SpringCloud微服务实战——搭建企业级开发框架（二十七）：集成多数据源+Seata分布式事务+读写分离+分库分表

读写分离：为了确保数据库产品的稳定性，很多数据库拥有双机热备功能。也就是，第一台数据库服务器，是对外提供增删改业务的生产服务器；第二台数据库服务器，主要进行读的操作。
目前有多种方式实现读写分离，一种是Mycat这种数据库中间件，需要单独部署服务，通过配置来实现读写分离，不侵入到业务代码中；还有一种是dynamic-datasource/shardingsphere-jdbc这种，需要在业务代码引入jar包进行开发。
本框架集成 dynamic-datasource（多数据源+读写分离+分库）+ druid（数据库连接池）+ seata（分布式事务）+ mybatis-plus+shardingsphere-jdbc(分库分表)， dynamic-datasource可以实现简单的分库操作，目前还不支持分表。复杂的分库分表需要用到shardingsphere-jdbc,本文参考dynamic-datasource中的实例，模拟用户下单，扣商品库存，扣用户余额操作，初步可分为订单服务+商品服务+用户服务。

一、Seata安装配置

1、我们将服务安装到CentOS环境上，所以这里我们下载tar.gz版本,下载地址:https://github.com/seata/seata/releases
seata-server-1.4.1.tar.gz
2、上传到CentOS服务器，执行解压命令

tar -zxvf seata-server-1.4.1.tar.gz

3、下载Seata需要的SQL脚本，新建Seata数据库并将需要使用的数据库脚本seata-1.4.1\seata-1.4.1\script\server\db\mysql.sql刷进去

4、修改Seata配置文件，将seata服务端的注册中心和配置中心设置为Nacos

vi /bigdata/soft_home/seata/conf/registry.conf

registry {
  # file 、nacos 、eureka、redis、zk、consul、etcd3、sofa
  type = "nacos"
  loadBalance = "RandomLoadBalance"
  loadBalanceVirtualNodes = 10

  nacos {
    application = "seata-server"
    serverAddr = "127.0.0.1:8848"
    group = "SEATA_GROUP"
    namespace = ""
    cluster = "default"
    username = "nacos"
    password = "nacos"
  }
  eureka {
    serviceUrl = "http://localhost:8761/eureka"
    application = "default"
    weight = "1"
  }
  redis {
    serverAddr = "localhost:6379"
    db = 0
    password = ""
    cluster = "default"
    timeout = 0
  }
  zk {
    cluster = "default"
    serverAddr = "127.0.0.1:2181"
    sessionTimeout = 6000
    connectTimeout = 2000
    username = ""
    password = ""
  }
  consul {
    cluster = "default"
    serverAddr = "127.0.0.1:8500"
  }
  etcd3 {
    cluster = "default"
    serverAddr = "http://localhost:2379"
  }
  sofa {
    serverAddr = "127.0.0.1:9603"
    application = "default"
    region = "DEFAULT_ZONE"
    datacenter = "DefaultDataCenter"
    cluster = "default"
    group = "SEATA_GROUP"
    addressWaitTime = "3000"
  }
  file {
    name = "file.conf"
  }
}

config {
  # file、nacos 、apollo、zk、consul、etcd3
  type = "nacos"

  nacos {
    serverAddr = "127.0.0.1:8848"
    namespace = ""
    group = "SEATA_GROUP"
    username = "nacos"
    password = "nacos"
  }
  consul {
    serverAddr = "127.0.0.1:8500"
  }
  apollo {
    appId = "seata-server"
    apolloMeta = "http://192.168.1.204:8801"
    namespace = "application"
    apolloAccesskeySecret = ""
  }
  zk {
    serverAddr = "127.0.0.1:2181"
    sessionTimeout = 6000
    connectTimeout = 2000
    username = ""
    password = ""
  }
  etcd3 {
    serverAddr = "http://localhost:2379"
  }
  file {
    name = "file.conf"
  }
}

5、在Nacos添加Seata配置文件，修改script/config-center/config.txt，将script目录上传到CentOS服务器，执行script/config-center/nacos/nacos-config.sh命令

service.vgroupMapping.gitegg_seata_tx_group=default

service.default.grouplist=127.0.0.1:8091

store.mode=db

store.db.url=jdbc:mysql://127.0.0.1:3306/seata?useUnicode=true
store.db.user=root
store.db.password=root

chmod 777 nacos-config.sh

sh nacos-config.sh -h 127.0.0.1 -p 8848

设置成功
6、在CentOS上进去到Seata安装目录的bin目录执行命令，启动Seata服务端

nohup ./seata-server.sh -h 127.0.0.1 -p 8091 >log.out 2>1 &

如果服务器有多网卡，存在多个ip地址，-h后面一定要加可以访问的ip地址

7、在Nacos上可以看到配置文件和服务已经注册成功

二、Seata安装成功后，我们需要在微服务中集成Seata客户端

1、因为我们在微服务中使用Seata，所以，我们将Seata客户端的依赖添加在gitegg-plaform-cloud中

        <!-- Seata 分布式事务管理 -->
        <dependency>
            <groupId>com.alibaba.cloud</groupId>
            <artifactId>spring-cloud-starter-alibaba-seata</artifactId>
        </dependency>

2、我们这里打算使用多数据源，所以这里也把动态多数据源组件Dynamic Datasource加入到gitegg-plaform-mybatis依赖中

        <!-- 动态数据源 -->
        <dependency>
            <groupId>com.baomidou</groupId>
            <artifactId>dynamic-datasource-spring-boot-starter</artifactId>
        </dependency>

3、配置Nacos数据库多数据源及Seata

spring:
  datasource: 
    druid:
      stat-view-servlet:
        enabled: true
        loginUsername: admin
        loginPassword: 123456
    dynamic:
      # 设置默认的数据源或者数据源组,默认值即为master
      primary: master
      # 设置严格模式,默认false不启动. 启动后在未匹配到指定数据源时候会抛出异常,不启动则使用默认数据源.
      strict: false
      # 开启seata代理，开启后默认每个数据源都代理，如果某个不需要代理可单独关闭
      seata: true
      #支持XA及AT模式,默认AT
      seata-mode: AT
      druid:
        initialSize: 1
        minIdle: 3
        maxActive: 20
        # 配置获取连接等待超时的时间
        maxWait: 60000
        # 配置间隔多久才进行一次检测，检测需要关闭的空闲连接，单位是毫秒
        timeBetweenEvictionRunsMillis: 60000
        # 配置一个连接在池中最小生存的时间，单位是毫秒
        minEvictableIdleTimeMillis: 30000
        validationQuery: select 'x'
        testWhileIdle: true
        testOnBorrow: false
        testOnReturn: false
        # 打开PSCache，并且指定每个连接上PSCache的大小
        poolPreparedStatements: true
        maxPoolPreparedStatementPerConnectionSize: 20
        # 配置监控统计拦截的filters，去掉后监控界面sql无法统计，'wall'用于防火墙
        filters: config,stat,slf4j
        # 通过connectProperties属性来打开mergeSql功能；慢SQL记录
        connectionProperties: druid.stat.mergeSql=true;druid.stat.slowSqlMillis=5000;
        # 合并多个DruidDataSource的监控数据
        useGlobalDataSourceStat: true
      datasource: 
        master: 
          url: jdbc:mysql://127.0.0.1/gitegg_cloud?zeroDateTimeBehavior=convertToNull&useUnicode=true&characterEncoding=utf8&all owMultiQueries=true&serverTimezone=Asia/Shanghai
          username: root
          password: root
        mall_user:
          url: jdbc:mysql://127.0.0.1/gitegg_cloud_mall_user?zeroDateTimeBehavior=convertToNull&useUnicode=true&characterEncoding=utf8&allowMultiQueries=true&serverTimezone=Asia/Shanghai
          username: root
          password: root
        mall_goods:
          url: jdbc:mysql://127.0.0.1/gitegg_cloud_mall_goods?zeroDateTimeBehavior=convertToNull&useUnicode=true&characterEncoding=utf8&allowMultiQueries=true&serverTimezone=Asia/Shanghai
          username: root
          password: root
        mall_order:
          url: jdbc:mysql://127.0.0.1/gitegg_cloud_mall_order?zeroDateTimeBehavior=convertToNull&useUnicode=true&characterEncoding=utf8&allowMultiQueries=true&serverTimezone=Asia/Shanghai
          username: root
          password: root
        mall_pay:
          url: jdbc:mysql://127.0.0.1/gitegg_cloud_mall_pay?zeroDateTimeBehavior=convertToNull&useUnicode=true&characterEncoding=utf8&alowMultiQueries=true&serverTimezone=Asia/Shanghai
          username: root
          password: root
seata:
  enabled: true
  application-id: ${spring.application.name}
  tx-service-group: gitegg_seata_tx_group
  # 一定要是false
  enable-auto-data-source-proxy: false
  service:
    vgroup-mapping:
      #key与上面的gitegg_seata_tx_group的值对应
      gitegg_seata_tx_group: default
  config:
    type: nacos
    nacos:
      namespace:
      serverAddr: 127.0.0.1:8848
      group: SEATA_GROUP
      userName: "nacos"
      password: "nacos"
  registry:
    type: nacos
    nacos:
      #seata服务端(TC)在nacos中的应用名称
      application: seata-server
      server-addr: 127.0.0.1:8848
      namespace:
      userName: "nacos"
      password: "nacos"

