tests. The
'ignition' test is the actual IoT 'device'; all it really does
is boot up, sit around, and wait to be provisioned. The 'server'
test first sets up a Zezere server, then logs into it, adds an
ssh key, claims the IoT device, provisions it, and connects to
it to create a special file which tells the 'ignition' test
everything worked and it can close out.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-07-23 18:01:06 -07:00
Adam Williamson
72edbfe991 Use qemu host IP 172.16.2.2 not 10.0.2.2
This is to make the infra folks happy, apparently using 10.0.x.x
and 10.1.x.x is causing conflicts since our actual infra network
uses those ranges too.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-07-23 16:40:45 -07:00
Adam Williamson
15eae21722 start_with_launcher: adjust for new overview behaviour
The annoying submenus in the overview app list now scroll right
not down :/ have to adapt this function for that. Had to move
get_release_number earlier because perl ordering.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-07-09 15:54:33 -07:00
Adam Williamson
a3943dc307 Sigh, fix missing brace
Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-07-09 14:31:44 -07:00
Adam Williamson
4a6cd8bcd5 Abstract out overview type-to-search bug workaround
And also use it in GNOME apps settings.pm test.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-07-09 14:25:22 -07:00
Adam Williamson
6232be32cf Sigh, move menu_launch_type after get_release_number
Stupid perl.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-07-08 17:09:45 -07:00
Adam Williamson
2f93645a55 Workaround GNOME overview search box not active bug
See https://pagure.io/background-logo-extension/issue/26 - in
current Rawhide, the search box in the overview is not active
when the overview is opened, so you can't just open the
overview and type, you have to click it first.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-07-08 17:07:48 -07:00
Adam Williamson
f70b8ec943 post_fail_hook: don't try ctrl-c on serial console
Can't do that.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-07-07 17:35:32 -07:00
Adam Williamson
ddea8e2169 post_fail_hook: hit ctrl-c if 'dnf -y installer tar' gets stuck
This happens in the iSCSI install test.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-07-06 16:38:12 -07:00
Adam Williamson
7ee071ca08 Add g-i-s timezone bug fix updates as workarounds
This should avoid the bug happening in upgrade tests (I already
built the fix into the base disk image, which should avoid it
happening in other tests).

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-06-24 09:31:40 -07:00
Adam Williamson
f96c14adc0 gnome-initial-setup: add one more check for that auth_required
Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-06-17 08:17:49 -07:00
Adam Williamson
fd26d347fe Workaround unexpected authentication required dialog in g-i-s
https://gitlab.gnome.org/GNOME/gnome-initial-setup/-/issues/106
This started happening after the data center move. Really not
sure why.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-06-12 12:28:18 -07:00
Adam Williamson
1b37200446 Work around GNOME Shell app submenu bug
See https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/2878 .
GNOME 3.37.2 seems to have a bug with submenus in the app menu;
the first time you open one you can't scroll through it using
the keyboard. On every open after the first it works fine. This
is a quick and dirty workaround - when we're dealing with a
submenu, open it then close it then open it again.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-06-11 16:00:56 -07:00
Adam Williamson
2e21cfac15 Update cockpit needles and select_cockpit_update for 220
In Cockpit 220, the Updates entry is off the bottom of the screen
so we need to scroll the left bar down before we can click it.
Also update some other needles.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-05-30 10:27:20 -07:00
Lukáš Růžička
4d4b2d7739 Finish the post review changes. 2020-05-28 13:32:42 -07:00
Lukáš Růžička
b5169e602b Use existing needles for fs selection. 2020-05-27 15:54:10 -07:00
Lukáš Růžička
9efb0e10e6 Create a test to resize LVM partitions using Blivet. 2020-05-27 15:54:09 -07:00
Adam Williamson
b59f875af1 Drop F32 workaround updates (both now stable)
Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-05-27 13:15:37 -07:00
Adam Williamson
acc46464f8 Revise release variable handling, prerelease checks, os-release
I started out trying to fix os-release for the recent change to
add "Prerelease" tags to the VERSION and PRETTY_NAME fields, then
things spiralled. It got me thinking about the awkward DEVELOPMENT
variable we use, so I decided to get rid of it and refactor the
few things that use it. I refactored the anaconda prerelease tag
check, and wrote a new giant comment that gives details about
exactly how anaconda decides whether to show those tags, to give
context to our choices about when to expect them. This check now
uses a new LABEL variable the scheduler now sets. I also wound up
creating new UP1REL and UP2REL vars to define the 'source' release
for upgrade tests, separate from CURRREL and PREVREL, which are
now never lies - they really are the current stable and previous
stable release, even for update upgrade tests.

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-05-07 15:42:08 -07:00
Adam Williamson
8eaddd8c3d Set OfferToSaveLogins policy to false in Firefox config
This seems to work around the Firefox 76 bug with password entry
sometimes breaking:
https://bugzilla.mozilla.org/show_bug.cgi?id=1635833

Signed-off-by: Adam Williamson <awilliam@redhat.com>
2020-05-06 13:25:46 -07:00