昨天跟各位介紹完一些關於 Deployment 的觀念後，今天不免俗的要再來跟各位介紹一些實作的技巧啦！

那我們廢話不多說，馬上開始我們今天的介紹吧！

首先我們一樣先建立一個屬於這個 Deployment 的 YAML 檔：

#test_deployment.yaml

apiVersion: apps/v1
kind: Deployment
metadata:
  name: ubuntu-deployment
  labels:
    app: test
spec:
  replicas: 3 # 建立三個有 ubuntu image 的 pod
  selector:
    matchLabels:
      app: test
  template:
    metadata:
      labels:
        app: test
    spec:
      containers:
      - name: ubuntu
        image: ubuntu:20.04
        args: [bash, -c, 'for ((i = 0; ; i++)); do echo "$i: $(date)"; sleep 100; done']

建立完 YAML 檔後，一樣將它建立於 Kubernetes 中：

kubectl apply -f test_deployment.yaml
deployment.apps/ubuntu-deployment created

這時我們可以來看看是否成功建立：

kubectl get deployment
NAME                READY   UP-TO-DATE   AVAILABLE   AGE
ubuntu-deployment   3/3     3            3           16s

kubectl get pod 
NAME                                 READY   STATUS    RESTARTS   AGE
ubuntu-deployment-55ffcc9677-85g9g   1/1     Running   0          5s
ubuntu-deployment-55ffcc9677-9bqgw   1/1     Running   0          5s
ubuntu-deployment-55ffcc9677-fpsvv   1/1     Running   0          5s

READY : 就緒個數/總數。
UP-TO-DATE : 有幾個 Pod 副本已經 onboard。
AVAILABLE : 目前有多少 Pod 副本可以運作。
Age : pod運行時間。

正常來說，出現以上畫面就代表你已經成功建立了！

我們在前一天有說過，他還有一個功能是滾動升級和回滾應用，這邊我們可以來嘗試看看：

kubectl set image deployment/ubuntu-deployment ubuntu=ubuntu:22.04
deployment.apps/ubuntu-deployment image updated

這段跑完後，他會顯示它已經更新了，這邊我們來驗證看看是否成功將 ubuntu 升級到 22.04 的版本。

# 確認已全部完成更新
ubuntu@ubuntu:~$ kubectl get deployment
NAME                READY   UP-TO-DATE   AVAILABLE   AGE
ubuntu-deployment   3/3     3            3           91m

# 查詢 Pod 的資訊
ubuntu@ubuntu:~$ kubectl describe deployment
Name:                   ubuntu-deployment
Namespace:              default
CreationTimestamp:      Tue, 13 Sep 2022 09:00:52 +0000
Labels:                 app=test
Annotations:            deployment.kubernetes.io/revision: 2
Selector:               app=test
Replicas:               3 desired | 3 updated | 3 total | 3 available | 0 unavailable
StrategyType:           RollingUpdate
MinReadySeconds:        0
RollingUpdateStrategy:  25% max unavailable, 25% max surge
Pod Template:
  Labels:  app=test
  Containers:
   ubuntu:
    Image:      ubuntu:22.04 
    Port:       <none>
    Host Port:  <none>
    Args:
      bash
      -c
      for ((i = 0; ; i++)); do echo "$i: $(date)"; sleep 100; done
    Environment:  <none>
    Mounts:       <none>
  Volumes:        <none>
Conditions:
  Type           Status  Reason
  ----           ------  ------
  Available      True    MinimumReplicasAvailable
  Progressing    True    NewReplicaSetAvailable
OldReplicaSets:  <none>
NewReplicaSet:   ubuntu-deployment-869bd6f4c (3/3 replicas created)
Events:
  Type    Reason             Age   From                   Message
  ----    ------             ----  ----                   -------
  Normal  ScalingReplicaSet  57s   deployment-controller  Scaled up replica set ubuntu-deployment-869bd6f4c to 1
  Normal  ScalingReplicaSet  46s   deployment-controller  Scaled down replica set ubuntu-deployment-55ffcc9677 to 2
  Normal  ScalingReplicaSet  46s   deployment-controller  Scaled up replica set ubuntu-deployment-869bd6f4c to 2
  Normal  ScalingReplicaSet  45s   deployment-controller  Scaled down replica set ubuntu-deployment-55ffcc9677 to 1
  Normal  ScalingReplicaSet  45s   deployment-controller  Scaled up replica set ubuntu-deployment-869bd6f4c to 3
  Normal  ScalingReplicaSet  44s   deployment-controller  Scaled down replica set ubuntu-deployment-55ffcc9677 to 0

