This post shows how to autoscale your pods based on cpu metrics.
Hope you get some help!.
I used to below jmeter site to stress wordpress pod(two) to increase.
https://jmeter.apache.org/usermanual/build-web-test-plan.html
1. Two wordpress pod.
oyj@Workstation-oyj-X555QG ~/apache-jmeter-5.1.1/bin$kubectl get po | grep wordpress
wordpress-6c7f4d4874-bd257 1/1 Running 1 2d21h
wordpress-6c7f4d4874-kxj5s 1/1 Running 1 2d21h
oyj@Workstation-oyj-X555QG ~/apache-jmeter-5.1.1/bin$
2.metrics to autoscale.(https://wnapdlf.blogspot.com/2019/07/ubuntu18kvmvagrantkubernetes-10metrics.html)
oyj@Workstation-oyj-X555QG ~/apache-jmeter-5.1.1/bin$kb get po -n kube-system | grep metrics
metrics-server-75cb7fd5d7-tq7l8 1/1 Running 14 10d
3.wordpress deploy yaml (resource is important,it resource no exists, then no autoscale is possible)
oyj@Workstation-oyj-X555QG ~/u18kvk8s/k8s/wordpress$cat wp-dp.yaml
---
apiVersion: apps/v1 # for versions before 1.9.0 use apps/v1beta2
kind: Deployment
metadata:
name: wordpress
labels:
app: wordpress
spec:
replicas: 2
selector:
matchLabels:
app: wordpress
strategy:
type: Recreate
template:
metadata:
labels:
app: wordpress
spec:
containers:
- image: wordpress:4.8-apache
name: wordpress
resources:
requests:
cpu: 100m
env:
- name: WORDPRESS_DB_HOST
#Previsous mariadb-master svc, I already created.!
value: mariadb-master
- name: WORDPRESS_DB_PASSWORD
valueFrom:
secretKeyRef:
name: mariadb-pass
key: password
ports:
- containerPort: 80
name: wordpress
volumeMounts:
- name: wordpress-persistent-storage
mountPath: /var/www/html
volumes:
- name: wordpress-persistent-storage
persistentVolumeClaim:
claimName: storage-claim-for-wp
#https://wnapdlf.blogspot.com/2019/07/ubuntu18kvmvagrantkubernetes-8.html
4. Create autoscale on wordpress deployment.
oyj@Workstation-oyj-X555QG ~/u18kvk8s/k8s/wordpress$kb autoscale deployment wordpress --cpu-percent=50 --min=2 --max=5
horizontalpodautoscaler.autoscaling/wordpress autoscaled
oyj@Workstation-oyj-X555QG ~/u18kvk8s/k8s/wordpress$kb get hpa
NAME REFERENCE TARGETS MINPODS MAXPODS REPLICAS AGE
wordpress Deployment/wordpress <unknown>/50% 2 5 0 4s
oyj@Workstation-oyj-X555QG ~/u18kvk8s/k8s/wordpress$kb describe hpa
Name: wordpress
Namespace: default
Labels: <none>
Annotations: <none>
CreationTimestamp: Wed, 24 Jul 2019 04:55:27 +0900
Reference: Deployment/wordpress
Metrics: ( current / target )
resource cpu on pods (as a percentage of request): 1% (1m) / 50%
Min replicas: 2
Max replicas: 5
Deployment pods: 2 current / 2 desired
Conditions:
Type Status Reason Message
---- ------ ------ -------
AbleToScale True ScaleDownStabilized recent recommendations were higher than current one, applying the highest recent recommendation
ScalingActive True ValidMetricFound the HPA was able to successfully calculate a replica count from cpu resource utilization (percentage of request)
ScalingLimited False DesiredWithinRange the desired count is within the acceptable range
Events: <none>
oyj@Workstation-oyj-X555QG ~/u18kvk8s/k8s/wordpress$kb get hpa
NAME REFERENCE TARGETS MINPODS MAXPODS REPLICAS AGE
wordpress Deployment/wordpress 1%/50% 2 5 2 27s
5. jmeter setting
oyj@Workstation-oyj-X555QG ~/u18kvk8s/k8s/wordpress$watch kubectl get hpa
oyj@Workstation-oyj-X555QG ~/u18kvk8s/k8s/wordpress$kubectl get po -l app=wordpress
Let's stress like below clip.
As a result pod and cpu usage is increasing.
As cpu usage goes down, then pods will be come back to original value.(2).
According to "https://kubernetes.io/docs/tasks/run-application/horizontal-pod-autoscale-walkthrough/"
"Roughly speaking, HPA will increase and decrease the number of replicas (via the deployment) to maintain an average CPU utilization across all Pods of 50% (since each pod requests 200 milli-cores by kubectl run, this means average CPU usage of 100 milli-cores"
Thanks for reading.
