精品seata综合示例：订单-库存-扣款

Posted 2022-10-18

有关seata的安装，请参看博客：https://blog.51cto.com/lianghecai/5759100

业务需求：下单—减库存—扣钱—改订单状态

当用户下单时，会在订单服务中创建一个订单，然后通过远程调用库存服务来扣减下单商品的库存；再通过远程调用账户服务来扣减用户账户里面的余额；最后订单服务中修改订单状态。

需要涉及到三个模块：

模块名	数据库名	port	Context-path	Application-name
fafu-account	seata_account	3001	fafu	Fafu-account-3001
fafu-storage	seata_storage	4001	fafu	Fafu-storage-4001
fafu-order	seata_order	5001	fafu	Fafu-order-5001

可以看出：该操作需要跨域三个数据库，有两次远程调用，需要用到分布式事务技术。

数据库准备

创建seata_order数据库，然后在其中创建表tb_order：

CREATE TABLE `tb_order`  (
  `id` bigint UNSIGNED NOT NULL COMMENT 订单编号,
  `user_id` bigint UNSIGNED NULL DEFAULT NULL COMMENT 用户编号,
  `sku_id` bigint UNSIGNED NULL DEFAULT NULL COMMENT 商品编号,
  `count` int UNSIGNED NULL DEFAULT NULL COMMENT 数量,
  `money` decimal(10, 3) UNSIGNED NULL DEFAULT NULL COMMENT 金额,
  `state` int UNSIGNED NULL DEFAULT NULL COMMENT 状态,
  PRIMARY KEY (`id`) USING BTREE
) ENGINE = InnoDB CHARACTER SET = utf8mb4 COLLATE = utf8mb4_0900_ai_ci ROW_FORMAT = Dynamic;

上面表的设计只是模拟性质的，大概对应的是拼多多的业务，一个订单中只能购买1种商品，但这种商品可以购买多个。

创建seata_storage数据库，然后在其中创建表tb_storage

CREATE TABLE `tb_storage`  (
  `id` bigint UNSIGNED NOT NULL COMMENT 编号,
  `sku_id` bigint UNSIGNED NULL DEFAULT NULL COMMENT 商品编号,
  `total` int UNSIGNED NULL DEFAULT NULL COMMENT 总库存,
  `used` int UNSIGNED NULL DEFAULT NULL COMMENT 已用库存,
  PRIMARY KEY (`id`) USING BTREE
) ENGINE = InnoDB CHARACTER SET = utf8mb4 COLLATE = utf8mb4_0900_ai_ci ROW_FORMAT = Dynamic;

创建seata_account数据库，然后在其中创建表tb_account：

CREATE TABLE `tb_account`  (
  `id` bigint UNSIGNED NOT NULL COMMENT 编号,
  `total` decimal(11, 3) UNSIGNED NULL DEFAULT NULL COMMENT 总额度,
  `used` decimal(11, 3) UNSIGNED NULL DEFAULT NULL COMMENT 已用额度,
  PRIMARY KEY (`id`) USING BTREE
) ENGINE = InnoDB CHARACTER SET = utf8mb4 COLLATE = utf8mb4_0900_ai_ci ROW_FORMAT = Dynamic;

依次在三个数据库创建undo_log 表，用于保存需要回滚的数据：

CREATE TABLE `undo_log`  (
  `id` bigint NOT NULL AUTO_INCREMENT,
  `branch_id` bigint NOT NULL,
  `xid` varchar(100) CHARACTER SET utf8 COLLATE utf8_general_ci NOT NULL,
  `context` varchar(128) CHARACTER SET utf8 COLLATE utf8_general_ci NOT NULL,
  `rollback_info` longblob NOT NULL,
  `log_status` int NOT NULL,
  `log_created` datetime NOT NULL,
  `log_modified` datetime NOT NULL,
  `ext` varchar(100) CHARACTER SET utf8 COLLATE utf8_general_ci NULL DEFAULT NULL,
  PRIMARY KEY (`id`) USING BTREE,
  UNIQUE INDEX `ux_undo_log`(`xid` ASC, `branch_id` ASC) USING BTREE
) ENGINE = InnoDB CHARACTER SET = utf8 COLLATE = utf8_general_ci ROW_FORMAT = Dynamic;

