I have installed nginx ingress controller. As I understand that one of the main reasons to use ingress is to save money by not creating multiple load balancers.
My kubesphere-console service yaml manifest looks like below:
cat kubesphere-console.yaml
apiVersion: v1
kind: Service
metadata:
annotations:
meta.helm.sh/release-name: ks-core
meta.helm.sh/release-namespace: kubesphere-system
creationTimestamp: "2022-05-30T04:51:22Z"
finalizers:
- service.kubernetes.io/load-balancer-cleanup
labels:
app: ks-console
app.kubernetes.io/managed-by: Helm
tier: frontend
version: v3.1.0
name: ks-console
namespace: kubesphere-system
resourceVersion: "785863"
uid: 8628c2d0-164b-499f-ac0c-254ac77aa48c
spec:
allocateLoadBalancerNodePorts: true
clusterIP: 10.0.29.29
clusterIPs:
- 10.0.29.29
externalTrafficPolicy: Cluster
internalTrafficPolicy: Cluster
ipFamilies:
- IPv4
ipFamilyPolicy: SingleStack
ports:
- name: nginx
nodePort: 30880
port: 80
protocol: TCP
targetPort: 8000
selector:
app: ks-console
tier: frontend
sessionAffinity: None
type: LoadBalancer
status:
loadBalancer:
ingress:
- ip: 20.248.217.111
The kubesphere-console ingress route yaml manifest is like below:
cat ingress-route-kubesphere.yaml
apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
name: kubesphere-console
annotations:
nginx.ingress.kubernetes.io/ssl-redirect: "false"
nginx.ingress.kubernetes.io/use-regex: "true"
nginx.ingress.kubernetes.io/rewrite-target: /$2
spec:
ingressClassName: nginx
rules:
- http:
paths:
- path: /
pathType: Prefix
backend:
service:
name: ks-console
port:
number: 80
- path: /(.*)
pathType: Prefix
backend:
service:
name: ks-console
port:
number: 80
---
I created ingress-controller on AKS cluster as like below:
NAMESPACE=ingress-basic
helm repo add ingress-nginx https://kubernetes.github.io/ingress-nginx
helm repo update
helm install ingress-nginx ingress-nginx/ingress-nginx \
--create-namespace \
--namespace $NAMESPACE \
--set controller.service.annotations."service\.beta\.kubernetes\.io/azure-load-balancer-health-probe-request-path"=/healthz
LB created for nginxingress-controller as like below
k get svc -ningress-basic
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
ingress-nginx-controller LoadBalancer 10.0.194.38 20.92.133.79 80:32703/TCP,443:31053/TCP 23h
ingress-nginx-controller-admission ClusterIP 10.0.189.8 <none> 443/TCP 23h
Kindly do note that I am not that great in Ingress resource/networking so I am still learning. My main intention is use one ingress controller to expose all applications
Example:
Kubesphere-console --> Port 80 and Endpoint should be ksconsole
tekton-pipelines ----> Port 80 and Endpoint should be tekton-dashboard
I shall sincerely appreciate any help. P.S: I investigated the ingress control pod logs but no luck there or I really don't what they mean....
k logs -f ingress-nginx-controller-6648b5dbb8-ntm8t -ningress-basic
10.240.0.4 - - [02/Jun/2022:12:29:42 0000] "GET /healthz HTTP/1.1" 302 43 "-" "Load Balancer Agent" 194 0.002 [kubesphere-system-ks-console-80] [] 10.240.0.41:8000 43 0.000 302 6ea21ca50e247cc5d5e56975c8946fcc
10.240.0.4 - - [02/Jun/2022:12:29:42 0000] "GET /healthz HTTP/1.1" 302 43 "-" "Load Balancer Agent" 194 0.003 [kubesphere-system-ks-console-80] [] 10.240.0.41:8000 43 0.004 302 1f2e1bfa0223e6b3b630a59fe7dba4d1
10.240.0.4 - - [02/Jun/2022:12:29:48 0000] "GET /healthz HTTP/1.1" 302 43 "-" "Load Balancer Agent" 194 0.004 [kubesphere-system-ks-console-80] [] 10.240.0.41:8000 43 0.000 302 dbb5af0fd6426f783c337da06912e427
10.240.0.4 - - [02/Jun/2022:12:29:48 0000] "GET /healthz HTTP/1.1" 302 43 "-" "Load Balancer Agent" 194 0.003 [kubesphere-system-ks-console-80] [] 10.240.0.41:8000 43 0.004 302 9fc084018c70c0f61f379e2266a056f8
