Deploy Microservices in Kubernetes¶

Now that we have a good understanding of kubernetes and related AWS services, let's deploy a few microservices in our EKS kubernetes cluster.

Prerequisite¶

To follow this tutorial, you'll require a domain and, additionally, an SSL certificate for the domain and its subdomains.

Register a Route 53 Domain

Go to AWS Console and register a Route 53 domain. You can opt for a cheaper TLD (top level domain) such as .link

Note

It usually takes about 10 minutes but it might take about an hour for the registered domain to become available.
Request a Public Certificate

Visit AWS Certificate Manager in AWS Console and request a public certificate for your domain and all the subdomains. For example, if you registered for a domain example.com then request certificate for example.com and *.example.com

Note

Make sure you request the certificate in the region where your EKS cluster is in.
Validate the Certificate

Validate the requested certificate by adding CNAME records in Route 53. It is a very simple process. Go to the certificate you created and click on Create records in Route 53. The CNAMEs will be automatically added to Route 53.

Note

It usually takes about 5 minutes but it might take about an hour for the certificate to be ready for use.

Now that you have everything you need, let's move on to the demonstration.

Docker Images¶

Here are the Docker Images used in this tutorial:

Note

reyanshkharga/nodeapp:mongo is a Node.js backend application that uses MongoDB to store and retrieve data.

Environment Variables:
- MONGODB_URI (Required)
- POD_NAME (Optional)
Tha app has the following routes:
- GET / Returns a JSON object containing Host and Version. If the POD_NAME environment variable is set, the value of the Host will be the value of the variable.
- GET /health Returns health status of the app
- GET /random Returns a randomly generated number between 1 and 10
- GET /books Returns the list of books
- POST /addBook Adds a book
- POST /updateBook Updates the copies_sold value for the given book id.
- POST /deleteBook Deletes the book record that matches the given id.
Sample body for POST /addBook:
```
{
  "id": 1,
  "title": "Book1",
  "copies_sold": 0
}
```
Sample body for POST /updateBook:
```
{
  "id": 1,
  "copies_sold": 100
}
```
Sample body for POST /deleteBook:
```
{
  "id": 1
}
```
reyanshkharga/reactapp:v1 is a React application. It's a frontend application that interacts with reyanshkharga/nodeapp:mongo backend to perform the CRUD operation.

Environment variables:
- REACT_APP_API_ENDPOINT (Optional)
The environment variable REACT_APP_API_ENDPOINT is optional. If provided, you will be able to do the CRUD operations.
mongo:5.0.2 is MongoDB database. Our backend will use it to store and retrieve data to perform CRUD operations.

Objective¶

We are going to deploy the following microservices on our EKS kubernetes cluster:

MongoDB Database microservice: uses docker image mongo:5.0.2
Node.js Backend microservice: uses docker image reyanshkharga/nodeapp:mongo
React Frontend microservice: uses docker image reyanshkharga/reactapp:v1

The following diagram illustrates the communication between microservices:

graph LR
  A(Frontend) ---> B(Backend);
  B -..-> C[(Database)];

Note

We will use the same load balancer for both backend and frontend microservices because using more load balancers will be expensive since load balancers are charged hourly. We can achieve this using IngressGroup.

Step 1: Deploy MongoDB Database Microservice¶

Let's create the kubernetes objects for our MongoDB database microservice as follows:

00-namespace.yml storageclass.yml pvc.yml deployment-and-service.yml

apiVersion: v1
kind: Namespace
metadata:
  name: mongodb

apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
  name: mongodb-storageclass
provisioner: ebs.csi.aws.com
parameters:
  type: gp3
  tagSpecification_1: "Name=eks-mongodb-storage"
  tagSpecification_2: "CreatedBy=aws-ebs-csi-driver"
reclaimPolicy: Delete

apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: mongodb-pvc
  namespace: mongodb
spec:
  accessModes:
    - ReadWriteOnce
  storageClassName: mongodb-storageclass
  resources:
    requests:
      storage: 4Gi

# Deployment
apiVersion: apps/v1
kind: Deployment
metadata:
  name: mongodb-deployment
  namespace: mongodb
spec:
  replicas: 1
  selector:
    matchLabels:
      app: mongodb
  template:
    metadata:
      labels:
        app: mongodb
    spec:
      containers:
      - name: mongodb
        image: mongo:5.0.2
        ports:
          - containerPort: 27017
        volumeMounts:
        - name: mongodb-storage
          mountPath: /data/db
      volumes:
      - name: mongodb-storage
        persistentVolumeClaim:
          claimName: mongodb-pvc
---
# Service
apiVersion: v1
kind: Service
metadata:
  name: mongodb-service
  namespace: mongodb
spec:
  type: ClusterIP
  selector:
    app: mongodb
  ports:
    - port: 27017
      targetPort: 27017

Assuming your folder structure looks like the one below:

|-- manifests
|   |-- mongodb
│   |   |-- 00-namespace.yml
│   |   |-- deployment-and-service.yml
│   |   |-- storageclass.yml
│   |   |-- pvc.yml

Let's apply the manifests to create the kubernetes objects for MongoDB database microservice:

kubectl apply -f mongodb/

This will create the following kubernetes objects:

A namespace named mongodb
A StorageClass (SC) for dynamic provisioning of persistent volume
A PersistentVolumeClaim (PVC) in the mongodb namespace
MongoDB deployment in the mongodb namespace
MongoDB service in the mongodb namespace

Note

The order in which yaml files are applied doesn't matter since every relation except namespace is handled by label selectors, so it fixes itself once all resources are deployed.

