Completek8s Classroomnotes 05/Oct/2023

AKS Contd

Scaling in AKS

• Lets experiment Scaling in AKS
• AKS cluster by default comes with metrics server in it
• We can scale manually by using kubectl scale to scale automatically have two options
  • HPA (Horizontal pod autoscaler)
  • VPA (Vertical Pod autoscaler)
• For this experiment we will be using a deployment with stress side car container which puts load

apiVersion: apps/v1
kind: Deployment
metadata:
  name: azure-vote-back
spec:
  replicas: 1
  selector:
    matchLabels:
      app: azure-vote-back
  template:
    metadata:
      labels:
        app: azure-vote-back
    spec:
      nodeSelector:
        "kubernetes.io/os": linux
      containers:
      - name: azure-vote-back
        image: mcr.microsoft.com/oss/bitnami/redis:6.0.8
        env:
        - name: ALLOW_EMPTY_PASSWORD
          value: "yes"
        resources:
          requests:
            cpu: 100m
            memory: 128Mi
          limits:
            cpu: 250m
            memory: 256Mi
        ports:
        - containerPort: 6379
          name: redis
---
apiVersion: v1
kind: Service
metadata:
  name: azure-vote-back
spec:
  ports:
  - port: 6379
  selector:
    app: azure-vote-back
---
apiVersion: apps/v1
kind: Deployment
metadata:
  name: azure-vote-front
spec:
  replicas: 2
  selector:
    matchLabels:
      app: azure-vote-front
  template:
    metadata:
      labels:
        app: azure-vote-front
    spec:
      nodeSelector:
        "kubernetes.io/os": linux
      containers:
      - name: azure-vote-front
        image: mcr.microsoft.com/azuredocs/azure-vote-front:v1
        resources:
          requests:
            cpu: 100m
            memory: 128Mi
          limits:
            cpu: 250m
            memory: 256Mi
        ports:
        - containerPort: 80
        env:
        - name: REDIS
          value: "azure-vote-back"
      - name: stress
        image: alpine
        resources:
          requests:
            cpu: 100m
            memory: 128Mi
        command:
          - apk update && apk add stress-ng && stress-ng --cpu 4 --vm 2 --hdd 1 --fork 8 --timeout 100m --metrics
---
apiVersion: v1
kind: Service
metadata:
  name: azure-vote-front
spec:
  type: LoadBalancer
  ports:
  - port: 80
  selector:
    app: azure-vote-front

• Refer Here for HPA docs
• We have created hpa

---
apiVersion: autoscaling/v2
kind: HorizontalPodAutoscaler
metadata:
  name: voting-hpa
spec:
  scaleTargetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: azure-vote-front
  maxReplicas: 10
  metrics:
    - type: Resource
      resource:
        name: cpu
        target:
          type: Utilization
          averageUtilization: 60

• The current state is we have enough nodes to meet the demand.
  
• this is where cluster autoscaler Refer Here

az aks nodepool update --enable-cluster-autoscaler -g myResourceGroup --cluster-name myAKSCluster -n nodepool1 --min-count 1 --max-count 3

• Now have a look at nodes
  
• use the following images in the stress container

venuchowgani/stress:latest
necindia/test-hpa:v1

• Refer Here for vertical pod autoscaler

Need to provide access for Azure AD (Microsoft Entra) user to access kubernetes

• Follow the video for setup
• Custom cluster role for a group

---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRole
metadata:
  name: viewers
rules:
  - apiGroups:
      - ""
    resources:
      - "pods"
    verbs:
      - "get"
      - "watch"
      - "list"
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
  name: viewersbinding
roleRef:
  kind: ClusterRole
  name: viewers
  apiGroup: rbac.authorization.k8s.io
subjects:
  - apiGroup: rbac.authorization.k8s.io
    kind: Group
    name: "711bf260-fd71-469f-801f-075a0da619f3"