zoukankan      html  css  js  c++  java
  • keepalive配置mysql自动故障转移

     

    转载

    原创

    一 Mysql主主复制搭建

    1.1 实验环境

    两台机器事先都已经装好了mysql单实例。


    二者的端口号需要保持一致,否则在最后用vip连接的时候,不能使用相同端口号连接。

    1.2 实验步骤

    1.2.1 修改配置文件

    修改master1:

    在[mysqld]下面添加:

    server-id = 1

    relay-log=/data/server/mysql_3307/binlog/ZabbixServer-relay-bin

    relay-log-index=/data/server/mysql_3307/binlog/ZabbixServer-relay-bin.index

    auto-increment-offset = 1

    auto-increment-increment = 2

    log-slave-updates=true

    修改master2:

    在[mysqld]下面添加:

    server-id = 3

    relay-log =/data/server/mysql/binlog/single-relay-bin

    relay-log-index=/data/server/mysql/binlog/single-relay-bin.index

    auto-increment-offset = 2

    auto-increment-increment = 2

    log-slave-updates=true

    添加auto-increment-offset那两项,是为了避免在MySQL INSERT时主键冲突。

    修改完后记得重启mysql

    1.2.2 建复制用户

    分别在两台mysql上执行

    GRANT REPLICATION SLAVE ON *.* TO 'RepUser'@'%'identified by 'beijing';

    1.2.3 指向master

    两台服务器均为新建立,且无其它写入操作,各服务器只需记录当前自己二进制日志文件及事件位置,以之作为另外的服务器复制起始位置即可。否则,需要先备份主库,在备库进行恢复,从而保持数据一致,然后再指向master。

    Master1:

    mysql> show master status;

    +------------------+----------+--------------+------------------+-------------------+

    | File | Position | Binlog_Do_DB | Binlog_Ignore_DB | Executed_Gtid_Set |

    +------------------+----------+--------------+------------------+-------------------+

    | mysql-bin.000001 | 302| | | |

    +------------------+----------+--------------+------------------+-------------------+

    1 row in set (0.00 sec)

    Master2:

    mysql> show master status;

    +------------------+----------+--------------+------------------+-------------------+

    | File | Position | Binlog_Do_DB | Binlog_Ignore_DB | Executed_Gtid_Set |

    +------------------+----------+--------------+------------------+-------------------+

    | mysql-bin.000001 | 120 | | | |

    +------------------+----------+--------------+------------------+-------------------+

    1 row in set (0.00 sec)

    #Master1指向Master2

    CHANGE MASTER TO MASTER_USER='RepUser',MASTER_HOST='192.168.1.21',MASTER_PASSWORD='beijing',MASTER_PORT=3307,MASTER_LOG_FILE='mysql-bin.000001',MASTER_LOG_POS=120;


    #Master2指向Master1

    CHANGE MASTER TO MASTER_USER='RepUser',MASTER_HOST='192.168.1.22',MASTER_PASSWORD='beijing', MASTER_PORT=3307,MASTER_LOG_FILE='mysql-bin.000001',MASTER_LOG_POS=302;

    1.2.4 分别启动slave

    start slave ;

    确保show slave status

    Slave_IO_Running: Yes

    Slave_SQL_Running: Yes

    测试两边是否同步,略。

    二:配置keepalived

    1 keepalive安装

    分别在master1master2上安装keepalive

    yum install -y popt-devel
    cd /usr/local/src
    wget http://www.keepalived.org/software/keepalived-1.2.2.tar.gz
    tar zxvf keepalived-1.2.2.tar.gz
    cd keepalived-1.2.2
    ./configure --prefix=/
    make
    make install

    #假如在执行./configure --prefix=/时报错: OpenSSL is not properly installed on your system !!!Can notinclude OpenSSL headers files,则yum install openssl-devel -y

    2 分别在master1,master2上新建检查mysql脚本

    vi /root/check_mysql.sh

    内容如下

    MYSQL=/usr/local/mysql/bin/mysql
    MYSQL_HOST=localhost
    MYSQL_USER=root
    MYSQL_PASSWORD=system@123
    
    
    $MYSQL -h $MYSQL_HOST -u $MYSQL_USER -p$MYSQL_PASSWORD -e "show status;" >/dev/null 2>&1
    #$mysqlclient --host=$host --port=$port --user=$user --password=$password  -e "show databases;" > /dev/null 2>&1
    if [ $? == 0 ]
    then
        echo " $host mysql login successfully "
        exit 0
    else
        #echo " $host mysql login faild"
        /etc/init.d/keepalived stop
        exit 2
    fi
    

    chmod +x /root/check_mysql.sh

    3 修改配置文件

    vi /etc/keepalived/keepalived.conf

    master1和master2配置文件内容相同。

    内容:

    #ConfigurationFile for keepalived
    global_defs {
    notification_email {                        ######定义接受邮件的邮箱
      wangjj@hrloo.com
            }
      notification_email_from jiankong@staff.tuge.com    ######定义发送邮件的邮箱
      smtp_server mail.tuge.com
      smtp_connect_timeout 10
    }
    vrrp_script check_mysql {                   ######定义监控mysql的脚本
         script "/root/check_mysql.sh"
         interval 2                             ######监控时间间隔
         weight 2                               ######负载参数
         }
    vrrp_instance vrrptest {                 ######定义vrrptest实例
            state BACKUP              ######服务器状态
      interface eth0                      ######使用的接口
            virtual_router_id 51                ######虚拟路由的标志,一组lvs的虚拟路由标识必须相同,这样才能切换
            priority 150                        ######服务启动优先级,值越大,优先级越高,BACKUP 不能大于MASTER
            advert_int 1                        ######服务器之间的存活检查时间
    authentication {
            auth_type PASS                      ######认证类型
            auth_pass ufsoft       ######认证密码,一组lvs 服务器的认证密码必须一致
    }
    track_script {                              ######执行监控mysql进程的脚本
         check_mysql
         }
    virtual_ipaddress {                         ######虚拟IP地址
             192.168.1.60
    }
    }

    这里state不配置MASTER,且优先级一样,是期望在MASTER1宕机后再恢复时,不主动将MASTER状态抢过来,避免MySQL服务的波动。

    由于不存在使用lvs进行负载均衡,不需要配置虚拟服务器virtual server,下同。

    4 vi /etc/sysconfig/iptables

    #注意,在两台机器上都要修改。

    添加:

    -A INPUT -d 192.168.1.60/32 -j ACCEPT

    -A INPUT -d 224.0.0.18 -j ACCEPT #添加VRRP通讯支持

    注意:第一行中的192.168.1.60需要改成你自己的vip。

    service iptables restart

    或者执行:iptables -I INPUT 4 -p vrrp -j ACCEPT

    CentOS7这样执行:

    sudo firewall-cmd --direct --permanent --add-rule ipv4 filter INPUT 0 --in-interface 网卡名称 --destination 224.0.0.18 --protocol vrrp -j ACCEPT
    sudo firewall-cmd --reload

    #检查修改是否生效
    sudo iptables -S | grep vrrp

    5 启动keepalived

    master1master2上分别启动:
    service keepalived start

    分别执行ip addr命令,可以在其中一台机器上看到虚拟IP.如:

    [root@slave1 keepalived]# ip addr
    1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
    inet6 ::1/128 scope host
    valid_lft forever preferred_lft forever
    2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_faststate UP qlen 1000
    link/ether 08:00:27:04:05:16 brd ff:ff:ff:ff:ff:ff
    inet 192.168.1.22/24 brd 192.168.1.255 scope global eth0
    inet 192.168.1.60/32 scope global eth0
    inet6 fe80::a00:27ff:fe04:516/64 scope link tentativedadfailed
    valid_lft forever preferred_lft forever

    说明虚拟vip连在了master1这台机器上。

    如果自动只连接到了master2,关闭master2keepalived,再启动,自动就连接到master1了。

    现在都可以ping通虚拟ip了。

    假如keepalive报错:one or more vip associated with vrid mismatch actual master advert...则可能是

    在同一网段中部署了令一套keepalived环境,导致virtual_router_id值跟我们部署的集群系统中该值冲突了,均使用了默认的值:51

    解决方法:

    修改其中一套集群中主从节点中keepalived的virtual_router_id 即可,主从需要保持一致

    6 测试

    停止master1服务器keepalived,检查VIP是否切换到master2服务器(用ip addr命令验证即可);

    三 测试高可用环境是否配置成功

    3.1 建允许远程访问的用户

    在master1,master2创建允许远程访问的用户:

    grant select,update,delete,insert on *.* to 'dandan' identified by 'dandan';

    3.2 访问虚拟IP

    用一台同网段的机器访问通过vip访问数据库:

    mysql -u dandan-pdandan -h 192.168.1.60 --port 3307

    停止master1服务器的mysqlVIP切换到了master2服务器。

    master2上查看:

    mysql> showprocesslist;
    +----+-------------+--------------------+------+---------+------+-----------------------------------------------------------------------------+------------------+
    | Id | User        | Host               | db   | Command | Time | State                                                                      | Info             |
    +----+-------------+--------------------+------+---------+------+-----------------------------------------------------------------------------+------------------+
    |  3 | root        | localhost          | dba | Query   |    0 | init                                                                       | show processlist |
    | 14 | systemuser |                    | NULL |Connect |  247 | Reconnecting after afailed master event read                               | NULL             |
    | 15 | systemuser |                    | NULL |Connect |  207 | Slave has read all relaylog; waiting for the slave I/O thread to update it | NULL             |
    | 90 |dandan      | 192.168.1.60:39995 |dba  | Sleep   |    8|                                                                            | NULL             |
    +----+-------------+--------------------+------+---------+------+-----------------------------------------------------------------------------+------------------+


    看到了dandan的连接信息。

  • 相关阅读:
    intellij idea 13
    tomcat加载web.xml
    tomcat-servlet-client
    请求由tomcat转到servlet的临界点
    传说用户发来的请求是在JIoEndpoint的accept函数中接收的,是tomact与外界交互的分界点
    java引用被设置为null的疑惑
    javascript之数组
    javascript学习笔记
    Http响应报文
    HTTP请求报文与响应报文
  • 原文地址:https://www.cnblogs.com/Zhao--C/p/12671687.html
Copyright © 2011-2022 走看看