Webhook 多区域部署与灰度发布：全球低延迟、金丝雀、蓝绿策略（2025 架构指南）

# 为每个区域创建 Record Set
# 记录类型: A (Alias 到 ALB)
# 路由策略: Latency
# 区域: us-east-1, eu-central-1, ap-southeast-1

# 等效 CLI
aws route53 change-resource-record-sets \
  --hosted-zone-id Z123456789 \
  --change-batch file://latency-routing.json

{
  "Changes": [
    {
      "Action": "CREATE",
      "ResourceRecordSet": {
        "Name": "webhooks.yourapp.com",
        "Type": "A",
        "SetIdentifier": "us-east-1",
        "Region": "us-east-1",
        "AliasTarget": {
          "HostedZoneId": "Z35SXDOTRQ7X7K",
          "DNSName": "us-east-alb-123.us-east-1.elb.amazonaws.com.",
          "EvaluateTargetHealth": true
        }
      }
    },
    {
      "Action": "CREATE",
      "ResourceRecordSet": {
        "Name": "webhooks.yourapp.com",
        "Type": "A",
        "SetIdentifier": "eu-central-1",
        "Region": "eu-central-1",
        "AliasTarget": {
          "HostedZoneId": "Z215JYRZR1TBD5",
          "DNSName": "eu-central-alb-456.eu-central-1.elb.amazonaws.com.",
          "EvaluateTargetHealth": true
        }
      }
    }
  ]
}

// 使用 Redis Cluster 实现跨区域去重
// 每个区域的 Gateway 连接到同一个 Redis Cluster

func isDuplicateCrossRegion(ctx context.Context, eventID string) (bool, error) {
    // Redis Cluster 自动路由到正确分片
    key := fmt.Sprintf("webhook:idempotency:%s", eventID)
    ok, err := rdb.SetNX(ctx, key, "1", 24*time.Hour).Result()
    if err != nil {
        return false, err
    }
    return !ok, nil // ok=true 表示 key 不存在，插入成功 → 非重复
}

# redis-cluster.yaml (简化)
redis-cluster:
  nodes:
    - name: redis-master-1
      region: us-east
    - name: redis-master-2
      region: eu-central
    - name: redis-master-3
      region: ap-southeast
  replicas: 1  # 每个主节点1个副本，分布在不同区域

// 基于租户 ID 的灰度路由
type CanaryRouter struct {
    rolloutPercentage int    // 5, 10, 25, 50, 100
    canaryVersion     string // "v2.1.0"
}

func (c *CanaryRouter) Route(tenantID string) string {
    // 一致性哈希：同一租户始终路由到同一版本
    hash := fnv1a(tenantID)
    if hash%100 < uint32(c.rolloutPercentage) {
        return c.canaryVersion
    }
    return "stable"
}

// FNV-1a 哈希（简单高效）
func fnv1a(s string) uint32 {
    var hash uint32 = 2166136261
    for i := 0; i < len(s); i++ {
        hash ^= uint32(s[i])
        hash *= 16777619
    }
    return hash
}

# blue-green-service.yaml
apiVersion: v1
kind: Service
metadata:
  name: webhook-gateway
  labels:
    version: blue  # 切换时改为 green
spec:
  selector:
    app: webhook-gateway
    version: blue  # 指向当前版本
  ports:
    - port: 80
      targetPort: 8080
---
# 同时部署 green 版本，但不加入 Service
apiVersion: apps/v1
kind: Deployment
metadata:
  name: webhook-gateway-green
spec:
  replicas: 3
  selector:
    matchLabels:
      app: webhook-gateway
      version: green
  template:
    metadata:
      labels:
        app: webhook-gateway
        version: green
    spec:
      containers:
        - name: gateway
          image: webhook-gateway:v2.1.0

kubectl patch service webhook-gateway \
  -p '{"spec":{"selector":{"version":"green"}}}'

func (c *CanaryRouter) RouteByProvider(provider string) string {
    // 白名单：这些 provider 永远走稳定版
    if provider == "stripe" || provider == "alipay" {
        return "stable"
    }
    // 其他 provider 按百分比灰度
    return c.RouteRandom()
}

type RegionHealth struct {
    Region     string
    LastCheck  time.Time
    SuccessRate float64
    LatencyP99  time.Duration
    Healthy    bool
}

type FailoverManager struct {
    regions map[string]*RegionHealth
    mu      sync.RWMutex
}

func (fm *FailoverManager) CheckHealth(region string) error {
    ctx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
    defer cancel()

    req, _ := http.NewRequestWithContext(ctx, "GET", 
        fmt.Sprintf("https://%s/webhooks/health", region), nil)
    resp, err := http.DefaultClient.Do(req)

    fm.mu.Lock()
    defer fm.mu.Unlock()

    rh := fm.regions[region]
    rh.LastCheck = time.Now()

    if err != nil || resp.StatusCode != 200 {
        rh.SuccessRate = rh.SuccessRate*0.9 // 指数衰减
        if rh.SuccessRate < 0.5 {
            rh.Healthy = false
        }
        return err
    }

    rh.SuccessRate = min(1.0, rh.SuccessRate*1.1+0.1)
    rh.Healthy = rh.SuccessRate > 0.8
    return nil
}

# 摘除故障区域
curl -X PATCH "https://api.cloudflare.com/client/v4/zones/$ZONE/load_balancers/$LB_ID" \
  -H "Authorization: Bearer $TOKEN" \
  -H "Content-Type: application/json" \
  --data '{
    "fallback_pool": "us-east-pool",
    "default_pools": ["eu-central-pool", "ap-southeast-pool"]
  }'

# Gateway 返回带缓存头的 200 OK
HTTP/1.1 200 OK
Cache-Control: public, max-age=60
Content-Type: application/json

{"status":"ack","timestamp":"2025-01-15T10:00:00Z"}