Skip to content

Latest commit

 

History

History
178 lines (132 loc) · 6.39 KB

expose-sample-app-with-ingress.md

File metadata and controls

178 lines (132 loc) · 6.39 KB
Raw

Expose a sample application with ingress

Now we have an Ingress Controller to satisfy an Ingress. We can configure load balancing for our sample web applications using an Ingress resource.

ingress for a sample application

Change back to lab project directory

  1. For this excerise we will change back to the lab project directory:
cd lab-kubernetes-ingress

1. Deploy the Cafe Applications

  1. Make sure our Coffee and Tea Application is running. Redeploy the the sample Coffee and Tea Application if needed
# Check the sample application is deployed and ready
k get deployment,service,pods -n cafe

NAME                     READY   UP-TO-DATE   AVAILABLE   AGE
deployment.apps/coffee   1/1     1            1           1d
deployment.apps/tea      1/1     1            1           1d

NAME                 TYPE        CLUSTER-IP       EXTERNAL-IP   PORT(S)   AGE
service/coffee-svc   ClusterIP   10.100.104.173   <none>        80/TCP    1d
service/tea-svc      ClusterIP   10.100.23.145    <none>        80/TCP    1d

NAME                          READY   STATUS    RESTARTS   AGE
pod/coffee-74c98dd7c8-5t6sq   1/1     Running   0          1d
pod/tea-7dcd6f57b6-kztrd      1/1     Running   0          1d

2. Deploy the Ingress Controller

  1. Make sure NGINX Plus Ingress Controller is installed. After following the instructions to deploying NGINX Plus Ingress Controller with the manifests or helm, the Ingress Controller should now be publicly accessible

3. Configure Load Balancing using Ingress

  1. Create a secret with a self-signed SSL certificate and a key in the cafe namespace:
kubectl create -f deployments/cafe-app/cafe-secret.yml
  1. Create an Ingress resource

Noteworthy points about cafe-ingress.yml:

  • A new ingressClassName field has been added to the Ingress spec in Kubernetes >= 1.18 and is used to reference the IngressClass that should be used to implement this Ingress. When using Kubernetes <1.18, remove or comment out the ingressClassName field in the YAML file
  • The ingress resource will configure a virtual server listening on the FQDN, cafe.example.com,
  • The ingress resource will configures URL path based routing, routing /tea to our tea service (tea-svc) and /coffee to our coffee service (coffee-svc)
# Create Ingress for cafe.example.com HTTP/S
kubectl create -f deployments/cafe-app/cafe-ingress.yml

# Check it has been deployed
kubectl get ingress -n cafe
NAME           CLASS   HOSTS              ADDRESS   PORTS     AGE
cafe-ingress   nginx   cafe.example.com             80, 443   67s

4. Access the Live Activity Monitoring

The Live Activity Monitoring Dashboard will allow us to see the real time activity metrics on our Ingress Controller upon scaling our applications (Pods)

Note: The NGINX Plus dashboard displaying metrics is from a a single Pod and a tool such as Grafana dashboards would need to be used to show aggregated metrics across all Pods in a Ingress Controller Cluster

  1. To get Access the Live Activity Monitoring Dashboard using kubectl port-forward on a single Ingress Controller pod, run the following:

Note: The port is only forwarded while the kubectl process is running, and so the following command will consume a terminal window

# Access the Live Activity Monitoring Dashboard
kubectl get pods --namespace=nginx-ingress
kubectl port-forward <nginx-plus-ingress-pod> 8080:8080 --namespace=nginx-ingress
  1. Open the NGINX Plus dashboard in a web browser on http://127.0.0.1:8080/dashboard.html

NGINX Plus Dashboard

5. Test the sample application with Ingress

  1. Save the public IP address of the Ingress controller into a shell variable. Refer to lab excerises to obtain the public IP address when using:
# save the IP Address of the Ingress controller into a shell variable
# This will be the LoadBalancer IP address when using service loadBalancer, or
# A worker node IP address for a daemonSet or when using service nodePort
IC_IP=XXX.XXX.XXX.XXX

# save the HTTPS Port of the Ingress controller into a shell variable:
# This will not be port 443 when using nodePort
IC_HTTPS_PORT=XXX
  1. Use curl to test external access to the sample coffee and the tea services. You may need to use curl's --insecure option to turn off certificate verification of our self-signed certificate and the --resolve option to set the Host header of a request with cafe.example.com

For example, accessing our sample application with Loadbalancer and Ingress Controller :

# Get External loadBalancer address
LB_FQDN=$(kubectl get services/nginx-ingress -n nginx-ingress -o jsonpath='{.status.loadBalancer.ingress[*].hostname}')
# Get loadBalancer IP address
IC_IP=$(dig +short $LB_FQDN A |  awk 'NR==1')

# Get coffee
curl https://cafe.example.com/coffee --insecure \
    --resolve cafe.example.com:80:$IC_IP \
    --resolve cafe.example.com:443:$IC_IP

# Get tea:
curl https://cafe.example.com/tea --insecure \
    --resolve cafe.example.com:80:$IC_IP \
    --resolve cafe.example.com:443:$IC_IP

You can also add a temporary host file entry and view the application in your web browser

echo "$IC_IP cafe.example.com" | sudo tee -a /etc/hosts
open https://cafe.example.com/coffee
open https://cafe.example.com/tea

6. Check "under the hood" of the Nginx Pod

  1. Look at the NGINX logs
# Find the Ingress controller pod name
kubectl get pods --namespace=nginx-ingress

# See the Ingress controller logs. -f for follow
kubectl logs <name> --namespace=nginx-ingress -f
  1. exec into the container and check out the nginx config and logs
# Connect into the container
kubectl exec -it <name> --namespace=nginx-ingress -- bash
kubectl exec -it <name> --namespace=nginx-ingress -- /bin/bash

# See full Nginx config
nginx -T