// the find
dotdc/grafana-dashboards-kubernetes
A set of modern Grafana dashboards for Kubernetes.
A collection of 8 Grafana dashboards covering Kubernetes from the global cluster view down to individual pods, plus add-ons like Prometheus and Trivy. Requires kube-state-metrics and prometheus-node-exporter. Targets anyone running kube-prometheus-stack who wants something that looks better and uses modern Grafana panel types than the default mixin dashboards.
Uses $__rate_interval correctly instead of hardcoded intervals, which is a common mistake in community dashboards that causes misleading graphs at non-default scrape intervals. Supports every sane deployment path: manual import, Helm values, ArgoCD, ConfigMaps, Terraform via kustomize, and Grafana Operator — no excuses for not fitting it into your stack. The nodename label mismatch is documented with concrete relabeling fixes for Prometheus, ServiceMonitor, and Grafana Alloy rather than just telling you to figure it out. Versioned with semver and automated releases via conventional commits, so you can pin a version and know what changed.
No Windows node support and the attempt to add it broke existing panels — this is a real gap if your cluster runs mixed workloads. The node IP address change issue on k8s-views-nodes is acknowledged with no fix, just a shrug; if your nodes are ephemeral or you run a cloud provider that recycles IPs aggressively, you will hit this. All dashboards are plain JSON files with no templating or parameterization, so customizing queries (e.g. adding a custom label filter) means forking and manually editing large JSON blobs that are painful to diff. No alerting rules included — you get visibility but still need to wire up your own alerts separately.