---
title: "Get cluster metrics"
method: GET
path: "/api/v1/k8s/clusters/{cluster_id}/metrics"
tags: ["Kubernetes"]
---

# Get cluster metrics

`GET /api/v1/k8s/clusters/{cluster_id}/metrics`

Live utilization (CPU, memory, pods, per-node breakdown) plus a 24-hour control-plane health summary. Values refresh every 60 seconds.

## Path parameters

- `cluster_id` string, required

## Headers

- `X-Project-ID` string, uuid, required

## Response `200`

Cluster metrics

- object
  - `success` boolean
  - `cpu_used_millicores` integer
  - `cpu_capacity_millicores` integer
  - `mem_used_bytes` integer
  - `mem_capacity_bytes` integer
  - `pod_count` integer
  - `pod_capacity` integer
  - `nodes_ready` integer
  - `nodes_total` integer
  - `nodes` K8sNodeMetrics[]
    - `name` string — Kubernetes node name
    - `internal_ip` string
    - `roles` string
    - `ready` boolean
    - `cpu_used_millicores` integer
    - `cpu_capacity_millicores` integer
    - `mem_used_bytes` integer
    - `mem_capacity_bytes` integer
  - `control_plane` K8sControlPlaneHealth — 24-hour control-plane health summary, sampled every minute. Absent until the first sample after the cluster starts running.
    - `sampled_at` string, date-time
    - `etcd_p99_ms` number, double — 99th-percentile etcd request latency over the last minute (healthy clusters sit under 15 ms; sustained values above 100 ms degrade the control plane)
    - `etcd_p99_ms_max_24h` number, double
    - `apiserver_5xx_24h` integer
    - `restarts_apiserver_24h` integer
    - `restarts_etcd_24h` integer
    - `restarts_scheduler_24h` integer
    - `restarts_controller_24h` integer
    - `restarts_cilium_op_24h` integer
    - `leader_changes_24h` integer — Scheduler + controller-manager restarts — lost leader leases
    - `not_ready_minutes_24h` integer
    - `nodes_not_ready` integer

## Other responses

- `404` — Resource not found
- `503` — Cluster unreachable

## Changes

- **2026-08-23** `72c803a128b2` — 1 info
  - endpoint added

[Change history](https://skmtc.dev/rafftechnologies/apis/raff-api/changes/api/v1/k8s/clusters/:cluster_id/metrics/get.md)

---

[API](https://skmtc.dev/rafftechnologies/apis/raff-api.md) · [All operations](https://skmtc.dev/rafftechnologies/apis/raff-api/llms.txt) · [OpenAPI document](https://skmtc-service-production.skmtc.workers.dev/v1/apis/rafftechnologies/raff-api/revisions/2f4df56be77c/schema)