Clean up kube-state-metrics noise so your dashboards mean something again

How to cut useless series, adapt to EndpointSlices, and separate state from usage without blinding the team.

kube-state-metrics is still valuable, but in 2026 it exposes more surface area, more stable metrics, and recent defaults such as EndpointSlices. If your dashboards filled up with irrelevant series, fragile joins, or duplicated states, this guide shows how to reduce noise at the source, fix your queries, and validate that the cleanup does not break alerts or troubleshooting.

Created: April 20, 2026

Published: April 20, 2026

Estimated time38 min

LevelIntermediate

Before you startRead access to Kubernetes and to Prometheus or kube-prometheus configuration

PlatformsLinux / Docker

WhatsApp X LinkedIn

Docker

Useful for validating queries and scrape reductions from a workstation or lab where Prometheus and kube-state-metrics endpoints are already exposed.

DockerReachable HTTP endpoints for Prometheus or kube-state-metrics

Query the Prometheus API for kube-state-metrics scrape volume

docker run --rm --network host curlimages/curl:8.7.1 -sG 'http://127.0.0.1:9090/api/v1/query' --data-urlencode 'query=sum(scrape_samples_post_metric_relabeling{job="kube-state-metrics"})'

Inspect a filtered kube-state-metrics endpoint response

docker run --rm --network host curlimages/curl:8.7.1 -s 'http://127.0.0.1:8080/metrics?resources=pods,deployments,statefulsets' | head -40

Compare the noisiest metric families by series count

docker run --rm --network host curlimages/curl:8.7.1 -sG 'http://127.0.0.1:9090/api/v1/query' --data-urlencode 'query=topk(20,count by (__name__) ({job="kube-state-metrics"}))'

Also available inLimpiar ruido de kube-state-metrics para que tus dashboards vuelvan a decir algo

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.