注：每个服务对应的业务数据库都需要创建这个日志表。

项目共同部分

Nacos中

在Nacos添加给客户端加载使用的配置seata-client-demo.yml：

• Data Id：seata-client-demo.yml
• Group：SEATA_GROUP

注意：上面配置中的group和nampespace都是在Nacos中创建命名空间时创建的。 上面配置的具体内容如下：

# seata配置
seata:
  enabled: true
  application-id: seata-cilent-deme
  # Seata 事务组编号，此处需于 seata 相同
  tx-service-group: default-tx-group
  config:
    type: nacos
    nacos:
      # nacos ip地址
      server-addr: 127.0.0.1:8848
      group: SEATA_GROUP
      namespace: 99970600-6f67-43fd-a2a2-4e9795947bf4
      data-id: seata-server.properties
  registry:
    type: nacos
    nacos:
      application: seata-server
      # nacos ip地址
      server-addr: 127.0.0.1:8848
      group: SEATA_GROUP
      namespace: 99970600-6f67-43fd-a2a2-4e9795947bf4

依赖

<dependency>
    <groupId>org.projectlombok</groupId>
    <artifactId>lombok</artifactId>
</dependency>

<dependency>
    <groupId>org.springframework.boot</groupId>
    <artifactId>spring-boot-starter-web</artifactId>
</dependency>
<dependency>
    <groupId>org.springframework.boot</groupId>
    <artifactId>spring-boot-starter-actuator</artifactId>
</dependency>
<dependency>
    <groupId>org.springframework.boot</groupId>
    <artifactId>spring-boot-devtools</artifactId>
    <scope>runtime</scope>
    <optional>true</optional>
</dependency>

<dependency>
    <groupId>org.springframework.boot</groupId>
    <artifactId>spring-boot-starter-test</artifactId>
    <scope>test</scope>
</dependency>

<!-- 持久层相关 -->
<dependency>
    <groupId>mysql</groupId>
    <artifactId>mysql-connector-java</artifactId>
</dependency>
<dependency>
    <groupId>com.alibaba</groupId>
    <artifactId>druid-spring-boot-starter</artifactId>
</dependency>
<dependency>
    <groupId>com.baomidou</groupId>
    <artifactId>mybatis-plus-boot-starter</artifactId>
</dependency>

<dependency>
    <groupId>org.springframework.cloud</groupId>
    <artifactId>spring-cloud-starter-openfeign</artifactId>
</dependency>
<!--  nacos自从2020版本之后不再整合的是Netflix，也就没有ribbon了 -->
<dependency>
    <groupId>org.springframework.cloud</groupId>
    <artifactId>spring-cloud-starter-loadbalancer</artifactId>
</dependency>
<dependency>
    <groupId>com.alibaba.cloud</groupId>
    <artifactId>spring-cloud-starter-alibaba-nacos-discovery</artifactId>
</dependency>
<dependency>
    <groupId>com.alibaba.cloud</groupId>
    <artifactId>spring-cloud-starter-alibaba-sentinel</artifactId>
</dependency>
<!-- sentinel配置持久化到Nacos -->
<dependency>
    <groupId>com.alibaba.csp</groupId>
    <artifactId>sentinel-datasource-nacos</artifactId>
</dependency>

<dependency>
    <groupId>io.seata</groupId>
    <artifactId>seata-spring-boot-starter</artifactId>
    <version>1.5.2</version>
</dependency>
<dependency>
    <groupId>com.alibaba.cloud</groupId>
    <artifactId>spring-cloud-starter-alibaba-seata</artifactId>
    <exclusions>
        <exclusion>
            <groupId>io.seata</groupId>
            <artifactId>seata-spring-boot-starter</artifactId>
        </exclusion>
    </exclusions>
</dependency>

Result

@Getter
@ToString
public class Result<T> 
    /**
     * 请求响应状态码
     */
    private int code;
    /**
     * 请求结果描述信息
     */
    private String msg;
    /**
     * 请求结果数据
     */
    private T data;
    public Result<T> setCode(int code) 
        this.code = code;
        return this;
    

    public Result<T> setMsg(String msg) 
        this.msg = msg;
        return this;
    

    public Result<T> setData(T data) 
        this.data = data;
        return this;
    

    public static void main(String[] args) 
        Result<Object> result = new Result<>().addData("a", "aaa", "b", 123, "c", new Date());
        System.out.println(result);
    

    /**
     * 将key-value形式的成对出现的参数转换为JSON
     *
     * @param objs
     * @return
     */
    public Result<T> addData(Object... objs) 
        if (objs.length % 2 != 0) 
            throw new RuntimeException("参数个数不对");
        
        for (int i = 0; i < objs.length; i += 2) 
            if (!(objs[i] instanceof String)) 
                throw new RuntimeException("奇数参数必须为字符串");
            
