Node not visible even after submariner deployment

1

Hi,
I have successfully deployed submariner on vmnode01with the below two commands but unable to see the node when I run kubectl get nodes

subctl deploy-broker --kubeconfig /home/ec2-user/.kube/config
subctl join --kubeconfig /home/ec2-user/.kube/config broker-info.subm --clustercidr "192.168.0.0/16" --servicecidr="10.0.0.0/24"

On, EKS CLuster, I unfortunately only see two nodes which are part of the cluster but not the third node/external node :frowning:

[ec2-user@ip-192-168-11-242 ~]$ kubectl get nodes
NAME                                                STATUS   ROLES    AGE     VERSION
ip-192-168-11-242.ap-southeast-2.compute.internal   Ready    <none>   5h51m   v1.17.11-eks-cfdc40
ip-192-168-92-167.ap-southeast-2.compute.internal   Ready    <none>   5h51m   v1.17.11-eks-cfdc40

On a positive note!, I do see submariner pods are running on eks cluster, wondering why I am unable to see the node.

[ec2-user@ip-192-168-11-242 ~]$ kubectl get pods --all-namespaces
NAMESPACE             NAME                                             READY   STATUS    RESTARTS   AGE
kube-system           aws-node-dfxl5                                   1/1     Running   0          5h58m
kube-system           aws-node-p2c5h                                   1/1     Running   0          5h58m
kube-system           coredns-76c9876f5-4nm7b                          1/1     Running   0          6h3m
kube-system           coredns-76c9876f5-rlv84                          1/1     Running   0          6h3m
kube-system           kube-proxy-9nllx                                 1/1     Running   0          5h58m
kube-system           kube-proxy-hnrvf                                 1/1     Running   0          5h58m
submariner-operator   submariner-gateway-28nd5                         1/1     Running   0          19m
submariner-operator   submariner-lighthouse-agent-579d9c687c-rldzr     1/1     Running   0          19m
submariner-operator   submariner-lighthouse-coredns-75ff486db4-f9gxr   1/1     Running   0          19m
submariner-operator   submariner-lighthouse-coredns-75ff486db4-v7n9p   1/1     Running   0          19m
submariner-operator   submariner-operator-7d6657767b-2qxz9             1/1     Running   0          19m
submariner-operator   submariner-routeagent-j6thm                      1/1     Running   0          19m
submariner-operator   submariner-routeagent-ltqwn                      1/1     Running   0          19m

What am I doing wrong? How to resolve this issue so I can see my node?

2

Submariner does not make nodes from the other (remote) clusters visible in your local cluster.

I would recommend running subctl verify in order to verify connectivity between your clusters. More information can be found here: https://submariner.io/deployment/subctl/#verify

3

Hi Oats87,

Perhaps I incorrectly understood then the use case of the submariner.

Is submariner not the tool to connect a VM(s) to an existing cluster?

On my newly created VM, I installed kubectl, configured aws and then performed eks update like below:

curl -LO https://storage.googleapis.com/kubernetes-release/release/$(curl -s https://storage.googleapis.com/kubernetes-release/release/stable.txt)/bin/linux/amd64/kubectl
chmod +x ./kubectl
sudo mv ./kubectl /usr/local/bin/kubectl
kubectl version
aws eks update-kubeconfig --name my-cluster82  --region ap-southeast-2

So, kubectl gives the number of worker nodes for EKS. I am thinking that if I install submariner on any VM(s), I should be able to join it to the existing EKS cluster by virtue of submariner.

Help me understand and correct if I am wrong here.

As in regards to subctl verify. It is asking for two kubeconfigs where I have only one kubeconfig not two. The kubeconfig of EKS cluster.

subctl verify
Error: Two kubeconfigs must be specified.
Usage:
  subctl verify <kubeConfig1> <kubeConfig2> [flags]

Kindly assist me.

thanks