首先配置主從的my.cnf文件
主

[mysqld]
log-bin=mysql-bin
server-id=1

從1和從2

[mysqld]
server-id=2

主配置從復(fù)制時使用的賬號密碼，%是全部ip，不建議

 CREATE USER 'copylocal'@'%' IDENTIFIED WITH mysql_native_password BY 'copyAll';
 CREATE USER 'copynew'@'%' IDENTIFIED WITH mysql_native_password BY 'copyAll';
GRANT REPLICATION SLAVE ON *.* TO 'copylocal'@'%';
GRANT REPLICATION SLAVE ON *.* TO 'copynew'@'%';
flush privileges;

查看一下狀態(tài)

mysql>show master status;

狀態(tài)

進(jìn)入從數(shù)據(jù)庫配置

mysql> CHANGE MASTER TO MASTER_HOST='192.168.1.101',  -- 主服務(wù)器IP
MASTER_USER='master',                                 -- 主服務(wù)器用戶
MASTER_PASSWORD='123456',                             -- 主服務(wù)器用戶密碼
MASTER_LOG_FILE='binlog.000001',                      -- 主服務(wù)器
MASTER_LOG_POS=0;                                 --  位置

這個位置，就是你的binlog的行號，從頭開始復(fù)制就是0，從現(xiàn)在為止開始復(fù)制就是上面查詢出來的行號
啟動復(fù)制：
mysql> start slave;
查看狀態(tài)
mysql> show slave status\G;
我遇到的問題是，IO不通

    Slave_IO_Running: Connecting
    Slave_SQL_Running: Yes

因為我的默認(rèn)端口不是3306，所以需要重新配置一下

--更新一下端口：
mysql>CHANGE MASTER TO MASTER_PORT = 3300;
--啟動服務(wù)
mysql>START SLAVE;
mysql>  show slave status\G

第二個問題

ERROR: 
No query specified

原因是show slave status\G加了分號，show slave status\G； 去掉就好了
第三個問題：server ids相同。大概是因為我用的同一款docker鏡像導(dǎo)致的。

 Last_IO_Errno: 13117
                Last_IO_Error: Fatal error: The slave I/O thread stops because master and slave have equal MySQL server ids; these ids must be different for replication to work (or the --replicate-same-server-id option must be used on slave but this does not always make sense; please check the manual before using it).

第四個問題：

 Last_SQL_Error: Coordinator stopped because there were error(s) in the worker(s). The most recent failure being: Worker 1 failed executing transaction 'ANONYMOUS' at master log binlog.000005, end_log_pos 3827262. See error log and/or performance_schema.replication_applier_status_by_worker table for more details about this failure or others, if any.
  Replicate_Ignore_Server_Ids: 

因為有錯誤導(dǎo)致的。這種要么選擇排查錯誤并修改log文件，要么干脆跳過錯誤

set global sql_replica_skip_counter=100;  數(shù)量可以設(shè)置大一點

雙方數(shù)據(jù)庫都重啟，ok。