Python implementation of CAUS
You can install everything using pip install -r requirements.txt (either inside a virtual env or globally).
Alternatively, you can install the needed libraries manually:
- Kuberentes Python Client
- homepage: https://github.com/kubernetes-client/python
pip install kubernetes
- Prometheus Api Client
- homepage: https://pypi.org/project/prometheus-api-client/
pip install prometheus_api_client.
Prometheus for Kubernetes. Metrics must be accessible from inside the cluster.
Kafka for Kubernetes
Adapt the config.ini config file to your needs.
Alternatively, you can set a custom path for your config file by setting the CAUS_CONFIG environment variable.
Run by calling python controller.py.
The caus-deployment.yaml has to be altered.
$imagename has to be replaced with the official name of the image you generate with the Dockerfile.
We used Dockerhub for saving the generated image and access it with the kubernetes deployment.
CAUS can then be applied using kubectl apply -f caus-deployment.yaml.
The CAUS currently uses black as its formatter.
If you followed the requirements above, it will already be downloaded and can be called using black *.py.
For docstrings, the google style is used.