        

        Map<String, Object> map = new HashMap<>();
        for (int i = 0; i < objs.length; i += 2) 
            map.put((String) objs[i], objs[i + 1]);
        

        this.data = (T) map;
        return this;
    

ResultUtil

public class ResultUtil 
    public static <T> Result<T> success() 
        Result<T> result = new Result<>();
        result.setCode(200);
        result.setMsg("成功");
        return result;
    

    public static <T> Result<T> success(Integer code, String msg) 
        Result<T> result = new Result<>();
        result.setCode(code);
        result.setMsg(msg);
        return result;
    

    public static <T> Result<T> success(T data) 
        Result<T> result = new Result<>();
        result.setCode(200);
        result.setMsg("成功");
        result.setData(data);
        return result;
    

    public static <T> Result<T> error() 
        Result<T> result = new Result<>();
        result.setCode(500);
        result.setMsg("失败");
        return result;
    

    public static <T> Result<T> error(Integer code, String msg) 
        Result<T> result = new Result<>();
        result.setCode(code);
        result.setMsg(msg);
        return result;
    


    public static void main(String[] args) 
        Result error = ResultUtil.error().addData("a", "aaaa", "b", LocalDateTime.now(), "c", true);
        System.out.println(error);

        //ResultBean res = ResultBeanUtil.success(DataUtil.build(UserVO.class));
        //System.out.println(res);
    

druid数据源代理配置

• MyBatis版本

@Configuration
public class DataSourceProxyConfig 
    @Bean
    @ConfigurationProperties(prefix = "spring.datasource")
    public DataSource druidDataSource() 
        DruidDataSource druidDataSource = new DruidDataSource();
        return druidDataSource;
    
    @Bean
    public SqlSessionFactory sqlSessionFactory(DataSource dataSource) throws Exception 
        SqlSessionFactoryBean factoryBean = new SqlSessionFactoryBean();
        factoryBean.setDataSource(dataSource);
        factoryBean.setMapperLocations(new PathMatchingResourcePatternResolver()
                .getResources("classpath*:/mapper/*.xml"));
        return factoryBean.getObject();
    

• MyBatisPlus版本

@Configuration
public class DataSourceProxyConfig 
    @Bean
    @ConfigurationProperties(prefix = "spring.datasource.druid")
    public DataSource druidDataSource() 
        DruidDataSource druidDataSource = new DruidDataSource();
        return druidDataSource;
    

在2.2.0.RELEASE及以后，数据源代理自动实现了，不需要再手动去配置一个代理类。

seata-account-3001

第一步；创建项目，添加依赖，修改application.yml文件：

server:
  port: 3001
  servlet:
    context-path: /seata

spring:
  application:
    name: seata-account-3001
  #配置数据源
  datasource:
    type: com.alibaba.druid.pool.DruidDataSource
    druid:
      driver-class-name: com.mysql.cj.jdbc.Driver
      url: jdbc:mysql://127.0.0.1:3306/seata_account?useSSL=false&serverTimezone=GMT%2B8&useUnicode=true&characterEncoding=UTF8&autoReconnect=true&failOverReadOnly=false
      username: root
      password: root
      # 连接池的配置信息:初始化大小，最小，最大
      initial-size: 8
      min-idle: 1
      max-active: 20
      # 配置获取连接等待超时的时间
      max-wait: 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
      time-between-eviction-runs-millis: 60000
      # 配置一个连接在池中最小生存的时间，单位是毫秒
      min-evictable-idle-time-millis: 300000
      #验证库是否正常sql
      validation-query: select x from dual
      #空闲时验证，防止连接断开
      test-while-idle: true
      test-on-borrow: false
      test-on-return: false
      # 打开PSCache，并且指定每个连接上PSCache的大小
      pool-prepared-statements: true
      max-open-prepared-statements: 20
      max-pool-prepared-statement-per-connection-size: 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: stat,wall,slf4j
      # 通过connectProperties属性来打开mergeSql功能；慢SQL记录
      connection-properties: druid.stat.mergeSql=true;druid.stat.slowSqlMillis=5000
      # 合并多个DruidDataSource的监控数据
      use-global-data-source-stat: true
  cloud:
    nacos:
      discovery:
        server-addr: 127.0.0.1:8848 # 配置nacos地址
    sentinel:
      transport:
        # 配置sentinel dashboard地址
        dashboard: 127.0.0.1:8080
        # 默认端口8719，假如被占用会自动从8719开始依次+1扫描，直至找到未被占用的端口
        port: 8719
  config:
    import:
      - optional:nacos:seata-client-demo.yml