這邊我們觀察 Pod Templete 裡面的 Image 會發現已成功變成 ubuntu:22.04，所以已成功的更新了。

可是如果我們在運作時，發現此版本相比上一版太不穩定，想回到上一版的話也可以，可以用以下的指令：

ubuntu@ubuntu:~$ kubectl rollout undo deployment/ubuntu-deployment
deployment.apps/ubuntu-deployment rolled back

# 再確認一次是否回到原本的20.04
ubuntu@ubuntu:~$ kubectl describe deployments
Name:                   ubuntu-deployment
Namespace:              default
CreationTimestamp:      Tue, 13 Sep 2022 09:00:52 +0000
Labels:                 app=test
Annotations:            deployment.kubernetes.io/revision: 3
Selector:               app=test
Replicas:               3 desired | 3 updated | 3 total | 3 available | 0 unavailable
StrategyType:           RollingUpdate
MinReadySeconds:        0
RollingUpdateStrategy:  25% max unavailable, 25% max surge
Pod Template:
  Labels:  app=test
  Containers:
   ubuntu:
    Image:      ubuntu:20.04
    Port:       <none>
    Host Port:  <none>
    Args:
      bash
      -c
      for ((i = 0; ; i++)); do echo "$i: $(date)"; sleep 100; done
    Environment:  <none>
    Mounts:       <none>
  Volumes:        <none>
Conditions:
  Type           Status  Reason
  ----           ------  ------
  Available      True    MinimumReplicasAvailable
  Progressing    True    NewReplicaSetAvailable
OldReplicaSets:  <none>
NewReplicaSet:   ubuntu-deployment-55ffcc9677 (3/3 replicas created)
Events:
  Type    Reason             Age                From                   Message
  ----    ------             ----               ----                   -------
  Normal  ScalingReplicaSet  15m                deployment-controller  Scaled up replica set ubuntu-deployment-869bd6f4c to 1
  Normal  ScalingReplicaSet  15m                deployment-controller  Scaled down replica set ubuntu-deployment-55ffcc9677 to 2
  Normal  ScalingReplicaSet  15m                deployment-controller  Scaled up replica set ubuntu-deployment-869bd6f4c to 2
  Normal  ScalingReplicaSet  15m                deployment-controller  Scaled down replica set ubuntu-deployment-55ffcc9677 to 1
  Normal  ScalingReplicaSet  15m                deployment-controller  Scaled up replica set ubuntu-deployment-869bd6f4c to 3
  Normal  ScalingReplicaSet  15m                deployment-controller  Scaled down replica set ubuntu-deployment-55ffcc9677 to 0
  Normal  ScalingReplicaSet  4s                 deployment-controller  Scaled up replica set ubuntu-deployment-55ffcc9677 to 1
  Normal  ScalingReplicaSet  3s                 deployment-controller  Scaled down replica set ubuntu-deployment-869bd6f4c to 2
  Normal  ScalingReplicaSet  3s                 deployment-controller  Scaled up replica set ubuntu-deployment-55ffcc9677 to 2
  Normal  ScalingReplicaSet  2s (x2 over 106m)  deployment-controller  Scaled up replica set ubuntu-deployment-55ffcc9677 to 3
  Normal  ScalingReplicaSet  2s                 deployment-controller  Scaled down replica set ubuntu-deployment-869bd6f4c to 1
  Normal  ScalingReplicaSet  1s                 deployment-controller  Scaled down replica set ubuntu-deployment-869bd6f4c to 0

這邊我們會發現 image 已成功回到了 20.04了。

還有當我們可能需要的 Pod 數量可能不夠或因為太多有點佔用到資源的話，我們也可以進行擴容和縮容，現在就讓我來為你們介紹。

如果希望 Pod 數可以增加的話，可以使用以下的指令：

ubuntu@ubuntu:~$ kubectl scale deployment ubuntu-deployment --replicas 5
deployment.apps/ubuntu-deployment scaled
# 查詢 Deployment 裡面含有的 Pod 數
ubuntu@ubuntu:~$ kubectl get deployment
NAME                READY   UP-TO-DATE   AVAILABLE   AGE
ubuntu-deployment   5/5     5            5           135m

這邊我們看到已成功擴充了 Pod 的數量了。
相對的，如果有縮減的話，就是直接修改後面 replicas 的參數就 ok 了。

到這邊，已經差不多把一些基本的 Deployment 的操作介紹完了，讀者們可以試著自己做做看，依照自己的需求來配置自己的 Deployment。

那我們今天就先介紹到這邊吧，大家掰掰！