kube-prometheus-stack Starter Config

kube-prometheus-stack — Skip the Config Rabbit Hole

The kube-prometheus-stack Helm chart has 1,500+ configurable values. This starter config answers the 20 questions everyone hits the first time: what storage class, how much retention, how to route alerts, which dashboards to enable out of the box.

What is Included

• Production values.yaml with annotated storage, retention, and resource settings
• Alertmanager routing config with Slack and PagerDuty receiver templates
• 8 Grafana dashboards pre-enabled: cluster overview, node exporter, pod resources, PVC usage, API server latency, etcd, CoreDNS, and Ingress
• PrometheusRule examples for common alerts (OOMKill, high CPU, disk pressure)

Developer Experience

Install the chart with one values file and get a working monitoring stack. Dashboards load on first login. Alert routing is pre-configured — just fill in your Slack webhook or PagerDuty integration key.

What You Get

• values-production.yaml and values-dev.yaml (lighter resource profile)
• alerting/ — Alertmanager config and PrometheusRule manifests
• dashboards/ — dashboard JSON for additional custom imports
• README.md covering upgrade notes and storage considerations

Requirements

• Kubernetes >= 1.22 with a StorageClass supporting ReadWriteOnce
• Helm >= 3.8
• ~4 GB RAM available in the monitoring namespace