Signed-off-by: Emanuele Aina <emanuele.aina@collabora.com>

Guillaume Tucker recently fixed the LAVA-side meaning of the sd-card device id
to actually point to the SD card on the ARM64 Renesas R-Car m3ulcb boards.

This solves some out-of-space errors but also means that flashing images takes
more time and sometimes 20 minutes are not enough. Raising the timeout to 60
minutes should be plenty.

Signed-off-by: Emanuele Aina <emanuele.aina@collabora.com>

Since the firewall test was removed and replaced by
iptables-basic in gitlab.apertis.org/tests/iptables-basic,
this folder is no longer relevant.

Signed-off-by: Frédéric Dalleau <frederic.dalleau@collabora.com>

Without the filename it can be cumbersome to find which file caused
the error
user@apertis:~/apertis-tests$ make check
misc/syntax-check.py
Traceback (most recent call last):
  File misc/syntax-check.py, line 56, in <module>
    check_jinja2_template(template)
  File misc/syntax-check.py, line 34, in check_jinja2_template
    expanded_file = env.get_template(filename).render()
  File /usr/lib/python3/dist-packages/jinja2/environment.py, line 812, in get_template
    return self._load_template(name, self.make_globals(globals))
  File /usr/lib/python3/dist-packages/jinja2/environment.py, line 774, in _load_template
    cache_key = self.loader.get_source(self, name)[1]
  File /usr/lib/python3/dist-packages/jinja2/loaders.py, line 175, in get_source
    contents = f.read().decode(self.encoding)
UnicodeDecodeError: 'utf-8' codec can't decode byte 0x87 in position 16: invalid start byte
Makefile:76: recipe for target 'check' failed
make: *** [check] Error 1

Replace that ugly traceback with a more comprehensive output:
user@apertis:~/apertis-tests$ ./misc/syntax-check.py
Exception 'utf-8' codec can't decode byte 0x87 in position 16: invalid start byte when checking .common-boot-tpl.yaml.swp

Signed-off-by: Frédéric Dalleau <frederic.dalleau@collabora.com>

The JSC testsuite has never been executed on LAVA due to resource
contraints. We are not going to work on the JavaScript engine in any
case so running the testsuite is of dubious value.

Since it has never been used, takes a lot of resources and doesn't
provide any obvious benefit, let's drop it.

Signed-off-by: Emanuele Aina <emanuele.aina@collabora.com>

Commit 3df21f3c moved the firewall test to the iptables-basic
repository but didn't drop the files from this repository.

The test definition here is now unused, with the new definition being
also available at https://qa.apertis.org/iptables-basic.html



Signed-off-by: Emanuele Aina <emanuele.aina@collabora.com>

Delete the YAML test definitions for the tests that are now shipped in
the https://gitlab.apertis.org/tests/apertis-test-cases repository
and published on https://qa.apertis.org/



Signed-off-by: Emanuele Aina <emanuele.aina@collabora.com>

This commit redirects the minimal standalone lava tests to use
the new test cases files.

Signed-off-by: Luis Araujo <luis.araujo@collabora.co.uk>

Signed-off-by: Luis Araujo <luis.araujo@collabora.co.uk>

This commit redirects the remaining tests in the LAVA groups
to use the new test cases.

Signed-off-by: Luis Araujo <luis.araujo@collabora.co.uk>

This commit redirects the apparmor lava jobs to use the new
test case repository.

Signed-off-by: Luis Araujo <luis.araujo@collabora.co.uk>

Luis Araujo authored 6 years ago and Frédéric Dalleau committed 6 years ago

This commit updates the ADE lava job templates to use the new
test case repository.

Signed-off-by: Luis Araujo <luis.araujo@collabora.co.uk>

This commit redirects the common, folks and modules lava tests
groups to use the new test cases repositories.

Signed-off-by: Luis Araujo <luis.araujo@collabora.co.uk>

This commit redirects the SDK lava group tests to use
the test cases repositories.

Signed-off-by: Luis Araujo <luis.araujo@collabora.co.uk>

This commit redirects the apparmor standalone lava group tests to use
the test cases repositories.

Signed-off-by: Luis Araujo <luis.araujo@collabora.co.uk>

This commit redirects the common standalone lava group tests to use
the test cases repositories.

Signed-off-by: Luis Araujo <luis.araujo@collabora.co.uk>

The test issue ended up being a dbus packaging issue
and should be sorted now.

Signed-off-by: Andre Moreira Magalhaes (andrunko) <andre.magalhaes@collabora.co.uk>

The debug-build tests shipped with dbus-tests are built with different
configure flags than the system dbus daemon, more specifically the debug-build tests
are built with --enable-verbose-mode, which makes the tests expect for the
DBUS_VERBOSE_INTERFACE which is only present if the dbus daemon is also
built with verbose mode enabled (not the case).

