Public | Automated Build

Last pushed: 2 months ago
Short Description
An increasingly intelligent method to test RH Satellite.
Full Description


An increasingly intelligent method to test RH Satellite.


pip install .
python install

Rizza will add a 'rizza' directory to your home directory, as well as a config directory with example config.
After installation copy/rename the example configuration, located in ~/rizza/config/ to ~/rizza/config/rizza.yaml


rizza [-h] {brute,genetic,config,list,test}

Brute Force Testing

Rizza's most basic, and time consuming, operation is a brute force method of testing entities. It will try every combination of an entity's methods, fields, arguments, and available input methods.
It is highly recommended that you limit the scope of this kind of test with --max-field, --max-inputs, and the exclude options available. An unlimited test can easily generate trillions of combinations and will likely take longer than the lifecycle of your product.
-i, --input and -o, --output use relative paths only.
Note: Using --async will load all tests into memory, so it is advised to limit the numbers to tests by either breaking them up into separate files, or limiting the scope of the methods and fields.

rizza brute --help

rizza brute -e Product -o tester.txt --max-fields 2 --max-inputs 1 --method-exclude raw search read get payload --async

rizza brute -i 10tests.txt -l stdout

Genetic Algorithm-Based Testing

Rizza is able to test entities (via their methods) using genetic algorithms to evolve toward a positive or negative goal. You can adjust the scoring criteria in config/rizza.yaml.
Rizza, by default, will recursively try to create entities it both does and doesn't know how to, in order to resolve dependencies. You can limit or turn this off both in the config or at run-time with cli args. Note that this recursive process adds a significant amount of time.
Once a test completes, it is saved in ~/rizza/data/genetic_tests/

rizza genetic --help

rizza genetic -e Organization -m create

rizza genetic -e Organization -m create --max-generations 100 --seek-bad --fresh --disable-recursion --async


Rizza's main configuration file is located in config/rizza.yaml. After cloning, you will need to copy the example file to rizza.yaml. Most of these configurations have CLI overrides, and there are even some limited support for environment variables (SATHOST, SATUSER, SATPASS, CONFILE).
Additionally, there is support for modifying some configuration options through rizza's cli.

rizza config --help

rizza config nailgun -p demo

rizza config nailgun --show


Rizza can tell you all the information it knows about your product plugin using the list command. These results are currently unfiltered, so will show anything that meets rizza's criteria for what constitutes an entity.

rizza list entities

rizza list methods -e Organization


Rizza is also able to test itself, using pytest. This is mainly useful for testing your container images, to verify everything is working before you begin using rizza.
You can even pass in pytest args by adding the --args flag.
rizza test

rizza test --args=-v


Rizza is also available with automatic builds on dockerhub.
You can either pull down the latest, or a specific released version.
Additionally, you can build your own image locally. You will want to mount the local rizza directory to provide your configuration and keep any data rizza creates.

docker build -t rizza .
docker pull jacobcallahan/rizza

docker run -it rizza brute --help

docker run -it -v $(pwd):/root/rizza/:Z rizza brute -e Product -o docker.txt --max-fields 2 --max-inputs 1 --method-exclude raw search read get payload

docker run --rm -v $(pwd):/root/rizza/:Z jacobcallahan/rizza genetic -e Organization


This project only explicitly supports python 3.6+

Docker Pull Command
Source Repository