Introduction

The deployment files have been tested on a Minikube cluster (for the moment). Other users have reported successful tests on rancher RKE, RKE2, OVH managed K8S, Kubernetes 1.20, 1.24...

Quick start on a Minikube instance

For instructions on how to install Minikube, please follow https://minikube.sigs.k8s.io/docs/start/.

• Start your cluster

minikube start

• Create a secret for the credentials to use to connect to PostgreSQL

kubectl create secret generic pg-credentials \
    --from-literal=POSTGRES_USER='stellio' \
    --from-literal=POSTGRES_PASS='stellio_password'

• Register the config maps

kubectl apply -f deployment/stellio-configmap.yaml \
    -f deployment/kafka-configmap.yaml

• Create the persistent volumes (in a production setup, they have been created beforehand by an admin)

kubectl apply -f deployment/stellio-pv.yaml

• Start all pods

kubectl apply -f deployment/api-gateway-deployment.yaml \
-f deployment/kafka-deployment.yaml \
-f deployment/postgres-deployment.yaml \
-f deployment/search-service-deployment.yaml \
-f deployment/subscription-service-deployment.yaml

• Start ingress

kubectl apply -f deployment/stellio-ingress.yaml # started in ingress-nginx namespace

• Enable and configure Ingress addon for Minikube

minikube addons enable ingress
minikube tunnel

• Check you can access Stellio with a sample HTTP request

http http://localhost/ngsi-ld/v1/entities?type=Device

• Add a json-ld context server using kubectl -k => kustomize

kubectl apply -k

• To follow the status of the pods

kubectl get pods

• To see the logs of a pod

kubectl logs -f <pod_name>