当前位置： 首页 > news >正文

Apache ShardingSphere Proxy5.5.0实现MySQL分库分表与读写分离

news 来源：原创 2024/9/20 7:35:47

1. 前提准备

1.1 主机IP:192.168.186.77        

version: '3.8'services:mysql-master:image: mysql:latestcontainer_name: mysql-masterenvironment:MYSQL_ROOT_PASSWORD: 123456MYSQL_USER: masterMYSQL_PASSWORD: 123456MYSQL_DATABASE: db1  ports:- "3306:3306"volumes:- mysql-master-data:/var/lib/mysqlcommand: --server-id=1 --log-bin=mysql-bin --binlog-format=ROWcap_add:- SYS_NICEsecurity_opt:- seccomp:unconfinedmysql-slave:image: mysql:latestcontainer_name: mysql-slaveenvironment:MYSQL_ROOT_PASSWORD: 123456MYSQL_USER: slaveMYSQL_PASSWORD: 123456MYSQL_DATABASE: db1ports:- "3307:3306"volumes:- mysql-slave-data:/var/lib/mysqlcommand: --server-id=2 --log-bin=mysql-bin --binlog-format=ROW --relay-log=relay-bin --relay-log-index=relay-bin.indexdepends_on:- mysql-mastercap_add:- SYS_NICEsecurity_opt:- seccomp:unconfinedvolumes:mysql-master-data:mysql-slave-data:

注：3306端口扮演master数据库角色，3307端口扮演salve数据库角色。 

1.2 主机IP:192.168.186.216

version: '3.8'services:mysql-master:image: mysql:latestcontainer_name: mysql-masterenvironment:MYSQL_ROOT_PASSWORD: 123456MYSQL_USER: masterMYSQL_PASSWORD: 123456MYSQL_DATABASE: db2  ports:- "3306:3306"volumes:- mysql-master-data:/var/lib/mysqlcommand: --server-id=3 --log-bin=mysql-bin --binlog-format=ROWcap_add:- SYS_NICEsecurity_opt:- seccomp:unconfinedmysql-slave:image: mysql:latestcontainer_name: mysql-slaveenvironment:MYSQL_ROOT_PASSWORD: 123456MYSQL_USER: slaveMYSQL_PASSWORD: 123456MYSQL_DATABASE: db2  ports:- "3307:3306"volumes:- mysql-slave-data:/var/lib/mysqlcommand: --server-id=4 --log-bin=mysql-bin --binlog-format=ROW --relay-log=relay-bin --relay-log-index=relay-bin.indexdepends_on:- mysql-mastercap_add:- SYS_NICEsecurity_opt:- seccomp:unconfinedvolumes:mysql-master-data:mysql-slave-data:

 1.3 主机IP:192.168.186.216

version: '3.8'services:mysql-master:image: mysql:latestcontainer_name: mysql-masterenvironment:MYSQL_ROOT_PASSWORD: 123456MYSQL_USER: masterMYSQL_PASSWORD: 123456MYSQL_DATABASE: db3  ports:- "3306:3306"volumes:- mysql-master-data:/var/lib/mysqlcommand: --server-id=5 --log-bin=mysql-bin --binlog-format=ROWcap_add:- SYS_NICEsecurity_opt:- seccomp:unconfinedmysql-slave:image: mysql:latestcontainer_name: mysql-slaveenvironment:MYSQL_ROOT_PASSWORD: 123456MYSQL_USER: slaveMYSQL_PASSWORD: 123456MYSQL_DATABASE: db3 ports:- "3307:3306"volumes:- mysql-slave-data:/var/lib/mysqlcommand: --server-id=6 --log-bin=mysql-bin --binlog-format=ROW --relay-log=relay-bin --relay-log-index=relay-bin.indexdepends_on:- mysql-mastercap_add:- SYS_NICEsecurity_opt:- seccomp:unconfinedvolumes:mysql-master-data:mysql-slave-data:

