文章展示了ShardingSphere的JDBC和Proxy配置示例，包括数据源设置、分库分表策略以及SNOWFLAKE主键生成。同时，详细讨论了ShardingSphere支持的分布式事务类型，如XA和SeataBASE事务，以及它们的启用和配置方法。
jdbc配置示例：
# 配置ds0 和ds1两个数据源
spring.shardingsphere.datasource.names=ds,ds0,ds1
#ds1 配置
spring.shardingsphere.datasource.ds.type=com.alibaba.druid.pool.DruidDataSource
spring.shardingsphere.datasource.ds.driver-class-name=com.mysql.cj.jdbc.Driver
spring.shardingsphere.datasource.ds.url=jdbc:mysql://localhost:3306/test?serverTimezone=UTC&useSSL=false&useUnicode=true&characterEncoding=UTF-8
#初始连接数
spring.shardingsphere.datasource.ds.initialSize=5
#最小空闲连接数
spring.shardingsphere.datasource.ds.minIdle=10
#最大连接数
spring.shardingsphere.datasource.ds.maxActive=30
spring.shardingsphere.datasource.ds.validationQuery=SELECT 1 FROM DUAL
spring.shardingsphere.datasource.ds.username=root
spring.shardingsphere.datasource.ds.password=root
#ds0 配置
spring.shardingsphere.datasource.ds0.type=com.alibaba.druid.pool.DruidDataSource
spring.shardingsphere.datasource.ds0.driver-class-name=com.mysql.cj.jdbc.Driver
spring.shardingsphere.datasource.ds0.url=jdbc:mysql://localhost:3306/test_ds_0?serverTimezone=UTC&useSSL=false&useUnicode=true&characterEncoding=UTF-8
spring.shardingsphere.datasource.ds0.initialSize=5
spring.shardingsphere.datasource.ds0.minIdle=10
spring.shardingsphere.datasource.ds0.maxActive=30
spring.shardingsphere.datasource.ds0.validationQuery=SELECT 1 FROM DUAL
spring.shardingsphere.datasource.ds0.username=root
spring.shardingsphere.datasource.ds0.password=root
#ds1 配置
spring.shardingsphere.datasource.ds1.type=com.alibaba.druid.pool.DruidDataSource
spring.shardingsphere.datasource.ds1.driver-class-name=com.mysql.cj.jdbc.Driver
spring.shardingsphere.datasource.ds1.url=jdbc:mysql://localhost:3306/test_ds_1?serverTimezone=UTC&useSSL=false&useUnicode=true&characterEncoding=UTF-8
spring.shardingsphere.datasource.ds1.initialSize=5
spring.shardingsphere.datasource.ds1.minIdle=10
spring.shardingsphere.datasource.ds1.maxActive=30
spring.shardingsphere.datasource.ds1.validationQuery=SELECT 1 FROM DUAL
spring.shardingsphere.datasource.ds1.username=root
spring.shardingsphere.datasource.ds1.password=root
# 对于没有做任何业务拆分的表，直接走本默认数据源即可
spring.shardingsphere.sharding.default-data-source-name=ds
spring.shardingsphere.sharding.default-database-strategy.hint.algorithm-class-name=com.tuling.tulingmall.sharding.OrderAllRangeHintAlgorithm
spring.shardingsphere.sharding.default-table-strategy.hint.algorithm-class-name=com.tuling.tulingmall.sharding.OrderAllRangeHintAlgorithm
# oms_order分片策略
# 节点 ds0.oms_order_0,ds0.oms_order_1,ds1.oms_order_0,ds1.oms_order_1
spring.shardingsphere.sharding.tables.oms_order.actual-data-nodes=ds$->{0..1}.oms_order_$->{0..1}
#分库策略
spring.shardingsphere.sharding.tables.oms_order.database-strategy.inline.sharding-column=id
spring.shardingsphere.sharding.tables.oms_order.database-strategy.inline.algorithm-expression=ds$->{id % 2}
# 分表策略
spring.shardingsphere.sharding.tables.oms_order.table-strategy.inline.sharding-column=id
# 注意下，对于除法，groovy会计算出浮点数，而不是整数。即 3/2=1.5，如果需要计算出整数 3.intdiv(2)=1
spring.shardingsphere.sharding.tables.oms_order.table-strategy.inline.algorithm-expression = oms_order_$->{((id+1) % 4).intdiv(2)}
