Setting up a Horizontal Pod Autoscaler (HPA)
3 minute read
Here are the steps involved in provisioning a Kubernetes cluster and setting up Horizontal Pod Autoscaling (HPA) with some testing and cleanup steps:
1) Provision a Cluster:
First, provision a Kubernetes cluster.
2) Deploy the Application:
deployment.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
name: hpa-demo-deployment
spec:
selector:
matchLabels:
run: hpa-demo-deployment
replicas: 1
template:
metadata:
labels:
run: hpa-demo-deployment
spec:
containers:
- name: hpa-demo-deployment
image: k8s.gcr.io/hpa-example
ports:
- containerPort: 80
resources:
limits:
cpu: 500m
requests:
cpu: 200m
Use kubectl apply -f deployment.yaml to create the deployment.
3) Create a Service:
service.yaml
apiVersion: v1
kind: Service
metadata:
name: hpa-demo-deployment
labels:
run: hpa-demo-deployment
spec:
ports:
- port: 80
selector:
run: hpa-demo-deployment
Use kubectl apply -f service.yaml to create the service.
4) Configure HPA:
hpa.yaml
apiVersion: autoscaling/v1
kind: HorizontalPodAutoscaler
metadata:
name: hpa-demo-deployment
spec:
scaleTargetRef:
apiVersion: apps/v1
kind: Deployment
name: hpa-demo-deployment
minReplicas: 1
maxReplicas: 10
targetCPUUtilizationPercentage: 50
Use kubectl apply -f hpa.yaml to create the Horizontal Pod Autoscaler.
5) Testing:
You can test the HPA by simulating load using various methods. The provided example uses kubectl run to simulate load:
kubectl run -i --tty load-simulation --rm --image=busybox --restart=Never -- /bin/sh -c "while sleep 0.01; do wget -q -O- http://hpa-demo-deployment; done"
6) Monitoring:
Monitor the HPA using kubectl and other monitoring tools to observe how it reacts to changes in load.
# after 30 seconds
$ kubectl -n default get pods && kubectl -n default get horizontalpodautoscaler
NAME READY STATUS RESTARTS AGE
hpa-demo-deployment-85bff8877-5b49k 1/1 Running 0 28s
hpa-demo-deployment-85bff8877-98qdq 1/1 Running 0 58s
hpa-demo-deployment-85bff8877-9z9p9 1/1 Running 0 2m51s
hpa-demo-deployment-85bff8877-sfdjh 1/1 Running 0 58s
hpa-demo-deployment-85bff8877-vtn48 1/1 Running 0 58s
load-simulation 1/1 Running 0 2m24s
NAME REFERENCE TARGETS MINPODS MAXPODS REPLICAS AGE
hpa-demo-deployment Deployment/hpa-demo-deployment 241%/50% 1 10 4 24m
# after 3 Minutes
$ kubectl -n default get pods && kubectl -n default get horizontalpodautoscaler
NAME READY STATUS RESTARTS AGE
hpa-demo-deployment-85bff8877-5b49k 1/1 Running 0 3m29s
hpa-demo-deployment-85bff8877-98qdq 1/1 Running 0 3m59s
hpa-demo-deployment-85bff8877-9z9p9 1/1 Running 0 6m42s
hpa-demo-deployment-85bff8877-gmvr2 1/1 Running 0 58s
hpa-demo-deployment-85bff8877-sfdjh 1/1 Running 0 3m59s
hpa-demo-deployment-85bff8877-vtn48 1/1 Running 0 3m59s
load-simulation 1/1 Running 0 5m25s
NAME REFERENCE TARGETS MINPODS MAXPODS REPLICAS AGE
hpa-demo-deployment Deployment/hpa-demo-deployment 56%/50% 1 10 6 27m
# after 5 Minutes
$ kubectl -n default get pods && kubectl -n default get horizontalpodautoscaler
NAME READY STATUS RESTARTS AGE
hpa-demo-deployment-85bff8877-5b49k 1/1 Running 0 5m48s
hpa-demo-deployment-85bff8877-98qdq 1/1 Running 0 6m18s
hpa-demo-deployment-85bff8877-9z9p9 1/1 Running 0 8m12s
hpa-demo-deployment-85bff8877-gmvr2 1/1 Running 0 3m17s
hpa-demo-deployment-85bff8877-sfdjh 1/1 Running. 0 6m18s
hpa-demo-deployment-85bff8877-vtn48 1/1 Running 0 6m18s
load-simulation 1/1 Running 0 7m44s
NAME REFERENCE TARGETS MINPODS MAXPODS REPLICAS AGE
hpa-demo-deployment Deployment/hpa-demo-deployment 47%/50% 1 10 6 30m
7) Remove Load:
To remove the load simulator, you can delete the corresponding pod:
kubectl delete pod load-simulation
8) Cleanup:
Finally, clean up the resources by deleting the Horizontal Pod Autoscaler, deployment, and service:
kubectl delete -f hpa.yaml
kubectl delete -f deployment.yaml
kubectl delete -f service.yaml
These steps provide an example of how to set up and test HPA in a Kubernetes cluster and then clean up the resources when done.