I created a load balancer service and the describe command returns the following:
Name: minio-service
Namespace: minio
Labels: app=minio
Annotations: <none>
Selector: app=minio
Type: LoadBalancer
IP Family Policy: SingleStack
IP Families: IPv4
IP: 10.43.150.49
IPs: 10.43.151.50
LoadBalancer Ingress: 192.168.31.12, 192.168.32.13, 192.168.33.14
Port: <unset> 9012/TCP
TargetPort: 9011/TCP
NodePort: <unset> 30777/TCP
Endpoints: 10.42.10.110:9011,10.42.10.111:9011
Session Affinity: None
External Traffic Policy: Cluster
Events: <none>
If I try curl http://192.168.31.12:9012 it returns:
curl: (7) Failed to connect to 192.168.31.12 port 9012: Connection timed out
Furthermore, I observed something strange.
kubectl get nodes -o wide
NAME STATUS ROLES AGE VERSION INTERNAL-IP EXTERNAL-IP OS-IMAGE KERNEL-VERSION CONTAINER-RUNTIME
antonis-dell Ready control-plane,master 4h42m v1.21.2+k3s1 192.168.31.12 <none> Ubuntu 18.04.1 LTS 4.15.0-147-generic containerd://1.4.4-k3s2
knodea Ready <none> 4h9m v1.21.2+k3s1 192.168.32.13 <none> Raspbian GNU/Linux 10 (buster) 5.10.17-v7l+ containerd://1.4.4-k3s2
knodeb Ready <none> 4h2m v1.21.2+k3s1 192.168.33.14 <none> Raspbian GNU/Linux 10 (buster) 5.4.51-v7l+ containerd://1.4.4-k3s2
which means that LoadBalancer Ingress ips are the same with internal ips of nodes in the cluster.
Does anyone know why I have three LoadBalancer Ingress ips which are the same as the internal node ips and how to fix this?