ShiYu

# Deployment

虽然ReplicaSet可以确保在任何给定时间运行的Pod副本达到指定的数量，但是Deployment（部署）是一个更高级的概念，它管理ReplicaSet并为Pod和ReplicaSet提供声明性更新以及许多其他有用的功能，所以建议在实际使用中，使用Deployment代替ReplicaSet。

如果在Deployment对象中描述了所需的状态，Deployment控制器就会以可控制的速率将实际状态更改为期望状态。也可以在Deployment中创建新的ReplicaSet，或者删除现有的Deployment并使用新的Deployment部署所用的资源。

# 创建Deployment

创建一个Deployment文件，并命名为dc-nginx.yaml，用于部署三个Nginx Pod：

apiVersion: apps/v1
kind: Deployment
metadata:
  name: nginx-deployment
  labels:
    app: nginx
spec:
  replicas: 3
  selector:
    matchLabels:
      app: nginx
  template:
    metadata:
      labels:
        app: nginx
    spec:
      containers:
      - name: nginx
        image: nginx:1.7.9
        ports:
        - containerPort: 80
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21

示例解析：

  • nginx-deployment：Deployment的名称
  • replicas： 创建Pod的副本数
  • selector：定义Deployment如何找到要管理的Pod，与template的label（标签）对应
  • template字段包含以下字段：
    • app: nginx使用label（标签）标记Pod
    • spec：表示Pod运行一个名字为nginx的容器
    • image：运行此Pod使用的镜像
    • Port：容器用于发送和接收流量的端口

使用kubectl create创建此Deployment：

kubectl create -f dc-nginx.yaml 
deployment.apps/nginx-deployment created
1
2

使用kubectl get或者kubectl describe查看此Deployment：

kubectl get deploy
NAME               DESIRED   CURRENT   UP-TO-DATE   AVAILABLE   AGE
nginx-deployment   3         3         3            1           60s
1
2
3

其中：

  • NAME：集群中Deployment的名称
  • DESIRED：应用程序副本数
  • CURRENT：当前正在运行的副本数
  • UP-TO-DATE：显示已达到期望状态的被更新的副本数
  • AVAILABLE：显示用户可以使用的应用程序副本数，当前为1，因为部分Pod仍在创建过程中
  • AGE：显示应用程序运行的时间

查看此时Deployment rollout的状态：

kubectl rollout status deployment/nginx-deployment
deployment "nginx-deployment" successfully rolled out
1
2

再次查看此Deployment：

kubectl get deploy
NAME               DESIRED   CURRENT   UP-TO-DATE   AVAILABLE   AGE
nginx-deployment   3         3         3            3           11m
1
2
3

查看此Deployment创建的ReplicaSet：

kubectl get rs
NAME                          DESIRED   CURRENT   READY   AGE
nginx-deployment-5c689d88bb   3         3         3       12m
1
2
3

查看此Deployment创建的Pod：

kubectl get pods --show-labels
NAME                                READY   STATUS    RESTARTS   AGE   LABELS
nginx-deployment-5c689d88bb-6b95k   1/1     Running   0          13m   app=nginx,pod-template-hash=5c689d88bb
nginx-deployment-5c689d88bb-9z5z2   1/1     Running   0          13m   app=nginx,pod-template-hash=5c689d88bb
nginx-deployment-5c689d88bb-jc8hr   1/1     Running   0          13m   app=nginx,pod-template-hash=5c689d88bb
1
2
3
4
5

# 更新Deployment

一般对应用程序升级或者版本迭代时，会通过Deployment对Pod进行滚动更新。

假如更新Nginx Pod的image使用nginx:1.9.1：

kubectl set image deployment nginx-deployment nginx=nginx:1.9.1 --record
deployment.extensions/nginx-deployment image updated
1
2

当然也可以直接编辑Deployment，效果相同：

kubectl edit deployment.v1.apps/nginx-deployment
deployment.apps/nginx-deployment edited
1
2

使用kubectl rollout status查看更新状态：

kubectl rollout status deployment.v1.apps/nginx-deployment
Waiting for deployment "nginx-deployment" rollout to finish: 1 out of 3 new replicas have been updated...
Waiting for deployment "nginx-deployment" rollout to finish: 2 out of 3 new replicas have been updated...
Waiting for deployment "nginx-deployment" rollout to finish: 2 out of 3 new replicas have been updated...
Waiting for deployment "nginx-deployment" rollout to finish: 2 out of 3 new replicas have been updated...
Waiting for deployment "nginx-deployment" rollout to finish: 1 old replicas are pending termination...
Waiting for deployment "nginx-deployment" rollout to finish: 1 old replicas are pending termination...
deployment "nginx-deployment" successfully rolled out
1
2
3
4
5
6
7
8

查看ReplicaSet：

kubectl get rs 
NAME                          DESIRED   CURRENT   READY   AGE
nginx-deployment-5c689d88bb   0         0         0       34m
nginx-deployment-6987cdb55b   3         3         3       5m14s
1
2
3
4

