管理设计篇之"部署升级策略" [2026重制版]

# recreate-deployment.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
  name: my-app-recreate
spec:
  replicas: 3
  strategy:
    type: Recreate  # 明确指定停机部署
  selector:
    matchLabels:
      app: my-app
  template:
    metadata:
      labels:
        app: my-app
        version: v2.0.0
    spec:
      containers:
        - name: my-app
          image: my-registry/my-app:v2.0.0
          ports:
            - containerPort: 8080
          resources:
            requests:
              cpu: 100m
              memory: 128Mi
            limits:
              cpu: 500m
              memory: 512Mi

# rolling-update-deployment.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
  name: my-app-rolling
spec:
  replicas: 6
  strategy:
    type: RollingUpdate
    rollingUpdate:
      maxSurge: 1          # 滚动更新时最多超出期望副本数
      maxUnavailable: 0   # 滚动更新时最多不可用副本数 (0 = 零宕机)
  selector:
    matchLabels:
      app: my-app
  template:
    metadata:
      labels:
        app: my-app
        version: v2.1.0
    spec:
      # ====== 关键：优雅关闭 ======
      terminationGracePeriodSeconds: 30  # 优雅关闭等待时间
      containers:
        - name: my-app
          image: my-registry/my-app:v2.1.0
          ports:
            - containerPort: 8080
              protocol: TCP
          env:
            - name: POD_NAME
              valueFrom:
                fieldRef:
                  fieldPath: metadata.name
          # ====== 就绪探针 ======
          readinessProbe:
            httpGet:
              path: /health/ready
              port: 8080
            initialDelaySeconds: 5
            periodSeconds: 3
            failureThreshold: 3
            successThreshold: 1
          # ====== 存活探针 ======
          livenessProbe:
            httpGet:
              path: /health/live
              port: 8080
            initialDelaySeconds: 15
            periodSeconds: 10
            failureThreshold: 3
          # ====== 生命周期钩子 ======
          lifecycle:
            preStop:
              exec:
                command: ["/bin/sh", "-c", "sleep 5 && curl -X POST localhost:8080/shutdown"]
          resources:
            requests:
              cpu: 100m
              memory: 128Mi
            limits:
              cpu: 500m
              memory: 512Mi

# blue-green-deployment.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
  name: my-app-blue
  labels:
    app: my-app
    env: blue
    version: v1.0.0
spec:
  replicas: 3
  selector:
    matchLabels:
      app: my-app
      env: blue
  template:
    metadata:
      labels:
        app: my-app
        env: blue
        version: v1.0.0
    spec:
      containers:
        - name: my-app
          image: my-registry/my-app:v1.0.0
          ports:
            - containerPort: 8080
---
apiVersion: apps/v1
kind: Deployment
metadata:
  name: my-app-green
  labels:
    app: my-app
    env: green
    version: v2.0.0
spec:
  replicas: 3
  selector:
    matchLabels:
      app: my-app
      env: green
  template:
    metadata:
      labels:
        app: my-app
        env: green
        version: v2.0.0
    spec:
      containers:
        - name: my-app
          image: my-registry/my-app:v2.0.0
          ports:
            - containerPort: 8080
---
# Service 指向蓝色（当前生产）
apiVersion: v1
kind: Service
metadata:
  name: my-app
spec:
  selector:
    app: my-app
    env: blue  # 当前指向蓝色环境
  ports:
    - port: 80
      targetPort: 8080
  type: ClusterIP

#!/bin/bash
# switch-to-green.sh — 切换流量到绿色环境

set -e

NAMESPACE="default"
APP_NAME="my-app"

echo "=== 开始蓝绿切换 ==="
echo "当前状态:"
kubectl get svc $APP_NAME -n $NAMESPACE -o wide

# 1. 将 Service 的 selector 从 blue 改为 green
echo ""
echo "步骤1: 切换 Service 选择器到绿色环境..."
kubectl patch svc $APP_NAME -n $NAMESPACE -p '{"spec":{"selector":{"env":"green"}}}'

# 2. 等待 Endpoint 更新
echo ""
echo "步骤2: 等待 Endpoint 更新..."
sleep 5

# 3. 验证切换结果
echo ""
echo "步骤3: 验证切换结果..."
kubectl get endpoints $APP_NAME -n $NAMESPACE -o wide

echo ""
echo "=== 蓝绿切换完成! ==="
echo "流量已切换至绿色环境 (v2.0.0)"
echo "如需回滚，执行: kubectl patch svc $APP_NAME -p '{\"spec\":{\"selector\":{\"env\":\"blue\"}}}'"

# canary-deployment.yaml
apiVersion: argoproj.io/v1alpha1
kind: Rollout
metadata:
  name: my-app-canary
spec:
  replicas: 10
  strategy:
    canary:
      # ====== 金丝雀配置 ======
      steps:
        # 步骤1: 暂停，设置金丝雀权重为 10%
        - setWeight: 10
        - pause: {duration: 5m}