注：每个 MySQL 实例必须有一个唯一的 server-id。这是必要的，因为在主从复制设置中，主服务器和从服务器需要能够互相识别，并避免循环复制和冲突。       

1.4 简单主从搭建过程

1.4.1 连接主服务器

1.4.2 连接从服务器 

1.4.3 配置主服务器
# 创建一个具有复制权限的用户
CREATE USER 'master_slave'@'%' IDENTIFIED BY '123456' REQUIRE SSL;
GRANT REPLICATION SLAVE ON *.* TO 'master_slave'@'%';
FLUSH PRIVILEGES;
SHOW MASTER STATUS;

1.4.4 配置从服务器
CHANGE MASTER TOMASTER_HOST ='192.168.186.216', # 主服务器的 IP 地址MASTER_USER ='master_slave', # 主服务器上配置的复制用户MASTER_PASSWORD ='123456', # 复制用户的密码MASTER_LOG_FILE ='mysql-bin.000003', # 主服务器的日志文件名MASTER_LOG_POS =920, # 日志文件的位置MASTER_SSL=1;
START SLAVE;
SHOW SLAVE STATUS;

        其余两台使用同样的方式进行配置。 

2. 安装 Apache ShardingSphere Proxy

2.1 安装合适的JDK版本

sudo apt install openjdk-8-jdk -y

2.2 验证JDK版本

java -version

liber@liber-VMware-Virtual-Platform:/home/sp$ java -version
openjdk version "1.8.0_412"
OpenJDK Runtime Environment (build 1.8.0_412-8u412-ga-1~24.04.2-b08)
OpenJDK 64-Bit Server VM (build 25.412-b08, mixed mode)

2.3 使用wget下载Apache ShardingSphere Proxy

sudo wget https://dlcdn.apache.org/shardingsphere/5.5.0/apache-shardingsphere-5.5.0-shardingsphere-proxy-bin.tar.gz

2.4 解压Apache ShardingSphere Proxy文件

 sudo tar -zxvf apache-shardingsphere-5.5.0-shardingsphere-proxy-bin.tar.gz

2.5 进入Apache ShardingSphere Proxy解压后的目录 

cd apache-shardingsphere-5.5.0-shardingsphere-proxy-bin

2.5 查看MySQL的版本 

        找台主服务器或者从服务器输入查看数据库的版本。

SELECT VERSION();

2.6 下载并安装MySQL Connector/J 8.0.27

sudo wget https://downloads.mysql.com/archives/get/p/3/file/mysql-connector-java-8.0.27.tar.gz

 

2.7 解压MySQL文件

sudo tar -zxvf mysql-connector-java-8.0.27.tar.gz

2.8 进入解压后的MySQL目录

cd mysql-connector-java-8.0.27

 2.9 将 JAR 文件放置到合适的位置

 需要将JAR文件复制到Apache ShardingSphere-Proxy解压目录的lib 目录中，因为我在Apache ShardingSphere-Proxy的解压目录中下载的MySQL压缩包，解压后只需要将JAR文件移动到上一级的lib目录即可。

sudo mv mysql-connector-java-8.0.27.jar  ../lib

  2.10 global.yaml

# 回退上一级目录
cd ..# 进入conf目录
cd conf# 编辑global.yaml文件
sudo nano global.yaml

      global.yaml 内容如下：

# 释放注解
authority:users:- user: root@%password: 123456- user: shardingpassword: 123456privilege:type: ALL_PERMITTED
sqlParser:sqlStatementCache:initialCapacity: 2000maximumSize: 65535parseTreeCache:initialCapacity: 128maximumSize: 1024
props:proxy-default-port: 3308 #修改端口，因为被占用了3307所以我修改了3308sql-show: true