We are using Amazon EBS to persist the MongoDB data. EBS is provisioned dynamically using AWS EBS-CSI driver.

With persistent volume even if the MongoDB pod goes down the data will remain intact. When the new pod comes up we'll have the access to the same data.

Verify if the resources were created successfully:

# List all resources in mongodb namespace
kubectl get all -n mongodb

# List StorageClass
kubectl get sc

# List PersistentVolume
kubectl get pv

# List PersistenvVolumeClaim
kubectl get pvc -n mongodb

Verify if MongoDB is working as expected:

# Start a shell session inside the mongodb container
kubectl exec -it <mongodb-pod-name> -n mongodb -- bash

# Start the mongo Shell to interact with MongoDB
mongo

# List Databases
show dbs

# Switch to a Database
use <db-name>

# List collections
show collections

Step 2: Deploy Node.js Backend Microservice¶

Let's create the kubernetes objects for our Node.js backend microservice as follows:

00-namespace.yml deployment-and-service.yml ingress.yml

apiVersion: v1
kind: Namespace
metadata:
  name: backend

# Deployment
apiVersion: apps/v1
kind: Deployment
metadata:
  name: backend-deployment
  namespace: backend
spec:
  replicas: 1
  selector:
    matchLabels:
      app: backend
  template:
    metadata:
      labels:
        app: backend
    spec:
      containers:
      - name: nodeapp
        image: reyanshkharga/nodeapp:mongo
        imagePullPolicy: Always
        ports:
          - containerPort: 5000
        env:
        - name: MONGODB_URI
          # <service-name>.<namespace>.svc.cluster.local
          value: mongodb://mongodb-service.mongodb.svc.cluster.local:27017
---
# Service
apiVersion: v1
kind: Service
metadata:
  name: backend-service
  namespace: backend
spec:
  type: ClusterIP
  selector:
    app: backend
  ports:
    - port: 5000
      targetPort: 5000

apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
  name: backend-ingress
  namespace: backend
  annotations:
    # Load Balancer Annotations
    alb.ingress.kubernetes.io/scheme: internet-facing
    alb.ingress.kubernetes.io/load-balancer-name: my-load-balancer
    alb.ingress.kubernetes.io/target-type: ip
    # Health Check Annotations
    alb.ingress.kubernetes.io/healthcheck-protocol: HTTP
    alb.ingress.kubernetes.io/healthcheck-port: traffic-port
    alb.ingress.kubernetes.io/healthcheck-path: /health
    alb.ingress.kubernetes.io/healthcheck-interval-seconds: '5'
    alb.ingress.kubernetes.io/healthcheck-timeout-seconds: '2'
    alb.ingress.kubernetes.io/success-codes: '200'
    alb.ingress.kubernetes.io/healthy-threshold-count: '2'
    alb.ingress.kubernetes.io/unhealthy-threshold-count: '2'
    # Listerner Ports Annotation
    alb.ingress.kubernetes.io/listen-ports: '[{"HTTP": 80}, {"HTTPS": 443}]'
    # SSL Redicrect Annotation
    alb.ingress.kubernetes.io/ssl-redirect: '443'
    # IngressGroup
    alb.ingress.kubernetes.io/group.name: my-group
spec:
  ingressClassName: alb
  rules:
  - host: backend.example.com
    http:
      paths:
      - path: /
        pathType: Prefix
        backend:
          service:
            name: backend-service
            port:
              number: 5000

Assuming your folder structure looks like the one below:

|-- manifests
|   |-- backend
│   |   |-- 00-namespace.yml
│   |   |-- deployment-and-service.yml
│   |   |-- ingress.yml

Let's apply the manifests to create the kubernetes objects for Node.js backend microservice:

kubectl apply -f backend/

This will create the following kubernetes objects:

A namespace named backend
Backend deployment in the backend namespace
Backend service in the backend namespace
Ingress for backend service

The ingress creates an internet-facing load balancer and the SSL certificate is attached to the load balancer.

Note that the certificate is automatically discovered with hostnames from the ingress resource. Also, a Route 53 record is added for the host. This is all done by the AWS Load Balancer Controller and ExternalDNS.

Verify if the resources were created successfully:

# List all resources in backend namespace
kubectl get all -n backend

# List ingress in backend namespace
kubectl get ing -n backend

Go to AWS console and verify if the load balancer was created and a record was added to Route 53 for the host specified in ingress.

Open any browser on your local host machine and hit the URL to access the backend service:

https://backend.example.com

Note

In real world it is best to have authorization and authentication in place for the backend services that is accessible on the internet. But for the sake of simplicity we have not used any authorization or authentication for the backend service.

