← Back to guidesGuide

Practical golden signals for APIs on Kubernetes without inflating the stack

Define metrics, dashboards and realistic response actions for latency, traffic, errors and saturation for HTTP services on Kubernetes

A technical, actionable guide to map golden signals to Prometheus metrics and PromQL, build Grafana panels, create alerts and follow a reproducible troubleshooting flow for Kubernetes APIs without adding unnecessary agents.

Created: April 15, 2026

Published: April 15, 2026

Estimated time45 min

LevelIntermediate

Before you startAccess to a Kubernetes cluster (kubeconfig with read permissions)

PlatformsLinux / Docker

WhatsApp X LinkedIn

Linux

Commands for local inspection of metrics and logs from a workstation with kubectl; includes PromQL examples and validation steps.

kubectl configured and connected to the clustercurl installedjq (optional) for JSON parsing

List API server pods

kubectl get pods -n kube-system -l component=kube-apiserver -o wide

View recent kube-apiserver logs (read-only)

kubectl logs -n kube-system -l component=kube-apiserver --tail=200

Port-forward the first apiserver pod to query /metrics locally

kubectl -n kube-system port-forward $(kubectl -n kube-system get pod -l component=kube-apiserver -o name | head -n1) 8001:8080 &
# then: curl -s http://127.0.0.1:8001/metrics | head -n 40

Check non-destructive GET to the API

kubectl proxy --port=8001 &
# then locally:
curl -sS http://127.0.0.1:8001/api/v1/namespaces/default/pods | jq '.items | length'

Query Prometheus (example) for request rate

curl -s 'http://PROMETHEUS_HOST:9090/api/v1/query?query=sum(rate(apiserver_request_total[1m]))' | jq

Also available inGolden signals prácticos para APIs en Kubernetes sin inflar el stack

Content locked

This guide requires both steps before full content is available.

○Click “Like” on this guide.
○Share on WhatsApp, X, LinkedIn, or copy the link.

Access is automatically unlocked as soon as both steps are completed.