三、数据库表设计

这里参考Dynamic Datasource官方提供的示例项目,并结合电商项目数据库设计,新建四个数据库，gitegg_cloud_mall_goods（商品数据库），gitegg_cloud_mall_order（订单数据库），gitegg_cloud_mall_pay（支付数据库）,gitegg_cloud_mall_user（账户数据库）四个数据库，下面是具体表结构和简要说明:

1、商品数据库表设计

表设计：

商品分类表：t_mall_goods_category
商品品牌表： t_mall_goods_brand
分类品牌关联关系表：t_mall_goods_category_brand
商品规格参数组表： t_mall_goods_spec_group
商品规格参数表：t_mall_goods_spec_param
商品SPU表： t_mall_goods_spu
商品SPU详情表： t_mall_goods_spu_detail
商品SKU表： t_mall_goods_sku

关系：

一个分类有多个品牌，一个品牌属于多个分类，所以是多对多
一个分类有多个规格组，一个规格组有多个规格参数，所以是一对多
一个分类下有多个SPU，所以是一对多
一个品牌下有多个SPU，所以是一对多

一个SPU下有多个SKU，所以是一对多

DROP TABLE IF EXISTS `t_mall_goods_brand`;
CREATE TABLE `t_mall_goods_brand`  (
`id` bigint(20) NOT NULL AUTO_INCREMENT COMMENT '品牌id',
`tenant_id` bigint(20) NOT NULL DEFAULT 0 COMMENT '租户id',
`name` varchar(64) CHARACTER SET utf8 COLLATE utf8_general_ci NOT NULL COMMENT '品牌名称',
`image` varchar(256) CHARACTER SET utf8 COLLATE utf8_general_ci NULL DEFAULT '' COMMENT '品牌图片地址',
`letter` char(1) CHARACTER SET utf8 COLLATE utf8_general_ci NULL DEFAULT '' COMMENT '品牌的首字母',
`create_time` datetime(0) NULL DEFAULT NULL COMMENT '创建时间',
`creator` bigint(20) NULL DEFAULT NULL COMMENT '创建者',
`update_time` datetime(0) NULL DEFAULT NULL COMMENT '更新时间',
`operator` bigint(20) NULL DEFAULT NULL COMMENT '更新者',
`del_flag` tinyint(2) NULL DEFAULT 0 COMMENT '1:删除 0:不删除',
PRIMARY KEY (`id`) USING BTREE
) ENGINE = InnoDB AUTO_INCREMENT = 1 CHARACTER SET = utf8 COLLATE = utf8_general_ci COMMENT = '品牌表，一个品牌下有多个商品（spu），一对多关系' ROW_FORMAT = DYNAMIC;

-- ----------------------------
-- Table structure for t_mall_goods_brand_category
-- ----------------------------
DROP TABLE IF EXISTS `t_mall_goods_brand_category`;
CREATE TABLE `t_mall_goods_brand_category`  (
`id` bigint(20) NOT NULL AUTO_INCREMENT COMMENT '品牌id',
`tenant_id` bigint(20) NOT NULL DEFAULT 0 COMMENT '租户id',
`brand_id` bigint(20) NOT NULL COMMENT '品牌id',
`category_id` bigint(20) NOT NULL COMMENT '商品类目id',
`create_time` datetime(0) NULL DEFAULT NULL COMMENT '创建时间',
`creator` bigint(20) NULL DEFAULT NULL COMMENT '创建者',
`update_time` datetime(0) NULL DEFAULT NULL COMMENT '更新时间',
`operator` bigint(20) NULL DEFAULT NULL COMMENT '更新者',
`del_flag` tinyint(2) NULL DEFAULT 0 COMMENT '1:删除 0:不删除',
PRIMARY KEY (`id`) USING BTREE,
INDEX `key_tenant_id`(`tenant_id`) USING BTREE,
INDEX `key_category_id`(`category_id`) USING BTREE,
INDEX `key_brand_id`(`brand_id`) USING BTREE
) ENGINE = InnoDB AUTO_INCREMENT = 1 CHARACTER SET = utf8 COLLATE = utf8_general_ci COMMENT = '商品分类和品牌的中间表，两者是多对多关系' ROW_FORMAT = DYNAMIC;

-- ----------------------------
-- Table structure for t_mall_goods_category
-- ----------------------------
DROP TABLE IF EXISTS `t_mall_goods_category`;
CREATE TABLE `t_mall_goods_category`  (
`id` bigint(20) NOT NULL AUTO_INCREMENT COMMENT '类目id',
`tenant_id` bigint(20) NOT NULL DEFAULT 0 COMMENT '租户id',
`name` varchar(32) CHARACTER SET utf8 COLLATE utf8_general_ci NOT NULL COMMENT '类目名称',
`parent_id` bigint(20) NOT NULL COMMENT '父类目id,顶级类目填0',
`is_parent` tinyint(2) NOT NULL COMMENT '是否为父节点，0为否，1为是',
`sort` tinyint(2) NOT NULL COMMENT '排序指数，越小越靠前',
`comments` varchar(255) CHARACTER SET utf8 COLLATE utf8_general_ci NULL DEFAULT NULL COMMENT '备注',
`create_time` datetime(0) NULL DEFAULT NULL COMMENT '创建时间',
`creator` bigint(20) NULL DEFAULT NULL COMMENT '创建者',
`update_time` datetime(0) NULL DEFAULT NULL COMMENT '更新时间',
`operator` bigint(20) NULL DEFAULT NULL COMMENT '更新者',
`del_flag` tinyint(2) NULL DEFAULT 0 COMMENT '1:删除 0:不删除',
PRIMARY KEY (`id`) USING BTREE,
INDEX `key_parent_id`(`parent_id`) USING BTREE
) ENGINE = InnoDB AUTO_INCREMENT = 1 CHARACTER SET = utf8 COLLATE = utf8_general_ci COMMENT = '商品类目表，类目和商品(spu)是一对多关系，类目与品牌是多对多关系' ROW_FORMAT = DYNAMIC;