2.11 创建物理表

use db1; # 每个主服务器都需要创建，只需要把db1，改db2，db3即可。CREATE TABLE t_order_0
(order_id   INT PRIMARY KEY,user_id    INT,order_date DATE,status     VARCHAR(15)
);CREATE TABLE t_order_1
(order_id   INT PRIMARY KEY,user_id    INT,order_date DATE,status     VARCHAR(15)
);

注： 每个主服务器的数据库都需要建立结构相同的物理表，只是库不一样db1，db2，db3。

2.12 database-sharding.yaml

sudo nano database-sharding.yaml
2.1.1 配置数据库源
dataSources:# 主库配置ds_77_master:url: jdbc:mysql://192.168.186.77:3306/db1?serverTimezone=UTC&useSSL=false&allowPublicKeyRetrieval=trueusername: root  # 数据库用户名password: 123456  # 数据库密码connectionTimeoutMilliseconds: 30000  # 连接超时时间，单位毫秒idleTimeoutMilliseconds: 60000  # 空闲连接超时时间，单位毫秒maxLifetimeMilliseconds: 1800000  # 连接最大生命周期，单位毫秒maxPoolSize: 50  # 连接池最大连接数minPoolSize: 1  # 连接池最小连接数# 从库配置ds_77_slave:url: jdbc:mysql://192.168.186.77:3307/db1?serverTimezone=UTC&useSSL=false&allowPublicKeyRetrieval=trueusername: slave  # 数据库用户名password: 123456  # 数据库密码connectionTimeoutMilliseconds: 30000  # 连接超时时间，单位毫秒idleTimeoutMilliseconds: 60000  # 空闲连接超时时间，单位毫秒maxLifetimeMilliseconds: 1800000  # 连接最大生命周期，单位毫秒maxPoolSize: 50  # 连接池最大连接数minPoolSize: 1  # 连接池最小连接数ds_216_master:url: jdbc:mysql://192.168.186.216:3306/db2?serverTimezone=UTC&useSSL=false&allowPublicKeyRetrieval=trueusername: rootpassword: 123456connectionTimeoutMilliseconds: 30000idleTimeoutMilliseconds: 60000maxLifetimeMilliseconds: 1800000maxPoolSize: 50minPoolSize: 1ds_216_slave:url: jdbc:mysql://192.168.186.216:3307/db2?serverTimezone=UTC&useSSL=false&allowPublicKeyRetrieval=trueusername: slavepassword: 123456connectionTimeoutMilliseconds: 30000idleTimeoutMilliseconds: 60000maxLifetimeMilliseconds: 1800000maxPoolSize: 50minPoolSize: 1ds_18_master:url: jdbc:mysql://192.168.186.18:3306/db3?serverTimezone=UTC&useSSL=false&allowPublicKeyRetrieval=trueusername: rootpassword: 123456connectionTimeoutMilliseconds: 30000idleTimeoutMilliseconds: 60000maxLifetimeMilliseconds: 1800000maxPoolSize: 50minPoolSize: 1ds_18_slave:url: jdbc:mysql://192.168.186.18:3307/db3?serverTimezone=UTC&useSSL=false&allowPublicKeyRetrieval=trueusername: slavepassword: 123456connectionTimeoutMilliseconds: 30000idleTimeoutMilliseconds: 60000maxLifetimeMilliseconds: 1800000maxPoolSize: 50minPoolSize: 1# 设置逻辑数据库名称
databaseName: my_database
2.1.2 读写分离规则
rules:- !READWRITE_SPLITTING# 定义用于读写分离的数据源配置dataSources:# 第一组读写分离规则，命名为 "readwrite_77"readwrite_77:writeDataSourceName: ds_77_master  # 指定写操作的数据源为 ds_77_master（主库）readDataSourceNames:- ds_77_slave  # 指定读操作的数据源列表，这里只包括 ds_77_slave（从库）transactionalReadQueryStrategy: PRIMARY  # 事务中的读查询策略设为 PRIMARY，即事务中所有读操作都将指向主库loadBalancerName: random  # 使用随机策略从可用的从库中选择readwrite_216:writeDataSourceName: ds_216_master  readDataSourceNames:- ds_216_slave  transactionalReadQueryStrategy: PRIMARY loadBalancerName: random  readwrite_18:writeDataSourceName: ds_18_master  # 主库readDataSourceNames:- ds_18_slave  transactionalReadQueryStrategy: PRIMARY  loadBalancerName: random  # 定义负载均衡器的配置loadBalancers:random:type: RANDOM  # 指定负载均衡器的类型为 RANDOM，随机选择读库
2.1.3 分库分表规则
