linux centos 6.10 编译安装mysql5.6.40版本，数据库主从

主机1 MYSQL配置：

/etc/my.cnf :

[mysqld]

server_id = 41
log_bin=mysql-bin
binlog-do-db =dbb221
binlog-do-db =dbb1000
log_bin_trust_function_creators=TRUE
binlog-ignore-db=mysql
binlog-ignore-db=information_schema
binlog-ignore-db=performance_schema
binlog-ignore-db=test
lower_case_table_names=1
skip-name-resolve
max_connections=2000
sql_mode=NO_ENGINE_SUBSTITUTION,STRICT_TRANS_TABLES

给从库添加复制用户：

grant replication slave on *.* to 'backup'@'192.168.1.250' identified by 'backup@123';

-------------------------------------------------------------------

从库配置：

/etc/my.cnf:

[mysqld]
server-id=42
basedir = /usr/local/mysql/
datadir = /usr/local/mysql/data/
replicate-do-db=dbb221
replicate-do-db=dbb1000
max_connections=2000
lower_case_table_names=1
wait_timeout=31536000
interactive_timeout=31536000
relay-log=/usr/local/mysql/logs/mysql-relay-bin.log
expire_logs_days=10

sql_mode=NO_ENGINE_SUBSTITUTION,STRICT_TRANS_TABLES

##修改从库的主库指向

change master to
master_host='192.168.1.40',
master_user='backup',
master_password='backup@123',
master_log_file='mysql-bin.000001',
master_log_pos=120;

#这个配置添加上就会自动保存住，下次重启不用重新配置

--------------------------------

#这里的，是主库挂掉了，主库与从库数据不一致的时候，当然我们有办法让他们一致，重建最省事

重建后涉及到日志问题，既然两边数据一致了，我们就省事点，直接重置重启完事，提高效率

flush tables with read lock;  #锁主库的表
unlock tables;  #解锁表

reset master; #重置master的同步位
reset slave; #重置slave的同部位

reset slave all; #重置从库配置，删除从库master指向

--------------------------------