10.240.0.4 - - [02/Jun/2022:12:29:54 0000] "GET /healthz HTTP/1.1" 302 43 "-" "Load Balancer Agent" 194 0.003 [kubesphere-system-ks-console-80] [] 10.240.0.41:8000 43 0.004 302 f9e4c75d81aa0e7ff03e402dfdca2d35
10.240.0.4 - - [02/Jun/2022:12:29:54 0000] "GET /healthz HTTP/1.1" 302 43 "-" "Load Balancer Agent" 194 0.003 [kubesphere-system-ks-console-80] [] 10.240.0.41:8000 43 0.000 302 187cc1fdb0fb1499c1ef040e8aeb5e01
10.240.0.4 - - [02/Jun/2022:12:30:00 0000] "GET /healthz HTTP/1.1" 302 43 "-" "Load Balancer Agent" 194 0.003 [kubesphere-system-ks-console-80] [] 10.240.0.41:8000 43 0.004 302 9b3b1ae4165184ed16fa974be51cd460
10.240.0.4 - - [02/Jun/2022:12:30:00 0000] "GET /healthz HTTP/1.1" 302 43 "-" "Load Balancer Agent" 194 0.003 [kubesphere-system-ks-console-80] [] 10.240.0.41:8000 43 0.004 302 eac2591d333af20ab1a3ee4052327291
10.240.0.4 - - [02/Jun/2022:12:30:06 0000] "GET /healthz HTTP/1.1" 302 43 "-" "Load Balancer Agent" 194 0.004 [kubesphere-system-ks-console-80] [] 10.240.0.41:8000 43 0.000 302 f08f30018bcfd229adb1ed36b6473563
10.240.0.4 - - [02/Jun/2022:12:30:06 0000] "GET /healthz HTTP/1.1" 302 43 "-" "Load Balancer Agent" 194 0.002 [kubesphere-system-ks-console-80] [] 10.240.0.41:8000 43 0.000 302 f157141e253c7fcc61008440ff047605
10.240.0.4 - - [02/Jun/2022:12:30:12 0000] "GET /healthz HTTP/1.1" 302 43 "-" "Load Balancer Agent" 194 0.003 [kubesphere-system-ks-console-80] [] 10.240.0.41:8000 43 0.004 302 6ba07067e1345b665e16a3cab03ee50c
10.240.0.4 - - [02/Jun/2022:12:30:12 0000] "GET /healthz HTTP/1.1" 302 43 "-" "Load Balancer Agent" 194 0.003 [kubesphere-system-ks-console-80] [] 10.240.0.41:8000 43 0.004 302 4978c44153381d1aac6bc959c4afa72d
10.240.0.4 - - [02/Jun/2022:12:30:18 0000] "GET /healthz HTTP/1.1" 302 43 "-" "Load Balancer Agent" 194 0.003 [kubesphere-system-ks-console-80] [] 10.240.0.41:8000 43 0.004 302 fc1e6e7fd4f4219ef6ca38689275035a
10.240.0.4 - - [02/Jun/2022:12:30:18 0000] "GET /healthz HTTP/1.1" 302 43 "-" "Load Balancer Agent" 194 0.003 [kubesphere-system-ks-console-80] [] 10.240.0.41:8000 43 0.000 302 1e286ef7a2b36a5b04521e59657e0433
CodePudding user response：
Usually, you can access your services directly through the external IP (wildcard) of the ingress-controller svc if you create an ingress without a specified host. You can set a domain name for the ingress after configuring DNS resolution correctly.
Here is an example of exposing both ks-apiserver and ks-console through nginx-ingress-controller.
nginx controller svc
root@master:~# kubectl -n ingress-nginx get svc
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
ingress-nginx-controller NodePort 10.233.11.24 <none> 80:30882/TCP,443:31054/TCP 15m
ingress-nginx-controller-admission ClusterIP 10.233.18.191 <none> 443/TCP 37m
ingress of ks-apiserver and ks-console
root@master:~# kubectl -n kubesphere-system get ing api -o yaml
apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
annotations:
kubesphere.io/creator: admin
creationTimestamp: "2022-06-02T08:48:48Z"
generation: 8
name: kubesphere
namespace: kubesphere-system
resourceVersion: "982261"
uid: f8f6cadf-874c-4996-8759-3467ac33d61c
spec:
ingressClassName: nginx
rules:
- host: api.172.31.50.170.nip.io
http:
paths:
- backend:
service:
name: ks-apiserver
port:
number: 80
path: /
pathType: ImplementationSpecific
- host: console.172.31.50.170.nip.io
http:
paths:
- backend:
service:
name: ks-console
port:
number: 80
path: /
pathType: ImplementationSpecific
status:
loadBalancer:
ingress:
- ip: 172.31.50.170
Verify the ingresses
$ curl http://api.172.31.50.170.nip.io:30882/
{
"kind": "Status",
"apiVersion": "v1",
"metadata": {
},
"status": "Failure",
"message": "forbidden: User \"system:anonymous\" cannot GET path \"/\"",
"reason": "Forbidden",
"details": {
},
"code": 403
}%
$ curl http://console.172.31.50.170.nip.io:30882/
Redirecting to <a href="/login">/login</a>.%