In deploying keepalived/haproxy for a k8s cluster by following this instruction (Creating Highly Available Clusters with kubeadm | Kubernetes),
should we make the frontend bind server:port consistent with the control plane endpoint? (as shown in the instructionsudo kubeadm init --control-plane-endpoint “LOAD_BALANCER_DNS:LOAD_BALANCER_PORT” --upload-certs )
But I tried and it always failed when they are consistent with the following error:
$ sudo kubeadm init --control-plane-endpoint “192.168.222.11:6443” --upload-certs
…
error execution phase preflight: [preflight] Some fatal errors occurred:
[ERROR Port-6443]: Port 6443 is in use
Cluster information:
Kubernetes version: 1.22
Cloud being used: (put bare-metal if not on a public cloud) Openstack
Installation method: kubeadm init
Host OS: CentOS7
CNI and version: Weave
CRI and version: N/A
haproxy.cfg
…
frontend frontserver
bind *:6443
stats uri /haproxy?stats
option tcplog
acl url_static path_beg -i /static /images /javascript /stylesheets
acl url_static path_end -i .jpg .gif .png .css .js
use_backend risc_cntrollers if url_static
default_backend risc_cntrollers
backend risc_cntrollers
option httpchk GET /healthz
http-check expect status 200
option ssl-hello-chk
balance roundrobin
server riscmaster1 192.168.222.201:6443 check
server riscmaster2 192.168.222.202:6443 check
server riscmaster3 192.168.222.203:6443 check
NAMESPACE NAME READY STATUS RESTARTS AGE
kube-system coredns-78fcd69978-cs97k 0/1 Pending 0 3d22h
kube-system coredns-78fcd69978-v994v 0/1 Pending 0 3d22h
kube-system etcd-riscmaster1 1/1 Running 0 3d22h
kube-system kube-apiserver-riscmaster1 1/1 Running 0 3d22h
kube-system kube-controller-manager-riscmaster1 1/1 Running 0 3d22h
kube-system kube-proxy-lrvqw 1/1 Running 0 3d22h
kube-system kube-scheduler-riscmaster1 1/1 Running 0 3d22h
NAMESPACE NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
default kubernetes ClusterIP 10.96.0.1 443/TCP 3d22h
kube-system kube-dns ClusterIP 10.96.0.10 53/UDP,53/TCP,9153/TCP 3d22h