-- ----------------------------
-- Table structure for t_mall_goods_sku
-- ----------------------------
DROP TABLE IF EXISTS `t_mall_goods_sku`;
CREATE TABLE `t_mall_goods_sku`  (
`id` bigint(20) NOT NULL AUTO_INCREMENT COMMENT '主键',
`tenant_id` bigint(20) NOT NULL DEFAULT 0 COMMENT '租户id',
`spu_id` bigint(20) NOT NULL COMMENT 'spu id',
`title` varchar(256) CHARACTER SET utf8 COLLATE utf8_general_ci NOT NULL COMMENT '商品标题',
`images` varchar(1024) CHARACTER SET utf8 COLLATE utf8_general_ci NULL DEFAULT '' COMMENT '商品的图片，多个图片以‘,’分割',
`stock` int(8) UNSIGNED NULL DEFAULT 0 COMMENT '库存',
`price` decimal(10, 2) NOT NULL DEFAULT 0.00 COMMENT '销售价格',
`indexes` varchar(32) CHARACTER SET utf8 COLLATE utf8_general_ci NULL DEFAULT '' COMMENT '特有规格属性在spu属性模板中的对应下标组合',
`own_spec` varchar(1024) CHARACTER SET utf8 COLLATE utf8_general_ci NULL DEFAULT '' COMMENT 'sku的特有规格参数键值对，json格式，反序列化时请使用linkedHashMap，保证有序',
`status` tinyint(1) NOT NULL DEFAULT 1 COMMENT '是否有效，0无效，1有效',
`create_time` datetime(0) NULL DEFAULT NULL COMMENT '创建时间',
`creator` bigint(20) NULL DEFAULT NULL COMMENT '创建者',
`update_time` datetime(0) NULL DEFAULT NULL COMMENT '更新时间',
`operator` bigint(20) NULL DEFAULT NULL COMMENT '更新者',
`del_flag` tinyint(2) NULL DEFAULT 0 COMMENT '1:删除 0:不删除',
PRIMARY KEY (`id`) USING BTREE,
INDEX `key_spu_id`(`spu_id`) USING BTREE
) ENGINE = InnoDB AUTO_INCREMENT = 3 CHARACTER SET = utf8 COLLATE = utf8_general_ci COMMENT = 'sku表,该表表示具体的商品实体' ROW_FORMAT = DYNAMIC;

-- ----------------------------
-- Table structure for t_mall_goods_spec_group
-- ----------------------------
DROP TABLE IF EXISTS `t_mall_goods_spec_group`;
CREATE TABLE `t_mall_goods_spec_group`  (
`id` bigint(20) NOT NULL AUTO_INCREMENT COMMENT '主键',
`tenant_id` bigint(20) NOT NULL DEFAULT 0 COMMENT '租户id',
`category_id` bigint(20) NOT NULL COMMENT '商品分类id，一个分类下有多个规格组',
`name` varchar(32) CHARACTER SET utf8 COLLATE utf8_general_ci NOT NULL COMMENT '规格组的名称',
`create_time` datetime(0) NULL DEFAULT NULL COMMENT '创建时间',
`creator` bigint(20) NULL DEFAULT NULL COMMENT '创建者',
`update_time` datetime(0) NULL DEFAULT NULL COMMENT '更新时间',
`operator` bigint(20) NULL DEFAULT NULL COMMENT '更新者',
`del_flag` tinyint(2) NULL DEFAULT 0 COMMENT '1:删除 0:不删除',
PRIMARY KEY (`id`) USING BTREE,
INDEX `key_tenant_id`(`tenant_id`) USING BTREE,
INDEX `key_category_id`(`category_id`) USING BTREE
) ENGINE = InnoDB AUTO_INCREMENT = 1 CHARACTER SET = utf8 COLLATE = utf8_general_ci COMMENT = '规格参数的分组表，每个商品分类下有多个规格参数组' ROW_FORMAT = DYNAMIC;

-- ----------------------------
-- Table structure for t_mall_goods_spec_param
-- ----------------------------
DROP TABLE IF EXISTS `t_mall_goods_spec_param`;
CREATE TABLE `t_mall_goods_spec_param`  (
`id` bigint(20) NOT NULL AUTO_INCREMENT COMMENT '主键',
`tenant_id` bigint(20) NOT NULL DEFAULT 0 COMMENT '租户id',
`category_id` bigint(20) NOT NULL COMMENT '商品分类id',
`group_id` bigint(20) NOT NULL COMMENT '所属组的id',
`name` varchar(128) CHARACTER SET utf8 COLLATE utf8_general_ci NOT NULL COMMENT '参数名',
`numeric` tinyint(1) NOT NULL COMMENT '是否是数字类型参数，true或false',
`unit` varchar(128) CHARACTER SET utf8 COLLATE utf8_general_ci NULL DEFAULT '' COMMENT '数字类型参数的单位，非数字类型可以为空',
`generic` tinyint(1) NOT NULL COMMENT '是否是sku通用属性，true或false',
`searching` tinyint(1) NOT NULL COMMENT '是否用于搜索过滤，true或false',
`segments` varchar(1024) CHARACTER SET utf8 COLLATE utf8_general_ci NULL DEFAULT '' COMMENT '数值类型参数，如果需要搜索，则添加分段间隔值，如CPU频率间隔：0.5-1.0',
`create_time` datetime(0) NULL DEFAULT NULL COMMENT '创建时间',
`creator` bigint(20) NULL DEFAULT NULL COMMENT '创建者',
`update_time` datetime(0) NULL DEFAULT NULL COMMENT '更新时间',
`operator` bigint(20) NULL DEFAULT NULL COMMENT '更新者',
`del_flag` tinyint(2) NULL DEFAULT 0 COMMENT '1:删除 0:不删除',
PRIMARY KEY (`id`) USING BTREE,
INDEX `key_tenant_id`(`tenant_id`) USING BTREE,
INDEX `key_category_id`(`category_id`) USING BTREE,
INDEX `key_group_id`(`group_id`) USING BTREE
) ENGINE = InnoDB AUTO_INCREMENT = 1 CHARACTER SET = utf8 COLLATE = utf8_general_ci COMMENT = '规格参数组下的参数名' ROW_FORMAT = DYNAMIC;

-- ----------------------------
-- Table structure for t_mall_goods_spu
-- ----------------------------
DROP TABLE IF EXISTS `t_mall_goods_spu`;
CREATE TABLE `t_mall_goods_spu`  (
`id` bigint(20) NOT NULL AUTO_INCREMENT COMMENT '主键',
`tenant_id` bigint(20) NOT NULL DEFAULT 0 COMMENT '租户id',
`brand_id` bigint(20) NOT NULL COMMENT '商品所属品牌id',
`category_id` bigint(20) NOT NULL COMMENT '商品分类id',
`name` varchar(256) CHARACTER SET utf8 COLLATE utf8_general_ci NOT NULL DEFAULT '' COMMENT '商品名称',
`sub_title` varchar(256) CHARACTER SET utf8 COLLATE utf8_general_ci NULL DEFAULT '' COMMENT '副标题，一般是促销信息',
`on_sale` tinyint(2) NOT NULL DEFAULT 1 COMMENT '是否上架，0下架，1上架',
`price` decimal(10, 2) NOT NULL DEFAULT 0.00 COMMENT '售价',
`use_spec` tinyint(2) NOT NULL DEFAULT 1 COMMENT '是否使用规格：0=不使用，1=使用',
`spec_groups` text CHARACTER SET utf8 COLLATE utf8_general_ci NOT NULL COMMENT '商品规格组',
`goods_stock` int(11) NOT NULL DEFAULT 0 COMMENT '商品库存',
`virtual_sales` int(11) NOT NULL DEFAULT 0 COMMENT '虚拟销售数量',
`confine_count` int(11) NOT NULL DEFAULT -1 COMMENT '购物数量限制',
`pieces` int(11) NOT NULL DEFAULT 0 COMMENT '满件包邮',
`forehead` decimal(10, 2) NOT NULL DEFAULT 0.00 COMMENT '满额包邮',
`freight_id` int(11) NOT NULL COMMENT '运费模板ID',
`give_integral` int(11) NOT NULL DEFAULT 0 COMMENT '赠送积分',
`give_integral_type` tinyint(2) NOT NULL DEFAULT 1 COMMENT '赠送积分类型1固定值 2百分比',
`deductible_integral` decimal(10, 2) NOT NULL DEFAULT 0.00 COMMENT '可抵扣积分',
`deductible_integral_type` tinyint(2) NOT NULL DEFAULT 1 COMMENT '可抵扣积分类型1固定值 2百分比',
`accumulative` tinyint(2) NOT NULL DEFAULT 0 COMMENT '允许多件累计折扣 0否 1是',
`individual_share` tinyint(2) NOT NULL DEFAULT 0 COMMENT '是否单独分销设置：0否 1是',
`share_setting_type` tinyint(2) NOT NULL DEFAULT 0 COMMENT '分销设置类型 0普通设置 1详细设置',
`share_commission_type` tinyint(1) NOT NULL DEFAULT 0 COMMENT '佣金配比 0 固定金额 1 百分比',
`membership_price` tinyint(2) NOT NULL DEFAULT 0 COMMENT '是否享受会员价购买',
`membership_price_single` tinyint(2) NOT NULL DEFAULT 0 COMMENT '是否单独设置会员价',
`share_image` varchar(255) CHARACTER SET utf8 COLLATE utf8_general_ci NOT NULL DEFAULT '' COMMENT '自定义分享图片',
`share_title` varchar(65) CHARACTER SET utf8 COLLATE utf8_general_ci NOT NULL DEFAULT '' COMMENT '自定义分享标题',
`is_default_services` tinyint(2) NOT NULL DEFAULT 1 COMMENT '默认服务 0否  1是',
`sort` int(11) NOT NULL DEFAULT 100 COMMENT '排序',
`create_time` datetime(0) NULL DEFAULT NULL COMMENT '创建时间',
`creator` bigint(20) NULL DEFAULT NULL COMMENT '创建者',
`update_time` datetime(0) NULL DEFAULT NULL COMMENT '更新时间',
`operator` bigint(20) NULL DEFAULT NULL COMMENT '更新者',
`del_flag` tinyint(2) NULL DEFAULT 0 COMMENT '1:删除 0:不删除',
PRIMARY KEY (`id`) USING BTREE,
INDEX `key_tenant_id`(`tenant_id`) USING BTREE,
INDEX `key_category_id`(`category_id`) USING BTREE
) ENGINE = InnoDB AUTO_INCREMENT = 1 CHARACTER SET = utf8 COLLATE = utf8_general_ci COMMENT = 'spu表，该表描述的是一个抽象性的商品' ROW_FORMAT = DYNAMIC;

