灰度与蓝绿：风险可控的发布——流量切分、指标回滚与版本管理策略

graph TD
    A[传统停机发布] --> B[蓝绿发布]
    B --> C[灰度发布]
    C --> D[功能开关发布]
    D --> E[影子测试]
    
    style A fill:#f9d5c8
    style B fill:#c8e6f5
    style C fill:#d4edda
    style D fill:#f0e6f5
    style E fill:#fff2cc

# 蓝色环境（当前生产版本）
apiVersion: apps/v1
kind: Deployment
metadata:
  name: app-blue
spec:
  replicas: 3
  selector:
    matchLabels:
      app: my-app
      version: blue  # 版本标识
  template:
    metadata:
      labels:
        app: my-app
        version: blue
    spec:
      containers:
      - name: app
        image: my-app:v1.0.0
        ports:
        - containerPort: 8080

# 绿色环境（新版本）
apiVersion: apps/v1
kind: Deployment
metadata:
  name: app-green
spec:
  replicas: 3
  selector:
    matchLabels:
      app: my-app
      version: green  # 版本标识
  template:
    metadata:
      labels:
        app: my-app
        version: green
    spec:
      containers:
      - name: app
        image: my-app:v1.1.0
        ports:
        - containerPort: 8080

# Service配置，通过修改selector实现切换
apiVersion: v1
kind: Service
metadata:
  name: app-service
spec:
  ports:
  - port: 80
    targetPort: 8080
  selector:
    app: my-app
    version: blue  # 初始指向蓝色环境
  type: LoadBalancer

# 从蓝色切换到绿色环境
kubectl patch service app-service -p '{"spec":{"selector":{"version":"green"}}}'

# 快速回滚到蓝色环境
kubectl patch service app-service -p '{"spec":{"selector":{"version":"blue"}}}'

# v1版本（现有版本）
apiVersion: apps/v1
kind: Deployment
metadata:
  name: app-v1
spec:
  replicas: 9  # 90%流量
  selector:
    matchLabels:
      app: my-app
      version: v1.0
  template:
    metadata:
      labels:
        app: my-app
        version: v1.0
    # ... 其他配置

# v2版本（新版本）
apiVersion: apps/v1
kind: Deployment
metadata:
  name: app-v2
spec:
  replicas: 1  # 10%流量
  selector:
    matchLabels:
      app: my-app
      version: v1.1
  template:
    metadata:
      labels:
        app: my-app
        version: v1.1
    # ... 其他配置

# Service配置，同时选择两个版本
apiVersion: v1
kind: Service
metadata:
  name: app-service
spec:
  ports:
  - port: 80
    targetPort: 8080
  selector:
    app: my-app  # 不指定版本，选择所有匹配的Pod
  type: LoadBalancer

apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
  name: app-canary-ingress
  annotations:
    nginx.ingress.kubernetes.io/canary: "true"
    nginx.ingress.kubernetes.io/canary-weight: "10"  # 10%流量到新版本
    nginx.ingress.kubernetes.io/canary-by-header: "X-Canary"  # 基于Header
    nginx.ingress.kubernetes.io/canary-by-header-value: "true"
spec:
  rules:
  - http:
      paths:
      - path: /
        pathType: Prefix
        backend:
          service:
            name: app-service
            port:
              number: 80

graph LR
    A[内部测试 1%] --> B[特定用户 5%]
    B --> C[小范围用户 20%]
    C --> D[半数用户 50%]
    D --> E[全量发布 100%]
    
    style A fill:#ffcccc
    style B fill:#ffebcc
    style C fill:#ffffcc
    style D fill:#ebffcc
    style E fill:#ccffcc

# 使用Istio进行权重配置
apiVersion: networking.istio.io/v1alpha3
kind: VirtualService
metadata:
  name: app-virtual-service
spec:
  hosts:
  - app.example.com
  http:
  - route:
    - destination:
        host: app-service
        subset: v1
      weight: 90  # 90%流量到v1
    - destination:
        host: app-service
        subset: v2
      weight: 10  # 10%流量到v2

# Kruise Rollout的自动化验收配置示例
apiVersion: rollouts.kruise.io/v1alpha1
kind: Rollout
metadata:
  name: app-rollout
spec:
  strategy:
    canary:
      steps:
      - weight: 10
        pause: {duration: 300}  # 暂停5分钟进行验证
      - weight: 30
        pause: {duration: 600}
      - weight: 100
        pause: {duration: 0}
      metrics:
      - name: error-rate
        threshold: "5"  # 错误率阈值5%
        interval: 60s   # 每60秒检查一次
      - name: p99-latency  
        threshold: "500"  # P99延迟阈值500ms
        interval: 60s