MySQL5.7 新特性 多源复制(多主一从) 配置全解MySQL 5.7.18版本之前的最常见的复制方式，一主一从或者一主多从的架构：
MySQL 5.7.18之后就可以实现多主一从的复制：
多主一从架构带来的好处（个人总结）：
一、在从服务器进行数据汇总，如果我们的主服务器进行了分库分表的操作，为了实现后期的一些数据统计功能，往往需要把数据汇总在一起再统计。
二、如果我们想在从服务器时时对主服务器的数据进行备份，在MySQL 5.7.18之前每一个主服务器都需要一个从服务器，这样很容易造成资源浪费，同时也加大了DBA的维护成本，但MySQL 5.7.18引入多源复制，可以把多个主服务器的数据同步到一个从服务器进行备份。
下面演示一下在MySQL 5.7.18下搭建多主一从的过程：
实验环境：
操作系统环境：
[root@node86 dump]# uname -a
Linux node86 3.10.0-514.el7.x86_64 #1 SMP Tue Nov 22 16:42:41 UTC 2016 x86_64 x86_64 x86_64 GNU/Linux
[root@node86 dump]# cat /etc/redhat-release
CentOS Linux release 7.3.1611 (Core)
数据库拓扑环境：
Master_1: 192.168.66.111
Master_2: 192.168.66.112
Slave_3: 192.168.66.113
MySQL 5.7.18的搭建过程省略
一、分别在Master_1和Master_2上导出需要同步的数据库：
在Master_1：
[root@Master_1 mysql]# mysqldump -uroot -p123456 --master-data=2 --single-transaction --databases --add-drop-database Order >Order.sql
在Master_2：
[root@Master_2 mysql]# mysqldump -uroot -p123456 --master-data=2 --single-transaction --databases --add-drop-database Order_2 >Order_2.sql
把分别把备份scp到Slave上：
[root@Master_1 mysql]# scp -P22 Order.sql 192.168.66.113:/data/service/mysql/
[root@Master_2 mysql]# scp -P22 Order_2.sql 192.168.66.113:/data/service/mysql/
二、在Master_1和Master_2上创建复制账号，这个操作跟MySQL 5.7.18之前版本一样：
在Master_1：
<Master_1>[(none)]> grant replication slave on *.* to 'repl'@'192.168.66.113' identified by '123456';
Query OK, 0 rows affected, 1 warning (0.00 sec)
在Master_2:
<Master_2> [(none)]> grant replication slave on *.* to 'repl'@'192.168.66.113' identified by '123456';
Query OK, 0 rows affected, 1 warning (0.02 sec)
三、分别Slave上把Master_1和Master_2的数据导入Slave服务器，在导入前先修改MySQL存储master-info和relay-info的方式，即从文件存储改为表存储，在my.cnf里添加以下选择：
master_info_repository=TABLE
relay_log_info_repository=TABLE
也可以在线修改，非常方便：
<Slave> [(none)]> stop slave;
Query OK, 0 rows affected (0.02 sec)
<Slave> [(none)]> SET GLOBAL master_info_repository = 'TABLE';
Query OK, 0 rows affected (0.00 sec)
<Slave> [(none)]> SET GLOBAL relay_log_info_repository = 'TABLE';
Query OK, 0 rows affected (0.00 sec)
<Slave> [(none)]>
下面进行数据导入：
[root@Slave mysql]# mysql -uroot -p <./Order.sql
[root@Slave mysql]# mysql -uroot -p123456 <./Order_2.sql
分别找出Master_1和Master_2的binlog位置和Pos位置：
[root@Slave mysql]# cat Order.sql |grep " CHANGE MASTER"
-- CHANGE MASTER TO MASTER_LOG_FILE='Master_1-bin.000001', MASTER_LOG_POS=1539;
[root@Slave mysql]# cat Order_2.sql |grep " CHANGE MASTER"
-- CHANGE MASTER TO MASTER_LOG_FILE='Master_2-bin.000003', MASTER_LOG_POS=630;
[root@Slave mysql]#
四、登录Slave进行同步操作，分别change master到两台Master服务器，后面以FOR CHANNEL 'CHANNEL_NAME'区分
<Slave> [(none)]> CHANGE MASTER TO MASTER_HOST='192.168.66.111',MASTER_USER='repl', MASTER_PASSWORD='123456',MASTER_LOG_FILE='Master_1-bin.000001',MASTER_LOG_POS=1539 FOR CHANNEL 'Master_1';
Query OK, 0 rows affected, 2 warnings (0.05 sec)
<Slave> [(none)]> CHANGE MASTER TO MASTER_HOST='192.168.66.112',MASTER_USER='repl', MASTER_PASSWORD='123456',MASTER_LOG_FILE='Master_2-bin.000003',MASTER_LOG_POS=630 FOR CHANNEL 'Master_2';
