Welcome to the new Linux Foundation Forum!

Lab 3.4: can't access service of the deployment

I create three ubuntu 18.04 box in google cloud. one master and two worker nodes have static internal ip address:
master: 10.240.0.50
node1:10.240.0.51
node2:10.240.0.52

K8s CNI is Calico as Lab 3.1

apply nginx image and expose service

NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
kubernetes ClusterIP 10.96.0.1 443/TCP 3h11m
nginx ClusterIP 10.99.93.30 80/TCP 87m

endpoints

NAME ENDPOINTS AGE
kubernetes 10.240.0.50:6443 3h12m
nginx 192.168.1.3:80 88m

pod is alive on node1: 10.240.0.51
pod ip: 192.168.1.3

nginx container port: 80

problem:

curl 192.168.1.3:80 inside node1 works
curl 10.99.90.30:80 inside node1 works

but cant curl them in 10.240.0.40 which kubectl is working fine.

Comments

  • simonshengsimonsheng Posts: 3

    reference previous Lab 3.3 questions, I turn on all ports, all protocols, all ranges. still doesn't work

  • chrispokornichrispokorni Posts: 240

    Hi @simonsheng,
    A ClusterIP type service is accessible from inside the cluster: master (.50), node1 (.51) and node2 (.52).
    Client node (.40) where you have kubectl installed, is not part of the cluster.
    Your cluster is formed by the nodes where kubeadm was installed and issued, such as 'kubeadm init' on master, and then 'kubeadm join' on node1 and node2.
    Regards,
    -Chris

  • simonshengsimonsheng Posts: 3

    Hi @chrispokorni,

    Ops, sorry for this unclear question, master couldn't access node1 either, but has been fixed by open ports to all.

    Thanks a lot to help me!

Sign In or Register to comment.