# Core Concept
###### tags: `CKA`

---

[TOC]

---
## [Overview](https://kubernetes.io/docs/concepts/overview/)

- Service discovery and load balancing
基本上就是可以將內部服務透過 DNS/IP 來提供給外部使用，同時具有負載平衡機制來分配流量，使整體服務較穩定
- Storage orchestration
想 mount 在哪就去哪~
mount local -> Yes
mount cloud (e.g., GCP, AWS, Azure) -> Hell Yeah
- Automated rollouts and rollbacks
躺著玩趴著玩坐著玩，就是好好玩
提供你一個依你需求進行部署及回滾的好架構
- Automatic bin packing
我有酒，你有故事嗎˙ˇ˙
建立一個完整的 cluster 後，要部署或是執行 job 的時候就可以直接透過給定所需的條件(CPU, Mem etc.)來讓 k8s 幫你部署到最適合的環境中
- Self-healing
壞了？~~換女友~~ 換新的；不理我？換新的
可以自動去檢測容器的狀態，然後進行重啟或刪除
- Secret and configuration management
誒嘿，躲貓貓
將一些機敏性的資料或是連線資訊額外儲存，讓他可以不受服務的下上影響單獨更新，也不用把這些資料裸奔在外層

## Architecture





在 k8s 中啟動服務的方式(kubeadm init)：
https://kubernetes.io/docs/reference/setup-tools/kubeadm/kubeadm-init/

---
### [ETCD](https://kubernetes.io/docs/concepts/overview/components/#etcd)
`Control Plane`

store k8s cluster data
record whole cluster status, info
Key value storage
:::warning
If your Kubernetes cluster uses etcd as its backing store, make sure you have a back up plan for those data.
:::


---
### [kube-apiserver](https://kubernetes.io/docs/concepts/overview/components/#kube-apiserver)

`Control Plane`

K8s front end ->
execute kubectl command
node 溝通的友善橋樑
Which means
:::danger
if apiserver crashed -> Boom Boom Boom
so, scale horizontally to avoid the bomb
:::



---
### [Kube Controller Manager](https://kubernetes.io/docs/concepts/overview/components/#kube-controller-manager)
`Control Plane`

Monitoring the status

簡單的幾個 controller
- Node controller
觀察並回報 node 的狀態
- Job controller
觀察是否有 job 的執行條件被滿足，若有的話就 schedule 一個 pod 來執行
- EndpointSlice controller
觀察 service 與 pod 之間的狀態
- ServiceAccount controller
管理 namespace 底下的 ServiceAccount，剛建好 namespace 時會產生一個 default 的 ServiceAccount

---
### [Cloud Controller Manager](https://kubernetes.io/docs/concepts/overview/components/#cloud-controller-manager)

結合 cloud provider 的 controller manager
:::warning
The cloud-controller-manager only runs controllers that are specific to your cloud provider.
:::

基本的 ccm
- Node controller
For checking the cloud provider to determine if a node has been deleted in the cloud after it stops responding
- Route controller
For setting up routes in the underlying cloud infrastructure
- Service controller
For creating, updating and deleting cloud provider load balancers

---
### Kube Scheduler
`Control Plane`

叫你去哪，你就去哪
依據不同的需求條件，來決定要被部署的 pod 應該要去哪個 node 上

詳細的內容，請看佩璇show

---
### Kubelet

船長，指引前路
有哥在，就安心了

apiserver 的接口，負責回報現在這個 node 上所有的狀況，並執行指令們

kebectl create pod
kubectl delete pod

---
### Kube Proxy

就是 network proxy，負責維護 node's network rules，來決定這個node 的網路要怎麼跑


---
### Container Runtime

The container runtime is the software that is responsible for running containers.

Kubernetes supports container runtimes such as containerd, CRI-O, and any other implementation of the Kubernetes CRI (Container Runtime Interface).