Skip to main content

Installation

Deploy the PlantD Operator​

The easiest way to set up operator is to use the bundle.yaml deployments.

# Install the K6 Operator
curl https://raw.githubusercontent.com/grafana/k6-operator/main/bundle.yaml | kubectl create -f -

# Install the Prometheus Operator
curl https://raw.githubusercontent.com/prometheus-operator/prometheus-operator/main/bundle.yaml | kubectl create -f -

# Install the PlantD Operator
curl https://raw.githubusercontent.com/CarnegieMellon-PlantD/PlantD-operator/main/bundle.yaml | kubectl create -f - 

# Get the Studio service hostname
kubectl get svc plantd-studio -n plantd-operator-system -o jsonpath='{.status.loadBalancer.ingress[0].hostname}'

Enable Thanos (Metrics Persistence)​

Prometheus metrics storage is ephemeral. To enable long-term metrics persistence, PlantD uses Thanos to store and retrieve metrics from an S3 Bucket.

Follow the listed steps to enable Thanos,

  1. Generate a set of AWS Credentials (AWS Access Key and AWS Secret Access Key Id) in AWS

  2. Execute the script to setup the Thanos infrastructure

    python3 setup_thanos_storage.py <bucket_name> <access_key> <secret_key>
  1. Set thanosEnabled: true in plantdcore CR spec and kubectl apply

If all steps are successful, a Thanos sidecar, querier and store should be deployed in the K8 cluster and metrics will be queried from Thanos.

Note​

  • It may take up to 2-3 minutes for the PlantD Studio to be available at the above hostname.
  • If running locally in minikube, services of type LoadBalancer can't be assigned a public IP address directly as they would in a cloud environment.
    • Use this command to expose the PlantD service frontend endpoint - minikube service plantd-studio-service -n plantd-operator-system.

Run Experiment with Test Pipeline (Optional)​

If you have deployed the test pipeline, you can use the YAML files under the plantd_yamls directory to setup an experiment.