Versioned Online Boutique Application achieved through SADP methodology.

Four different variants are available:

• Normal High Performance
• Normal Low Power
• Basic High Performance
• Basic Low Power

Those variants can be deployed on a kubernetes cluster using the respective shell scripts contained in the release folder.

Loadgenerator10/50/100 generated a respective load of users into the application deployed.

Cluster setup on GCE (Google Compute Engine):

• Enter the Kubernetes Engine section
• Create cluster
• GKE Standard
• Cluster basics: name, region (eu-westx-x), release channel (default)
• Node Pools: name, number of nodes (3), Nodes -> image type (Container-Optimized OS with containerd (cos_containerd))
• Series -> N1, Machine Type -> n1-standard-2
• Networking -> Enable HTTP load balancing
• Features -> Enable Cloud Logging (System and Workloads), Enable Cloud Monitoring (System and Workloads)

Cloud Storage Bucket:

• Create bucket
• name, region (europe-west-x), standard, uniform, data encryption (none)

Kubernetes setup after cluster is created:

• Enter the Kubernentes Engine section
• Click connect on the right side of the cluster information
• Authorize console, and press enter to get the credentials for the cluster
• Istio setup:
  • curl -L https://istio.io/downloadIstio | sh -
  • cd istio-1.13.3
  • export PATH=$PWD/bin:$PATH
  • istioctl install --set profile=demo -y
  • kubectl label namespace default istio-injection=enabled
  • kubectl apply -f https://raw.githubusercontent.com/istio/istio/release-1.13/samples/addons/prometheus.yaml
  • kubectl apply -f https://raw.githubusercontent.com/istio/istio/release-1.13/samples/addons/grafana.yaml
• go back into root folder with cd ..
• open a second terminal (Google Cloud Shell) with the + button
• in the new terminal execute 'istioctl dashboard prometheus'
• it should open prometheus on port 9090
• go back into the first terminal
• boutique SADP download:
  • git clone https://github.com/yensky95/boutique-basic.git
• cd boutique-basic/release
• launch one of the three loadgenerator variants with:
  • kubectl apply -f loadgenerator10/50/100.yaml
    
• launch the autodeploy decision system:
  • bash autodeploy.sh

Docker images build:

• open a terminal inside the path of the microservice folder we want to build (e.g. boutique-basic/src/recommendationservice)
• execute the docker build command:
  • docker build -t dockerhub_username/desired_name .
• this will create a local image in docker
• push this image to dockerhub (connect docker with dockerhub) with the button in the GUI or with command line

YAML files image change:

• open the manifest file
• search for the desired microservice for which we desire to change image
• e.g. below
  

apiVersion: apps/v1
kind: Deployment
metadata:
  name: loadgenerator
spec:
  selector:
    matchLabels:
      app: loadgenerator
  replicas: 1
  template:
    metadata:
      labels:
        app: loadgenerator
      annotations:
        sidecar.istio.io/rewriteAppHTTPProbers: "true"
    spec:
      serviceAccountName: default
      terminationGracePeriodSeconds: 5
      restartPolicy: Always
      initContainers:
      - command:
        - /bin/sh
        - -exc
        - |
          echo "Init container pinging frontend: ${FRONTEND_ADDR}..."
          STATUSCODE=$(wget --server-response http://${FRONTEND_ADDR} 2>&1 | awk '/^  HTTP/{print $2}')
          if test $STATUSCODE -ne 200; then
              echo "Error: Could not reach frontend - Status code: ${STATUSCODE}"
              exit 1
          fi
        name: frontend-check
        image: busybox:latest
        env:
        - name: FRONTEND_ADDR
          value: "frontend:80"
      containers:
      - name: main
        image: yensky/loadgeneratortenusers:latest  <------ HERE (always put dockerhub_username/name:version_tag)
        env:
        - name: FRONTEND_ADDR
          value: "frontend:80"
        - name: USERS
          value: "10"
        resources:
          requests:
            cpu: 300m
            memory: 256Mi
          limits:
            cpu: 500m
            memory: 512Mi

Transfer a file from VM to Cloud Storage bucket:

• open the Google Cloud Shell
• gsutil cp file_name.xxx gs://bucket_name
• go in the Google Cloud Storage bucket and download