- !SHARDING# 分片表的配置tables:t_order:# 定义分片表t_order在各个数据源的具体分布actualDataNodes: readwrite_77.t_order_${0..1}, readwrite_216.t_order_${0..1}, readwrite_18.t_order_${0..1}# 表的分片策略配置tableStrategy:standard:shardingColumn: order_id  # 使用订单ID作为分片键shardingAlgorithmName: t_order_inline  # 分片算法名称，指向后面定义的内联算法# 主键生成策略，用于插入操作时自动生成主键keyGenerateStrategy:column: order_id  # 主键列keyGeneratorName: snowflake  # 使用雪花算法生成主键# 默认数据库分片策略defaultDatabaseStrategy:standard:shardingColumn: user_id  # 使用用户ID作为分片键shardingAlgorithmName: database_inline  # 数据库分片使用的算法名称，指向后面定义的内联算法# 默认的表分片策略，此处未定义分片策略defaultTableStrategy:none:# 绑定表组，确保相互关联的表在同一数据库分片中bindingTables:- t_order  # 把t_order表标记为绑定表# 定义使用的分片算法shardingAlgorithms:# 定义数据库分片的内联算法database_inline:type: INLINEprops:algorithm-expression: "readwrite_${(user_id % 3 == 0) ? '77' : ((user_id % 3 == 1) ? '216' : '18')}"# 根据用户ID的值进行模3运算来决定数据分配到哪个数据源# 定义表分片的内联算法t_order_inline:type: INLINEprops:algorithm-expression: "t_order_${order_id % 2}"# 根据订单ID的值进行模2运算来决定数据存储在哪个分表# 主键生成器配置，指定使用雪花算法生成主键keyGenerators:snowflake:type: SNOWFLAKEprops:worker-id: 123  # 设置雪花算法的工作节点ID
2.1.4 完整配置信息 
dataSources:ds_77_master:url: jdbc:mysql://192.168.186.77:3306/db1?serverTimezone=UTC&useSSL=false&allowPublicKeyRetrieval=trueusername: rootpassword: 123456connectionTimeoutMilliseconds: 30000idleTimeoutMilliseconds: 60000maxLifetimeMilliseconds: 1800000maxPoolSize: 50minPoolSize: 1ds_77_slave:url: jdbc:mysql://192.168.186.77:3307/db1?serverTimezone=UTC&useSSL=false&allowPublicKeyRetrieval=trueusername: slavepassword: 123456connectionTimeoutMilliseconds: 30000idleTimeoutMilliseconds: 60000maxLifetimeMilliseconds: 1800000maxPoolSize: 50minPoolSize: 1ds_216_master:url: jdbc:mysql://192.168.186.216:3306/db2?serverTimezone=UTC&useSSL=false&allowPublicKeyRetrieval=trueusername: rootpassword: 123456connectionTimeoutMilliseconds: 30000idleTimeoutMilliseconds: 60000maxLifetimeMilliseconds: 1800000maxPoolSize: 50minPoolSize: 1ds_216_slave:url: jdbc:mysql://192.168.186.216:3307/db2?serverTimezone=UTC&useSSL=false&allowPublicKeyRetrieval=trueusername: slavepassword: 123456connectionTimeoutMilliseconds: 30000idleTimeoutMilliseconds: 60000maxLifetimeMilliseconds: 1800000maxPoolSize: 50minPoolSize: 1ds_18_master:url: jdbc:mysql://192.168.186.18:3306/db3?serverTimezone=UTC&useSSL=false&allowPublicKeyRetrieval=trueusername: rootpassword: 123456connectionTimeoutMilliseconds: 30000idleTimeoutMilliseconds: 60000maxLifetimeMilliseconds: 1800000maxPoolSize: 50minPoolSize: 1ds_18_slave:url: jdbc:mysql://192.168.186.18:3307/db3?serverTimezone=UTC&useSSL=false&allowPublicKeyRetrieval=trueusername: slavepassword: 123456connectionTimeoutMilliseconds: 30000idleTimeoutMilliseconds: 60000maxLifetimeMilliseconds: 1800000maxPoolSize: 50minPoolSize: 1
databaseName: my_database
rules:- !READWRITE_SPLITTINGdataSources:readwrite_77:writeDataSourceName: ds_77_masterreadDataSourceNames:- ds_77_slavetransactionalReadQueryStrategy: PRIMARYloadBalancerName: randomreadwrite_216:writeDataSourceName: ds_216_masterreadDataSourceNames:- ds_216_slavetransactionalReadQueryStrategy: PRIMARYloadBalancerName: randomreadwrite_18:writeDataSourceName: ds_18_masterreadDataSourceNames:- ds_18_slavetransactionalReadQueryStrategy: PRIMARYloadBalancerName: randomloadBalancers:random:type: RANDOM- !SHARDINGtables:t_order:actualDataNodes: readwrite_77.t_order_${0..1}, readwrite_216.t_order_${0..1}, readwrite_18.t_order_${0..1}tableStrategy:standard:shardingColumn: order_idshardingAlgorithmName: t_order_inlinekeyGenerateStrategy:column: order_idkeyGeneratorName: snowflakedefaultDatabaseStrategy:standard:shardingColumn: user_idshardingAlgorithmName: database_inlinedefaultTableStrategy:none:bindingTables:- t_ordershardingAlgorithms:database_inline:type: INLINEprops:algorithm-expression: "readwrite_${(user_id % 3 == 0) ? '77' : ((user_id % 3 == 1) ? '216' : '18')}"t_order_inline:type: INLINEprops:algorithm-expression: "t_order_${order_id % 2}"keyGenerators:snowflake:type: SNOWFLAKEprops:worker-id: 123