-- ----------------------------
-- Table structure for t_mall_goods_spu_detail
-- ----------------------------
DROP TABLE IF EXISTS `t_mall_goods_spu_detail`;
CREATE TABLE `t_mall_goods_spu_detail`  (
`id` bigint(20) NOT NULL AUTO_INCREMENT COMMENT '主键',
`tenant_id` bigint(20) NOT NULL DEFAULT 0 COMMENT '租户id',
`spu_id` bigint(20) NOT NULL,
`description` text CHARACTER SET utf8 COLLATE utf8_general_ci NULL COMMENT '商品描述信息',
`generic_spec` varchar(2048) CHARACTER SET utf8 COLLATE utf8_general_ci NOT NULL DEFAULT '' COMMENT '通用规格参数数据',
`special_spec` varchar(1024) CHARACTER SET utf8 COLLATE utf8_general_ci NOT NULL COMMENT '特有规格参数及可选值信息，json格式',
`packing_list` varchar(1024) CHARACTER SET utf8 COLLATE utf8_general_ci NULL DEFAULT '' COMMENT '包装清单',
`after_service` varchar(1024) CHARACTER SET utf8 COLLATE utf8_general_ci NULL DEFAULT '' COMMENT '售后服务',
`create_time` datetime(0) NULL DEFAULT NULL COMMENT '创建时间',
`creator` bigint(20) NULL DEFAULT NULL COMMENT '创建者',
`update_time` datetime(0) NULL DEFAULT NULL COMMENT '更新时间',
`operator` bigint(20) NULL DEFAULT NULL COMMENT '更新者',
`del_flag` tinyint(2) NULL DEFAULT 0 COMMENT '1:删除 0:不删除',
PRIMARY KEY (`id`) USING BTREE,
INDEX `key_tenant_id`(`tenant_id`) USING BTREE
) ENGINE = InnoDB AUTO_INCREMENT = 1 CHARACTER SET = utf8 COLLATE = utf8_general_ci ROW_FORMAT = DYNAMIC;

-- ----------------------------
-- Table structure for undo_log
-- ----------------------------
DROP TABLE IF EXISTS `undo_log`;
CREATE TABLE `undo_log`  (
`branch_id` bigint(20) NOT NULL COMMENT 'branch transaction id',
`xid` varchar(100) CHARACTER SET utf8 COLLATE utf8_general_ci NOT NULL COMMENT 'global transaction id',
`context` varchar(128) CHARACTER SET utf8 COLLATE utf8_general_ci NOT NULL COMMENT 'undo_log context,such as serialization',
`rollback_info` longblob NOT NULL COMMENT 'rollback info',
`log_status` int(11) NOT NULL COMMENT '0:normal status,1:defense status',
`log_created` datetime(6) NOT NULL COMMENT 'create datetime',
`log_modified` datetime(6) NOT NULL COMMENT 'modify datetime',
UNIQUE INDEX `ux_undo_log`(`xid`, `branch_id`) USING BTREE
) ENGINE = InnoDB CHARACTER SET = utf8 COLLATE = utf8_general_ci COMMENT = 'AT transaction mode undo table' ROW_FORMAT = Dynamic;

SET FOREIGN_KEY_CHECKS = 1;

