> 本文作者：丁辉

# Deployment的使用

[官方文档](https://kubernetes.io/zh-cn/docs/tutorials/kubernetes-basics/deploy-app/deploy-intro/)

## 命令部署一个应用

```bash
kubectl create deployment nginx --image=nginx:alpine
```

## Yaml 部署一个 Deployment 应用

1. 部署应用(最小化 Yaml)

   ```yaml
   cat <<EOF | kubectl apply -f -
   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:alpine
           ports:
           - containerPort: 80
   EOF
   ```

2. 验证

   ```bash
   kubectl get deployments
   ```

## 更新和回滚

> Deployment 支持两种更新策略：滚动更新（RollingUpdate，默认）和删除式更新（Recreate，又称单批次更新）。

- ReCreate

  **删除式更新（Recreate）**：更新时先删除所有运行中的 Pod，待其彻底终止后创建新 ReplicaSet 及对应 Pod，更新期间服务暂不可用。

  ```yaml
  spec:
    strategy:
      type: Recreate
  ```

- RollingUpdate

  **滚动更新（RollingUpdate）**：分批更新 Pod，待一批更新后的 Pod 就绪后再更新下一批，实现服务不中断；更新过程中会存在新老版本应用共存的情况，不同客户端可能获取不同版本响应。

  ```yaml
  spec:
    strategy:
      type: RollingUpdate
  ```

### 更新

1. 更新镜像

   在这里，`deployment/nginx-deployment` 表明 Deployment 的名称，`nginx` 表明需要进行更新的容器， 而 `nginx:latest` 则表示镜像的新版本以及它的标签。

   ```bash
   kubectl set image deployment/nginx-deployment nginx=nginx:latest
   ```

2. 查看进度

   ```bash
   kubectl rollout status deployment/nginx-deployment
   ```

3. 查看状态

   ```bash
   kubectl get rs
   ```

### 回滚

1. 查看 Deployment 修订历史

   ```bash
   kubectl rollout history deployment/nginx-deployment
   ```

   设置 `CHANGE-CAUSE` 消息

   ```bash
   kubectl annotate deployment/nginx-deployment kubernetes.io/change-cause="image updated to latest"
   ```

2. 查看修订历史的详细信息

   ```bash
   kubectl rollout history deployment/nginx-deployment --revision=2
   ```

3. 回滚

   ```bash
   kubectl rollout undo deployment/nginx-deployment
   ```

   或回滚到特定版本

   ```bash
   kubectl rollout undo deployment/nginx-deployment --to-revision=2
   ```

### 重启

```bash
kubectl rollout restart deployment deployment/nginx-deployment
```

### 副本数调整

```bash
kubectl scale deployment/nginx-deployment --replicas=3
```

## 暂停、恢复 Deployment 的上线过程

1. 暂停上线

   ```bash
   kubectl rollout pause deployment/nginx-deployment
   ```

2. 更新 Deployment 镜像

   ```bash
   kubectl set image deployment/nginx-deployment nginx=nginx:latest
   ```

   你可以根据需要执行很多更新操作，例如，可以要使用的资源

   ```bash
   kubectl set resources deployment/nginx-deployment -c=nginx --limits=cpu=200m,memory=512Mi
   ```

3. 最终，恢复 Deployment 上线并观察新的 ReplicaSet 的创建过程，其中包含了所应用的所有更新

   ```bash
   kubectl rollout resume deployment/nginx-deployment
   ```

4. 监视上线的状态，直到完成

   ```bash
   watch kubectl get rs
   ```

## 更新失败标记为 **"Failed"** 状态

```bash
kubectl patch deployment/nginx-deployment -p '{"spec":{"progressDeadlineSeconds":600}}'
```

`progressDeadlineSeconds`

- **作用**: 定义 Deployment 滚动更新（Rolling Update）的最大允许时间
- **值 600**: 表示 600 秒（1 分钟）
- **默认值**: 通常为 600 秒，但显式设置可确保明确性
- **工作逻辑**:
  1. 当 Deployment 开始更新（如镜像版本升级）时启动计时
  2. 如果更新过程超过 600 秒仍未完成
  3. Kubernetes 会将 Deployment 标记为 **"Failed"** 状态
  4. 并自动回滚到之前的稳定版本（如果配置了回滚策略）

