kube-state-metrics
kube-state-metrics is a simple service that produces metrics for the state of objects by listening to the Kubernetes API server. Instead of the health of individual components, it focuses on the state of objects such as deployment, node, and pod. Exports metric through HTTP endpoint as raw and unmodified data.
kubernetes · Observability / Datadog Alternatives
Common commercial products in this category: Datadog, New Relic, Dynatrace
- You can listen to object states on the Kubernetes API server and generate metrics from raw data.
- You can consume the metric from your HTTP endpoint with a Prometheus compatible scraper.
- You can examine the current state of the cluster objects via metric endpoint.
- ↓ NegativeGitHub Issue
kube-state-metrics consuming too much memory
“kube-state-metrics is using >400mb of RAM. It is also very slow when I query `/metrics`. The kubernetes cluster has 2700 job objects. It seems surprising that this would consume 400mb of RAM for metrics aggregation. Below is a pprof top trace. This is running the latest git…”