        # 步骤2: 增加到 30%
        - setWeight: 30
        - pause: {duration: 10m}

        # 步骤3: 增加到 50%
        - setWeight: 50
        - pause: {duration: 10m}

        # 步骤4: 全量发布
        - setWeight: 100

      # ====== 分析模板（用于自动判断是否推进）=====
      analysis:
        templates:
          - templateName: success-rate
        startingStep: 1  # 从第一步开始分析
        args:
          - name: service-name
            value: my-app-canary

      # ====== 流量路由 ======
      canaryService: my-app-canary-canary  # 金丝雀 Service
      stableService: my-app-canary-stable  # 稳定版 Service

  revisionHistoryLimit: 2
  selector:
    matchLabels:
      app: my-app
  template:
    metadata:
      labels:
        app: my-app
    spec:
      containers:
        - name: my-app
          image: my-registry/my-app:v2.2.0
          ports:
            - containerPort: 8080
          readinessProbe:
            httpGet:
              path: /health/ready
              port: 8080
            initialDelaySeconds: 5
            periodSeconds: 5
---
# AnalysisTemplate — 自动化质量门禁
apiVersion: argoproj.io/v1alpha1
kind: AnalysisTemplate
metadata:
  name: success-rate
spec:
  metrics:
    - name: success-rate
      successCondition: result[0] >= 0.95  # 成功率 >= 95%
      provider:
        prometheus:
          query: |
            sum(rate(http_requests_total{service="my-app-canary",code="200"}[5m]))
            /
            sum(rate(http_requests_total{service="my-app-canary"}[5m]))
      failureLimit: 3  # 连续失败 3 次则中止发布

    - name: error-rate
      successCondition: result[0] <= 0.01  # 错误率 <= 1%
      provider:
        prometheus:
          query: |
            sum(rate(http_requests_total{service="my-app-canary",code=~"5.."}[5m]))
            /
            sum(rate(http_requests_total{service="my-app-canary"}[5m]))
      failureLimit: 3

    - name: p95-latency
      successCondition: result[0] < 500  # P95 延迟 < 500ms
      provider:
        prometheus:
          query: histogram_quantile(0.95,
            sum(rate(http_request_duration_seconds_bucket{service="my-app-canary"}[5m])) by (le))
      failureLimit: 3
---
# 金丝雀 Service
apiVersion: v1
kind: Service
metadata:
  name: my-app-canary-canary
spec:
  selector:
    app: my-app
    # Argo Rollouts 会自动添加 rollouts-pod-template-hash label
  ports:
    - port: 80
      targetPort: 8080
---
# 稳定版 Service
apiVersion: v1
kind: Service
metadata:
  name: my-app-canary-stable
spec:
  selector:
    app: my-app
  ports:
    - port: 80
      targetPort: 8080

# 安装 ArgoCD
kubectl create namespace argocd
kubectl apply -n argocd -f https://raw.githubusercontent.com/argoproj/argo-cd/stable/manifests/install.yaml

# 登录 ArgoCD UI（先做端口转发）
kubectl port-forward svc/argocd-server -n argocd 8080:443
argocd login localhost:8080

# 创建应用（声明式）
kubectl apply -f - <<EOF
apiVersion: argoproj.io/v1alpha1
kind: Application
metadata:
  name: my-app-production
  namespace: argocd
spec:
  project: default
  source:
    repoURL: https://github.com/my-org/app-deployments.git
    targetRevision: main
    path: overlays/production
  destination:
    server: https://kubernetes.default.svc
    namespace: production
  syncPolicy:
    automated:
      prune: true
      selfHeal: true
      allowEmpty: false
    syncOptions:
      - CreateNamespace=true
      - PrunePropagationPolicy=foreground
    retry:
      limit: 5
      backoff:
        duration: 5s
        factor: 2
        maxDuration: 3m
EOF

# overlays/production/kustomization.yaml — 生产环境叠加配置
namespace: production
resources:
  - ../../base
labels:
  - pairs:
      environment: production
      team: platform

configMapGenerator:
  - name: app-config
    behavior: merge
    literals:
      - LOG_LEVEL=INFO
      - DB_POOL_SIZE=20
      - CACHE_TTL=3600

images:
  - name: my-app
    newTag: v2.2.0

replicas: 6

┌─────────────────────────────────────────────┐
│           2026 推荐部署技术栈                  │
├─────────────────────────────────────────────┤
│  编排引擎: Kubernetes 1.35+                   │
│  GitOps 工具: ArgoCD 3.5+ 或 Flux 2.9+      │
│  发布策略: Argo Rollouts (金丝雀)             │
│  配置管理: Kustomize / Helm                  │
│  秘钥管理: External Secrets Operator         │
│  监控告警: Prometheus + Grafana + AlertManager │
│  日志收集: Loki / ELK                        │
│  追踪系统: Jaeger / Tempo                    │
│  混沌工程: Chaos Mesh / Litmus               │
└─────────────────────────────────────────────┘