2、订单数据库表设计

-- ----------------------------
-- Table structure for t_mall_order
-- ----------------------------
DROP TABLE IF EXISTS `t_mall_order`;
CREATE TABLE `t_mall_order`  (
`id` bigint(20) NOT NULL AUTO_INCREMENT COMMENT '主键',
`tenant_id` bigint(20) NOT NULL DEFAULT 0 COMMENT '租户id',
`user_id` bigint(20) NOT NULL COMMENT '主键',
`store_id` int(11) NOT NULL DEFAULT 0 COMMENT '店铺id',
`order_no` varchar(255) CHARACTER SET utf8mb4 COLLATE utf8mb4_general_ci NULL DEFAULT '' COMMENT '订单号',
`total_price` decimal(10, 2) NULL DEFAULT NULL COMMENT '订单总金额(含运费)',
`total_pay_price` decimal(10, 2) NOT NULL COMMENT '实际支付总费用(含运费）',
`express_original_price` decimal(10, 2) NOT NULL COMMENT '运费',
`express_price` decimal(10, 2) NULL DEFAULT NULL COMMENT '修改后运费',
`total_goods_original_price` decimal(10, 2) NULL DEFAULT NULL COMMENT '订单商品总金额',
`total_goods_price` decimal(10, 2) NULL DEFAULT NULL COMMENT '优惠后订单商品总金额',
`store_discount_price` decimal(10, 2) NULL DEFAULT 0.00 COMMENT '商家改价优惠',
`member_discount_price` decimal(10, 2) NULL DEFAULT NULL COMMENT '会员优惠价格',
`coupon_id` int(11) NULL DEFAULT NULL COMMENT '优惠券id',
`coupon_discount_price` decimal(10, 2) NULL DEFAULT NULL COMMENT '优惠券优惠金额',
`integral` int(11) NULL DEFAULT NULL COMMENT '使用的积分数量',
`integral_deduction_price` decimal(10, 2) NULL DEFAULT NULL COMMENT '积分抵扣金额',
`name` varchar(65) CHARACTER SET utf8mb4 COLLATE utf8mb4_general_ci NULL DEFAULT '' COMMENT '收件人姓名',
`mobile` varchar(255) CHARACTER SET utf8mb4 COLLATE utf8mb4_general_ci NULL DEFAULT '' COMMENT '收件人手机号',
`address` varchar(255) CHARACTER SET utf8mb4 COLLATE utf8mb4_general_ci NULL DEFAULT '' COMMENT '收件人地址',
`comments` varchar(255) CHARACTER SET utf8mb4 COLLATE utf8mb4_general_ci NULL DEFAULT '' COMMENT '用户订单备注',
`order_form` longtext CHARACTER SET utf8mb4 COLLATE utf8mb4_general_ci NULL COMMENT '自定义表单（JSON）',
`leaving_message` varchar(255) CHARACTER SET utf8mb4 COLLATE utf8mb4_general_ci NULL DEFAULT '' COMMENT '留言',
`store_comments` varchar(255) CHARACTER SET utf8mb4 COLLATE utf8mb4_general_ci NULL DEFAULT '' COMMENT '商家订单备注',
`pay_status` tinyint(2) NULL DEFAULT 0 COMMENT '是否支付：0.未支付 1.已支付',
`pay_type` tinyint(2) NULL DEFAULT 1 COMMENT '支付方式：1.在线支付 2.货到付款 3.余额支付',
`pay_time` timestamp(0) NULL DEFAULT '0000-00-00 00:00:00' COMMENT '支付时间',
`deliver_status` tinyint(2) NULL DEFAULT 0 COMMENT '是否发货：0.未发货 1.已发货',
`deliver_time` timestamp(0) NULL DEFAULT '0000-00-00 00:00:00' COMMENT '发货时间',
`express` varchar(65) CHARACTER SET utf8mb4 COLLATE utf8mb4_general_ci NULL DEFAULT '' COMMENT '物流公司',
`express_no` varchar(255) CHARACTER SET utf8mb4 COLLATE utf8mb4_general_ci NULL DEFAULT '' COMMENT '物流订单号',
`confirm_receipt` tinyint(2) NULL DEFAULT 0 COMMENT '收货状态：0.未收货 1.已收货',
`confirm_receipt_time` timestamp(0) NULL DEFAULT '0000-00-00 00:00:00' COMMENT '确认收货时间',
`cancel_status` tinyint(2) NULL DEFAULT 0 COMMENT '订单取消状态：0.未取消 1.已取消 2.申请取消',
`cancel_time` timestamp(0) NULL DEFAULT '0000-00-00 00:00:00' COMMENT '订单取消时间',
`recycle_status` tinyint(2) NULL DEFAULT 0 COMMENT '是否加入回收站 0.否 1.是',
`offline` tinyint(2) NULL DEFAULT 0 COMMENT '是否到店自提：0.否 1.是',
`offline_code` varchar(255) CHARACTER SET utf8mb4 COLLATE utf8mb4_general_ci NULL DEFAULT '' COMMENT '核销码',
`verifier_id` int(11) NULL DEFAULT 0 COMMENT '核销员ID',
`verifier_store_id` int(11) NULL DEFAULT 0 COMMENT '自提门店ID',
`support_pay_types` longtext CHARACTER SET utf8mb4 COLLATE utf8mb4_general_ci NULL COMMENT '支持的支付方式',
`evaluation_status` tinyint(2) NULL DEFAULT 0 COMMENT '是否评价 0.否 1.是',
`evaluation_time` timestamp(0) NULL DEFAULT '0000-00-00 00:00:00',
`after_sales_out` tinyint(2) NULL DEFAULT 0 COMMENT '是否过售后时间 0.否 1.是',
`after_sales_status` tinyint(2) NULL DEFAULT 0 COMMENT '是否申请售后 0.否 1.是',
`status` tinyint(2) NULL DEFAULT 1 COMMENT '订单状态 1.已完成 0.进行中',
`auto_cancel_time` timestamp(0) NULL DEFAULT NULL COMMENT '自动取消时间',
`auto_confirm_verifier_time` timestamp(0) NULL DEFAULT NULL COMMENT '自动确认收货时间',
`auto_after_sales_time` timestamp(0) NULL DEFAULT NULL COMMENT '自动售后时间',
`create_time` datetime(0) NULL DEFAULT NULL COMMENT '创建时间',
`creator` bigint(20) NULL DEFAULT NULL COMMENT '创建者',
`update_time` datetime(0) NULL DEFAULT NULL COMMENT '更新时间',
`operator` bigint(20) NULL DEFAULT NULL COMMENT '更新者',
`del_flag` tinyint(1) NOT NULL DEFAULT 0 COMMENT '1:删除 0:不删除',
PRIMARY KEY (`id`) USING BTREE,
INDEX `INDEX_TENANT_ID`(`tenant_id`) USING BTREE,
INDEX `INDEX_USER_ID`(`user_id`) USING BTREE,
INDEX `INDEX_STORE_ID`(`store_id`) USING BTREE,
INDEX `INDEX_ORDER_NO`(`order_no`) USING BTREE
) ENGINE = InnoDB AUTO_INCREMENT = 10 CHARACTER SET = utf8mb4 COLLATE = utf8mb4_general_ci ROW_FORMAT = DYNAMIC;

-- ----------------------------
-- Table structure for t_mall_order_sku
-- ----------------------------
DROP TABLE IF EXISTS `t_mall_order_sku`;
CREATE TABLE `t_mall_order_sku`  (
`id` bigint(20) NOT NULL AUTO_INCREMENT COMMENT '主键',
`tenant_id` bigint(20) NOT NULL DEFAULT 0 COMMENT '租户id',
`order_id` bigint(20) NOT NULL COMMENT '订单id',
`goods_sku_id` bigint(20) NULL DEFAULT NULL COMMENT '购买商品id',
`goods_sku_number` int(11) NULL DEFAULT NULL COMMENT '购买商品数量',
`goods_sku_price` decimal(10, 2) NULL DEFAULT NULL COMMENT '商品单价',
`total_original_price` decimal(10, 2) NULL DEFAULT NULL COMMENT '商品总价',
`total_price` decimal(10, 2) NULL DEFAULT NULL COMMENT '优惠后商品总价',
`member_discount_price` decimal(10, 2) NULL DEFAULT NULL COMMENT '会员优惠金额',
`store_discount_price` decimal(10, 2) NULL DEFAULT 0.00 COMMENT '商家改价优惠',
`goods_sku_info` longtext CHARACTER SET utf8mb4 COLLATE utf8mb4_general_ci NULL COMMENT '购买商品信息',
`refund_status` tinyint(1) NULL DEFAULT 0 COMMENT '是否退款',
`after_sales_status` tinyint(1) NULL DEFAULT 0 COMMENT '售后状态 0--未售后 1--售后中 2--售后结束',
`create_time` datetime(0) NULL DEFAULT NULL COMMENT '创建时间',
`creator` bigint(20) NULL DEFAULT NULL COMMENT '创建者',
`update_time` datetime(0) NULL DEFAULT NULL COMMENT '更新时间',
`operator` bigint(20) NULL DEFAULT NULL COMMENT '更新者',
`del_flag` tinyint(1) NOT NULL DEFAULT 0 COMMENT '1:删除 0:不删除',
PRIMARY KEY (`id`) USING BTREE,
INDEX `INDEX_TENANT_ID`(`tenant_id`) USING BTREE,
INDEX `INDEX_ORDER_ID`(`order_id`) USING BTREE,
INDEX `INDEX_GOODS_SKU_ID`(`goods_sku_id`) USING BTREE
) ENGINE = InnoDB AUTO_INCREMENT = 15 CHARACTER SET = utf8mb4 COLLATE = utf8mb4_general_ci ROW_FORMAT = DYNAMIC;

-- ----------------------------
-- Table structure for undo_log
-- ----------------------------
DROP TABLE IF EXISTS `undo_log`;
CREATE TABLE `undo_log`  (
`branch_id` bigint(20) NOT NULL COMMENT 'branch transaction id',
`xid` varchar(100) CHARACTER SET utf8 COLLATE utf8_general_ci NOT NULL COMMENT 'global transaction id',
`context` varchar(128) CHARACTER SET utf8 COLLATE utf8_general_ci NOT NULL COMMENT 'undo_log context,such as serialization',
`rollback_info` longblob NOT NULL COMMENT 'rollback info',
`log_status` int(11) NOT NULL COMMENT '0:normal status,1:defense status',
`log_created` datetime(6) NOT NULL COMMENT 'create datetime',
`log_modified` datetime(6) NOT NULL COMMENT 'modify datetime',
UNIQUE INDEX `ux_undo_log`(`xid`, `branch_id`) USING BTREE
) ENGINE = InnoDB CHARACTER SET = utf8 COLLATE = utf8_general_ci COMMENT = 'AT transaction mode undo table' ROW_FORMAT = Dynamic;

SET FOREIGN_KEY_CHECKS = 1;

