Sign In

Curriculum 10: Working with Namespaces

Resource Quotas

15 min · 15 XP

Resource Quotas

Resource quotas prevent any single namespace from consuming more than its fair share of cluster resources. They enforce hard limits on compute usage and object counts.

What Is a ResourceQuota?

A ResourceQuota is a Kubernetes object that constrains the total resources a namespace can consume. Once a quota is in place, requests that would exceed it are rejected by the API server.

Setting Compute Quotas

Compute quotas limit CPU and memory across all pods in a namespace:

apiVersion: v1
kind: ResourceQuota
metadata:
  name: compute-limits
  namespace: development
spec:
  hard:
    requests.cpu: "4"
    requests.memory: 8Gi
    limits.cpu: "8"
    limits.memory: 16Gi

kubectl apply -f compute-quota.yaml

With this quota active, every pod in the development namespace must declare resource requests and limits, or the API server will reject it.

Setting Object Count Quotas

Object count quotas cap the number of resources that can exist in a namespace:

apiVersion: v1
kind: ResourceQuota
metadata:
  name: object-counts
  namespace: development
spec:
  hard:
    pods: "20"
    services: "10"
    configmaps: "30"
    secrets: "30"
    persistentvolumeclaims: "5"

This prevents runaway resource creation from filling up a namespace.

Inspecting Quotas

Check what quotas are in place and how much has been consumed:

# List quotas in a namespace
kubectl get resourcequota -n development

# See detailed usage vs limits
kubectl describe resourcequota compute-limits -n development

The output shows Used versus Hard for each constrained resource, making it easy to spot when a namespace is approaching its limits.

Combining Multiple Quotas

You can apply multiple ResourceQuota objects to the same namespace. Kubernetes enforces all of them independently -- a request must satisfy every quota to succeed.

Key Takeaways

  • ResourceQuotas enforce hard limits per namespace
  • Compute quotas constrain total CPU and memory requests/limits
  • Object quotas cap the number of pods, services, secrets, and other resources
  • Pods must specify resource requests and limits when a compute quota is active
  • Use kubectl describe resourcequota to monitor usage against limits