注：分片规则根据 user_id 的值对3取模，决定将数据分片到哪个数据库， 根据 order_id 的值对2取模，决定将数据分片到哪个表。

2.1.5 参考教程文献

 数据分片 :: ShardingSphere

 读写分离 :: ShardingSphere

 混合规则 :: ShardingSphere

2.13 启动程序

# 回退上一级
cd ..
# 进入bin目录
cd bin
# 启动程序
sudo ./start.sh

liber@liber-VMware-Virtual-Platform:/home/sp/apache-shardingsphere-5.5.0-shardingsphere-proxy-bin/conf$ cd ..
liber@liber-VMware-Virtual-Platform:/home/sp/apache-shardingsphere-5.5.0-shardingsphere-proxy-bin$ cd bin
liber@liber-VMware-Virtual-Platform:/home/sp/apache-shardingsphere-5.5.0-shardingsphere-proxy-bin/bin$ sudo ./start.sh
/usr/bin/java
we find java version: java8, full_version=1.8.0_412, full_path=/usr/bin/java
The classpath is /home/sp/apache-shardingsphere-5.5.0-shardingsphere-proxy-bin/conf:/home/sp/apache-shardingsphere-5.5.0-shardingsphere-proxy-bin/conf:.:/home/sp/apache-shardingsphere-5.5.0-shardingsphere-proxy-bin/lib/*:/home/sp/apache-shardingsphere-5.5.0-shardingsphere-proxy-bin/ext-lib/*
main class org.apache.shardingsphere.proxy.Bootstrap -1 /home/sp/apache-shardingsphere-5.5.0-shardingsphere-proxy-bin/conf 0.0.0.0 false
Starting the ShardingSphere-Proxy ... PID: 128452
Please check the STDOUT file: /home/sp/apache-shardingsphere-5.5.0-shardingsphere-proxy-bin/logs/stdout.log #日志路径可以通过cat进行查看
liber@liber-VMware-Virtual-Platform:/home/sp/apache-shardingsphere-5.5.0-shardingsphere-proxy-bin/bin$ 