seata:
  service:
    vgroup-mapping:
      default_tx_group: default
    grouplist:
      default: 127.0.0.1:8091

management:
  endpoints:
    web:
      exposure:
        include: * #暴露出所有的端点

#MyBatisPlus相关配置
mybatis-plus:
  #mapper映射文件位置，多个目录用逗号或者分号分隔（告诉 Mapper 所对应的 XML 文件位置）
  mapper-locations: classpath:mapper/*.xml
  #实体扫描，多个package用逗号或者分号分隔
  typeAliasesPackage: com.seata.account.domain
  #  以下配置均有默认值,可以不设置
  global-config:
    db-config:
      #主键类型
      id-type: auto

      table-underline: true

      #数据库大写下划线转换
      capital-mode: true
  configuration:
    # 是否开启自动驼峰命名规则映射:从数据库列名到Java属性驼峰命名的类似映射
    map-underscore-to-camel-case: true

    cache-enabled: false
    #配置JdbcTypeForNull
    jdbc-type-for-null: null
    # 如果查询结果中包含空值的列，则 MyBatis 在映射的时候，不会映射这个字段
    call-setters-on-nulls: true
    # 将执行的sql打印出来，在开发或测试的时候可以用
    log-impl: org.apache.ibatis.logging.stdout.StdOutImpl

第二步：项目主启动类：

@EnableDiscoveryClient
@SpringBootApplication
public class SeataAccountApplication 
    public static void main(String[] args) 
        SpringApplication.run(SeataAccountApplication.class, args);
    

第三步：利用MyBatisCodeHelperPro生成Account表对应的实体类、Mapper、Service代码。

第四步：控制器AccountController

@RestController
@RequestMapping("/account")
public class AccountController 
    @Resource
    private AccountService accountService;

    @PostMapping("/decrease")
    Result decrease(@RequestParam("userId") Long userId, @RequestParam("money") BigDecimal money) 
        //模拟异常，全局事务回滚
        // System.out.println(3/0);      //--------------------①
        accountService.update(Wrappers.<Account>lambdaUpdate()
                .eq(Account::getId, userId)
                .setSql("used = used + " + money)
        );
        return ResultUtil.success().setMsg("扣款成功");
    

启动项目，在Nacos中会看到：

post方式请求：http://localhost:3001/seata/account/decrease?userId=1&money=15
能够正常访问。

seata-storage-4001

第一步；创建项目，添加依赖，修改application.yml文件：

server:
  port: 4001
  servlet:
    context-path: /seata

spring:
  application:
    name: seata-storage-4001
  #配置数据源
  datasource:
    type: com.alibaba.druid.pool.DruidDataSource
    druid:
      driver-class-name: com.mysql.cj.jdbc.Driver
      url: jdbc:mysql://127.0.0.1:3306/seata_storage?useSSL=false&serverTimezone=GMT%2B8&useUnicode=true&characterEncoding=UTF8&autoReconnect=true&failOverReadOnly=false
      username: root
      password: root
      # 连接池的配置信息:初始化大小，最小，最大
      initial-size: 8
      min-idle: 1
      max-active: 20
      # 配置获取连接等待超时的时间
      max-wait: 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
      time-between-eviction-runs-millis: 60000
      # 配置一个连接在池中最小生存的时间，单位是毫秒
      min-evictable-idle-time-millis: 300000
      #验证库是否正常sql
      validation-query: select x from dual
      #空闲时验证，防止连接断开
      test-while-idle: true
      test-on-borrow: false
      test-on-return: false
      # 打开PSCache，并且指定每个连接上PSCache的大小
      pool-prepared-statements: true
      max-open-prepared-statements: 20
      max-pool-prepared-statement-per-connection-size: 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: stat,wall,slf4j
      # 通过connectProperties属性来打开mergeSql功能；慢SQL记录
      connection-properties: druid.stat.mergeSql=true;druid.stat.slowSqlMillis=5000
      # 合并多个DruidDataSource的监控数据
      use-global-data-source-stat: true
  cloud:
    nacos:
      discovery:
        server-addr: 127.0.0.1:8848 # 配置nacos地址
    sentinel:
      transport:
        # 配置sentinel dashboard地址
        dashboard: 127.0.0.1:8080
        # 默认端口8719，假如被占用会自动从8719开始依次+1扫描，直至找到未被占用的端口
        port: 8719
  config:
    import:
      - optional:nacos:seata-client-demo.yml

