Mysql 主从数据库同步(centos篇)

环境:

主服务器:centos 5.2 mysql 5.1.35 源码 ip:192.168.1.22

从服务器:centos 5.2 mysql 5.1.35 源码  ip:192.168.1.33

配置:

一、主服务器

    1.1、创建一个复制用户，具有replication slave 权限。

mysql>grant replication slave on *.* to 'repl'@'192.168.1.22' identified by 'repl';

    1.2、编辑my.cnf文件

vi /etc/my.cnf
添加
server-id=1
并开启log-bin二进制日志文件
log-bin=mysql-bin

注:需要把默认的server-id=1去掉

    1.3、启动mysql数据库

mysqld_safe --user=mysql &

    1.4、设置读锁

mysql>flush tables with read lock;

    1.5、得到binlog日志文件名和偏移量

mysql>show master status;

+------------------+----------+--------------+------------------+
| file             | position | binlog_do_db | binlog_ignore_db |
+------------------+----------+--------------+------------------+
| mysql-bin.0000010 |      106|              |                  |
+------------------+----------+--------------+------------------+

    1.6、备份要同步的数据库

mysqldump test > test.sql

    1.7、解锁

mysql>unlock tables;

 

二、从服务器

        2.1、编辑my.cnf文件

vi /etc/my.cnf
添加
server-id=2

注:需要把默认的server-id=1去掉

    2.2、启动从数据库

mysqld_safe --user=mysql &

        2.3、对从数据库进行相应设置

mysql> change master to
        -> master_host='192.168.1.22'
        -> master_user='repl'
        -> master_password='repl'
        -> master_log_file='mysql-bin.0000010'
        -> master_log_pos=106;

        2.4、启动从服务器slave线程

mysql>start slave;

执行show processlist命令显示以下进程:

mysql>show processlist\g
*************************** 2. row ***************************
         id: 2
     user: system user
     host:    
         db: null
command: connect
     time: 2579
    state: has read all relay log; waiting for the slave i/o thread to update it
     info: null

表示slave已经连接上master，开始接受并执行日志

         2.5、查看slave线程状态

mysql>show slave status;

*************************** 1. row ***************************
               slave_io_state: waiting for master to send event
                  master_host: 192.168.1.22

                  master_user: repl
                  master_port: 3306
                connect_retry: 60
              master_log_file: mysql-bin.0000010
          read_master_log_pos: 106
               relay_log_file: centos-relay-bin.000002
                relay_log_pos: 529
        relay_master_log_file: mysql-bin.0000010
             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: 106
              relay_log_space: 830
              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:
1 row in set (0.00 sec)

 

验证是否配置正确

在从服务器上执行

show slave status\g;
waiting for master to send event
slave_io_running: yes
slave_sql_running: yes

如以上二行同时为yes 说明配置成功

 

测试

1、在主服务器test数据库中创建user表

mysql>use test;
mysql>create table user(id int);

2、在从服务器中查看user表

mysql>use test;
mysql> show tables like 'user';
+----------------------+
| tables_in_test(user) |
+----------------------+
| user                 |    
+----------------------+
1 row in set (0.00 sec)

说明主从数据同步成功。

 

问题?

1.在从数据库中查看slave状态时出现

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)

说明方从服务器里my.cnf中的server-id有相同的。

解决办法:

修改my.cnf里的server-id，并重启数据库服务。my.cnf文件默认有server-id=1

 

其它说明

主服务器my.cnf

#binlog-do-db=需要备份的数据库名，可写多行
#binlog-ignore-db=不需要备份的数据库名，可写多行
从服务器my.cnf

# replicate-do-db=test  需要备份的数据库名
# replicate-ignore-db=mysql 忽略的数据库
# master-connect-retry=60 如果从服务器发现主服务器断掉，重新连接的时间差(秒)

以下设置也可直接修改my.cnf配置文件

log-bin=mysql-bin
master-host=192.168.1.22

master-user=repl

master-password=repl
master-port=3306

主从服务器同步维护

由于各种原因，导致主从数据不一致，在负载低的时候，进行手动同步.

在主服务器上执行

 

mysql>flush tables with read lock;

query ok,rows affected (0.01 sec)

mysql>show master status;

+------------------+----------+--------------+------------------+
| file             | position | binlog_do_db | binlog_ignore_db |
+------------------+----------+--------------+------------------+
| mysql-bin.0000011 |      260|              |                  |
+------------------+----------+--------------+------------------+

在从服务器上执行

先得到当前主服务器的二进制文件名和偏移量,执行命令使从服务器与主服务器同步

mysql>select master_pos_wait('mysql-bin.0000011','260');
+--------------------------------------------------+
| master_pos_wait('mysql-bin.0000011','260')            |
+--------------------------------------------------+
|  0                                                                   |   
+--------------------------------------------------+
1 row in set (0.01 sec)

同步完成后，在主服务器上执行解锁

mysql>unlock tables;

 

切换主从服务器

 

当主服务器出现故障时，可将从服务器当主服务器来使用.步骤如下:

1、保证所有从数据库都已经执行了relay log中的全部更新，在从服务器中执行

stop slave io_thread,用show processlist检查，查看状态是否是has read all relay log,表示更新完成.

mysql>stop slave io_thread;
query ok,0 affected (0.00 sec)
mysql>show processlist\g;

*************************** 2. row ***************************
     id: 2
   user: system user
   host:
     db: null
command: connect
   time: 4757
  state: has read all relay log; waiting for the slave i/o thread to update it
   info: null

2、在从服务器上执行stop slave，reset master命令，重置成主数据库

mysql>stop slave;
query ok,0 affected (0.00 sec)
mysql>reset master;
query ok,0 affected (0.00 sec)

3、删除新的主服务器数据库目录中的master.info和relay-log.info文件，否则下次重启时还会按照从服务器来启动.