Step 3: Deploy React Frontend Microservice¶

Let's create the kubernetes objects for our React frontend microservice as follows:

00-namespace.yml deployment-and-service.yml ingress.yml

apiVersion: v1
kind: Namespace
metadata:
  name: frontend

# Deployment
apiVersion: apps/v1
kind: Deployment
metadata:
  name: frontend-deployment
  namespace: frontend
spec:
  replicas: 1
  selector:
    matchLabels:
      app: frontend
  template:
    metadata:
      labels:
        app: frontend
    spec:
      containers:
      - name: reactapp
        image: reyanshkharga/reactapp:v1
        imagePullPolicy: Always
        ports:
          - containerPort: 3000
        env:
        - name: REACT_APP_API_ENDPOINT
          value: https://backend.example.com
---
# Service
apiVersion: v1
kind: Service
metadata:
  name: frontend-service
  namespace: frontend
spec:
  type: ClusterIP
  selector:
    app: frontend
  ports:
    - port: 3000
      targetPort: 3000

apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
  name: frontend-ingress
  namespace: frontend
  annotations:
    # Load Balancer Annotations
    alb.ingress.kubernetes.io/scheme: internet-facing
    alb.ingress.kubernetes.io/load-balancer-name: my-load-balancer
    alb.ingress.kubernetes.io/target-type: ip
    # Health Check Annotations
    alb.ingress.kubernetes.io/healthcheck-protocol: HTTP
    alb.ingress.kubernetes.io/healthcheck-port: traffic-port
    alb.ingress.kubernetes.io/healthcheck-path: /
    alb.ingress.kubernetes.io/healthcheck-interval-seconds: '5'
    alb.ingress.kubernetes.io/healthcheck-timeout-seconds: '2'
    alb.ingress.kubernetes.io/success-codes: '200'
    alb.ingress.kubernetes.io/healthy-threshold-count: '2'
    alb.ingress.kubernetes.io/unhealthy-threshold-count: '2'
    # Listerner Ports Annotation
    alb.ingress.kubernetes.io/listen-ports: '[{"HTTP": 80}, {"HTTPS": 443}]'
    # SSL Redicrect Annotation
    alb.ingress.kubernetes.io/ssl-redirect: '443'
    # IngressGroup
    alb.ingress.kubernetes.io/group.name: my-group
spec:
  ingressClassName: alb
  rules:
  - host: frontend.example.com
    http:
      paths:
      - path: /
        pathType: Prefix
        backend:
          service:
            name: frontend-service
            port:
              number: 3000

Assuming your folder structure looks like the one below:

|-- manifests
|   |-- frontend
│   |   |-- 00-namespace.yml
│   |   |-- deployment-and-service.yml
│   |   |-- ingress.yml

Let's apply the manifests to create the kubernetes objects for React frontend microservice:

kubectl apply -f frontend/

This will create the following kubernetes objects:

A namespace named frontend
Frontend deployment in the frontend namespace
Frontend service in the frontend namespace
Ingress for frontend service

The ingress creates an internet-facing load balancer and the SSL certificate is attached to the load balancer.

Note that the certificate is automatically discovered with hostnames from the ingress resource. Also, a Route 53 record is added for the host. This is all done by the AWS Load Balancer Controller and ExternalDNS.

Note

The frontend microservice uses the API provided by the backend microservice to perform the CRUD operations.

Verify if the resources were created successfully:

# List all resources in frontend namespace
kubectl get all -n frontend

# List ingress in frontend namespace
kubectl get ing -n frontend

Go to AWS console and verify if the load balancer was created and a record was added to Route 53 for the host specified in ingress.

Open any browser on your local host machine and hit the URL to access the frontend service:

https://frontend.example.com

Step 4: Perform CRUD Operations and Verify Data in MongoDB¶

Access the frontend service from your browser and perform some CRUD operations as follows:

Get Books
Add a Book
Update a Book
Delete a Book
Check Health

Verify if the records have been inserted into the MongoDB database:

# Start a shell session inside the mongodb container
kubectl exec -it <mongodb-pod-name> -n mongodb -- bash

# Start the mongo Shell to interact with MongoDB
mongo

# List Databases
show dbs

# Switch to mydb database
use mydb

# List collections
show collections

# List items in the books collection
db.books.find()

Clean Up¶

Assuming your folder structure looks like the one below:

|-- manifests
|   |-- mongodb
│   |   |-- 00-namespace.yml
│   |   |-- deployment-and-service.yml
│   |   |-- storageclass.yml
│   |   |-- pvc.yml
|   |-- backend
│   |   |-- 00-namespace.yml
│   |   |-- deployment-and-service.yml
│   |   |-- ingress.yml
|   |-- frontend
│   |   |-- 00-namespace.yml
│   |   |-- deployment-and-service.yml
│   |   |-- ingress.yml

Let's delete all the resources we created:

kubectl delete -f manifests/

All AWS resources, such as load balancers, Route 53 records, etc., created by AWS Load Balancer Controller via ingress or service objects, will also be deleted.

References:

Kubernetes Resource Creation Order