I have a k3s (light weighted k8s) cluster running on my Raspberry PI. So, I am not using any cloud hosted cluster but a bear metal one on my Raspberry PI.
I have deployed a application with this manifest:
apiVersion: apps/v1
kind: Deployment
metadata:
name: hello-world
namespace: myapp
spec:
replicas: 3
selector:
matchLabels:
app: hello-world
template:
metadata:
labels:
app: hello-world
spec:
containers:
- name: hello-world
image: bashofmann/rancher-demo:1.0.0
imagePullPolicy: Always
resources:
requests:
cpu: 200m
ports:
- containerPort: 8080
name: web
protocol: TCP
I also created a service to forward traffic to the application pod. Its manifest is:
apiVersion: v1
kind: Service
metadata:
name: demo-app-svc
namespace: myapp
spec:
selector:
app: hello-world
ports:
- name: web
protocol: TCP
port: 31113
targetPort: 8080
Then, I created a Ingress for the routing rules:
apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
name: myapp-ing
namespace: myapp
annotations:
nginx.ingress.kubernetes.io/rewrite-target: /
spec:
ingressClassName: nginx
rules:
- host: myapp.com
http:
paths:
- pathType: Prefix
path: /
backend:
service:
name: demo-app-svc
port:
number: 31113
I successfully deployed above application pod, service & Ingress to my k3s cluster. Like the manifests indicate, they are under namespace myapp.
The next thing I would like to do is to deploy the Kubernetes Nginx Ingress Controller in order to have the clients outside the cluster be able to access the deployed application.
So, I deployed it by :
kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/controller-v1.1.2/deploy/static/provider/cloud/deploy.yaml
The above command successfully deployed Ingress Controller under namespace ingress-nginx along with other objects as shown below with command k get all -n ingress-nginx:
As you can see above, the LoadBalancer type service external IP is with value <pending>. So, client outside the cluster still can not access the application pod.
Why is that & what do I miss deploying the Nginx Ingress Controller on a bear metal machine? The goal is to have an external IP that can be used to access the application pod from outside cluster, how can I achieve that?
===== Update =====
Based on the answer below from @Dawid Kruk , I decided to use the k3s default Traefik Ingress Controller.
So, I deleted all the deployed Nginx Ingress Controller resources by k delete all --all -n ingress-nginx .
Then, I checked the Traefik Ingress related LoadBalancer type service:
The external IP of that Traefik service is exactly my Raspberry PI's IP address!
So, added this IP to /etc/hosts to map it to the hostname defined in my Ingress object:
192.168.10.203 myapp.com
I opened browser & use address http://myapp.com, with the routing rules defined in my Ingress object (see the manifest for my ingress above), I hoped I could see my deployed web application now. But get 404 Page Not Found. What am I missing now to access my deployed application?
Another side question: I noticed when I check the deployed Ingress object, its IP address is empty, I wonder am I supposed to see an IP address for this object or not when the Traefik Ingress Controller takes effect?
Another issue: Now, when I re-deploy my ingress manifest by k apply -f ingress.yaml, I get error:
Resource: "networking.k8s.io/v1, Resource=ingresses", GroupVersionKind: "networking.k8s.io/v1, Kind=Ingress"
...
for: "ingress.yaml": error when patching "ingress.yaml": Internal error occurred: failed calling webhook "validate.nginx.ingress.kubernetes.io": failed to call webhook: Post "https://ingress-nginx-controller-admission.ingress-nginx.svc:443/networking/v1/ingresses?timeout=10s": service "ingress-nginx-controller-admission" not found
It looks like even I decided to use Traefik Ingress Controller, I still need to instal Nginx Ingress Controller. I get confused now, anyone can explain it?
