In some external test suites, what is defined as a "test" can not be
run by itself and need a specific runner script/tool.

This introduces the concept of an "inner runner", that is, a custom
test runner or tool that can deal with custom tests that do follow
the Avocado SIMPLE test definition.

This implements the Trello card:

 https://trello.com/c/TpXecE2n/486-introduce-inner-runner-to-avocado-run

More information can be found in the man page section about it.

Changes from v1:
 * removed "level" from "inner level runner" mentions
 * replicated docs from man page to "main" docs
Signed-off-by: NCleber Rosa <crosa@redhat.com>

avocado.core.output: Use stdout to log test statuses

avocado.core.test: Use GDB-friendly srcdir

Throbber (test progress) supports two colors. Currently it turns green
only when new status arrives, which is currently only used when we
interrupt the test (GDB) or when the test finishes. Then the throbber
turns green for one iteration and back to yellow in next iteration.

This patch changes the behavior so the throbber is green when the test
process is alive and yellow when it's interrupted.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

The human output prints the test name, progress and pass/skip test
statuses into stdout and fail/error/interrupt ones as stderr. This
results in unreadable output. This patch logs all of them in stdout.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

Avocado allows the usage of GDB, but our default self.srcdir is
incompatible with it causing many tests to fail. This patch simply
replaces the ':' in basename to '_'. It's already created in a temporary
directory, so there shouldn't be any name clashes.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

We can't set signal handlers inside a thread. In such
cases, let's not try to set the signal handler. This
fixes a problem where subprocesses executed with
`avocado.utils.process` APIs would fail if executed
inside threads.

One example of a thread that runs subprocesses can
be found on our avocado-vt plugin: The virt related
tests run a screendump thread, that will produce
screendumps on a specific directory. These screendumps
will be encoded later into a video. The libvirt VM
implementation takes screendumps by executing
process.run('virsh screendump'). So that's one of
the possible examples where process APIs can be used
inside of threads, and we should support such examples.
Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

VM plugin cleanup bug

Fix example test "linuxbuild.py"

GNU make (mis)understands paths that contain ":". The reason is that they look
like static patterns:

 https://www.gnu.org/software/make/manual/html_node/Static-Usage.html#Static-Usage

The exact error message one would see would be similar to:

 Makefile:XXX: *** multiple target patterns.  Stop.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

The `avocado.utils.kernel_build.KernelBuild` utility class can now
build a Linux kernel without an external configuration file, so let's
add this possibility to the test itself.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

The Linux kernel compilation instructions suggest that a separate
directory is used for building the kernel and not the source tree
itself. Let's make this "exemplary" helper class follow those
guidelines.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

And let's use a generated config by default, that is, let's drop the
custom config file we ship with the test.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

On most recent kernel make dep is deprecated and doesn't add any value
to the build routine itself.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

A Linux kernel can be built without an external configuration file
because it can generate a default one for the current target
platform. So, let's make that parameter optional.

Also, if a user does provide a config file, we use a completely silent
target for updating that config file, using defaults for missing
configuration keys.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

It's reasonable to let the kernel build class to pick a work directory
itself, thus making that parameter optional.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

It's more common to have variables with all caps defined at the class
level.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

Signed-off-by: NCleber Rosa <crosa@redhat.com>

Since it's not possible (yet?) to depend on existing virtual machines
to run tests for some of the Avocado features, including many of the
VM Plugin features, let's add a test as an example that can be run
manually instead. While they will not be part (at this time?) of the
functional test suite, it may serve as a reference/example in the
future.

The approach chosen here is to have a first run of Avocado making sure that
a flag file does not exist. A second run of Avocado, now executed with an
additional `--vm-cleanup` parameter runs a test that creates the flag file.
Finally, the third and last execution of re-runs the same test as the first
run, checking that the flag file does *not* exist, that is, the clean up of
the VM did work.

Because this test requires a libvirt domain and the hostname/address
of that virtual machine, these parameters assume no defaults for
safety and/or security causes. The test will be skipped if these are
not configured properly.  Please edit the vm-cleanup.yaml file with
the appropriate parameters.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

Let's be practical and simply destroy (pull the plug on the VM) that
will have its state (snapshot reverted).

Not shutting down (or powering off/destroying) the VM has the side
effect of not reverting the state of the (still running) VM on a
couple of platforms, including EL7.

This fixed issue #396.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

The given dest parameters actually reflect the default generated by
argparse. So, let's clean it up.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

avocado.core.runner: Make getting the timeout from test plugins more …

Some test loader plugins will return the 'timeout' parameter
as a string, so this will cause avocado to break, given that
it will add the timeout to another float value later in the
runner code. Let's ensure that value will be converted to
a float before the addition is made.
Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

They are necessary for the GDB functionality in avocado,
so let's bring them in.
Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

trinity example test: don't output ANSI characters to the logs

Reconsider when the `latest` log symlink is created

The logs from the trinity test are full of ANSI characters that add
colors to messages, which makes it difficult to read/manipulate
them.

Fix that by passing -m to trinity. From trinity --help:

   '--monochrome,-m: don't output ANSI codes'

avocado.spec: Add python-mock to BuildRequires

With one new unittest requiring mock, now python-mock
is necessary as a build require.
Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

Instead of creating at the end of the job. This allows users running
long (time wise) jobs to be monitored by "following" the
$AVOCADO_RESULT/latest/job.log file.

This implements the card described at:

 https://trello.com/c/QJ4wCD8x/447-reconsider-when-the-latest-log-symlink-is-createdSigned-off-by: NCleber Rosa <crosa@redhat.com>

The `update_latest_job_logs_dir` is defined in the `data_dir` module
but it's only used in the `job` module. So, let's keep this code right
at where it's used.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

This adds a safety check and prevents Avocado from accidentaly
deleting a regular file, not a symlink, that is named `latest`.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

sysinfo: fix duplicated collection of current mounted filesystems

By default Avocado's sysinfo collects both the output of the `mount`
command and the content of the /proc/mounts file. This is really the
same information with a very minor formatting change (the addition of
parentheses around mount options).

Since reading a file is supposed to be faster than running a command,
let's remove the `mount` command execution.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

We probably don't want to log the full CDROM content...
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

requirements: Update inspektor requirement