microservices-deployment

Sample microservices deployment on k8s with reverse proxy, monitoring, logging features...

Introduction

In this project you'll use following tools and applications:

• K8s
• Helm
• Prometheus + Grafana
• Nginx
• 10+ microservices
• K8s dashboard

Now let's start the journey!

1. Deploy the namespace

# monitor
kubectl create ns monitoring
# app
kubectl create ns demoapp

# Check namespaces
kubectl get ns

2. Deploy K8s dashboard

• Docs:

  • https://kubernetes.io/docs/tasks/access-application-cluster/web-ui-dashboard/
  • https://github.com/kubernetes/dashboard/blob/master/docs/user/access-control/creating-sample-user.md

• Deploy dashboard

kubectl apply -f https://raw.githubusercontent.com/kubernetes/dashboard/v2.7.0/aio/deploy/recommended.yaml

# Check/Verify
kubectl get all -n kubernetes-dashboard

• Forward port so we can access the dashboard

# Note: You can also replace 8443 to the working PORT in your PC
kubectl port-forward service/kubernetes-dashboard -n kubernetes-dashboard 8443:443

• Create admin user

kubectl apply -f manifest/k8s-dashboard-user/admin.yaml

• Access the dashboard at: https://localhost:8443/, then you can use token to login to dashboard
• You can get the token to access dashboard via: kubectl get secret admin-user -n kubernetes-dashboard -o jsonpath={".data.token"} | base64 -d

3. Deploy the application

In this project we use the sample microservices application from GoogleCloudPlatform/microservices-demo

kubectl apply -f https://raw.githubusercontent.com/GoogleCloudPlatform/microservices-demo/main/release/kubernetes-manifests.yaml -n demoapp
# To debug, run: kubectl get services -n  demoapp

4. Deploy Nginx proxy

Deploy the Nginx proxy prior to fontend service, using manifest/nginx

kubectl apply -f manifest/nginx/nginx.yaml -n demoapp
# To debug, run: kubectl get services -n  demoapp

5. Expose Nginx service

Open another terminal and run:

# Note: You can also replace 8090 to the working PORT in your PC
kubectl port-forward service/nginx-ingress -n demoapp 8090:80

Now visit http://localhost:8090/ to access your application locally:

6. Prometheus + Grafana

In section #4 we've successfully deploy our microservices application on k8s, deploy the Nginx proxy infont of the fontend app. Now let's monitor our application with Prometheus + Grafana

6.1. Deploy Prometheus/Grafana stack

• We will use Helm to deploy the Prometheus stack from prometheus-community:

# Prepare
helm repo add prometheus-community https://prometheus-community.github.io/helm-charts
helm repo update
# Deploy
helm install kube-prometheus-stack prometheus-community/kube-prometheus-stack --namespace monitoring

• Verify installation

# Pod checking
kubectl --namespace monitoring get pods -l "release=kube-prometheus-stack"
# All resources checking
kubectl --namespace monitoring get all

6.2. Access the dashboard

• Expose Grafana

kubectl port-forward svc/kube-prometheus-stack-grafana -n monitoring 4000:80

• Expose Prometheus

kubectl port-forward svc/kube-prometheus-stack-prometheus -n monitoring 4001:9090

• Now we can login to http://localhost:4000 (The default username/password for Grafana is admin/prom-operator)

6.3. Explore the Grafana

• Choose your dashboard

• Select the namespace

• Keep exploring the dashboard your way to see the metrics stats of our application on Kubernetes

7. Check K8s dashboard

• Visit the k8s dashboard (https://localhost:8443/), login and choose the demoapp namespace, you can see the workload and k8s object (deployment, pod, sevice, ...) statuses:

Cleanup!

In case you want to cleanup the resources:

# Cleanup application
kubectl delete namespace demoapp

# Cleanup other namespace
kubectl delete namespace <namespace_name>

Reference

• https://artifacthub.io
• https://prometheus-community.github.io/helm-charts
• https://helm.sh/docs/
• https://github.com/GoogleCloudPlatform/microservices-demo
• https://nginx.org/en/docs/