dp_aware_demon

README

du_aware_demon

1. flowcontrol
2. ipfilter
3. service register/deregister
4. configure dynamic get/upgrade
5. dynamic upgrade flowcontrol/ipfilter rule
6. service client side load balance

架构

config.yaml

⚠️ 如果有批量查询切批量查询被计算入日查询量限制中，需要业务添加如下代码

 num = 10 // num 记录本次批量查询换算成几次日查询量限制
 c.Set("queries", num)

• 场景一（id 映射）

resource-param: cid # cid需要从url参数中获取的需要配置，cid在url路径中直接带的，⚠️不需要配置
flow-control-rules:
  - resource: bigdata
    threshold: 100  # qps 限制
    queriesPerDay: 10000 # 日查询量限制
    queryBlock: false # 是否 block，默认 false
  - resource: test
    threshold: 50
    queriesPerDay: 10000
    queryBlock: false
  - resource: ads
    threshold: 1000
    queriesPerDay: 10000
    queryBlock: false
ip-filter-rules:
  allowed:
    - 127.0.0.1
    - localhost
    - 172.17.130.223
  blocked:
    - 10.10.1.21
    - 192.168.1.24
  urlPath: /q
  urlParam: cid # cid需要从url参数中获取的需要配置，cid在url路径中直接带的，⚠️不需要配置
  blockedDefault: false
  authorized:
    - resource: "test"
      ips:
        - 127.0.0.1
        - localhost
        - 172.17.130.223
    - resource: "bigdata"
      ips:
        - 127.0.0.1
        - localhost
        - 172.17.130.223

• 场景二（设备活跃）

resource-param:  # cid需要从url参数中获取的需要配置，cid在url路径中直接带的，⚠️不需要配置
flow-control-rules:
  - resource: bigdata
    threshold: 100   # qps 限制
    queriesPerDay: 10000 # 日查询量限制
    queryBlock: false # 是否 block，默认配置 false
  - resource: test
    threshold: 50
    queriesPerDay: 10000
    queryBlock: false
  - resource: ads
    threshold: 1000
    queriesPerDay: 10000
    queryBlock: false
ip-filter-rules:
  allowed:
    - 127.0.0.1
    - localhost
    - 172.17.130.223
  blocked:
    - 10.10.1.21
    - 192.168.1.24
  urlPath: /active
  urlParam:  # cid需要从url参数中获取的需要配置，cid在url路径中直接带的，⚠️不需要配置
  blockedDefault: false
  authorized:
    - resource: "test"
      ips:
        - 127.0.0.1
        - localhost
        - 172.17.130.223
    - resource: "bigdata"
      ips:
        - 127.0.0.1
        - localhost
        - 172.17.130.223

init awarent

ServiceName: 服务名字 

Port： 服务端口 

Nacos： nacos配置 包括 IP，Port 

Group： 组，服务所在组 例如 DDV_TEST,DDV_DEV,DDV_PROD 

ConfigID: nasco 配置 dataid 

RuleID： IP Filter， 流量控制规则ID 

	aware, err := awarent.InitAwarent(awarent.Config{
		ServiceName: "ddv",
		Port:        8080,
		Nacos: awarent.Nacos{
			IP:   "192.168.1.71",
			Port: 8848,
		},
		Group: "DDV_TEST",
		RuleID: "DDV_RULES",// 可选，如果RuleID为空 则 限流和IP过滤功能将不可用
	})

gin 使用 awarent

    e := gin.New()
	e.Use(gin.Recovery())
	//gin 使用 IP过滤middleware
	e.Use(aware.IPFilter())
	//gin 使用 限流middleware
	e.Use(aware.Sentinel())
	//gin 使用prometheus监控 包含限流统计
	e.GET("/awarent", awarent.PromHandler)
	//获取配置  
	content, _ := aware.GetConfig("DDV_CONFIG")
	fmt.Printf("content:%s", content)
	//监听配置更新 动态配置
	aware.ConfigOnChange("DDV_CONFIG", func(data string) {
		fmt.Printf("config updated:%s\n", data)
	})
	e.GET("/q", handlers.GetDDV)
	srv := &http.Server{
		Addr:    "0.0.0.0:8080",
		Handler: e,
	}

	go func() {
		if err := srv.ListenAndServe(); err != nil {
			fmt.Printf("start server error:%v\n", err)
		}
	}()
	quit := make(chan os.Signal)
	signal.Notify(quit, syscall.SIGINT, syscall.SIGTERM, syscall.SIGKILL)
	<-quit
	//服务注销
	aware.Deregister()

nacos docker-compose 安装

• Clone 项目 并且进入项目根目录

  git clone https://github.com/paderlol/nacos-docker.git
  cd nacos-docker
  

• 单机mysql8

  docker-compose -f example/standalone-mysql-8.yaml up -d