学习seata分布式事务管理AT模式的笔记:用到的组件分别是 springcloud-eureka-feign-mybatis-seata
参考及源码:https://github.com/seata/seata-samples/tree/master/springcloud-eureka-feign-mybatis-seata
引用:https://github.com/seata/seata-samples/blob/master/springcloud-eureka-feign-mybatis-seata/README.md
业务原型:
订单微服务(order-server):添加订单(本地调用)、减少库存(远程调用storage-server)、修改账户余额(远程调用account-server)
仓库微服务(storage-server):减少库存
账户微服务(account-server):修改帐户余额、修改订单状态(远程调用order-server ---->成环调用)
注意:
1、水平线以下是我复制的官网的操作文档内容稍作修改,自己尝试过整理,但始终不如下面的文档精简,干脆直接用原文!
2、在进行分布式事务的学习之前建议下载源码后,先将下面的注解注释掉,修改对应的mybatis的配置,先用最简单的spring-boot的方式,把每个服务跑起来!
3.别忘了建数据库,sql脚本在类路径下的resorce文件夹里。
4.如果能正常跑起来,那么恭喜你,你可以进入下面的学习,下面的步骤学会,分布式事务也就基本会了!
1.整合seata的demo,此demo都配置好了,拉下来按照步骤,直接可以跑起来观察效果。
2.自己项目整合Seata,主要步骤如下:
关于调用成环和seata-server HA,见最后部分
注册中心:eureka
服务间调用:feign
持久层:mybatis
数据库:mysql 5.7.20
Springboot:2.1.7.RELEASE
Springcloud:Greenwich.SR2
jdk:1.8
seata:0.8
使用不同组件,配置情况不同,可参考其他sample;
demo分为四个项目,单独启动。
order服务关键代码如下:
@Override
@GlobalTransactional(name = "fsp-create-order",rollbackFor = Exception.class) //此注解开启全局事务
public void create(Order order) {
//本地方法 创建订单
orderDao.create(order);
//远程方法 扣减库存
storageApi.decrease(order.getProductId(),order.getCount());
//远程方法 扣减账户余额 可在accountServiceImpl中模拟异常
accountApi.decrease(order.getUserId(),order.getMoney());
}
seata-server中,/conf目录下,有两个配置文件,需要结合自己的情况来修改:
1.file.conf
里面有事务组配置,锁配置,事务日志存储等相关配置信息,由于此demo使用db存储事务信息,我们这里要修改store中的配置:
## transaction log store
store {
## store mode: file、db
mode = "db" 修改这里,表明事务信息用db存储
## file store 当mode=db时,此部分配置就不生效了,这是mode=file的配置
file {
dir = "sessionStore"
# branch session size , if exceeded first try compress lockkey, still exceeded throws exceptions
max-branch-session-size = 16384
# globe session size , if exceeded throws exceptions
max-global-session-size = 512
# file buffer size , if exceeded allocate new buffer
file-write-buffer-cache-size = 16384
# when recover batch read size
session.reload.read_size = 100
# async, sync
flush-disk-mode = async
}
## database store mode=db时,事务日志存储会存储在这个配置的数据库里
db {
## the implement of javax.sql.DataSource, such as DruidDataSource(druid)/BasicDataSource(dbcp) etc.
datasource = "dbcp"
## mysql/oracle/h2/oceanbase etc.
db-type = "mysql"
driver-class-name = "com.mysql.jdbc.Driver"
url = "jdbc:mysql://116.62.62.26/seat-server" 修改这里
user = "root" 修改这里
password = "root" 修改这里
min-conn = 1
max-conn = 3
global.table = "global_table"
branch.table = "branch_table"
lock-table = "lock_table"
query-limit = 100
}
}
由于此demo我们使用db模式存储事务日志,所以,我们要创建三张表:global_table,branch_table,lock_table,建表sql在seata/seata/script/server/db/mysql.sql;
由于存储undo_log是在业务库中,所以在每个业务库中,还要创建undo_log表,建表sql在/conf/db_undo_log.sql中。
由于我自定义了事务组名称,所以这里也做了修改:
service {
#vgroup->rgroup
vgroup_mapping.fsp_tx_group = "default" 修改这里,fsp_tx_group这个事务组名称是我自定义的,一定要与client端的这个配置一致!否则会报错!
