Mysql主从复制排错案例一

MYSQL主从复制排错案例一：

问题：主从无法同步
现象：MASTER: mysql> show master status;
              Empty set (0.00 sec)
      SLAVE:  mysql> show slave status \G;
              Slave_IO_Running: Connecting
              Slave_SQL_Running: Yes
              Seconds_Behind_Master: NULL
              Last_IO_Errno: 1045
              Last_IO_Error: error connecting to master ‘rep@192.168.254.253:3306‘
              - retry-time: 60  retries: 86400
问题排查过程：

一 、 MASTER :[root@localhost ~]# egrep "log-bin|server" /etc/my.cnf 
                      # The MySQL server
                       server-id       = 1
                         #log-bin=mysql-bin    //没有开启
            SLAVE: [root@localhost ~]# egrep "log-bin|server" /etc/my.cnf
                          # The MySQL server
                           server-id       = 3
                           #log-bin=mysql-bin
 二、.开启MASTART log-bin日志，重新启动mysql数据库
                              MASTER: mysql> show master status;
                                mysql> show master status;
+------------------+----------+--------------+------------------+
| File             | Position | Binlog_Do_DB | Binlog_Ignore_DB |
+------------------+----------+--------------+------------------+
| mysql-bin.000001 |      107 |              |                  |
+------------------+----------+--------------+------------------+
                            SLAVE:     mysql> show slave status \G;
                             Slave_IO_Running: Connecting
                                 Slave_SQL_Running: Yes
   三、查看master.info 和relay-log.info信息

                  SLAVE:  [root@localhost data]# cat master.info 
                        18
                          mysql-bin.000001
                            348
                            192.168.254.253
                              rep
                             oldboy123
                              3306
                                60
                                 0
 
                             [root@localhost data]# cat relay-log.info 
                           ./localhost-relay-bin.000005
                             4
                              mysql-bin.000001
                                348
                              8

MASTER上的log-bin的POS点107，SLAVE上的log-bin的POS点348,两边不一致。
 四、. 在master.info中用户名为rep,在MASTER端没有找到该用户注册的信息。
                 mysql> select host,user from mysql.user;
 五、.重新注册用于同步的用户
                 grant replication slave on *.* to ‘rep‘@‘192.168.254.250‘ identified by ‘oldboy123‘;
 六、 在SLAVE重新注册CHANGE MASTER TO

                    mysql> CHANGE MASTER TO 
                     -> MASTER_HOST=‘192.168.254.253‘,
                      -> MASTER_PORT=3306,
                        -> MASTER_USER=‘rep‘,
                       -> MASTER_PASSWORD=‘oldboy123‘,
                        -> MASTER_LOG_FILE=‘mysql-bin.000002‘,
                         -> MASTER_LOG_POS=107;
                          mysql> start slave;

七、.mysql> show slave status \G;
                    Slave_IO_Running: Yes
                    Slave_SQL_Running: Yes
八、.测试：MASTER:  mysql> create database mama;
                          Query OK, 1 row affected (0.00 sec)
                     SLAVE: mysql> show databases;
                                         mama
问题：由于虚拟机测试环境，没有把MASTER数据库变为只读，备份主库。直接同步到从库，这样后果丢失数据。在出现问题之前创建新的数据库没有同步过来。 生产环境下要先备份主库。
                         
        

本文出自 “技术在手，天下我有” 博客，请务必保留此出处http://xin521long.blog.51cto.com/11884590/1832143

原文：http://xin521long.blog.51cto.com/11884590/1832143