3、支付数据库表设计

DROP TABLE IF EXISTS `t_mall_pay_record`;
CREATE TABLE `t_mall_pay_record`  (
`id` bigint(20) NOT NULL AUTO_INCREMENT,
`tenant_id` bigint(20) NOT NULL DEFAULT 0 COMMENT '租户id',
`user_id` bigint(20) NOT NULL COMMENT '用户id',
`order_no` varchar(32) CHARACTER SET utf8mb4 COLLATE utf8mb4_general_ci NOT NULL DEFAULT '0',
`amount` decimal(9, 2) NOT NULL,
`pay_status` tinyint(2) NOT NULL DEFAULT 0 COMMENT '支付状态：0=未支付，1=已支付, 2=已退款',
`pay_type` tinyint(2) NOT NULL DEFAULT 3 COMMENT '支付方式：1=微信支付，2=货到付款，3=余额支付，4=支付宝支付,  5=银行卡支付',
`title` varchar(128) CHARACTER SET utf8mb4 COLLATE utf8mb4_general_ci NULL DEFAULT '',
`refund` decimal(9, 2) NOT NULL DEFAULT 0.00 COMMENT '已退款金额',
`comments` varchar(255) CHARACTER SET utf8mb4 COLLATE utf8mb4_general_ci NOT NULL DEFAULT '' COMMENT '备注',
`create_time` datetime(0) NULL DEFAULT NULL COMMENT '创建时间',
`creator` bigint(20) NULL DEFAULT NULL COMMENT '创建者',
`update_time` datetime(0) NULL DEFAULT NULL COMMENT '更新时间',
`operator` bigint(20) NULL DEFAULT NULL COMMENT '更新者',
`del_flag` tinyint(2) NULL DEFAULT 0 COMMENT '1:删除 0:不删除',
PRIMARY KEY (`id`) USING BTREE,
INDEX `INDEX_TENANT_ID`(`tenant_id`) USING BTREE
) ENGINE = InnoDB AUTO_INCREMENT = 9 CHARACTER SET = utf8mb4 COLLATE = utf8mb4_general_ci ROW_FORMAT = DYNAMIC;

-- ----------------------------
-- Table structure for undo_log
-- ----------------------------
DROP TABLE IF EXISTS `undo_log`;
CREATE TABLE `undo_log`  (
`branch_id` bigint(20) NOT NULL COMMENT 'branch transaction id',
`xid` varchar(100) CHARACTER SET utf8 COLLATE utf8_general_ci NOT NULL COMMENT 'global transaction id',
`context` varchar(128) CHARACTER SET utf8 COLLATE utf8_general_ci NOT NULL COMMENT 'undo_log context,such as serialization',
`rollback_info` longblob NOT NULL COMMENT 'rollback info',
`log_status` int(11) NOT NULL COMMENT '0:normal status,1:defense status',
`log_created` datetime(6) NOT NULL COMMENT 'create datetime',
`log_modified` datetime(6) NOT NULL COMMENT 'modify datetime',
UNIQUE INDEX `ux_undo_log`(`xid`, `branch_id`) USING BTREE
) ENGINE = InnoDB CHARACTER SET = utf8 COLLATE = utf8_general_ci COMMENT = 'AT transaction mode undo table' ROW_FORMAT = Dynamic;

SET FOREIGN_KEY_CHECKS = 1;

4、账户数据库表设计

-- ----------------------------
-- Table structure for t_mall_user_account
-- ----------------------------
DROP TABLE IF EXISTS `t_mall_user_account`;
CREATE TABLE `t_mall_user_account`  (
`id` bigint(20) NOT NULL AUTO_INCREMENT COMMENT '主键',
`tenant_id` bigint(20) NOT NULL DEFAULT 0 COMMENT '租户id',
`user_id` bigint(20) NOT NULL COMMENT '用户id',
`integral` bigint(20) NOT NULL DEFAULT 0 COMMENT '积分',
`balance` decimal(10, 2) NOT NULL DEFAULT 0.00 COMMENT '余额',
`account_status` tinyint(2) NULL DEFAULT 1 COMMENT '账户状态 \'0\'禁用，\'1\' 启用',
`comments` varchar(255) CHARACTER SET utf8 COLLATE utf8_general_ci NULL DEFAULT NULL COMMENT '备注',
`create_time` datetime(0) NULL DEFAULT NULL COMMENT '创建时间',
`creator` bigint(20) NULL DEFAULT NULL COMMENT '创建者',
`update_time` datetime(0) NULL DEFAULT NULL COMMENT '更新时间',
`operator` bigint(20) NULL DEFAULT NULL COMMENT '更新者',
`del_flag` tinyint(2) NULL DEFAULT 0 COMMENT '1:删除 0:不删除',
PRIMARY KEY (`id`) USING BTREE,
INDEX `INDEX_TENANT_ID`(`tenant_id`) USING BTREE,
INDEX `INDEX_USER_ID`(`user_id`) USING BTREE
) ENGINE = InnoDB AUTO_INCREMENT = 2 CHARACTER SET = utf8 COLLATE = utf8_general_ci COMMENT = '用户账户表' ROW_FORMAT = DYNAMIC;

-- ----------------------------
-- Table structure for t_mall_user_balance_record
-- ----------------------------
DROP TABLE IF EXISTS `t_mall_user_balance_record`;
CREATE TABLE `t_mall_user_balance_record`  (
`id` bigint(20) NOT NULL AUTO_INCREMENT COMMENT '主键',
`tenant_id` bigint(20) NOT NULL DEFAULT 0 COMMENT '租户id',
`user_id` bigint(20) NOT NULL COMMENT '用户id',
`type` tinyint(2) NOT NULL COMMENT '类型：1=收入，2=支出',
`amount` decimal(10, 2) NOT NULL COMMENT '变动金额',
`comments` varchar(255) CHARACTER SET utf8mb4 COLLATE utf8mb4_general_ci NOT NULL DEFAULT '' COMMENT '备注',
`create_time` datetime(0) NULL DEFAULT NULL COMMENT '创建时间',
`creator` bigint(20) NULL DEFAULT NULL COMMENT '创建者',
`update_time` datetime(0) NULL DEFAULT NULL COMMENT '更新时间',
`operator` bigint(20) NULL DEFAULT NULL COMMENT '更新者',
`del_flag` tinyint(2) NULL DEFAULT 0 COMMENT '1:删除 0:不删除',
PRIMARY KEY (`id`) USING BTREE,
INDEX `INDEX_TENANT_ID`(`tenant_id`) USING BTREE,
INDEX `INDEX_USER_ID`(`user_id`) USING BTREE
) ENGINE = InnoDB AUTO_INCREMENT = 17 CHARACTER SET = utf8mb4 COLLATE = utf8mb4_general_ci ROW_FORMAT = DYNAMIC;

-- ----------------------------
-- Table structure for undo_log
-- ----------------------------
DROP TABLE IF EXISTS `undo_log`;
CREATE TABLE `undo_log`  (
`branch_id` bigint(20) NOT NULL COMMENT 'branch transaction id',
`xid` varchar(100) CHARACTER SET utf8 COLLATE utf8_general_ci NOT NULL COMMENT 'global transaction id',
`context` varchar(128) CHARACTER SET utf8 COLLATE utf8_general_ci NOT NULL COMMENT 'undo_log context,such as serialization',
`rollback_info` longblob NOT NULL COMMENT 'rollback info',
`log_status` int(11) NOT NULL COMMENT '0:normal status,1:defense status',
`log_created` datetime(6) NOT NULL COMMENT 'create datetime',
`log_modified` datetime(6) NOT NULL COMMENT 'modify datetime',
UNIQUE INDEX `ux_undo_log`(`xid`, `branch_id`) USING BTREE
) ENGINE = InnoDB CHARACTER SET = utf8 COLLATE = utf8_general_ci COMMENT = 'AT transaction mode undo table' ROW_FORMAT = Dynamic;

SET FOREIGN_KEY_CHECKS = 1;

5、上面的脚本中，每个数据都需要刷入了Seata分布式事务回滚需要的表脚本,在下载Seata包的seata-1.4.1\seata-1.4.1\script\client\at\db路径下

-- ----------------------------
-- Table structure for undo_log
-- ----------------------------
DROP TABLE IF EXISTS `undo_log`;
CREATE TABLE `undo_log`  (
`branch_id` bigint(20) NOT NULL COMMENT 'branch transaction id',
`xid` varchar(100) CHARACTER SET utf8 COLLATE utf8_general_ci NOT NULL COMMENT 'global transaction id',
`context` varchar(128) CHARACTER SET utf8 COLLATE utf8_general_ci NOT NULL COMMENT 'undo_log context,such as serialization',
`rollback_info` longblob NOT NULL COMMENT 'rollback info',
`log_status` int(11) NOT NULL COMMENT '0:normal status,1:defense status',
`log_created` datetime(6) NOT NULL COMMENT 'create datetime',
`log_modified` datetime(6) NOT NULL COMMENT 'modify datetime',
UNIQUE INDEX `ux_undo_log`(`xid`, `branch_id`) USING BTREE
) ENGINE = InnoDB CHARACTER SET = utf8 COLLATE = utf8_general_ci COMMENT = 'AT transaction mode undo table' ROW_FORMAT = Dynamic;