Signed-off-by: Andre Moreira Magalhaes (andrunko) <andre.magalhaes@collabora.co.uk>

Apertis: T5424

Signed-off-by: Andrej Shadura <andrew.shadura@collabora.co.uk>

The `group-modules-tpl.yaml` template is always included after
`group-common-tpl.yaml` which already runs the
`common/boot-no-crashes.yaml` test, so it would just be a duplicate.

To make things worse, since commit 1f9a12de the
`group-modules-ported-tpl.yaml` template is included at the top of
`group-modules-tpl.yaml`, so the Canterbury tests run before checking
for crashes in the log: this means that any crash in the Canterbury
testsuite would mistakenly trigger the `boot-no-crashes` check.

Signed-off-by: Emanuele Aina <emanuele.aina@collabora.com>

The test-dbus-daemon test fails due to some hard limits set in our
default dbus configuration, so lets mark it to skip until the issue
with limits is properly investigated.

See also https://phabricator.apertis.org/T5565

 to track this issue/fix.

Signed-off-by: Andre Moreira Magalhaes (andrunko) <andre.magalhaes@collabora.co.uk>

Some of the tests executed by the dbus-installed-tests have shell counterparts that
do some setup work (e.g. load apparmor profiles) before executing the actual binaries,
so lets skip executing the binaries directly when a shell counterpart is available.

Signed-off-by: Andre Moreira Magalhaes (andrunko) <andre.magalhaes@collabora.co.uk>

Signed-off-by: Ritesh Raj Sarraf <rrs@debian.org>

Andre Moreira Magalhaes authored 6 years ago and Frédéric Dalleau committed 6 years ago

The tumbler-thumbnailing test has now moved gitlab.apertis.org/tests/tumbler-thumbnailing.git.
The Apertis test plan needs to be updated to use those tests in order to
test on immutable rootfs setups.

Signed-off-by: Andre Moreira Magalhaes (andrunko) <andre.magalhaes@collabora.co.uk>

The idea is that the name should communicate what they are instead
of what we did.

Signed-off-by: Frédéric Dalleau <frederic.dalleau@collabora.com>

Signed-off-by: Frédéric Dalleau <frederic.dalleau@collabora.com>

The apparmor-tumbler test has now moved gitlab.apertis.org/tests/apparmor-tumbler.git.
The Apertis test plan needs to be updated to use those tests in order to
test on immutable rootfs setups.

Signed-off-by: Andre Moreira Magalhaes (andrunko) <andre.magalhaes@collabora.co.uk>

The base SDK does not ship any of the packages in the `modules` group
(Canterbury, Ribchester, Didcot, Traprain, etc.) so it should not run
those tests.

Signed-off-by: Emanuele Aina <emanuele.aina@collabora.com>

The apparmor test has now moved gitlab.apertis.org/tests/apparmor.git.
The Apertis test plan needs to be updated to use those tests in order to
test on immutable rootfs setups.

Signed-off-by: Frédéric Dalleau <frederic.dalleau@collabora.com>

Andre Moreira Magalhaes authored 6 years ago and Frédéric Dalleau committed 6 years ago

The apparmor-pulseaudio test has now moved gitlab.apertis.org/tests/apparmor-pulseaudio.git.
The Apertis test plan needs to be updated to use those tests in order to
test on immutable rootfs setups.

Signed-off-by: Andre Moreira Magalhaes (andrunko) <andre.magalhaes@collabora.co.uk>

The only images running in QEMU are SDK images and they
already contain all the required repositories, so this test
adding repos to the image is not needed there.

This add-repo test was also causing errors because of multiple
entries in the sources.list of the SDK images.

Signed-off-by: Luis Araujo <luis.araujo@collabora.co.uk>

Andre Moreira Magalhaes authored 6 years ago and Frédéric Dalleau committed 6 years ago

The rfkill-toggle test has now moved gitlab.apertis.org/tests/rfkill-toggle.git.

Signed-off-by: Andre Moreira Magalhaes (andrunko) <andre.magalhaes@collabora.co.uk>

Signed-off-by: Ritesh Raj Sarraf <ritesh.sarraf@collabora.com>

Ritesh Raj Sarraf authored 6 years ago and Ritesh Raj Sarraf committed 6 years ago

Also invoke ade job through qemu sdk image

Signed-off-by: Ritesh Raj Sarraf <ritesh.sarraf@collabora.com>

Both `common-*boot-tpl.yaml` and `group-*-tpl.yaml` are running the
sanity-check tests due to historical reasons that date back to LAVAv1
where a reboot was needed between the two set of tests.

Nowadays running the test twice is no longer needed, so drop the
sanity-check in `group-*-tpl.yaml` and rename `sanity-check-initial` to
`sanity-check` in `common-*boot-tpl.yaml`.

Signed-off-by: Emanuele Aina <emanuele.aina@collabora.com>

Signed-off-by: Ritesh Raj Sarraf <ritesh.sarraf@collabora.com>

Signed-off-by: Frédéric Dalleau <frederic.dalleau@collabora.com>