 2.15 查看日志

cat /home/sp/apache-shardingsphere-5.5.0-shardingsphere-proxy-bin/logs/stdout.log
#显示该信息代表成功
[INFO ] 2024-07-25 15:39:09.724 [main] o.a.s.d.p.c.l.PipelineContextManagerLifecycleListener - mode type is not Cluster, mode type='Standalone', ignore
[INFO ] 2024-07-25 15:39:09.933 [main] o.a.s.p.v.ShardingSphereProxyVersion - Database type is `MySQL`, version is `8.0.27`, database name is `my_database`
[INFO ] 2024-07-25 15:39:09.966 [main] o.a.s.p.frontend.ssl.ProxySSLContext - Proxy frontend SSL/TLS is not enabled.
[INFO ] 2024-07-25 15:39:12.683 [main] o.a.s.p.frontend.ShardingSphereProxy - ShardingSphere-Proxy Standalone mode started successfully

2.16 测试连接 

2.17 分库分表验证 

        在ShardingSphere-Proxy连接的客户端进行操作，因为t_order是ShardingSphere-Proxy创建的逻辑表，所以实际上通过第三方连接工具是看不到存在的实体表，比如我使用的是IDEA自带的MySQL的连接管理工具。

use my_database;INSERT INTO t_order (order_id, user_id, order_date, status) VALUES (1, 10, '2023-07-01', 'NEW');
INSERT INTO t_order (order_id, user_id, order_date, status) VALUES (2, 20, '2023-07-02', 'SHIPPED');
INSERT INTO t_order (order_id, user_id, order_date, status) VALUES (3, 30, '2023-07-03', 'DELIVERED');
INSERT INTO t_order (order_id, user_id, order_date, status) VALUES (14, 40, '2023-07-04', 'RETURNED');

运行日志：

[INFO ] 2024-07-25 15:43:21.874 [ShardingSphere-Command-1] ShardingSphere-SQL - Logic SQL: /* ApplicationName=IntelliJ IDEA 2024.1 */ INSERT INTO t_order (order_id, user_id, order_date, status) VALUES (1, 10, '2023-07-01', 'NEW')
[INFO ] 2024-07-25 15:43:21.874 [ShardingSphere-Command-1] ShardingSphere-SQL - Actual SQL: ds_216_master ::: /* ApplicationName=IntelliJ IDEA 2024.1 */ INSERT INTO t_order_1 (order_id, user_id, order_date, status) VALUES (1, 10, '2023-07-01', 'NEW')
解释：分片规则根据 user_id 的值对3取模，决定将数据分片到哪个数据库， 根据 order_id 的值对2取模，决定将数据分片到哪个表。数据库取模顺序：77=>0,216=>1,18=>2
user_id: 10%3=1，order_id: 1%2=1，所以分片到第2个数据库ds_216_master，t_order_1表。 [INFO ] 2024-07-25 15:43:21.976 [ShardingSphere-Command-1] ShardingSphere-SQL - Logic SQL: /* ApplicationName=IntelliJ IDEA 2024.1 */ INSERT INTO t_order (order_id, user_id, order_date, status) VALUES (2, 20, '2023-07-02', 'SHIPPED')
[INFO ] 2024-07-25 15:43:21.976 [ShardingSphere-Command-1] ShardingSphere-SQL - Actual SQL: ds_18_master ::: /* ApplicationName=IntelliJ IDEA 2024.1 */ INSERT INTO t_order_0 (order_id, user_id, order_date, status) VALUES (2, 20, '2023-07-02', 'SHIPPED')
解释：分片规则根据 user_id 的值对3取模，决定将数据分片到哪个数据库， 根据 order_id 的值对2取模，决定将数据分片到哪个表。数据库取模顺序：77=>0,216=>1,18=>2
user_id: 20%3=2，order_id: 2%2=0，所以分片到第3个数据库ds_18_master，t_order_0表。
[INFO ] 2024-07-25 15:43:22.108 [ShardingSphere-Command-1] ShardingSphere-SQL - Logic SQL: /* ApplicationName=IntelliJ IDEA 2024.1 */ INSERT INTO t_order (order_id, user_id, order_date, status) VALUES (3, 30, '2023-07-03', 'DELIVERED')
[INFO ] 2024-07-25 15:43:22.108 [ShardingSphere-Command-1] ShardingSphere-SQL - Actual SQL: ds_77_master ::: /* ApplicationName=IntelliJ IDEA 2024.1 */ INSERT INTO t_order_1 (order_id, user_id, order_date, status) VALUES (3, 30, '2023-07-03', 'DELIVERED')
解释：分片规则根据 user_id 的值对3取模，决定将数据分片到哪个数据库， 根据 order_id 的值对2取模，决定将数据分片到哪个表。数据库取模顺序：77=>0,216=>1,18=>2
user_id: 30%3=0，order_id: 3%2=1，所以分片到第1个数据库ds_77_master，t_order_1表。

