QUICK-INSTALLATION.md

Installation

In this document we show how to set up a test/minimal cluster instance. For more details, please refer to the full Installation.

1. Kubernetes

Install a light weight cluster by k3s (minimum 20GB of storage).

 curl -sfL https://get.k3s.io | sh -

 sudo chmod +r /etc/rancher/k3s/k3s.yaml
 export KUBECONFIG=/etc/rancher/k3s/k3s.yaml 

 sudo apt install open-iscsi

2. Helm 3

 curl https://get.helm.sh/helm-v3.1.2-linux-amd64.tar.gz | tar xfz -
 sudo mv linux-amd64/helm /usr/local/bin/helm

3. Cloning project

Clone the project repository and cd into it:

 git clone https://github.com/Smile-SA/rating-operator.git

4. Storage provider

Install Longhorn

git clone https://github.com/longhorn/longhorn ./quickstart/longhorn/longhorn

kubectl apply -f ./quickstart/longhorn/longhorn/deploy/longhorn.yaml

Wait a minute, then verify everything is working as expected by running:

 kubectl get pods -n longhorn-system

5. Prometheus

 helm repo add prometheus-community https://prometheus-community.github.io/helm-charts

 helm repo update

./quickstart/prometheus/install.sh

Wait a minute, then verify everything is working as expected by running:

 kubectl get pods -n monitoring

6. Rating Operator

6.1 Installing as an operator

 RATING_NAMESPACE=rating hack/install.sh

To check if everything is running correctly:

 kubectl -n rating get pods

6.2. Accessing rating operator

While inside the rating operator repo, and inside the rating namespace, run:

 sudo kubectl config set-context --current --namespace=rating

Then we can access rating operator components:

Rating-api

 ./hack/forward-api

for more details about the API, please consult API endpoints

Prometheus

 ./hack/forward-prometheus

Grafana

 ./hack/forward-grafana