kube-apiserver¶

cluster info¶

detailed information about the overall health of the cluster

kubectl cluster-info
kubectl cluster-info dump

check kube-apiserver is running and hasn't crashed¶

https://discuss.kubernetes.io/t/the-connection-to-the-server-localhost-8080-was-refused-did-you-specify-the-right-host-or-port/1464/3

docker ps | grep kube-apiserver

check kube-apiserver SSL cert¶

https://stackoverflow.com/questions/67132520/kubelet-service-cant-access-kube-apiserver-at-port-6443-with-https-due-to-error

curl -kv https://172.24.4.159:6443

check kube-apiserver health¶

https://kubernetes.io/docs/reference/using-api/health-checks/

curl -k https://localhost:6443/livez?verbose

Why disable swap on kubernetes¶

https://serverfault.com/questions/881517/why-disable-swap-on-kubernetes

• Support for swap is non-trivial.

• Guaranteed pods should never require swap.

• Burstable pods should have their requests met without requiring swap.

• BestEffort pods have no guarantee.

• The kubelet right now lacks the smarts to provide the right amount of predictable behavior here across pods.

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

https://discuss.kubernetes.io/t/the-connection-to-the-server-host-6443-was-refused-did-you-specify-the-right-host-or-port/552

https://www.reddit.com/r/kubernetes/comments/10aya7n/master_node_not_accessible_after_a_few_minutes

• certs expired

  kubeadm alpha certs check-expiration
  

• has config setup correctly?

  mkdir -p $HOME/.kube
  sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
  sudo chown $(id -u):$(id -g) $HOME/.kube/config
  export KUBECONFIG=$HOME/.kube/config
  

• disable SWAP swapoff -a

  #to make it permanent go to /etc/fstab
  sudo -i
  swapoff -a
  exit
  strace -eopenat kubectl version
  sudo systemctl restart kubelet.service
  

  further steps?

  #remove the && and \ and copy and paste each line in the command line, this will ensure swap is disabled and persistent across a reboot.
  sudo dphys-swapfile swapoff && \
  sudo dphys-swapfile uninstall && \
  sudo systemctl disable dphys-swapfile
  

• check if docker was running

• checked if kubelet was running, any log errors?

• disk is under 20GB but it is 100GB already df -H

• reboot the master

• firewall blocks the port

  sudo systemctl status firewalld #redhat centos
  sudo systemctl stop firewalld #redhat, centos
  sudo ufw status verbose #ubuntu
  sudo ufw disable #ubuntu
  

• looking with netstat to see if something is running on port 6443

• kubeadm reset on master and worker nodes?