Grafana Labs has released version 4 of its Kubernetes Monitoring Helm chart, marking the most significant update since its launch. Key changes include converting destinations and collectors from lists to maps for stable, order-independent referencing — solving silent misconfiguration bugs in GitOps workflows with Argo CD, Terraform, and Flux. Backing services like Node Exporter and kube-state-metrics are now deployed explicitly via a new `telemetryServices` key, eliminating surprise duplicate deployments. Cluster metrics are split into three separate features (`clusterMetrics`, `hostMetrics`, `costMetrics`), and the pod log pipeline now requires explicit label promotion instead of bulk allocation, reducing memory usage in Alloy. A migration tool is available to convert v3 values files to v4 format.
Sort: