zoukankan      html  css  js  c++  java
  • mysql 主从部署

    主从配置mysql

    grant all privileges on *.* to 'root'@'%' identified by 'root';
    flush privileges;

    》》》》》》主服务器配置 10.10.6.169

    grant replication slave on *.* to 'testuser'@'10.10.6.141' identified by '12345678';
    flush privileges;

    vi /etc/my.cnf
    [mysqld]下添加以下参数,若文件中已经存在,则不用添加
    server-id=1
    log-bin=mysql-bin #启动MySQL二进制日志系统,
    binlog-do-db=ourneeddb #需要同步的数据库
    binlog-ignore-db=mysql #不同步mysql系统数据库,若还有其它不想同步的,继续添加
    重启数据库

    mysql> show master status;
    +------------------+----------+--------------+------------------+
    | File | Position | Binlog_Do_DB | Binlog_Ignore_DB |
    +------------------+----------+--------------+------------------+
    | mysql-bin.000008 | 107 | bak | mysql |
    +------------------+----------+--------------+------------------+

    》》》》》导出数据库
    flush tables with read lock; #数据库只读锁定命令,防止导出数据库的时候有数据写入
    unlock tables; #解除锁定
    导出数据库结构及数据:mysqldump -uroot -p ourneeddb > /home/ourneeddb.sql
    导出存储过程及函数 :mysqldump -uroot -p -ntd -R ourneeddb > ourneeddb_func.sql
    tips:-ntd导出存储过程、-R导出函数

    》》》》》从服务器配置 10.10.6.141

    [mysqld]下添加以下参数,若文件中已经存在,则不用添加
    server-id=2 #设置从服务器id,必须于主服务器不同
    log-bin=mysql-bin #启动MySQ二进制日志系统
    replicate-do-db=ourneeddb #需要同步的数据库名
    replicate-ignore-db=mysql #不同步mysql系统数据库

    [root@localhost~ ]/etc/init.d/mysqld restart #重启服务
    use mysql
    stop slave;
    change master to
    master_host='10.10.6.169',
    master_user='testuser',
    master_password='12345678',
    master_log_file='mysql-bin.000008',
    master_log_pos=107; #log_file与log_pos是主服务器master状态下的File与Position
    start slave;
    show slave statusG;

    主主复制

    GRANT ALL PRIVILEGES ON *.* TO 'root'@'%' IDENTIFIED BY 'root' WITH GRANT OPTION;
    
    rpm安装方式
    复制配置文件
    cp /usr/share/mysql/my-default.cnf /etc/my.cnf
    
    105mysql>>>my.cnf
    #开启二进制日志
    log_bin=mysql-bin 
    binlog_format=mixed
    #数据存放文件夹
    datadir =/var/lib/mysql 
    port =3306
    #与另一台机器不同的ID
    server_id = 1
    socket =/var/lib/mysql/mysql.sock
    # 步进值,根据mysql机器数量决定
    auto_increment_increment=2
    #起始值。一般填第n台主MySQL。此时为第一台主MySQL
    auto_increment_offset=1
    # 忽略mysql库,一般不需要写
    binlog-ignore-db=mysql
    binlog-ignore-db=information_schema
    #要同步的数据库,默认所有库
    replicate-do-db=test
    sql_mode=NO_ENGINE_SUBSTITUTION,STRICT_TRANS_TABLES
    <<<
    创建同步用户
    CREATE USER 'mysqluser'@'10.10.6.106' IDENTIFIED BY 'root';
    GRANT REPLICATION SLAVE ON *.* TO 'mysqluser'@'10.10.6.106' IDENTIFIED BY 'root';
    flush privileges;
    
    主从复制配置
    主机105运行
    show master status; 
    从机106运行 
    change master to master_host='10.10.6.105', master_user='mysqluser', master_password='root', master_log_file='mysql-bin.000003', master_log_pos=1412;
    start slave;
    show slave statusG
    
    Last_SQL_Errno: 0
    Last_SQL_Error:
    Master_Host: 主库地址
    Master_User: 连接主库使用的 用户
    Master_Log_File: 主库二进制文件
    Read_Master_Log_Pos: 主库二进制文件坐标
    Slave_IO_Running: 与主库 IO 通信状态,需要为 Yes 方可
    Slave_SQL_Running: 负责自己的 Slave MySQL 进行状态,需要为 Yes 方可
    Replicate_Do_DB: 同步的 DB
    单台105主106从复制配置完毕
    
    106>>>my.cnf
    log_bin=mysql-bin
    binlog_format=mixed
    # These are commonly set, remove the # and set as required.
    # basedir = .....
    datadir =/var/lib/mysql
    port =3306
    server_id = 2
    socket =/var/lib/mysql/mysql.sock
    auto_increment_increment=2
    auto_increment_offset=2
    
    创建同步用户
    CREATE USER 'mysqluser'@'10.10.6.105' IDENTIFIED BY 'root';
    GRANT REPLICATION SLAVE ON *.* TO 'mysqluser'@'10.10.6.105' IDENTIFIED BY 'root';
    flush privileges;
    主从复制配置
    主机106运行
    show master status; 
    从机105运行 
    change master to master_host='10.10.6.106', master_user='mysqluser', master_password='root', master_log_file='mysql-bin.000003', master_log_pos=1412;
    start slave;
    show slave statusG
  • 相关阅读:
    Linux各目录及每个目录的详细介绍
    centos7 + mysql5.7 tar包解压安装
    Hive2.0的新特性介绍
    Hadoop HIVE
    PIG执行MR时报Connection refused错误
    Zookeeper启动Permission denied
    Hadoop Pig
    Hadoop组件之-HDFS(HA实现细节)
    Datanode启动问题 FATAL org.apache.hadoop.hdfs.server.datanode.DataNode: Initialization failed for Block pool <registering>
    HDFS Federation
  • 原文地址:https://www.cnblogs.com/maoxianfei/p/9477850.html
Copyright © 2011-2022 走看看