Query OK, 0 rows affected, 2 warnings (0.04 sec)
进行启动slave操作，可以通过start slave的方式去启动所有的复制，也可以通过启动单个复制源的方式，下面进行单个复制源的启动进行演示(停止也是一样)：
<Slave> [(none)]> start slave for CHANNEL 'Master_1';
Query OK, 0 rows affected (0.01 sec)
<Slave> [(none)]> start slave for CHANNEL 'Master_2';
Query OK, 0 rows affected (0.02 sec)
正常启动后，可以查看同步的状态：执行SHOW SLAVE STATUS FOR CHANNEL 'channel_name'\G
查看复制源Master_1的同步状态：
<Slave> [(none)]> SHOW SLAVE STATUS FOR CHANNEL 'Master_1'\G
*************************** 1. row ***************************
Slave_IO_State: Waiting for master to send event
Master_Host: 192.168.66.111
Master_User: repl
Master_Port: 3306
Connect_Retry: 60
Master_Log_File: Master_1-bin.000001
Read_Master_Log_Pos: 1987
Relay_Log_File: localhost-relay-bin-master_1.000002
Relay_Log_Pos: 771
Relay_Master_Log_File: Master_1-bin.000001
Slave_IO_Running: Yes
Slave_SQL_Running: Yes
Replicate_Do_DB:
Replicate_Ignore_DB:
Replicate_Do_Table:
Replicate_Ignore_Table:
Replicate_Wild_Do_Table:
Replicate_Wild_Ignore_Table:
Last_Errno: 0
Last_Error:
Skip_Counter: 0
Exec_Master_Log_Pos: 1987
Relay_Log_Space: 991
Until_Condition: None
Until_Log_File:
Until_Log_Pos: 0
Master_SSL_Allowed: No
Master_SSL_CA_File:
Master_SSL_CA_Path:
Master_SSL_Cert:
Master_SSL_Cipher:
Master_SSL_Key:
Seconds_Behind_Master: 0
Master_SSL_Verify_Server_Cert: No
Last_IO_Errno: 0
Last_IO_Error:
Last_SQL_Errno: 0
Last_SQL_Error:
Replicate_Ignore_Server_Ids:
Master_Server_Id: 100128
Master_UUID: 44b653d4-8843-11e5-b97e-000c29dfaaf7
Master_Info_File: mysql.slave_master_info
SQL_Delay: 0
SQL_Remaining_Delay: NULL
Slave_SQL_Running_State: Slave has read all relay log; waiting for more updates
Master_Retry_Count: 86400
Master_Bind:
Last_IO_Error_Timestamp:
Last_SQL_Error_Timestamp:
Master_SSL_Crl:
Master_SSL_Crlpath:
Retrieved_Gtid_Set:
Executed_Gtid_Set:
Auto_Position: 0
Replicate_Rewrite_DB:
Channel_Name: master_1
Master_TLS_Version:
1 row in set (0.00 sec)
查看复制源Master_2的同步状态：
<Slave> [(none)]> SHOW SLAVE STATUS FOR CHANNEL 'Master_2'\G
*************************** 1. row ***************************
Slave_IO_State: Waiting for master to send event
Master_Host: 192.168.66.112
Master_User: repl
Master_Port: 3306
Connect_Retry: 60
Master_Log_File: Master_2-bin.000003
Read_Master_Log_Pos: 1078
Relay_Log_File: localhost-relay-bin-master_2.000002
Relay_Log_Pos: 771
Relay_Master_Log_File: Master_2-bin.000003
Slave_IO_Running: Yes
Slave_SQL_Running: Yes
Replicate_Do_DB:
Replicate_Ignore_DB:
Replicate_Do_Table:
Replicate_Ignore_Table:
Replicate_Wild_Do_Table:
Replicate_Wild_Ignore_Table:
Last_Errno: 0
Last_Error:
Skip_Counter: 0
Exec_Master_Log_Pos: 1078
Relay_Log_Space: 991
Until_Condition: None
Until_Log_File:
Until_Log_Pos: 0
Master_SSL_Allowed: No
Master_SSL_CA_File:
Master_SSL_CA_Path:
Master_SSL_Cert:
Master_SSL_Cipher:
Master_SSL_Key:
Seconds_Behind_Master: 0
Master_SSL_Verify_Server_Cert: No
