MongoDB CDC 导入 Elasticsearch

一、docker-compose

复制代码
version: '3'
services:
  mongo:
    image: "mongo:4.0-xenial"
    command: --replSet rs0 --smallfiles --oplogSize 128
    ports:
      - "27017:27017"
    environment:
      - MONGO_INITDB_ROOT_USERNAME=mongouser
      - MONGO_INITDB_ROOT_PASSWORD=mongopw
  elasticsearch:
    image: elastic/elasticsearch:7.6.0
    environment:
      - cluster.name=docker-cluster
      - bootstrap.memory_lock=true
      - "ES_JAVA_OPTS=-Xms512m -Xmx512m"
      - discovery.type=single-node
    ports:
      - "9200:9200"
      - "9300:9300"
    ulimits:
      memlock:
        soft: -1
        hard: -1
      nofile:
        soft: 65536
        hard: 65536
  kibana:
    image: elastic/kibana:7.6.0
    ports:
      - "5601:5601"

二、进入 MongoDB 容器,初始化副本集和数据

复制代码
docker-compose exec mongo /usr/bin/mongo -u mongouser -p mongopw

// 1. 初始化副本集
rs.initiate();
rs.status();

// 2. 切换数据库
use mgdb;

// 3. 初始化数据
db.orders.insertMany([
  {
    order_id: 101,
    order_date: ISODate("2020-07-30T10:08:22.001Z"),
    customer_id: 1001,
    price: NumberDecimal("50.50"),
    product: {
      name: 'scooter',
      description: 'Small 2-wheel scooter'
    },
    order_status: false
  },
  {
    order_id: 102, 
    order_date: ISODate("2020-07-30T10:11:09.001Z"),
    customer_id: 1002,
    price: NumberDecimal("15.00"),
    product: {
      name: 'car battery',
      description: '12V car battery'
    },
    order_status: false
  },
  {
    order_id: 103,
    order_date: ISODate("2020-07-30T12:00:30.001Z"),
    customer_id: 1003,
    price: NumberDecimal("25.25"),
    product: {
      name: 'hammer',
      description: '16oz carpenter hammer'
    },
    order_status: false
  }
]);

db.customers.insertMany([
  { 
    customer_id: 1001, 
    name: 'Jark', 
    address: 'Hangzhou' 
  },
  { 
    customer_id: 1002, 
    name: 'Sally',
    address: 'Beijing'
  },
  { 
    customer_id: 1003,
    name: 'Edward',
    address: 'Shanghai'
  }
]);

下载链接只对已发布的版本有效, SNAPSHOT 版本需要本地编译

复制代码
-- Flink SQL
-- 设置间隔时间为3秒                       
Flink SQL> SET execution.checkpointing.interval = 3s;

-- 设置本地时区为 Asia/Shanghai
Flink SQL> SET table.local-time-zone = Asia/Shanghai;

Flink SQL> CREATE TABLE orders (
   _id STRING,
   order_id INT,
   order_date TIMESTAMP_LTZ(3),
   customer_id INT,
   price DECIMAL(10, 5),
   product ROW<name STRING, description STRING>,
   order_status BOOLEAN,
   PRIMARY KEY (_id) NOT ENFORCED
 ) WITH (
   'connector' = 'mongodb-cdc',
   'hosts' = 'localhost:27017',
   'username' = 'mongouser',
   'password' = 'mongopw',
   'database' = 'mgdb',
   'collection' = 'orders'
 );
 
 Flink SQL> CREATE TABLE customers (
   _id STRING,
   customer_id INT,
   name STRING,
   address STRING,
   PRIMARY KEY (_id) NOT ENFORCED
 ) WITH (
   'connector' = 'mongodb-cdc',
   'hosts' = 'localhost:27017',
   'username' = 'mongouser',
   'password' = 'mongopw',
   'database' = 'mgdb',
   'collection' = 'customers'
 );

Flink SQL> CREATE TABLE enriched_orders (
   order_id INT,
   order_date TIMESTAMP_LTZ(3),
   customer_id INT,
   price DECIMAL(10, 5),
   product ROW<name STRING, description STRING>,
   order_status BOOLEAN,
   customer_name STRING,
   customer_address STRING,
   PRIMARY KEY (order_id) NOT ENFORCED
 ) WITH (
     'connector' = 'elasticsearch-7',
     'hosts' = 'http://localhost:9200',
     'index' = 'enriched_orders'
 );

Flink SQL> INSERT INTO enriched_orders
 SELECT o.order_id,
        o.order_date,
        o.customer_id,
        o.price,
        o.product,
        o.order_status,
        c.name,
        c. address
   FROM orders AS o
   LEFT JOIN customers AS c ON o.customer_id = c.customer_id;

五、修改 MongoDB 里面的数据,观察 elasticsearch 里的结果

复制代码
db.orders.insert({
    order_id: 104,
    order_date: ISODate("2020-07-30T12:00:30.001Z"),
    customer_id: 1004,
    price: NumberDecimal("25.25"),
    product: {
        name: 'rocks',
        description: 'box of assorted rocks'
    },
    order_status: false
});

db.customers.insert({
    customer_id: 1004,
    name: 'Jacob',
    address: 'Shanghai'
});

db.orders.updateOne(
    { order_id: 104 },
    { $set: { order_status: true } }
);

db.orders.deleteOne(
    { order_id : 104 }
);
相关推荐
C语言小火车9 小时前
C++ 堆排序深度精讲:基于完全二叉树的选择排序进化,最坏情况 O(n log n) 的稳定王者
开发语言·c++·算法·排序算法·堆排序
北冥you鱼9 小时前
abigen 最佳实践:从入门到精通,高效生成 Go 语言合约绑定
开发语言·golang·区块链
ALex_zry10 小时前
C++26 std::complex 结构化绑定详解:auto [re, im] = c
c语言·开发语言·c++
shushangyun_11 小时前
2026智能采购商城系统选型指南:如何引领企业数字化采购升级
java·大数据·数据库·人工智能·机器学习
她说..11 小时前
Java 默认值设置方式
java·开发语言·后端·springboot
忧郁的紫菜11 小时前
基础实现:单篇 Markdown 转 Word
开发语言·c#·word
甜美的小天鹅11 小时前
Swifter C#之inline还是不inline,这是个问题
开发语言·c#
学渣超11 小时前
记一次分布式事务数据不一致的排查之旅:从超时到索引,层层剥茧
java·后端·架构
小短腿的代码世界11 小时前
Qt Bluetooth源码深度解析:从HCI协议到跨平台API的完整架构
开发语言·qt·架构
掉鱼的猫11 小时前
Agent Harness 实战指南:构建生产级 AI Agent 的"马具"框架
java·llm·aigc