-- Installs the helm client on the host where the script is run (if not already installed)
-- Creates a service account in the k8s cluster to be used by tiller, with sufficient permissions to be able to deploy kubernetes objects.
-- Installs the helm server part (tiller) and assigns to tiller the previously created service account (if not already installed)
-- Creates a namespace (monitoring) where all the components that are part of the OSM deployment monitoring `pack` will be installed.
-- Installs prometheus-operator using the `stable/prometheus-operator` chart which is located at the default helm repository (<https://kubernetes-charts.storage.googleapis.com/>). This installs a set of basic metrics for CPU, memory, etc. of hosts and pods. It also includes grafana and dashboards.
-- Installs an exporter for mongodb using the `stable/prometheus-mongodb-exporter` chart, which is located at the default helm repository (<https://kubernetes-charts.storage.googleapis.com/>).
-- Adds a dashboard for mongodb to grafana through a local yaml file.
-- Installs an exporter for mysql using the `stable/prometheus-mysql-exporter` chart which is located at the default helm repository (<https://kubernetes-charts.storage.googleapis.com/>).
-- Adds a dashboard for mysql to grafana through a local yaml file.
-- Installs an exporter for kafka using a custom-build helm chart with a deployment and its corresponding service and service monitor with local yaml files. We take the kafka exporter from <https://hub.docker.com/r/danielqsj/kafka-exporter>.
-- Add a dashboard for kafka to grafana through a local yaml file.
+Kubernetes cluster metrics (for nodes, pods, deployments, etc.) are stored in the dedicated Prometheus instance and accessible using Grafana.