Deploy a MEAN Web App with Google Kubernetes Engine, Portworx

In this tutorial, we will deploy and manage a Node.js web application and a MongoDB database in Google Kubernetes Engine. To achieve high availability for MongoDB, we will use a Portworx storage cluster deployed on GKE.

Mar 1st, 2019 9:12am by Janakiram MSV

Featued image for: Deploy a MEAN Web App with Google Kubernetes Engine, Portworx

In one of my previous articles, I introduced Portworx as the container-native storage platform. In this tutorial, we will deploy and manage a Node.js web application and a MongoDB database in Google Kubernetes Engine. To achieve high availability for MongoDB, we will use a Portworx storage cluster deployed on GKE.

Launching a GKE Cluster

Let’s launch a three node GKE cluster based on Ubuntu OS with an SSD-based disk of 50GB attached to each node. Replace the PROJECT with your own GCP project id.

export PROJECT=’<Your GCP Project ID>’
gcloud container --project $PROJECT clusters create "tns-demo" \
--zone "asia-south1-a" \
--username "admin" \
--cluster-version "1.11.7-gke.4" \
--machine-type "n1-standard-4" \
--image-type "UBUNTU" \
--disk-type "pd-ssd" \
--disk-size "50" \
--scopes "https://www.googleapis.com/auth/compute","https://www.googleapis.com/auth/devstorage.read_only","https://www.googleapis.com/auth/logging.write","https://www.googleapis.com/auth/monitoring","https://www.googleapis.com/auth/servicecontrol","https://www.googleapis.com/auth/service.management.readonly","https://www.googleapis.com/auth/trace.append" \
--num-nodes "3" \
--enable-cloud-logging \
--enable-cloud-monitoring \
--network "default" \
--addons HorizontalPodAutoscaling,HttpLoadBalancing,KubernetesDashboard

The below command updates kubeconfig with the credentials and endpoint of the cluster.

gcloud container clusters get-credentials tns-demo \
--zone asia-south1-a \
--project $PROJECT

Let’s add the current user to role cluster-admin

kubectl create clusterrolebinding cluster-admin-binding \
--clusterrole cluster-admin \
--user $(gcloud config get-value account)

Verify that the cluster is up and running.

Installing Portworx Storage Cluster

Portworx is installed as a DaemonSet on each node of GKE. We can install it by generating the YAML spec through an online tool. Visit the Portworx documentation page to get started. Get the version of Kubernetes with the following command. The installation tool needs to know the exact version of the distribution.

kubectl version --short | awk -Fv '/Server Version: / {print $3}'

Portworx relies on etcd to store the metadata and cluster state. For this demo, we will use the built-in etcd cluster.

Under the Storage tab, choose GKE and populate the information for the spec. We are choosing an SSD disk with 20GB as dedicated block storage for Portworx. Since we have three nodes, we will get aggregate storage of 60GB.

Choose defaults for the Network tab and click Next.

In the last tab, choose Google Kubernetes Engine and click Finish.

We are ready to install Portworx based on the generated specification. You can either download the spec or copy it.

Switch to the terminal and run the command copied from the spec generator. It will take a few minutes for Portworx cluster to get installed. Verify the cluster by checking the Portworx Pods running in kube-system namespace. All the Pods should be running.

Deploying MongoDB

We will create a StorageClass for Portworx with a replication factor of 3. This ensures that the data is redundantly available on multiple nodes.

kind: StorageClass
apiVersion: storage.k8s.io/v1beta1
metadata:
    name: px-sc
provisioner: kubernetes.io/portworx-volume
parameters:
   repl: "3"

kubectl create -f px-sc.yaml

With the StorageClass in place, we will create a Persistent Volume Control (PVC) that will be used by MongoDB Pod.

kind: PersistentVolumeClaim
apiVersion: v1
metadata:
   name: px-mongo-pvc
   annotations:
     volume.beta.kubernetes.io/storage-class: px-sc
spec:
   accessModes:
     - ReadWriteOnce
   resources:
     requests:
       storage: 1Gi

kubectl create -f px-mongo-pvc.yaml

The storage backend for MongoDB Pod is now ready.

Let’s go ahead and create the MongoDB Pod.

apiVersion: extensions/v1beta1
kind: Deployment
metadata:
  name: db
  labels:
    name: mongo
    app: todoapp  
spec:
  strategy:
    rollingUpdate:
      maxSurge: 1
      maxUnavailable: 1
    type: RollingUpdate
  replicas: 1
  template:
    metadata:
      labels:
        name: mongo
        app: todoapp
    spec:
      schedulerName: stork
      containers:
      - name: mongo
        image: mongo
        imagePullPolicy: "Always"
        ports:
        - containerPort: 27017
        volumeMounts:
        - mountPath: /data/db
          name: mongodb
      volumes:
      - name: mongodb
        persistentVolumeClaim:
          claimName: px-mongo-pvc

kubectl create -f db-pod.yaml

To make the database Pod accessible to the web application, we will expose it through a ClusterIP-based Service.

apiVersion: v1
kind: Service
metadata:
  name: db
  labels:
    name: mongo
    app: todoapp
spec:
  selector:
    name: mongo
  type: ClusterIP
  ports:
    - name: db
      port: 27017
      targetPort: 27017

kubectl create -f db-svc.yaml

The database backend is now ready. It’s time to deploy the web application.

Deploying Node.js Web Application

The web application is a simple todo task list that persists changes to MongoDB. Create the Deployment for the web app with replicas.

apiVersion: apps/v1
kind: Deployment
metadata:
  name: web
  labels:
    name: web
    app: todoapp
spec:
  replicas: 3
  selector:
    matchLabels:
      name: web
  template:
    metadata:
      labels:
        name: web
    spec:
      containers:
      - name: web
        image: janakiramm/todo
        ports:
        - containerPort: 3000

Finally, we will expose the web application through a load balancer.

apiVersion: v1
kind: Service
metadata:
  name: web
  labels:
    name: web
    app: todoapp
spec:
  selector:
    name: web 
  type: LoadBalancer
  ports:
   - name: http
     port: 80
     targetPort: 3000
     protocol: TCP

Check the Service created for the web app to get the IP address of the load balancer.

Accessing the web app shows the below UI.

In the next part of this series, I will show you how to perform failover of MongoDB database running within GKE.

Portworx is a sponsor of The New Stack.

Feature image via Pixabay.

Janakiram MSV (Jani) is a practicing architect, research analyst, and advisor to Silicon Valley startups. He focuses on the convergence of modern infrastructure powered by cloud-native technology and machine intelligence driven by generative AI. Before becoming an entrepreneur, he spent...