Last_IO_Errno: 0
Last_IO_Error:
Last_SQL_Errno: 0
Last_SQL_Error:
Replicate_Ignore_Server_Ids:
Master_Server_Id: 100129
Master_UUID: 583f5433-43ef-11e5-8958-000c29d5bdfa
Master_Info_File: mysql.slave_master_info
SQL_Delay: 0
SQL_Remaining_Delay: NULL
Slave_SQL_Running_State: Slave has read all relay log; waiting for more updates
Master_Retry_Count: 86400
Master_Bind:
Last_IO_Error_Timestamp:
Last_SQL_Error_Timestamp:
Master_SSL_Crl:
Master_SSL_Crlpath:
Retrieved_Gtid_Set:
Executed_Gtid_Set:
Auto_Position: 0
Replicate_Rewrite_DB:
Channel_Name: master_2
Master_TLS_Version:
1 row in set (0.00 sec)
也可以通过查看performance_schema相关的表查看同步状态，执行命令：SELECT * FROM performance_schema.replication_connection_status; 监控复制状态。
+--------------+------------+--------------------------------------+-----------+---------------+---------------------------+--------------------------+--------------------------+-------------------+--------------------+----------------------+
| CHANNEL_NAME | GROUP_NAME | SOURCE_UUID | THREAD_ID | SERVICE_STATE |COUNT_RECEIVED_HEARTBEATS | LAST_HEARTBEAT_TIMESTAMP |RECEIVED_TRANSACTION_SET | LAST_ERROR_NUMBER | LAST_ERROR_MESSAGE |LAST_ERROR_TIMESTAMP |
+--------------+------------+--------------------------------------+-----------+---------------+---------------------------+--------------------------+--------------------------+-------------------+--------------------+----------------------+
| master_1 | | 44b653d4-8843-11e5-b97e-000c29dfaaf7| 34 | ON | 184 | 2015-08-14 08:06:10 | | 0 | | 0000-00-00 00:00:00 |
| master_2 | | 583f5433-43ef-11e5-8958-000c29d5bdfa| 36 | ON | 183 | 2015-08-14 08:06:24 | | 0 | | 0000-00-00 00:00:00 |
+--------------+------------+--------------------------------------+-----------+---------------+---------------------------+--------------------------+--------------------------+-------------------+--------------------+----------------------+
2 rows in set (0.00 sec)
<Slave> [(none)]>
五、验证数据是否同步
在Master_1上插入两条数据：
<Master_1>[Order]> insert into tb1(name) values ('user1'),('user2');
Query OK, 2 rows affected (0.01 sec)
在Master_2上插入两条数据：
<Master_2> [Order_2]> insert into tb2(name) values ('user3'),('user4');
Query OK, 2 rows affected (0.04 sec)
回到Slave上查看数据是否正常把数据同步过来了：
<Slave> [(none)]> show databases;
+--------------------+
| Database |
+--------------------+
| information_schema |
| mysql |
| performance_schema |
| sys |
| Order |
| Order_2 |
+--------------------+
6 rows in set (0.03 sec)
<Slave> [(none)]> select * from Order.tb1;
+----+-------+
| id | name |
+----+-------+
| 1 | user1 |
| 2 | user2 |
+----+-------+
2 rows in set (0.00 sec)
<Slave> [(none)]> select * from Order_2.tb2;
+----+-------+
| id | name |
+----+-------+
| 1 | user3 |
| 2 | user4 |
+----+-------+
2 rows in set (0.00 sec)
成功的实现了多主一从的环境搭建，*。*
总结：
一、MySQL 5.7.18的多源复制，能有效的解决分库分表的数据统计问题，同时也可以实现在一台从服务器对多台主服务器的数据备份。
二、MySQL 5.7.18的多源复制的出现，我们就不需要使用MariaDB 的多主一从的架构了，让很多小伙伴又看到了新的希望。
