Helm chart for deploying Prometheus and Grafana with full built in Kubernetes pod and infrastructure dashboard
- Kubernetes cluster up and running
- Helm
- Kubectl
You will need to install helm and ensure it has sufficient deployment access. See helm-rbac.yaml
Please note that this will create a helm service account with admin rights.
You might want to reduce the scope of helm account based on your security requirements.
kubectl apply -f helm-rbac.yaml
helm init --service-account helm
Note: for existing helm installs, use --upgrade
to add new service account
Make sure you have kubernetes up and running and a kubectl
context pointing to your target cluster.
cd helm
kubectl create namespace monitoring
helm install --name kube-prometheus prometheus -f ../values.yaml --namespace monitoring
helm install --name kube-grafana grafana -f ../values.yaml --namespace monitoring
kubectl get pods -n monitoring
Once everything is running, you can port-forward to check grafana:
kubectl port-forward --namespace=monitoring $(kubectl get pods --namespace=monitoring --selector=app=grafana --output=jsonpath='{.items[*].metadata.name}') 3000:3000