MySQL高可用性之Keepalived+Mysql(双主热备)
环境描述:
OS:CentOS6.5_X64
MySQL-VIP:192.168.1.200
MySQL-master1:192.168.1.201
MySQL-master2:192.168.1.202
1、配置两台Mysql主主同步
[root@master ~]# yum install mysql-server mysql -y[root@master ~]# service mysqld start[root@master ~]# mysqladmin -u root password 123.com[root@master ~]# vi /etc/my.cnf #开启二进制日志,设置id[mysqld]server-id = 1 #backup这台设置2log-bin = mysql-binbinlog-ignore-db = mysql,information_schema #忽略写入binlog日志的库auto-increment-increment = 2 #字段变化增量值auto-increment-offset = 1 #初始字段ID为1slave-skip-errors = all #忽略所有复制产生的错误 [root@master ~]# service mysqld restart
如果是Centos7,则卸载mariadb,安装mysql,参考文章:https://blog.51cto.com/sf1314/2073389
安装mysql,参考文章:
mysql5.6.37 https://blog.51cto.com/sf1314/2058754
# mysql -uroot -p'密码'
先查看下log bin日志和pos值位置
master配置如下:
[root@ master ~]# mysql -u root -p123.commysql> GRANT REPLICATION SLAVE ON *.* TO 'replication'@'192.168.0.%' IDENTIFIED BY 'replication';mysql> flush privileges;mysql> change master to -> master_host='192.168.0.203', -> master_user='replication', -> master_password='replication', -> master_log_file='mysql-bin.000002', -> master_log_pos=106; #对端状态显示的值mysql> start slave; #启动同步
backup配置如下:
[root@backup ~]# mysql -u root -p123.commysql> GRANT REPLICATION SLAVE ON *.* TO 'replication'@'192.168.0.%' IDENTIFIED BY 'replication';mysql> flush privileges;mysql> change master to -> master_host='192.168.0.202', -> master_user='replication', -> master_password='replication', -> master_log_file='mysql-bin.000002', -> master_log_pos=106;mysql> start slave;
#主主同步配置完毕,查看同步状态Slave_IO和Slave_SQL是YES说明主主同步成功。
在master插入数据测试下:
在backup查看是否同步成功:
可以看到已经成功同步过去,同样在backup插入到user表数据,一样同步过去,双主就做成功了。
2、配置keepalived实现热备
[root@backup ~]# yum install -y pcre-devel openssl-devel popt-devel #安装依赖包
[root@master ~]# wget http://www.keepalived.org/software/keepalived-1.2.7.tar.gz [root@master ~]# tar -xf keepalived-1.2.7.tar.gz -C /usr/local/src/[root@master ~]# cd /usr/local/[root@master ~]# ln -sv /usr/local/src/keepalived-1.2.7/ keepalived #根据实际情况修改 [root@master ~]# cd keepalived[root@master ~]#./configure --prefix=/usr/local/keepalivedmake && make install
#将keepalived配置成系统服务
[root@master ~]# cp /usr/local/etc/rc.d/init.d/keepalived /etc/init.d/chmod +x /etc/init.d/keepalived [root@master ~]# cp /usr/local/etc/sysconfig/keepalived /etc/sysconfig/[root@master ~]# chkconfig --add keepalived [root@master ~]# chkconfig keepalived on [root@master ~]# mkdir /etc/keepalived[root@master ~]# cp /usr/local/etc/keepalived/keepalived.conf /etc/keepalived/[root@master ~]# cp /usr/local/sbin/keepalived /usr/sbin/ #或者ln -s /usr/local/sbin/keepalived /usr/sbin/
配置keepalived
我们自己在新建一个配置文件,默认情况下keepalived启动时会去/etc/keepalived目录下找配置文件