#only support single node
default.grouplist = "127.0.0.1:8091" 此配置作用参考:https://blog.csdn.net/weixin_39800144/article/details/100726116
#degrade current not support
enableDegrade = false
#disable
disable = false
#unit ms,s,m,h,d represents milliseconds, seconds, minutes, hours, days, default permanent
max.commit.retry.timeout = "-1"
max.rollback.retry.timeout = "-1"
}
其他的可以先使用默认值。
2.registry.conf
registry{}中是注册中心相关配置,config{}中是配置中心相关配置。seata中,注册中心和配置中心是分开实现的,是两个东西。
我们这里用eureka作注册中心,所以,只用修改registry{}中的:
registry {
# file 、nacos 、eureka、redis、zk、consul、etcd3、sofa
type = "eureka" 修改这里,指明注册中心使用什么
nacos {
serverAddr = "localhost"
namespace = ""
cluster = "default"
}
eureka {
serviceUrl = "http://localhost:8761/eureka" 修改这里
application = "default"
weight = "1"
}
redis {
serverAddr = "localhost:6379"
db = "0"
}
zk {
cluster = "default"
serverAddr = "127.0.0.1:2181"
session.timeout = 6000
connect.timeout = 2000
}
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"
}
}
其他的配置可以暂时使用默认值。
如果是在windows下启动seata-server,现在已经完成配置修改了,等eureka启动后,就可以启动seata-server了:执行/bin/seata-server.bat即可。
1.普通配置
client端的几个服务,都是普通的springboot整合了springCloud组件的正常服务,所以,你需要配置eureka,数据库,mapper扫描等,即使不使用seata,你也需要做,这里不做特殊说明,看代码就好。
2.特殊配置
1.application.yml
以order服务为例,除了常规配置外,这里还要配置下事务组信息:
spring:
application:
name: order-server
cloud:
alibaba:
seata:
tx-service-group: fsp_tx_group 这个fsp_tx_group自定义命名很重要,server,client都要保持一致
2.file.conf
自己新建的项目是没有这个配置文件的,copy过来,修改下面配置:
service {
#vgroup->rgroup
vgroup_mapping.fsp_tx_group = "default" 这个fsp_tx_group自定义命名很重要,server,client都要保持一致
#only support single node
default.grouplist = "127.0.0.1:8091"
#degrade current not support
enableDegrade = false
#disable
disable = false
disableGlobalTransaction = false
}
3.registry.conf
使用eureka做注册中心,仅需要修改eureka的配置即可:
registry {
# file 、nacos 、eureka、redis、zk
type = "eureka" 修改这里
nacos {
serverAddr = "localhost"
namespace = "public"
cluster = "default"
}
eureka {
serviceUrl = "http://localhost:8761/eureka" 修改这里
application = "default"
weight = "1"
}
redis {
serverAddr = "localhost:6381"
db = "0"
}
zk {
cluster = "default"
serverAddr = "127.0.0.1:2181"
session.timeout = 6000
connect.timeout = 2000
}
file {
name = "file.conf"
}
}
其他的使用默认值就好。
3.数据源代理
这个是要特别注意的地方,seata对数据源做了代理和接管,在每个参与分布式事务的服务中,都要做如下配置:
/**
* 数据源代理
* @author wangzhongxiang
*/
@Configuration
public class DataSourceConfiguration {
@Bean
@ConfigurationProperties(prefix = "spring.datasource")
public DataSource druidDataSource(){
DruidDataSource druidDataSource = new DruidDataSource();
return druidDataSource;
}
@Primary
@Bean("dataSource")
public DataSourceProxy dataSource(DataSource druidDataSource){
return new DataSourceProxy(druidDataSource);
}
@Bean
public SqlSessionFactory sqlSessionFactory(DataSourceProxy dataSourceProxy)throws Exception{
SqlSessionFactoryBean sqlSessionFactoryBean = new SqlSessionFactoryBean();
sqlSessionFactoryBean.setDataSource(dataSourceProxy);
sqlSessionFactoryBean.setMapperLocations(new PathMatchingResourcePatternResolver()
.getResources("classpath*:/mapper/*.xml"));
sqlSessionFactoryBean.setTransactionFactory(new SpringManagedTransactionFactory());
return sqlSessionFactoryBean.getObject();
}
}
然后可以模拟正常情况,异常情况,超时情况等,观察数据库即可。
这个demo,未做各种优化,如果压测,需要修改和优化一些配置,压测出错了,不一定是seata的锅,自己先排查,再去群里问问。
正常情况:
1.order
2019-09-06 15:44:33.536 INFO 53904 --- [io-8080-exec-10] i.seata.tm.api.DefaultGlobalTransaction : Begin new global transaction [192.168.158.133:8091:2021468859]
2019-09-06 15:44:33.536 INFO 53904 --- [io-8080-exec-10] c.j.order.service.OrderServiceImpl : ------->交易开始
2019-09-06 15:44:34.376 INFO 53904 --- [io-8080-exec-10] c.j.order.service.OrderServiceImpl : ------->交易结束
