Nodeport Question

General Discussions
,
1

I run three VM’s with Ubuntu 22.04 and Kubernetes: v1.26.2

I’ve created a nginx service as follow:

apiVersion: v1
kind: Service
metadata:
  creationTimestamp: "2023-03-13T12:42:03Z"
  labels:
    app: nginx
  name: nginx
  namespace: default
  resourceVersion: "6147648"
  uid: 586c1b14-bd4f-45c8-913a-edfaa71f1168
spec:
  clusterIP: 10.100.238.2
  clusterIPs:
  - 10.100.238.2
  externalTrafficPolicy: Cluster
  internalTrafficPolicy: Cluster
  ipFamilies:
  - IPv4
  ipFamilyPolicy: SingleStack
  ports:
  - nodePort: 32714
    port: 80
    protocol: TCP
    targetPort: 80
  selector:
    app: nginx
  sessionAffinity: None
  type: NodePort
status:
  loadBalancer: {}

Now as the thing spins up I see:

n@k8s-control:~$ kubectl describe service nginx
Name:                     nginx
Namespace:                default
Labels:                   app=nginx
Annotations:              <none>
Selector:                 app=nginx
Type:                     NodePort
IP Family Policy:         SingleStack
IP Families:              IPv4
IP:                       10.100.238.2
IPs:                      10.100.238.2
Port:                     <unset>  80/TCP
TargetPort:               80/TCP
NodePort:                 <unset>  32714/TCP
Endpoints:                10.0.2.243:80
Session Affinity:         None
External Traffic Policy:  Cluster
Events:                   <none>

My machines are on a 192.168.10.0/24 Network. The ClusterIP (see above) is: 10.100.238.2
But this IP is (obviously) not reachable from my subnet, so all curl’s would fail.
However, reaching the Service via (e.g.) 192.168.10.11:80 works perfectly.

Why ?