[INFO ] 2024-07-25 15:43:22.248 [ShardingSphere-Command-1] ShardingSphere-SQL - Logic SQL: /* ApplicationName=IntelliJ IDEA 2024.1 */ INSERT INTO t_order (order_id, user_id, order_date, status) VALUES (14, 40, '2023-07-04', 'RETURNED')
[INFO ] 2024-07-25 15:43:22.248 [ShardingSphere-Command-1] ShardingSphere-SQL - Actual SQL: ds_216_master ::: /* ApplicationName=IntelliJ IDEA 2024.1 */ INSERT INTO t_order_0 (order_id, user_id, order_date, status) VALUES (14, 40, '2023-07-04', 'RETURNED')
解释：分片规则根据 user_id 的值对3取模，决定将数据分片到哪个数据库， 根据 order_id 的值对2取模，决定将数据分片到哪个表。数据库取模顺序：77=>0,216=>1,18=>2
user_id: 40%3=1，order_id: 14%2=0，所以分片到第2个数据库ds_216_master，t_order_0表。

2.18 读写分离验证 

select *from t_order;
运行日志：
[INFO ] 2024-07-25 15:57:22.968 [ShardingSphere-Command-2] ShardingSphere-SQL - Logic SQL: /* ApplicationName=IntelliJ IDEA 2024.1 */ select *from t_order
[INFO ] 2024-07-25 15:57:22.969 [ShardingSphere-Command-2] ShardingSphere-SQL - Actual SQL: ds_77_slave ::: /* ApplicationName=IntelliJ IDEA 2024.1 */ select *from t_order_0 UNION ALL /* ApplicationName=IntelliJ IDEA 2024.1 */ select *from t_order_1
[INFO ] 2024-07-25 15:57:22.969 [ShardingSphere-Command-2] ShardingSphere-SQL - Actual SQL: ds_216_slave ::: /* ApplicationName=IntelliJ IDEA 2024.1 */ select *from t_order_0 UNION ALL /* ApplicationName=IntelliJ IDEA 2024.1 */ select *from t_order_1
[INFO ] 2024-07-25 15:57:22.973 [ShardingSphere-Command-2] ShardingSphere-SQL - Actual SQL: ds_18_slave ::: /* ApplicationName=IntelliJ IDEA 2024.1 */ select *from t_order_0 UNION ALL /* ApplicationName=IntelliJ IDEA 2024.1 */ select *from t_order_1