2019-09-06 15:44:34.593 INFO 53904 --- [io-8080-exec-10] i.seata.tm.api.DefaultGlobalTransaction : [192.168.158.133:8091:2021468859] commit status:Committed
2019-09-06 15:44:35.296 INFO 53904 --- [atch_RMROLE_6_8] i.s.core.rpc.netty.RmMessageListener : onMessage:xid=192.168.158.133:8091:2021468859,branchId=2021468861,branchType=AT,resourceId=jdbc:mysql://116.62.62.26/seat-order,applicationData=null
2019-09-06 15:44:35.297 INFO 53904 --- [atch_RMROLE_6_8] io.seata.rm.AbstractRMHandler : Branch committing: 192.168.158.133:8091:2021468859 2021468861 jdbc:mysql://116.62.62.26/seat-order null
2019-09-06 15:44:35.297 INFO 53904 --- [atch_RMROLE_6_8] io.seata.rm.AbstractRMHandler : Branch commit result: PhaseTwo_Committed
2.storage
2019-09-06 15:44:33.776 INFO 9704 --- [nio-8082-exec-1] c.j.storage.service.StorageServiceImpl : ------->扣减库存开始
2019-09-06 15:44:34.030 INFO 9704 --- [nio-8082-exec-1] c.j.storage.service.StorageServiceImpl : ------->扣减库存结束
2019-09-06 15:44:35.422 INFO 9704 --- [atch_RMROLE_5_8] i.s.core.rpc.netty.RmMessageListener : onMessage:xid=192.168.158.133:8091:2021468859,branchId=2021468864,branchType=AT,resourceId=jdbc:mysql://116.62.62.26/seat-storage,applicationData=null
2019-09-06 15:44:35.423 INFO 9704 --- [atch_RMROLE_5_8] io.seata.rm.AbstractRMHandler : Branch committing: 192.168.158.133:8091:2021468859 2021468864 jdbc:mysql://116.62.62.26/seat-storage null
2019-09-06 15:44:35.423 INFO 9704 --- [atch_RMROLE_5_8] io.seata.rm.AbstractRMHandler : Branch commit result: PhaseTwo_Committed
3.account
2019-09-06 15:44:34.039 INFO 36556 --- [nio-8081-exec-5] c.j.account.service.AccountServiceImpl : ------->扣减账户开始
2019-09-06 15:44:34.039 INFO 36556 --- [nio-8081-exec-5] c.j.account.service.AccountServiceImpl : ------->扣减账户结束
2019-09-06 15:44:35.545 INFO 36556 --- [atch_RMROLE_3_8] i.s.core.rpc.netty.RmMessageListener : onMessage:xid=192.168.158.133:8091:2021468859,branchId=2021468867,branchType=AT,resourceId=jdbc:mysql://116.62.62.26/seat-account,applicationData=null
2019-09-06 15:44:35.545 INFO 36556 --- [atch_RMROLE_3_8] io.seata.rm.AbstractRMHandler : Branch committing: 192.168.158.133:8091:2021468859 2021468867 jdbc:mysql://116.62.62.26/seat-account null
2019-09-06 15:44:35.545 INFO 36556 --- [atch_RMROLE_3_8] io.seata.rm.AbstractRMHandler : Branch commit result: PhaseTwo_Committed
在AccountServiceImpl中模拟异常情况,然后可以查看日志
/**
* 扣减账户余额
* @param userId 用户id
* @param money 金额
*/
@Override
public void decrease(Long userId, BigDecimal money) {
LOGGER.info("------->扣减账户开始");
// try {
// Thread.sleep(30*1000);
// } catch (InterruptedException e) {
// e.printStackTrace();
// }
LOGGER.info("------->扣减账户结束");
accountDao.decrease(userId,money);
}
前面的调用链为order->storage->account; 这里测试的成环是指order->storage->account->order, 这里的account服务又会回头去修改order在前面添加的数据。 经过测试,是支持此种场景的。
/**
* 扣减账户余额
* @param userId 用户id
* @param money 金额
*/
@Override
public void decrease(Long userId, BigDecimal money) {
LOGGER.info("------->扣减账户开始account中");
//模拟超时异常,全局事务回滚
// try {
// Thread.sleep(30*1000);
// } catch (InterruptedException e) {
// e.printStackTrace();
// }
accountDao.decrease(userId,money);
LOGGER.info("------->扣减账户结束account中");
//修改订单状态,此调用会导致调用成环
LOGGER.info("修改订单状态开始");
String mes = orderApi.update(userId, money.multiply(new BigDecimal("0.09")),0);
LOGGER.info("修改订单状态结束:{}",mes);
}
在最初的order会创建一个订单,然后扣减库存,然后扣减账户,账户扣减完,会回头修改订单的金额和状态,这样调用就成环了。
下载seata server包,地址:https://github.com/seata/seata/releases;
部署集群,第一台和第二台配置相同,在server端的registry.conf中,注意:
registry {
# file 、nacos 、eureka、redis、zk、consul、etcd3、sofa
type = "eureka"
......
eureka {
serviceUrl = "http://192.168.xx.xx:8761/eureka" //两台tcc相同,注册中心的地址
application = "default" //两台tc相同
weight = "1" //权重,截至0.9版本,暂时不支持此参数
}
......
注意上述配置和client的配置要一致,2台和多台情况相同。