Velox is an extension library for the Kolibri project with a purpose to complete Kolibri by allowing setting up reproducible tests to simulate various use-case scenarios and measure their performance.

The project is being actively developed, which means that you may run into potential issues and that you will most probably have to set up a few things before being able to successfully run Velox in your environment.

You can follow the online developer documentation to set up the development environment.

To be able to run tests in the PostgreSQL context, you will have to install PostgreSQL and PostgresSQL related dependencies by running the following command (using the Kolibri virtualenv):

pip install -r requirements/postgres.txt

• Clone the Velox repository
• Set up the virtualenv
• Activate the virtualenv
• run pip install -r requirements.txt to retrieve dependencies

Velox library consist of two main scripts which can be used independently:

• bootstrap.py - used to prepare the channels data for the actual tests
• velox.py - used to run tests

velox.py implicitly calls bootstrap.py with all the necessary arguments, so there may be not be a need to call bootstrap.py independently, but nevertheless, it is possible.

You can run python src/velox.py -h or python src/bootstrap.py -h to see more info on the available command line options.

It should be possible to run velox script simply by:

python src/velox.py

as there are certain internal defaults defined.

To specify which testing scenario to run, -t or --test command line option can be used, e.g.:

python src/velox.py -t multiple_clients_multiple_resources

If no tests are specified as command line options, all tests within the scenarios directory will be ran.

All testing scenarios should be located within the scenarios directory. Each test scenario should be a python file and at minimum contain a function with the following syntax:

def run(base_url='http://kolibridemo.learningequality.org', learners=3):
    pass

Once ran, velox will try to load this module and execute its run function to launch the tests.

Command line arguments have the highest precedence, but it is also possible to enable Velox settings module to be able to configure settings on a more permanent basis.

1. command line arguments
2. settings module
3. internally set defaults

To enable the settings module simply copy the settings.example.py, rename it to settings.py and configure as per your development environment settings.

{
    'kolibri_dev': '',
    'kolibri_venv': os.path.join(os.path.expanduser('~'), os.path.join('.venvs', 'kolibri')),
    'kolibri_exec': '',
    'database': 'sqlite',
    'channel': 'multiple',
    'learners': 30,
    'classrooms': 1,
    'test': 'all',
    'iterations': 3,
    'db_postgresql_name': '',
    'db_postgresql_user': '',
    'db_postgresql_password': '',
    'db_postgresql_host': '127.0.0.1'
}