seata:
  service:
    vgroup-mapping:
      default_tx_group: default
    grouplist:
      default: 127.0.0.1:8091

management:
  endpoints:
    web:
      exposure:
        include: * #暴露出所有的端点

#MyBatisPlus相关配置
mybatis-plus:
  #mapper映射文件位置，多个目录用逗号或者分号分隔（告诉 Mapper 所对应的 XML 文件位置）
  mapper-locations: classpath:mapper/*.xml
  #实体扫描，多个package用逗号或者分号分隔
  typeAliasesPackage: com.seata.storage.domain
  #  以下配置均有默认值,可以不设置
  global-config:
    db-config:
      #主键类型
      id-type: auto

      table-underline: true

      #数据库大写下划线转换
      capital-mode: true
  configuration:
    # 是否开启自动驼峰命名规则映射:从数据库列名到Java属性驼峰命名的类似映射
    map-underscore-to-camel-case: true

    cache-enabled: false
    #配置JdbcTypeForNull
    jdbc-type-for-null: null
    # 如果查询结果中包含空值的列，则 MyBatis 在映射的时候，不会映射这个字段
    call-setters-on-nulls: true
    # 将执行的sql打印出来，在开发或测试的时候可以用
    log-impl: org.apache.ibatis.logging.stdout.StdOutImpl

第二步：项目主启动类添加注解

@EnableFeignClients
@EnableDiscoveryClient
@SpringBootApplication(exclude = DataSourceAutoConfiguration.class)//取消数据源自动创建
public class SeataStorageApplication 
    public static void main(String[] args) 
        SpringApplication.run(SeataStorageApplication.class, args);
    

第三步：利用MyBatisCodeHelperPro生成Storage表对应的实体类、Mapper、Service代码。

第四步：控制器StorageController

@RestController
@RequestMapping("/storage")
public class StorageController 
    @Resource
    private StorageService storageService;

    @PostMapping("/decrease")
    Result decrease(@RequestParam("skuId") Long skuId, @RequestParam("count") Integer count) 
        storageService.update(Wrappers.<Storage>lambdaUpdate()
                .eq(Storage::getSkuId, skuId)
                .setSql("used= used+" + count)
        );
        return ResultUtil.success().setMsg("库存扣减成功");
    

启动项目，在Nacos中会看到：

post方式请求：http://localhost:4001/seata/storage/decrease?skuId=1&money=3
能够正常访问。

seata-order-5001

第一步；创建项目，添加依赖，修改application.yml文件：

server:
  port: 5001
  servlet:
    context-path: /seata

spring:
  application:
    name: seata-order-5001
  #配置数据源
  datasource:
    type: com.alibaba.druid.pool.DruidDataSource
    druid:
      driver-class-name: com.mysql.cj.jdbc.Driver
      url: jdbc:mysql://127.0.0.1:3306/seata_order?useSSL=false&serverTimezone=GMT%2B8&useUnicode=true&characterEncoding=UTF8&autoReconnect=true&failOverReadOnly=false
      username: root
      password: root
      # 连接池的配置信息:初始化大小，最小，最大
      initial-size: 8
      min-idle: 1
      max-active: 20
      # 配置获取连接等待超时的时间
      max-wait: 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
      time-between-eviction-runs-millis: 60000
      # 配置一个连接在池中最小生存的时间，单位是毫秒
      min-evictable-idle-time-millis: 300000
      #验证库是否正常sql
      validation-query: select x from dual
      #空闲时验证，防止连接断开
      test-while-idle: true
      test-on-borrow: false
      test-on-return: false
      # 打开PSCache，并且指定每个连接上PSCache的大小
      pool-prepared-statements: true
      max-open-prepared-statements: 20
      max-pool-prepared-statement-per-connection-size: 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: stat,wall,slf4j
      # 通过connectProperties属性来打开mergeSql功能；慢SQL记录
      connection-properties: druid.stat.mergeSql=true;druid.stat.slowSqlMillis=5000
      # 合并多个DruidDataSource的监控数据
      use-global-data-source-stat: true
  cloud:
    nacos:
      discovery:
        server-addr: 127.0.0.1:8848 # 配置nacos地址
    sentinel:
      transport:
        # 配置sentinel dashboard地址
        dashboard: 127.0.0.1:8080
        # 默认端口8719，假如被占用会自动从8719开始依次+1扫描，直至找到未被占用的端口
        port: 8719
  config:
    import:
      - optional:nacos:seata-client-demo.yml

