Signed-off-by: NCaio Carrara <ccarrara@redhat.com>

Signed-off-by: NCaio Carrara <ccarrara@redhat.com>

In an effort to make jobs more self contained, and given that dry-run
jobs created temporary content that is not supposed to be kept after
the job is run, this introduces the following changes:

 * Jobs with "dry-run" enabled will now, by default, cleanup their
   own data.  Basically, this means that the "base log directory"
   created during job initialization, and that for non "dry-run"
   jobs is usually "~/avocado/job-results", will also be removed
   at the job's "cleanup()" phase.

 * A new "--dry-run-no-cleanup" command line option, that only makes
   sense when jobs with dry-run enabled start to clean up data for
   themselves

 * Because some result plugins will, by default, write to the job
   results directory, results plugins are now run before a job cleanup
   is done.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

By making it inherit from object instead, now that all children classes
are boiler-plate free and all setup a "iso" attribute, we can remove
the check and make only them (children classes) inherit from
unittest.TestCase.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

As per the last change, let's move the test into the base class, and
just check if the ISO has been setup for this test to work with, and
if not, skip it.  This removes a lot of the boiler plate code.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

The division of responsiblities in the ISO9660 test classes is one that,
because the children classes need to setup the "iso" instance, the
"mnt_dir_worflow" is not named as a method that is considered a test
in the base class.

The problems with that are:

 1) a lot of boiler plate code in the children classes just to call
    the "mnt_dir_workflow" method.
 2) skip*() decorators do not work in generic methods.

To simplify issue 1, and fix 2, let's rename that method to become
a test method, and just add a condition that, unless "self.iso" is
prepared, the test is going to be skipped.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

Signed-off-by: NCaio Carrara <ccarrara@redhat.com>

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

Signed-off-by: NCaio Carrara <ccarrara@redhat.com>

When using the "process.kill_process_tree" it might be useful to wait
for all of the pids to die.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

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

Signed-off-by: NCaio Carrara <ccarrara@redhat.com>

Signed-off-by: NCaio Carrara <ccarrara@redhat.com>

Signed-off-by: NPlamen Dimitrov <pdimitrov@pevogam.com>

Signed-off-by: NPlamen Dimitrov <pdimitrov@pevogam.com>

In particular, add parameter types, exceptions, and fixes. Also,
single line docstrings are recommended to be one liners in terms
of quotation as well.
Signed-off-by: NPlamen Dimitrov <pdimitrov@pevogam.com>

Signed-off-by: NPlamen Dimitrov <pdimitrov@pevogam.com>

lvconvert instructs the kernel to do the LVM volume merge
and then polls every fifteen seconds if the conversion finished.

Using a ramdisk, the merge is usually done below one second,
so we wait 14 seconds for nothing.
Signed-off-by: NPlamen Dimitrov <pdimitrov@pevogam.com>

Since we're running unittests, let's keep it honest and use the true
list of tests for both the parallel and non-parallel checks.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

So that external users can pick and choose which components
of the test suite to include.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

It will be useful to choose which tests to put into the suite we'll
run.  For instance, when running tests in parallel, we pick tests
from each of the directories at a time.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

Which can be used elsewhere (such as in parallel testing code).
Signed-off-by: NCleber Rosa <crosa@redhat.com>

There are at least two places in which the same functionality is
present.  Let's make it a single utility function instead.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

According to the references given below, Travis workers now run
on VMs that are sudo capable, and the sudo configuration should
now be removed.

Reference: https://docs.travis-ci.com/user/reference/overview/
Reference: https://blog.travis-ci.com/2018-11-19-required-linux-infrastructure-migrationSigned-off-by: NCleber Rosa <crosa@redhat.com>

There are a number of backends and requirements for iso9660 tests
to run.  Let's add those as build requirements to that as many
as possible tests and code paths get executed during the checks
in the RPM build.  With this, two other test cases get executed:

Before:
 - (Python 2): Ran 600 tests in 198.076s OK (skipped=42)
 - (Python 3): Ran 600 tests in 196.489s OK (skipped=45)

After:
 - (Python 2): Ran 600 tests in 180.514s OK (skipped=40)
 - (Python 3): Ran 600 tests in 188.545s OK (skipped=43)
Signed-off-by: NCleber Rosa <crosa@redhat.com>

Currently RHEL provides function accepts just filename, absolute
path of file is not required to be passed. But provides function
for Ubuntu expects users to send the absolute path. This commit
makes changes to Ubuntu provides code to make it generic. So
the consumers just can call provides function with filename.
Signed-off-by: NSrikanth Aithal <sraithal@linux.vnet.ibm.com>

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

On a number of situations, sudo is being missed as a parameter for
command executions that need to run privileged.  Examples include when
writing to mountpoint, using a Python interpreter, and when trying to
listing open files during a forceful unmount of the disk.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

Signed-off-by: NCaio Carrara <ccarrara@redhat.com>

Signed-off-by: NCaio Carrara <ccarrara@redhat.com>

Signed-off-by: NCaio Carrara <ccarrara@redhat.com>

Some distros don't create `/bin/python` and only ship with
`/bin/python2` or `/bin/python3`.  As we are already detecting
`python3` let's also look for `python2` explicitly.

Also, let's fallback to unversioned `python` if no other is found, as
unlikely as it may be.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

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

Signed-off-by: NCaio Carrara <ccarrara@redhat.com>

Signed-off-by: NPlamen Dimitrov <pdimitrov@pevogam.com>

And issue a deprecation warning.  It will then be removed in a couple
of Avocado releases.

Reference: https://trello.com/c/7TBqcAkR/1447-thinlvcreate-deprecate-the-utility-functionSigned-off-by: NCleber Rosa <crosa@redhat.com>

To minimize brekage of existing test code.

Even though the API may looks better on the last commit, it's better
to add the new parameters to the end. That is, force_flag should
continue being the third parameter, so users using positional
parameters won't be surprised by the order change.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

The thin volume creation can be refactored and created with just
a few different function arguments. A similar but a bit more
elaborate approach can then be used to take thin rather than
regular shapshots from thin or regular logical volumes.
Signed-off-by: NPlamen Dimitrov <pdimitrov@pevogam.com>

The test names are based on the executable output with '-l', but
it should be treated as text, and not as bytes.

This fixes issue #2927.

Reference: https://github.com/avocado-framework/avocado/issues/2927Signed-off-by: NCleber Rosa <crosa@redhat.com>

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