 

3. 总结 

        Ubtun24.04 TLS，JDK8，shardingsphere proxy/5.5.0，MySQL Connector/J 8.0.27，仅供学习交流使用。

相关文章：

  • 北京网站建设多少钱？
  • 辽宁网页制作哪家好_网站建设
  • 高端品牌网站建设_汉中网站制作
  • Halcon学习之边缘扩展
  • Java代理模式详解
  • React 的 KeepAlive 实战指南：深度解析组件缓存机制
  • 【网络爬虫技术】（1·绪论）
  • 深度学习高效性网络
  • 2024钉钉杯B题医疗门诊患者及用药数据案例分析
  • SolidWorks设计库的应用
  • 基于Golang+Vue3快速搭建的博客系统
  • 顺序表和单链表的代码实现
  • Ubuntu22.04安装Go语言的几种方式
  • Nginx系列-12 Nginx使用Lua脚本进行JWT校验
  • 【第三天】计算机网络 HTTP请求中常见的状态码 什么是强缓存和协商缓存
  • Spark进化论:从RDD到DataFrame,揭秘Spark SQL如何成为性能引擎的幕后英雄
  • 【数据结构】排序
  • Linux 安装 GDB （无Root 权限）
  • 【划重点】MySQL技术内幕：InnoDB存储引擎
  • CSS相对定位
  • ReactNative开发常用的三方模块
  • Shadow DOM 内部构造及如何构建独立组件
  • sublime配置文件
  • Terraform入门 - 3. 变更基础设施
  • 给github项目添加CI badge
  • 关于使用markdown的方法（引自CSDN教程）
  • 漫谈开发设计中的一些“原则”及“设计哲学”
  • 那些被忽略的 JavaScript 数组方法细节
  • 我建了一个叫Hello World的项目
  • 移动端解决方案学习记录
  • 移动互联网+智能运营体系搭建=你家有金矿啊！
  • 用quicker-worker.js轻松跑一个大数据遍历
  • AI又要和人类“对打”，Deepmind宣布《星战Ⅱ》即将开始 ...
  • k8s使用glusterfs实现动态持久化存储
  • Semaphore
  • ​草莓熊python turtle绘图代码（玫瑰花版）附源代码
  • #define MODIFY_REG(REG, CLEARMASK, SETMASK)
  • #mysql 8.0 踩坑日记
  • $.extend({},旧的,新的)；合并对象，后面的覆盖前面的
  • (1/2) 为了理解 UWP 的启动流程，我从零开始创建了一个 UWP 程序
  • （2024，LoRA，全量微调，低秩，强正则化，缓解遗忘，多样性）LoRA 学习更少，遗忘更少
  • (env: Windows,mp,1.06.2308310； lib: 3.2.4) uniapp微信小程序
  • （分布式缓存）Redis哨兵
  • （附源码）spring boot车辆管理系统 毕业设计 031034
  • （三维重建学习）已有位姿放入colmap和3D Gaussian Splatting训练
  • （四）linux文件内容查看
  • (原創) 如何使用ISO C++讀寫BMP圖檔? (C/C++) (Image Processing)
  • (转)shell中括号的特殊用法 linux if多条件判断
  • （自用）learnOpenGL学习总结-高级OpenGL-抗锯齿
  • .net core 连接数据库，通过数据库生成Modell
  • .net core 使用js，.net core 使用javascript，在.net core项目中怎么使用javascript
  • .NET Core中如何集成RabbitMQ
  • .NET HttpWebRequest、WebClient、HttpClient
  • .NET Remoting Basic(10)-创建不同宿主的客户端与服务器端
  • /bin/rm: 参数列表过长＂的解决办法
  • /var/lib/dpkg/lock 锁定问题
  • @JsonFormat 和 @DateTimeFormat 的区别
  • @RequestBody与@RequestParam