The connection to the server x.x.x.x:6443 was refused – did you specify the right host or port?

skatakam

Whenever I start my VMs(ubuntu master and worker running on Virtualbox) and run kubectl get nodes, this is the error I see
The connection to the server x.x.x.x:6443 was refused – did you specify the right host or port?

I tried re joining after kubeadm reset but I lose all my lab progress by doing so. Kindly suggest a fix for this.

devopsappsec

I'd recommend checking to make sure your IPs in your VMs aren't changing. I was using vmware workstation and using NAT, and you can actually set a static ip for your lab nodes. I'm not sure what the process is for virtualbox, but I'm sure there is an equivalent.

Something to check, anyway--might not be your issue.

chrispokorni

Hi @skatakam,

Aside form assigning static IP addresses to your vbox hosts, I'd also recommend that you modify the default IP subnet of vbox to prevent overlapping with the Pod IP network 192.168.0.0/16 managed by calico. Such an overlap will cause cluster DNS issues.

Regards,
-Chris

hemant1405

I too keep getting below.. I have tried all solutions provided by google, but nothing worked. I am using aws ec2

The connection to the server ...:6443 was refused - did you specify the right host or port?

chrispokorni

Hi @hemant1405,

It sounds to be a networking related issue, such as a firewall blocking access to a port.
When provisioning the EC2 instances, did you happen to follow the set up video guide from the intro chapter? It provides step by step instructions on how to configure VM resources, and most importantly networking and firewalls.

Regards,
-Chris

serewicz

Hello,

Could you let us know what command you were running when you got the error? Were you on the control plane, or trying to access from a different system?

I noticed someone mentioned kubeadm reset. This would cause the cluster to be destroyed. It does not destroy the .kube/config file. So after you built a new cluster you may have an issue with TLS keys causing a lack of access, even if the IP address is the same.

Regards,

paristiz

Hi, this has happened to me. In VM you must turn off swap: "sudo swapoff -a"

soramaru

Hello,

This is happening to me as well.
I am also running an AWS EC2 instance, and I did follow all the steps in the setup guide.
I get the message after running the following command.

kubectl apply -f calico.yaml

I have already ran swapoff -a, and I already checked the .yaml files for errors and could not find any.

alihasanahmedk

Please perform the below steps on the master node. It works like charm.
1. sudo -i
2. swapoff -a
3. remove the swap.img from root directory rm -rf /swap.img, if using swap partition exists.
4. exit
5. strace -eopenat kubectl version

Then wait for a few minutes so that all the pods get in running states.

soramaru

@alihasanahmedk said:
Please perform the below steps on the master node. It works like charm.
1. sudo -i
2. swapoff -a
3. remove the swap.img from root directory rm -rf /swap.img, if using swap partition exists.
4. exit
5. strace -eopenat kubectl version

Then wait for a few minutes so that all the pods get in running states.

@alihasanahmedk

Thank you for your input.
Unfortunately that did not work.
I had seen something similar in my search for a solution.
Most of the answers I run across center around SWAP, but I currently don't even have SWAP spaces set up my EC2 instance.

soramaru

kubectl apply -f calico.yaml

5. strace -eopenat kubectl version

When I run both of the above commands, I still get the connection refused message, and they both yield and exit code of 1

alihasanahmedk

No problem @soramaru . I can view your issue on call If you want to. Because that's works for me pretty much fine.

soramaru

@alihasanahmedk

That would be great. What is the best way, and time to get a hold of you ?

alihasanahmedk

@soramaru mail me your free slots. I will let you know then.
alihasanahmedkhan@gmail.com

soramaru

@alihasanahmedk

I am usually free from 1800-2200 on weekdays; however, I am currently located in Japan and operating on GMT +9 time zone. I can also be reached at caspar76@gmail.com

alihasanahmedk

@soramaru
I am located in Pakistan and Japan is 4 hours ahead of Pakistan. I will be available from 1900 to onwards PKT which will be 2300 JST time. Let me know you want to do this on weekends or on weekdays.

soramaru

@alihasanahmedk
I sent you a message to alihasanahmedkhan@gmail.com

alihasanahmedk

@soramaru
sure.

chrispokorni

Hi @soramaru,

Have you created a custom VPC with one SG to allow ingress traffic from all sources, all protocols, and to all ports?

Also, if the cluster happened to be re-initialized, did you manage to copy again the /etc/kubernetes/admin.conf file to your $HOME/.kube directory as instructed by step 16 of Exercise 3.1?

Regards,
-Chris

soramaru

@chrispokorni

Your first suggestion may have been the cause to my problem.

I ended up terminating my EC2 instance and starting a new one, and everything worked fine.

I must have made a mistake when setting up my EC2 instance.