## 配置 ReplicaSet 清理策略

你可以在 Deployment 中设置 `.spec.revisionHistoryLimit` 字段以指定保留此 Deployment 的多少个旧有 ReplicaSet。其余的 ReplicaSet 将在后台被垃圾回收。 默认情况下，此值为 10。

```bash
kubectl patch deployment/nginx-deployment -p '{"spec":{"revisionHistoryLimit":10}}'
```

## 金丝雀部署

[官方文档](https://kubernetes.io/zh-cn/docs/concepts/workloads/controllers/deployment/#canary-deployment)

### 通过副本数控制

**操作步骤：**

1. 先部署稳定版本（v1，9个副本）
2. 部署金丝雀版本（v2，1个副本）
3. Service会自动将约9%的流量（1/10）导向v2
4. 观察监控，如果v2正常，逐步增加v2副本数，减少v1副本数

1. 部署 Pod-1

   ```yaml
   cat <<EOF | kubectl apply -f -
   apiVersion: apps/v1
   kind: Deployment
   metadata:
     name: nginx-deployment-1
     labels:
       app: nginx
   spec:
     replicas: 9
     strategy:
       type: RollingUpdate
     selector:
       matchLabels:
         app: nginx
     template:
       metadata:
         labels:
           app: nginx
           version: v1
       spec:
         containers:
         - name: nginx
           image: registry.cn-hangzhou.aliyuncs.com/offends/demo:nginx-alpine-v1
           ports:
           - containerPort: 80
   EOF
   ```

2. 部署 Pod-2

   ```yaml
   cat <<EOF | kubectl apply -f -
   apiVersion: apps/v1
   kind: Deployment
   metadata:
     name: nginx-deployment-2
     labels:
       app: nginx
   spec:
     replicas: 1
     strategy:
       type: RollingUpdate
     selector:
       matchLabels:
         app: nginx
     template:
       metadata:
         labels:
           app: nginx
           version: v2
       spec:
         containers:
         - name: nginx
           image: registry.cn-hangzhou.aliyuncs.com/offends/demo:nginx-alpine-v2
           ports:
           - containerPort: 80
   EOF
   ```

3. 部署 Service（同时选择v1和v2的Pod）

   ```yaml
   cat <<EOF | kubectl apply -f -
   apiVersion: v1
   kind: Service
   metadata:
     name: nginx-deployment-service
   spec:
     selector:
       app: nginx  # 同时选择两个版本的Pod
     ports:
     - port: 80
       targetPort: 80
   EOF
   ```

4. 修改权重将流量导向 v2

   - 增加 v2

     ```bash
     kubectl scale deployment/nginx-deployment-2 --replicas=10
     ```

   - 降低 v1

     ```bash
     kubectl scale deployment/nginx-deployment-1 --replicas=0
     ```

### 通过 Ingress 实现基于权重的金丝雀

方法类似于前者，只是对象从副本数变换为 Ingress控制

1. 创建两个 Service

   ```bash
   cat <<EOF | kubectl apply -f -
   apiVersion: v1
   kind: Service
   metadata:
     name: nginx-deployment-service-1
   spec:
     selector:
       app: nginx
       version: v1
     ports:
     - port: 80
       targetPort: 80
   ---
   apiVersion: v1
   kind: Service
   metadata:
     name: nginx-deployment-service-2
   spec:
     selector:
       app: nginx
       version: v2
     ports:
     - port: 80
       targetPort: 80
   EOF
   ```

2. 创建 Ingress 资源

   ```yaml
   cat <<EOF | kubectl apply -f -
   # 金丝雀 Ingress 对象为 v1 版本
   apiVersion: networking.k8s.io/v1
   kind: Ingress
   metadata:
     name: nginx-ingress-1
   spec:
     rules:
     - host: demo.offends.cn
       http:
         paths:
         - path: /
           pathType: Prefix
           backend:
             service:
               name: nginx-deployment-service-1
               port:
                 number: 80
   ---
   # 金丝雀 Ingress 对象为 v2 版本
   apiVersion: networking.k8s.io/v1
   kind: Ingress
   metadata:
     name: nginx-ingress-2
     annotations:
       nginx.ingress.kubernetes.io/canary: "true"
       nginx.ingress.kubernetes.io/canary-weight: "10" # 10%流量到金丝雀
   spec:
     rules:
     - host: demo.offends.cn
       http:
         paths:
         - path: /
           pathType: Prefix
           backend:
             service:
               name: nginx-deployment-service-2
               port:
                 number: 80
   EOF
   ```

3. 慢慢调整 nginx-ingress-2 流量到 100%，即完成更新