Skip to content

Latest commit

 

History

History
118 lines (76 loc) · 3.43 KB

README.md

File metadata and controls

118 lines (76 loc) · 3.43 KB
Raw

Description

This is a very simple kubernetes demo application intended to show most of the benefits of using F5 Technology for your application delivery and security. It is composed of multiple technology frameworks.

This app will generate a sentence :)

alt text

alt text

App Documentation

Every WORD pod delivers a list of WORDS. Then, the GENERATOR select one WORD per POD, and generates a SENTENCE in a JSON format

alt text

{
    "adjectives": "proud",
    "animals": "lion",
    "colors": "blue",
    "locations": "park"
}

Then, the frontend web application will display all the words in a sentence. If one micro-service is not deployed, the word is not displayed.

In term of micro-services, this is how there are used by the Webapp frontend.

alt text

App Installation

Install Nginx OSS Ingress Controller

In your k8S, install an Nginx OSS Ingress Controller (helm is the way to do it - install helm first)

kubectl create ns ingress

helm repo add nginx-stable https://helm.nginx.com/stable
helm repo update

helm install nginx-ingress nginx-stable/nginx-ingress --namespace ingress --set controller.service.type=NodePort

Deploy all the pods manifests (in directoy /k8s-manifests)

WORDS pods are deployed in api namespace

FRONTEND pod is deployed in webapp-nginx namespace

kubectl create ns api
kubectl create ns webapp-nginx

kubectl apply -f sentence-api-all.yaml -n api
kubectl apply -f sentence-nginx-webapp.yaml -n webapp-nginx

Deploy the Ingress Virtual Server resource

Now, we must route the traffic either to the Nginx WebApp Frontend pod, if path is /, or route to WORDS pods is path starts with /api

To do so, we will Ingress Virtual Server Route resource (better than an Ingress resource)

kubectl apply -f ingress-vs-master.yaml
kubectl apply -f ingress-vs-route-api.yaml -n api
kubectl apply -f ingress-vs-route-frontend.yaml -n webapp-nginx

For Minikube, an extra step is required :

$ kubectl get service -n ingress
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
nginx-ingress-nginx-ingress NodePort 10.100.80.212 80:31859/TCP,443:30898/TCP 14m

$ curl -H "Host: api.sentence.com" http://127.0.0.1:31859/api/sentence
curl: (7) Failed to connect to 127.0.0.1 port 31859 after 0 ms: Connection refused

$ minikube service nginx-ingress-nginx-ingress --url -n ingress
http://192.168.49.2:31859
http://192.168.49.2:30898

$ curl -H "Host: api.sentence.com" http://192.168.49.2:31859/api/sentence
{"sentence": {"adjectives": "kind", "animals": "mouse", "colors": "green", "locatio...}

Test the app

First, find the NodePort used by the ingress.

kubectl get service -n ingress

NAME                          TYPE       CLUSTER-IP     EXTERNAL-IP   PORT(S)                      AGE
nginx-ingress-nginx-ingress   NodePort   10.43.129.50   <none>        80:32061/TCP,443:32553/TCP   77m

Collect the port used for port 80. Here, it is the port 32061

Test your app :)

curl -H "Host: api.sentence.com" http://127.0.0.1:32061/api/colors
curl -H "Host: api.sentence.com" http://127.0.0.1:32061/api/locations
curl -H "Host: api.sentence.com" http://127.0.0.1:32061/api/sentence

You can also test your app in a browser http://api.sentence.com:32061/

Courtesy of:

Thanks to https://www.npmjs.com/package/json-server for the zero coding JSON Server.