[root@master ~]# vi /etc/keepalived/keepalived.conf global_defs { notification_email { 862572301@qq.com } notification_email_from 862572301@qq.com smtp_server 127.0.0.1 smtp_connect_timeout 30 router_id MYSQL_HA #标识,双主相同 } vrrp_instance VI_1 { state BACKUP #两台配置此处均是BACKUP interface eth0 virtual_router_id 51 #主备相同 priority 100 #优先级,另一台backup改为90 advert_int 1 nopreempt #不抢占,只在优先级高master的机器上设置即可,优先级低backup的机器不设置 authentication { auth_type PASS auth_pass 1111 } virtual_ipaddress { 192.168.1.200 } } virtual_server 192.168.1.200 3306 { delay_loop 2 #每个2秒检查一次real_server状态 #lb_algo wrr #LVS算法,用不到,我们就关闭了 #lb_kind DR #LVS模式,如果不关闭,备用服务器不能通过VIP连接主MySQL persistence_timeout 60 #会话保持时间,同一IP的连接60秒内被分配到同一台真实服务器 protocol TCP real_server 192.168.1.201 3306 { #检测本地mysql,backup也要写检测本地mysql weight 3 notify_down /usr/local/keepalived/mysql.sh #当mysq服down时,执行此脚本,杀死keepalived实现切换 TCP_CHECK { connect_timeout 10 #连接超时时间 nb_get_retry 3 #重连次数 delay_before_retry 3 #重连间隔时间 connect_port 3306 #健康检查端口 } } }
编写检测服务down后所要执行的脚本[root@master ~]# vi /usr/local/keepalived/mysql.sh#!/bin/bashpkill keepalived[root@master ~]# chmod +x /usr/local/keepalived/mysql.sh[root@master ~]# /etc/init.d/keepalived start
注:此脚本是上面配置文件notify_down选项所用到的,keepalived使用notify_down选项来检查real_server 的服务状态,当发现real_server服务故障时,便触发此脚本;我们可以看到,脚本就一个命令,通过pkill keepalived强制杀死keepalived进程,从而实现了MySQL故障自动转移。另外,我们不用担心两个MySQL会同时提供数据更新操作, 因为每台MySQL上的keepalived的配置里面只有本机MySQL的IP+VIP,而不是两台MySQL的IP+VIP
启动keepalived
[root@master ~]# /usr/local/keepalived/sbin/keepalived -D 或者/etc/init.d/keepalived start[root@master ~]# ps -aux | grep keepalived
测试
找一台局域网PC,然后去ping MySQL的VIP,这时候MySQL的VIP是可以ping的通的
停止MySQL服务,看keepalived健康检查程序是否会触发我们编写的脚本
#backup服务器只修改priority为90、nopreempt不设置、real_server设置本地IP。
#授权两台Mysql服务器允许root远程登录,用于在其他服务器登陆测试!
mysql> grant all on *.* to'root'@'192.168.0.%' identified by '123.com';
mysql> flush privileges;
3、测试高可用性
1、通过Mysql客户端通过VIP连接,看是否连接成功。
2、停止master这台mysql服务,是否能正常切换过去,可通过ip addr命令来查看VIP在哪台服务器上。
3、可通过查看/var/log/messges日志,看出主备切换过程
4、master服务器故障恢复后,是否主动抢占资源,成为活动服务器。
附:keepalived-1.2.7 keepalived实现服务高可用 http://down.51cto.com/data/2440924
---------------------------------------------------------------------------------------------
针对网卡做bond的做VIP: https://blog.51cto.com/sf1314/2073519
[root@master ~]# vi /etc/keepalived/keepalived.conf global_defs { notification_email { 862572301@qq.com } notification_email_from 862572301@qq.com smtp_server 127.0.0.1 smtp_connect_timeout 30 router_id MYSQL_HA #标识,双主相同 } vrrp_instance VI_1 { state BACKUP #两台配置此处均是BACKUP interface bond0.101 #------->这边指定配置的聚合网卡bond0.101 virtual_router_id 51 #主备相同 priority 100 #优先级,另一台backup改为90 advert_int 1 nopreempt #不抢占,只在优先级高master的机器上设置即可,优先级低backup的机器不设置 authentication { auth_type PASS auth_pass 1111 } virtual_ipaddress { 10.104.101.13/24 #------->这边设置虚拟的VIP地址 } } virtual_server 10.104.101.13/24 3306 { #--------->指定虚拟VIP地址的配置信息 delay_loop 2 #每个2秒检查一次real_server状态 #lb_algo wrr #LVS算法,用不到,我们就关闭了 #lb_kind DR #LVS模式,如果不关闭,备用服务器不能通过VIP连接主MySQL persistence_timeout 60 #会话保持时间,同一IP的连接60秒内被分配到同一台真实服务器 protocol TCP real_server 10.104.101.12 3306 { #10.104.101.12指本地配置的IP地址,检测本地mysql,backup也要写检测本地mysql weight 3 notify_down /usr/local/keepalived/mysql.sh #当mysq服down时,执行此脚本,杀死keepalived实现切换 TCP_CHECK { connect_timeout 10 #连接超时时间 nb_get_retry 3 #重连次数 delay_before_retry 3 #重连间隔时间 connect_port 3306 #健康检查端口 } } }
附:修改mysql账户下的replication的密码可以参考本文章:https://blog.51cto.com/sf1314/2094562