SET FOREIGN_KEY_CHECKS = 1;

三、测试代码

在GitEgg-Cloud工程下，新建gitegg-mall和gitegg-mall-client子工程，client子工程用于fegin调用
1、订单服务

  @DS("mall_order")//每一层都需要使用多数据源注解切换所选择的数据库
  @Transactional(propagation = Propagation.REQUIRES_NEW)
  @GlobalTransactional //重点 第一个开启事务的需要添加seata全局事务注解
  @Override
  public void order(List<OrderSkuDTO> orderSkuList, Long userId) {

      //获取商品的详细信息
      Result<Object> goodsSkuResult = mallGoodsFeign.queryByIds(orderSkuList.stream()
              .map(OrderSkuDTO::getGoodsSkuId)
              .collect(Collectors.toList()));
      List<Object> resultSkuList = (List<Object>) goodsSkuResult.getData();
      List<GoodsSkuDTO> goodsSkuList = new ArrayList<>();
      if(CollectionUtils.isEmpty(resultSkuList) || resultSkuList.size() != orderSkuList.size()) {
          throw new BusinessException("商品不存在");
      }
      else {
          resultSkuList.stream().forEach(goodsSku -> {
              GoodsSkuDTO goodsSkuDTO = BeanUtil.fillBeanWithMap((Map<?, ?>) goodsSku, new GoodsSkuDTO(), false);
              goodsSkuList.add(goodsSkuDTO);
          });
      }

      //扣商品库存
      List<ReduceStockDTO> reduceStockDtoList = orderSkuList.stream()
              .map(t -> new ReduceStockDTO(t.getGoodsSkuId(),t.getGoodsSkuNumber()))
              .collect(Collectors.toList());
      mallGoodsFeign.reduceStock(reduceStockDtoList);

//        //支付
      BigDecimal totalMoney = new BigDecimal(0.0d);
      for(OrderSkuDTO orderSkuDTO: orderSkuList) {
          for(GoodsSkuDTO goodsSkuDTO: goodsSkuList) {
              if(orderSkuDTO.getGoodsSkuId().equals(goodsSkuDTO.getId())) {
                  BigDecimal skuNumber = new BigDecimal(orderSkuDTO.getGoodsSkuNumber());
                  totalMoney = totalMoney.add(goodsSkuDTO.getPrice().multiply(skuNumber));
                  break;
              }
          }
      }

      mallPayFeign.pay(userId, totalMoney);

      //主订单表插入数据
      Order order = new Order();
      order.setTotalPrice(totalMoney);
      order.setTotalPayPrice(totalMoney);
      order.setExpressOriginalPrice(totalMoney);
      order.setStatus(1);
      order.setUserId(userId);
      this.save(order);

      //子订单表插入数据
      ArrayList<OrderSku> orderSkus = new ArrayList<>();
      orderSkuList.forEach(payOrderReq -> {
          OrderSku orderSku = new OrderSku();
          orderSku.setOrderId(order.getId());
          orderSku.setGoodsSkuNumber(payOrderReq.getGoodsSkuNumber());
          orderSku.setGoodsSkuId(payOrderReq.getGoodsSkuId());
          for(GoodsSkuDTO goodsSkuDTO : goodsSkuList) {
              if(payOrderReq.getGoodsSkuId().equals(goodsSkuDTO.getId())) {
                  orderSku.setGoodsSkuPrice(goodsSkuDTO.getPrice());
                  break;
              }
          }
          orderSkus.add(orderSku);
      });
      orderSkuService.saveBatch(orderSkus);
  }

2、商品服务

  @DS("mall_goods")
  @Override
  public List<GoodsSku> queryGoodsByIds(List<Long> idList) {
      return goodsSkuMapper.queryGoodsByIds(idList);
  }

  /**
   * 事务传播特性设置为 REQUIRES_NEW 开启新的事务  重要！！！！一定要使用REQUIRES_NEW
   */
  @DS("mall_goods")
  @Transactional(propagation = Propagation.REQUIRES_NEW)
  @Override
  public void reduceStock(List<ReduceStockDTO> reduceStockReqList) {
      reduceStockReqList.forEach(sku -> {
          Integer line = goodsSkuMapper.reduceStock(sku.getNumber(), sku.getSkuId());
          if(line == null || line == 0) {
              throw new BusinessException("商品不存在或库存不足");
          }
      });
  }

3、支付服务

  /**
   * 事务传播特性设置为 REQUIRES_NEW 开启新的事务    重要！！！！一定要使用REQUIRES_NEW
   */
  @DS("mall_pay")
  @Transactional(propagation = Propagation.REQUIRES_NEW)
  @Override
  public Long pay(Long userId, BigDecimal payMoney) {

      //调用gitegg-mall-user的账户扣除余额接口
      mallUserFeign.accountDeduction(userId, payMoney);

      // 插入支付记录表
      PayRecord payRecord = new PayRecord();
      payRecord.setUserId(userId);
      payRecord.setAmount(payMoney);
      payRecord.setPayStatus(GitEggConstant.Number.ONE);
      payRecord.setPayType(GitEggConstant.Number.FIVE);
      payRecordService.save(payRecord);
      return payRecord.getId();
  }

4、账户服务

  /**
   * 事务传播特性设置为 REQUIRES_NEW 开启新的事务    重要！！！！一定要使用REQUIRES_NEW
   */
  @DS("mall_user")
  @Transactional(propagation = Propagation.REQUIRES_NEW)
  @Override
  public void deduction(Long userId, BigDecimal amountOfMoney) {
      //查看账户余额是否满足扣款
      QueryUserAccountDTO queryUserAccountDTO = new QueryUserAccountDTO();
      queryUserAccountDTO.setUserId(userId);
      UserAccountDTO userAccount = this.queryUserAccount(queryUserAccountDTO);

      if(userAccount == null) {
          throw new BusinessException("用户未开通个人账户");
      }

      if(amountOfMoney.compareTo(userAccount.getBalance()) > GitEggConstant.Number.ZERO) {
          throw new BusinessException("账户余额不足");
      }
      //执行扣款
      userAccountMapper.deductionById(userAccount.getId(), amountOfMoney);

      //加入账户变动记录
      UserBalanceRecord userBalanceRecord = new UserBalanceRecord();
      userBalanceRecord.setUserId(userId);
      userBalanceRecord.setAmount(amountOfMoney);
      userBalanceRecord.setType(GitEggConstant.Number.TWO);
      userBalanceRecordService.save(userBalanceRecord);
  }

5、使用Postman测试，发送请求，然后查看数据库是否都增加了数据，正常情况下，几个数据库的表都有新增或更新
请求头
请求参数
6、测试异常情况，在代码中抛出异常，然后进行debug，查看在异常之前数据库数据是否入库，异常之后，入库数据是否已回滚。同时可观察undo_log表的数据情况。

# 在订单服务中添加
throw new BusinessException("测试异常回滚");

四、整合数据库分库分表

首先在我们整合dynamic-datasource和shardingsphere-JDBC之前，需要了解它们的异同点:dynamic-datasource从字面意思可以看出，它是动态多数据源，其主要功能是支持多数据源及数据源动态切换不支持数据分片，shardingsphere-jdbc主要功能是数据分片、读写分离，当然也支持多数据源，但是到目前为止如果要支持多数据源动态切换的话，需要自己实现，所以，这里结合两者的优势，使用dynamic-datasource的动态多数据源切换+shardingsphere-jdbc的数据分片、读写分离。
1、在gitegg-platform-bom和gitegg-platform-db中引入shardingsphere-jdbc的依赖,重新install。（注意这里使用了5.0.0-alpha版本，正式环境请使用最新发布版。）


      <!-- shardingsphere-jdbc -->
      <sharding.jdbc.version>5.0.0-alpha</sharding.jdbc.version>