# 复合分片算法
#spring.shardingsphere.sharding.tables.oms_order.table-strategy.complex.sharding-columns=id
#spring.shardingsphere.sharding.tables.oms_order.table-strategy.complex.algorithm-class-name = com.sharding.OrderComplexShardingAlgorithm
# 使用SNOWFLAKE算法生成主键
spring.shardingsphere.sharding.tables.oms_order.key-generator.column=id
spring.shardingsphere.sharding.tables.oms_order.key-generator.type=SNOWFLAKE
spring.shardingsphere.sharding.tables.oms_order.key-generator.props.worker.id=123
# 使用自定义主键生成策略
#spring.shardingsphere.sharding.tables.oms_order.key-generator.type=CUSTOM
#spring.shardingsphere.sharding.tables.oms_order.key-generator.props.redis.prefix=order:id:prefix:
# 节点 ds0.oms_order_item_0,ds0.oms_order_item_1,ds1.oms_order_item_0,ds1.oms_order_item_1
spring.shardingsphere.sharding.tables.oms_order_item.actual-data-nodes=ds$->{0..1}.oms_order_item_$->{0..1}
# 分库策略 按order_id分片
spring.shardingsphere.sharding.tables.oms_order_item.database-strategy.inline.sharding-column=order_id
spring.shardingsphere.sharding.tables.oms_order_item.database-strategy.inline.algorithm-expression=ds$->{order_id % 2}
# 分表策略
spring.shardingsphere.sharding.tables.oms_order_item.table-strategy.inline.sharding-column=order_id
spring.shardingsphere.sharding.tables.oms_order_item.table-strategy.inline.algorithm-expression=oms_order_item_$->{((order_id+1) % 4).intdiv(2)}
# 使用SNOWFLAKE算法生成主键
spring.shardingsphere.sharding.tables.oms_order_item.key-generator.column=id
spring.shardingsphere.sharding.tables.oms_order_item.key-generator.type=SNOWFLAKE
spring.shardingsphere.sharding.tables.oms_order_item.key-generator.props.worker.id=123
# 配置绑定表，防止笛卡尔乘积
spring.shardingsphere.sharding.binding-tables[0]=oms_order,oms_order_item
# 打印SQL语句
spring.shardingsphere.props.sql.show=true
proxy配置示例：shardingProxy的config-sharding.yaml
schemaName: sharding_db
dataSources:
ds:
url: jdbc:mysql://localhost:3306/test?serverTimezone=UTC&useSSL=false&useUnicode=true&characterEncoding=UTF-8
username: root
password: root
ds0:
url: jdbc:mysql://localhost:3306/test_ds_0?serverTimezone=UTC&useSSL=false&useUnicode=true&characterEncoding=UTF-8
username: root
password: root
ds1:
url: jdbc:mysql://localhost:3306/test_ds_1?serverTimezone=UTC&useSSL=false&useUnicode=true&characterEncoding=UTF-8
username: root
password: root
shardingRule:
tables:
oms_order:
actualDataNodes: ds$->{0..1}.oms_order_$->{0..1}
databaseStrategy:
inline:
shardingColumn: id
algorithmExpression: ds$->{id % 2}
tableStrategy:
inline:
shardingColumn: id
algorithmExpression: oms_order_$->{((id+1) % 4).intdiv(2)}
keyGenerator:
column: id
type: SNOWFLAKE
props:
worker.id: 123
oms_order_item:
actualDataNodes: ds$->{0..1}.oms_order_item_$->{0..1}
databaseStrategy:
inline:
shardingColumn: order_id
algorithmExpression: ds$->{id % 2}
tableStrategy:
inline:
shardingColumn: order_id
algorithmExpression: oms_order_item_$->{((order_id+1) % 4).intdiv(2)}
keyGenerator:
column: id
type: SNOWFLAKE
props:
worker.id: 123
bindingTables:
- oms_order,oms_order_item
defaultDataSourceName: ds