通过describe查看Deployment的详细信息：

kubectl describe deploy nginx-deployment
Name:                   nginx-deployment
Namespace:              default
CreationTimestamp:      Thu, 24 Jan 2019 15:15:15 +0800
Labels:                 app=nginx
Annotations:            deployment.kubernetes.io/revision: 2
                        kubernetes.io/change-cause: kubectl set image deployment nginx-deployment nginx=nginx:1.9.1 --record=true
Selector:               app=nginx
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=nginx
  Containers:
   nginx:
    Image:        nginx:1.9.1
    Port:         80/TCP
    Host Port:    0/TCP
    Environment:  <none>
    Mounts:       <none>
  Volumes:        <none>
Conditions:
  Type           Status  Reason
  ----           ------  ------
  Available      True    MinimumReplicasAvailable
  Progressing    True    NewReplicaSetAvailable
OldReplicaSets:  <none>
NewReplicaSet:   nginx-deployment-6987cdb55b (3/3 replicas created)
Events:
  Type    Reason             Age    From                   Message
  ----    ------             ----   ----                   -------
  Normal  ScalingReplicaSet  36m    deployment-controller  Scaled up replica set nginx-deployment-5c689d88bb to 3
  Normal  ScalingReplicaSet  7m16s  deployment-controller  Scaled up replica set nginx-deployment-6987cdb55b to 1
  Normal  ScalingReplicaSet  5m18s  deployment-controller  Scaled down replica set nginx-deployment-5c689d88bb to 2
  Normal  ScalingReplicaSet  5m18s  deployment-controller  Scaled up replica set nginx-deployment-6987cdb55b to 2
  Normal  ScalingReplicaSet  4m35s  deployment-controller  Scaled down replica set nginx-deployment-5c689d88bb to 1
  Normal  ScalingReplicaSet  4m34s  deployment-controller  Scaled up replica set nginx-deployment-6987cdb55b to 3
  Normal  ScalingReplicaSet  3m30s  deployment-controller  Scaled down replica set nginx-deployment-5c689d88bb to 0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39

在describe中可以看出，第一次创建时，它创建了一个名为nginx-deployment-5c689d88bb的ReplicaSet，并直接将其扩展为3个副本。更新部署时，它创建了一个新的ReplicaSet，命名为nginx-deployment-6987cdb55b，并将其副本数扩展为1，然后将旧的ReplicaSet缩小为2，这样至少可以有2个Pod可用，最多创建了4个Pod。以此类推，使用相同的滚动更新策略向上和向下扩展新旧ReplicaSet，最终新的ReplicaSet可以拥有3个副本，并将旧的ReplicaSet缩小为0

# 回滚Deployment

当新版本不稳定时，可以对其进行回滚操作，默认情况下，所有Deployment的rollout历史都保留在系统中，可以随时回滚。

假设我们又进行了几次更新：

kubectl set image deployment nginx-deployment nginx=dotbalo/canary:v1 --record
kubectl set image deployment nginx-deployment nginx=dotbalo/canary:v2 --record
1
2

使用kubectl rollout history查看部署历史：

kubectl rollout history deployment/nginx-deployment
deployment.extensions/nginx-deployment 
REVISION  CHANGE-CAUSE
1         <none>
2         kubectl set image deployment nginx-deployment nginx=nginx:1.9.1 --record=true
3         kubectl set image deployment nginx-deployment nginx=dotbalo/canary:v1 --record=true
4         kubectl set image deployment nginx-deployment nginx=dotbalo/canary:v2 --record=true
1
2
3
4
5
6
7

查看Deployment某次更新的详细信息，使用--revision指定版本号：

kubectl rollout history deployment.v1.apps/nginx-deployment --revision=3
deployment.apps/nginx-deployment with revision #3
Pod Template:
  Labels:	app=nginx
	pod-template-hash=645959bf6b
  Annotations:	kubernetes.io/change-cause: kubectl set image deployment nginx-deployment nginx=dotbalo/canary:v1 --record=true
  Containers:
   nginx:
    Image:	dotbalo/canary:v1
    Port:	80/TCP
    Host Port:	0/TCP
    Environment:	<none>
    Mounts:	<none>
  Volumes:	<none>
1
2
3
4
5
6
7
8
9
10
11
12
13
14

使用kubectl rollout undo回滚到上一个版本：

kubectl rollout undo deployment.v1.apps/nginx-deployment
deployment.apps/nginx-deployment
1
2

再次查看更新历史，发现REVISION5回到了canary:v1：

kubectl rollout history deployment/nginx-deployment
deployment.extensions/nginx-deployment 
REVISION  CHANGE-CAUSE
1         <none>
2         kubectl set image deployment nginx-deployment nginx=nginx:1.9.1 --record=true
4         kubectl set image deployment nginx-deployment nginx=dotbalo/canary:v2 --record=true
5         kubectl set image deployment nginx-deployment nginx=dotbalo/canary:v1 --record=true
1
2
3
4
5
6
7

