This patch simplifies the concept of WARN test. In case user uses the
self.log.warn or self.log.warning methods, test is marked as dirty and
in case everything else worked fine it finishes with TestWarn exception.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

This example demonstrate simple usage of GDB API.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

A few contributions to examples: bind_cpu0, cabort and raise [V2].

Manual Test Plan

Add new test `raise.py` to raise signals to itself and check results.
Signed-off-by: NRudá Moura <rmoura@redhat.com>

Add new test `cabort.py` to call abort() from lib C.
Signed-off-by: NRudá Moura <rmoura@redhat.com>

Add new wrapper script `bind_cpu0.sh` to bind process to CPU 0.
So it will force the process (and its children) to run in one CPU.
Signed-off-by: NRudá Moura <rmoura@redhat.com>

This test is using the location of the source file as the build
location, instead of the `srcdir` provided by the Test class.
Signed-off-by: NRudá Moura <rmoura@redhat.com>

This is an example test plan of what could/should be run before the
release of a new avocado version.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

Some test routines can not yet be easily automated in Avocado, and for those, we
rather test things manually then not at all.

The idea behind this tool is to use a predefined test plan (a simple JSON file)
that has test names and descriptions with how to perform the test and what to
expect. The tool will ask for the results for each test (either PASS, FAIL or
SKIP) and possibly additional notes. These results will then be saved to a JSON
file or to a human readable report.

The human readable report can, for instance, be part of a release commit message,
or if preferred, the JSON result file can be committed to keep track of what is
working and what is known to be broken at a given time.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

Also, change the code a little to tell the difference
between stdout and stderr.
Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

Verify that errors in the list command really go to stderr.
Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

This way commands that do use the paginator will correctly
redirect error messages to stderr:

scripts/avocado list sbrubles > /dev/null
Cannot access 'sbrubles': File not found
Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

Make sure the paginator is cleaned up after displaying
an error.
Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

Following the [ObjectSingular][List] convention.
Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

Let's standardize the class names for plugins that
list things in [ObjectSingular][List].
Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

Following the [object_singular]_[list] convention.
Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

Let's standardize the file names for plugins that
list things in [object_singular]_[list].
Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

More checks to make sure we are not outputting errors to
the wrong loggers.
Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

Create an additional logging handler for the avocado application,
that outputs to stderr, and two filters, that filter messages
per logging level. This way, normal information (app output)
goes to stdout and error messages go to stderr. The unittests
were updated accordingly.
Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

fiotest: fix async engine name and update to latest upstream version

This simply updates fiotest to use the latest release of fio.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

It looks like the name of the loadable engine on fio itself has changed
from simpler "aio" to either a Linux or POSIX version.

Based on my tests on build attempts, it looks like POSIX is more dependant
when it comes to building out of the box. Let's default on that version.

This fixes the following error when running fio:

 fio: engine aio not loadable
 fio: failed to load engine aio
 ...
 fio: file:ioengines.c:99, func=dlopen, error=aio: cannot open shared object file: No such file or directory

That resulted in a test failure.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

Documentation: simple fixes/improvements to the Get Started Guide

doublefree test: fix the temporary directories being used

This test is using the location of the source file as the build
location, instead of the `srcdir` provided by the Test class.

Also, since `srcdir` is a temporary directory, I believe we should
not have to manually clean it up.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

Even though internally we have some more descendants of avocado.test.Test, to
a user, AFAICT, only instrumented and simple tests are available.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

It's obviously not true that Avocado only allows to run tests on "your laptop".
Also let's keep the running of tests and the outcome all in the present tense.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

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

Add a new config section 'runner.output', with a key 'enable' to
enable/disable colored output in the avocado ui. With this, users
troubled with terminal color schemas may turn off colors to make
things more readable.
Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

Make lines fit into 80 columns. No functional changes here.
Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

HTML result fixes

Modify the VM test info in order to correct paths of
test directories. This will fix a long standing bug with
regards to broken HTML report links.
Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

Add the fields 'logdir', 'logfile' and 'fail_reason'.
Those fields will be necessary to expand the HTML
reports to correctly exhibit failure information
from remotely run tests.
Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>