Kubernetes

HPA Commands

Automatically scale applications based on metrics. Learn to configure HPA for CPU, memory, and custom metrics to handle varying load while optimizing resource usage.

3 commands

Pro Tips

Set resource requests on pods - HPA uses requests to calculate current utilization percentage.

Use 'kubectl get hpa' to see current replicas, targets, and whether scaling is working.

Configure stabilization windows to prevent rapid scale up/down oscillation during traffic spikes.

Common Mistakes

HPA requires metrics-server installed in the cluster. Check 'kubectl top pods' works first.

Don't set HPA minReplicas to 0 in production - you'll have cold start latency when traffic returns.

Commands

List HPAs

$ kubectl get hpa

List all horizontal pod autoscalers.

Describe HPA

$ kubectl describe hpa <name>

Show detailed HPA information.

Create HPA

$ kubectl autoscale deployment/<name> --min=2 --max=10 --cpu-percent=80

Create autoscaler for a deployment.