使用--to-revision参数回到指定版本：

kubectl rollout undo deployment/nginx-deployment --to-revision=2
deployment.extensions/nginx-deployment
1
2

# 扩展Deployment

当公司访问量变大，三个Pod已无法支撑业务时，可以对其进行扩展。

使用kubectl scale动态调整Pod的副本数，比如增加Pod为5个：

kubectl scale deployment.v1.apps/nginx-deployment --replicas=5
deployment.apps/nginx-deployment scaled
1
2

查看Pod，此时Pod已经变成了5个：

kubectl get po
NAME                                READY   STATUS    RESTARTS   AGE
nginx-deployment-5f89547d9c-5r56b   1/1     Running   0          90s
nginx-deployment-5f89547d9c-htmn7   1/1     Running   0          25s
nginx-deployment-5f89547d9c-nwxs2   1/1     Running   0          99s
nginx-deployment-5f89547d9c-rpwlg   1/1     Running   0          25s
nginx-deployment-5f89547d9c-vlr5p   1/1     Running   0          95s
1
2
3
4
5
6
7

# 暂停和恢复Deployment更新

Deployment支持暂停更新，用于对Deployment进行多次修改操作。

使用kubectl rollout pause暂停Deployment更新：

kubectl rollout pause deployment/nginx-deployment
deployment.extensions/nginx-deployment paused
1
2

然后对Deployment进行相关更新操作，比如更新镜像，然后对其资源进行限制：

# 更新镜像
kubectl set image deployment.v1.apps/nginx-deployment nginx=nginx:1.9.1
deployment.apps/nginx-deployment image updated
# 资源进行限制
kubectl set resources deployment.v1.apps/nginx-deployment -c=nginx --limits=cpu=200m,memory=512Mi
deployment.apps/nginx-deployment resource requirements updated
1
2
3
4
5
6

通过rollout history可以看到没有新的更新：

kubectl rollout history deployment.v1.apps/nginx-deployment
deployment.apps/nginx-deployment 
REVISION  CHANGE-CAUSE
1         <none>
5         kubectl set image deployment nginx-deployment nginx=dotbalo/canary:v1 --record=true
7         kubectl set image deployment nginx-deployment nginx=dotbalo/canary:v2 --record=true
8         kubectl set image deployment nginx-deployment nginx=dotbalo/canary:v2 --record=true
1
2
3
4
5
6
7

使用kubectl rollout resume恢复Deployment更新：

kubectl rollout resume deployment.v1.apps/nginx-deployment
deployment.apps/nginx-deployment resumed
1
2

可以查看到恢复更新的Deployment创建了一个新的RS（复制集）：

kubectl get rs
NAME                          DESIRED   CURRENT   READY   AGE
nginx-deployment-57895845b8   5         5         4       11s
1
2
3

可以查看Deployment的image（镜像）已经变为nginx:1.9.1

kubectl describe deploy nginx-deployment
Name:                   nginx-deployment
Namespace:              default
CreationTimestamp:      Thu, 24 Jan 2019 15:15:15 +0800
Labels:                 app=nginx
Annotations:            deployment.kubernetes.io/revision: 9
                        kubernetes.io/change-cause: kubectl set image deployment nginx-deployment nginx=dotbalo/canary:v2 --record=true
Selector:               app=nginx
Replicas:               5 desired | 5 updated | 5 total | 5 available | 0 unavailable
StrategyType:           RollingUpdate
MinReadySeconds:        0
RollingUpdateStrategy:  25% max unavailable, 25% max surge
Pod Template:
  Labels:  app=nginx
  Containers:
   nginx:
    Image:      nginx:1.9.1
    Port:       80/TCP
    Host Port:  0/TCP
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19

# 更新Deployment的注意事项

清理策略：

在默认情况下，revision保留10个旧的ReplicaSet，其余的将在后台进行垃圾回收，可以在.spec.revisionHistoryLimit设置保留ReplicaSet的个数。当设置为0时，不保留历史记录

更新策略：

  • .spec.strategy.type==Recreate，表示重建，先删掉旧的Pod再创建新的Pod
  • .spec.strategy.type==RollingUpdate，表示滚动更新，可以指定maxUnavailable和maxSurge来控制滚动更新过程
    • .spec.strategy.rollingUpdate.maxUnavailable，指定在回滚更新时最大不可用的Pod数量，可选字段，默认为25%，可以设置为数字或百分比，如果maxSurge为0，则该值不能为0
    • .spec.strategy.rollingUpdate.maxSurge可以超过期望值的最大Pod数，可选字段，默认为25%，可以设置成数字或百分比，如果maxUnavailable为0，则该值不能为0

← StatefulSet →