主键生成策略：
默认支持UUID和雪花算法，也可以自定义一个id生成策略。
配置spring.shardingsphere.sharding.tables.oms_order.key-generator.type=CUSTOM，然后在项目的Resource/META-INF.service/路径下新建一个properties文件，文件名以自定义的实现策略类的全路径命名，再将类的全路径copy到文件中，并在该类中定义type属性等于CUSTOM，代码片段如下：
public class OrderPreciseShardingAlgorithm implements PreciseShardingAlgorithm {
@Override
public String doSharding(Collection collection, PreciseShardingValue preciseShardingValue) {
// collection中保存所有真实节点
collection.stream().forEach((item)->{
log.info("actual node table:{}",item);
});
final Long columnValue = (Long)preciseShardingValue.getValue();
final String columnName = preciseShardingValue.getColumnName();
final String logicTableName = preciseShardingValue.getLogicTableName();
//preciseShardingValue中保存分片逻辑，包含逻辑表，分片字段和查询的条件值
log.info("logic table name:{},rout column:{}", logicTableName, columnName);
log.info("column value:{}", columnValue);
//采用自定义的ID生成器后，ID的前四位数字就是年份
final String year = columnValue.toString().substring(0, 4);
RedisOpsUtil redisOpsUtil = TulingmallOrderApplication.getBean("redisOpsUtil");
final Integer shardingKey = redisOpsUtil.get("shardingYear:" + year, Integer.class);
String actualTableName = logicTableName+"_"+year+"_"+(columnValue % shardingKey);
//返回的结构就是具体的实际表。sharddingsphere会把逻辑表改成这个实际表
return actualTableName;
}
}
关于ShardingSphere的扩展点：
ShardingSphere基于SPI机制提供了非常多的扩展点，具体可以参见估官方文档，开发者手册部分。
ShardingAlgorithm扩展点就列出了ShardingSphere默认提供的多种分片策略：
InlineShardingAlgorithm：基于⾏表达式的分⽚算法
ModShardingAlgorithm：基于取模的分⽚算法
HashModShardingAlgorithm：基于哈希取模的分⽚算法
FixedIntervalShardingAlgorithm：基于固定时间范围的分⽚算法
MutableIntervalShardingAlgorithm：基于可变时间范围的分⽚算法
VolumeBasedRangeShardingAlgorithm：基于分⽚容量的范围分⽚算法
BoundaryBasedRangeShardingAlgorithm：基于分⽚边界的范围分⽚算法
分库分表的分布式事务
XA事务和BASE柔性事务
ShardingSphere基于SPI扩展分布式事务管理器。官方提供了两种实现方式， XA事务和Seata AT事务。
XA事务
需要引入maven依赖
然后在事务开始的方法上打开@Transactional注解 和 @ShardingTransactionType(TransactionType.XA) 注解。
Seata AT事务
首先需要提前部署好Seata，采用nacos配置中心，把Seata的一大堆配置项（1.5.1版本之后优化成一个文件了）都提前导入到nacos中。
使用时，需要引入依赖以下依赖，并且把XA事务的依赖去掉。
org.apache.shardingsphere sharding-transaction-base-seata-at ${sharding-sphere.version} 然后需要在resource下添加seata.conf文件。文件内容：client {
application.id = example ## 应用唯一id
transaction.service.group = my_test_tx_group ## 所属事务组
}
然后需要在每个数据分片建立undo_log表
CREATE TABLE IF NOT EXISTS undo_log
(
id BIGINT(20) NOT NULL AUTO_INCREMENT COMMENT ‘increment id’,
branch_id BIGINT(20) NOT NULL COMMENT ‘branch transaction id’,
xid VARCHAR(100) NOT NULL COMMENT ‘global transaction id’,
context VARCHAR(128) 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 NOT NULL COMMENT ‘create datetime’,
log_modified DATETIME NOT NULL COMMENT ‘modify datetime’,
PRIMARY KEY (id),
UNIQUE KEY ux_undo_log (xid, branch_id)
) ENGINE = InnoDB
AUTO_INCREMENT = 1
DEFAULT CHARSET = utf8 COMMENT =‘AT transaction mode undo table’;
五、SharingProxy的分布式事务
对于ShardingProxy，由于他的功能并没有ShardingJDBC那么灵活，所以在ShardingProxy中使用分布式事务的方式就简单很多。
对于XA事务：可以直接在config配置文件中将事务直接配置为XA。默认就会采用Atomikos来实现，当然也可以自行选择其他的实现框架，但是需要手动替换jar包。。而如果要使用seata的BASE柔性事务，就需要自行将jar包拷贝到lib目录下，比较麻烦，一般项目中很少用到。
扫一扫
232
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
