资料来源: https://github.com/walidshaari/Kubernetes-Certified-Administrator

先执行 source (kubectl completion bash)

---

提示

• 练习、练习、再练习

• 熟悉文档，主要包括 concepts 和 tasks
  https://kubernetes.io/docs/concepts/
  https://kubernetes.io/docs/tasks/

• 尽量熟悉 kubectl 命令，例如

  kubectl explain --recursive
  source <(kubectl completion bash)
  alias k=kubectl
  complete -F __start_kubectl k
  kubectl get pod -A -owide --show-labels --sort-by='{.metadata.name}'
  kubectl describe pod {podname}
  kubectl logs {podname}
  kubectl run testpod --image=nginx --dry-run=client -oyaml
  

---

Cluster Architecture, Installation, and Configuration 25%

etcd backup and restore brief

export ETCDCTL_API=3  # needed to specify etcd api versions, not sure if it is needed anylonger with k8s 1.19+ 
etcdctl snapshot save -h   #find save options
etcdctl snapshot restore -h  #find restore options

# possible example of save, options will change depending on cluster context, as TLS is used need to give ca,crt, and key paths
etcdctl snapshot save /backup/snapshot.db  \
  --cert=/etc/kubernetes/pki/etcd/server.crt  \
  --key=/etc/kubernetes/pki/etcd/server.key \
  --cacert=/etc/kubernetes/pki/etcd/ca.crt

evicting pods/nodes and bringing back node back to cluster

kubectl drain  <node> # to drain a node
kubectl uncordon  <node> # to return a node after updates back to the cluster from unscheduled state to Ready
kubectl cordon  <node>   # to not schedule new pods on a node

upgrade kubernetes worker node

kubectl drain <node>
apt-get upgrade -y kubeadm=<k8s-version-to-upgrade>
apt-get upgrade -y kubelet=<k8s-version-to-upgrade>
kubeadm upgrade node config --kubelet-version <k8s-version-to-upgrade>
systemctl restart kubelet
kubectl uncordon <node>

kubeadm upgrade steps

kubeadm upgrade plan
kubeadm upgrade apply

---

Troubleshooting – 30%

如何监控应用？

https://kubernetes.io/docs/tasks/debug-application-cluster/resource-usage-monitoring/

应用为什么失败？

https://kubernetes.io/docs/tasks/debug-application-cluster/debug-application/

集群为什么出问题了？

https://kubernetes.io/docs/tasks/debug-application-cluster/debug-cluster/

DNS为什么不能解析了？

https://kubernetes.io/docs/tasks/administer-cluster/dns-debugging-resolution/