          <!-- Shardingsphere-jdbc -->
          <dependency>
              <groupId>org.apache.shardingsphere</groupId>
              <artifactId>shardingsphere-jdbc-core-spring-boot-starter</artifactId>
              <version>${shardingsphere.jdbc.version}</version>
          </dependency>
          <dependency>
              <groupId>org.apache.shardingsphere</groupId>
              <artifactId>shardingsphere-jdbc-core-spring-namespace</artifactId>
              <version>${shardingsphere.jdbc.version}</version>
          </dependency>

2、在gitegg-platform-db中，新建DynamicDataSourceProviderConfig类，自定义DynamicDataSourceProvider完成与shardingsphere的集成

/**
 * @author GitEgg
 * @date 2021-04-23 19:06:51
 **/
@Configuration
@AutoConfigureBefore(DynamicDataSourceAutoConfiguration.class)
public class DynamicDataSourceProviderConfig {

  @Resource
  private DynamicDataSourceProperties properties;

  /**
   * shardingSphereDataSource
   */
  @Lazy
  @Resource(name = "shardingSphereDataSource")
  private DataSource shardingSphereDataSource;

  @Bean
  public DynamicDataSourceProvider dynamicDataSourceProvider() {
      Map<String, DataSourceProperty> datasourceMap = properties.getDatasource();
      return new AbstractDataSourceProvider() {
          @Override
          public Map<String, DataSource> loadDataSources() {
              Map<String, DataSource> dataSourceMap = createDataSourceMap(datasourceMap);
              dataSourceMap.put("sharding", shardingSphereDataSource);
              return dataSourceMap;
          }
      };
  }

  /**
   * 将动态数据源设置为首选的
   * 当spring存在多个数据源时, 自动注入的是首选的对象
   * 设置为主要的数据源之后，就可以支持shardingsphere-jdbc原生的配置方式了
   */
  @Primary
  @Bean
  public DataSource dataSource(DynamicDataSourceProvider dynamicDataSourceProvider) {
      DynamicRoutingDataSource dataSource = new DynamicRoutingDataSource();
      dataSource.setPrimary(properties.getPrimary());
      dataSource.setStrict(properties.getStrict());
      dataSource.setStrategy(properties.getStrategy());
      dataSource.setProvider(dynamicDataSourceProvider);
      dataSource.setP6spy(properties.getP6spy());
      dataSource.setSeata(properties.getSeata());
      return dataSource;
  }
}

3、新建用来分库的数据库表gitegg_cloud_mall_order0和gitegg_cloud_mall_order1，复制gitegg_cloud_mall_order中的表结构。
4、在Nacos中分别配置shardingsphere-jdbc和多数据源

# shardingsphere 配置
shardingsphere:
  props:
    sql:
      show: true
  datasource:
    common:
      type: com.alibaba.druid.pool.DruidDataSource
      validationQuery: SELECT 1 FROM DUAL        
    names: shardingorder0,shardingorder1
    shardingorder0:
      type: com.alibaba.druid.pool.DruidDataSource
      url: jdbc:mysql://127.0.0.1/gitegg_cloud_mall_order0?zeroDateTimeBehavior=convertToNull&useUnicode=true&characterEncoding=utf8&all owMultiQueries=true&serverTimezone=Asia/Shanghai
      username: root
      password: root
    shardingorder1:
      type: com.alibaba.druid.pool.DruidDataSource
      url: jdbc:mysql://127.0.0.1/gitegg_cloud_mall_order1?zeroDateTimeBehavior=convertToNull&useUnicode=true&characterEncoding=utf8&all owMultiQueries=true&serverTimezone=Asia/Shanghai
      username: root
      password: root
  rules:
    sharding:
      tables:
        t_mall_order:
          actual-data-nodes: shardingorder$->{0..1}.t_mall_order$->{0..1}
          # 配置分库策略
          databaseStrategy:
            standard:
              shardingColumn: id
              shardingAlgorithmName: database-inline
          table-strategy:
            standard:
              sharding-column: id
              sharding-algorithm-name: table-inline-order
          key-generate-strategy:
            column: id
            key-generator-name: snowflake
        t_mall_order_sku:
          actual-data-nodes: shardingorder$->{0..1}.t_mall_order_sku$->{0..1}
          # 配置分库策略
          databaseStrategy:
            standard:
              shardingColumn: id
              shardingAlgorithmName: database-inline
          table-strategy:
            standard:
              sharding-column: id
              sharding-algorithm-name: table-inline-order-sku
          key-generate-strategy:
            column: id
            key-generator-name: snowflake
      sharding-algorithms:
        database-inline:
          type: INLINE
          props:
            algorithm-expression: shardingorder$->{id % 2}
        table-inline-order:
          type: INLINE
          props:
            algorithm-expression: t_mall_order$->{id % 2}
        table-inline-order-sku:
          type: INLINE
          props:
            algorithm-expression: t_mall_order_sku$->{id % 2}
      key-generators:
        snowflake:
          type: SNOWFLAKE
          props:
            worker-id: 123

5、修改OrderServiceImpl.java的下单方法order注解，数据源选择sharding

  @DS("sharding")//每一层都需要使用多数据源注解切换所选择的数据库
  @Transactional(propagation = Propagation.REQUIRES_NEW)
  @GlobalTransactional //重点 第一个开启事务的需要添加seata全局事务注解
  @Override
  public void order(List<OrderSkuDTO> orderSkuList, Long userId) {
     ......
  }

6、postman模拟测试调用下单接口，观察数据库gitegg_cloud_mall_order0和gitegg_cloud_mall_order1里面的order表数据变化，我们发现，数据记录根据id取余存放到对应的库和表。这里的配置是使用order表的id，在实际生产环境中，需要根据实际情况来选择合适的分库分表策略。
订单数据根据分库分表策略存储

7、测试引入shardingsphere-jdbc后分布式事务是否正常，在OrderServiceImpl.java的下单方法order中的最后主动抛出异常，saveBatch之后打断点，使用postman模拟测试调用下单接口，到达断点时，查看数据是否入库，放开断点，抛出异常，然后再查看数据是否被回滚。

      orderSkuService.saveBatch(orderSkus);
      throw new BusinessException("测试异常");

gitegg-mall-pay最新数据记录已入库
抛出异常后gitegg-mall-pay入库数据被回滚

备注：
1、sharding-jdbc启动时报错java.sql.SQLFeatureNotSupportedException: isValid
解决：这个是4.x版本的问题，官方会在5.x结局这个问题，目前解决方案是关闭sql健康检查。

本文源码在https://gitee.com/wmz1930/GitEgg 的chapter-27（未使用shardingsphere-jdbc分库分表）和chapter-27-shardingsphere-jdbc（使用shardingsphere-jdbc分库分表）分支。

GitEgg-Cloud是一款基于SpringCloud整合搭建的企业级微服务应用开发框架，开源项目地址:

Gitee: https://gitee.com/wmz1930/GitEgg

GitHub: https://github.com/wmz1930/GitEgg

SpringCloud微服务实战——搭建企业级开发框架（二十七）：集成多数据源+Seata分布式事务+读写分离+分库分表

一、Seata安装配置

如果服务器有多网卡，存在多个ip地址，-h后面一定要加可以访问的ip地址

二、Seata安装成功后，我们需要在微服务中集成Seata客户端

三、数据库表设计

1、商品数据库表设计

2、订单数据库表设计

3、支付数据库表设计

4、账户数据库表设计

5、上面的脚本中，每个数据都需要刷入了Seata分布式事务回滚需要的表脚本,在下载Seata包的seata-1.4.1\seata-1.4.1\script\client\at\db路径下

三、测试代码

四、整合数据库分库分表

GitEgg-Cloud是一款基于SpringCloud整合搭建的企业级微服务应用开发框架，开源项目地址:

全栈程序猿

引用和评论

SpringCloud微服务实战——搭建企业级开发框架（四十四）：Actuator实现监控告警系统

在Java程序中监听mysql的binlog

Mysql数据库sql优化案例

前端微服务跨域配置解决办法，devServer为例

Go 程序如何实现优雅退出？来看看 K8s 是怎么做的——上篇

MySQL 回收表碎片实践教程

高效工程师的七个习惯