seata:
  service:
    vgroup-mapping:
      default_tx_group: default
    grouplist:
      default: 127.0.0.1:8091

management:
  endpoints:
    web:
      exposure:
        include: * #暴露出所有的端点

#MyBatisPlus相关配置
mybatis-plus:
  #mapper映射文件位置，多个目录用逗号或者分号分隔（告诉 Mapper 所对应的 XML 文件位置）
  mapper-locations: classpath:mapper/*.xml
  #实体扫描，多个package用逗号或者分号分隔
  typeAliasesPackage: com.seata.order.domain
  #  以下配置均有默认值,可以不设置
  global-config:
    db-config:
      #主键类型
      id-type: auto

      table-underline: true

      #数据库大写下划线转换
      capital-mode: true
  configuration:
    # 是否开启自动驼峰命名规则映射:从数据库列名到Java属性驼峰命名的类似映射
    map-underscore-to-camel-case: true

    cache-enabled: false
    #配置JdbcTypeForNull
    jdbc-type-for-null: null
    # 如果查询结果中包含空值的列，则 MyBatis 在映射的时候，不会映射这个字段
    call-setters-on-nulls: true
    # 将执行的sql打印出来，在开发或测试的时候可以用
    log-impl: org.apache.ibatis.logging.stdout.StdOutImpl

第二步：项目主启动类上添加注解：

@EnableFeignClients
@EnableDiscoveryClient
@SpringBootApplication(exclude = DataSourceAutoConfiguration.class)//取消数据源自动创建
public class SeataOrderApplication 
    public static void main(String[] args) 
        SpringApplication.run(SeataOrderApplication.class, args);
    

第三步：利用MyBatisCodeHelperPro生成Order表对应的实体类、Mapper、Service代码。

第四步：在feign包下创建：

• AccountFeignService.java

@FeignClient(name = "seata-account-3001", path = "/seata")
@RequestMapping("/account")
public interface AccountFeignService 
    /**
     * 将skuId对应的商品的库存减少count
     * @param userId
     * @param money
     * @return
     */
    @PostMapping("/decrease")
    Result decrease(@RequestParam("userId") Long userId, @RequestParam("money") BigDecimal money);

• StorageFeignService.java

@FeignClient(name = "seata-storage-4001", path = "/seata")
@RequestMapping("/storage")
public interface StorageFeignService 
    /**
     * 将skuId对应的商品的库存减少count
     * @param skuId
     * @param count
     * @return
     */
    @PostMapping("/decrease")
    Result decrease(@RequestParam("skuId") Long skuId, @RequestParam("count") Integer count);

第五步：修改OrderServiceImpl的代码如下所示：

@Service
public class OrderServiceImpl extends ServiceImpl<OrderMapper, Order> implements OrderService 
    @Resource
    private OrderMapper orderMapper;
    @Resource
    private StorageFeignService storageFeignService;
    @Resource
    private AccountFeignService accountFeignService;

    @Override
    //  @GlobalTransactional(rollbackFor = Exception.class)  //全局异常   //----------------②
    public void create(Order order) 
        System.out.println("user建订单");
        orderMapper.insert(order);

        System.out.println("扣减库存");
        storageFeignService.decrease(order.getSkuId(),order.getCount());

        System.out.println("账户余额扣减");
        accountFeignService.decrease(order.getUserId(),order.getMoney());

        System.out.println("修改订单的状态");
        //1表示完成状态
        orderMapper.updateStateByUserId(order.getUserId(),1);

        System.out.println("下单结束");
    

第六步：控制器OrderController

@RestController
@RequestMapping("/order")
public class OrderController 
    @Resource
    private OrderService orderService;

    @GetMapping("/create")
    public Result create(Order order)
        orderService.create(order);
        return ResultUtil.success().setMsg("下单成功！");
    

启动项目：

测试

url：http://localhost:5001/seata/order/create?userId=1&skuId=1&count=10&money=30

• 保持编号①和编号②处于注释状态：请求url，tb_order表增加了一条记录，用户账户用掉30，库存用掉10。
• 保持编号②处于注释状态，去掉编号①前的注释：请求url，tb_order表增加了一条记录但状态值为null，用户账户没有发生变化，库存用掉10。没有能够保证数据的完整性
• 去掉编号①和编号②前的注释：请求url，三个表的数据都没有发生变化。

结论： 微服务中使用alibaba的seata技术实现全局事务控制只需要在相应的Service实现类中添加@GlobalTransactional注解就可以了。