软件：
mysql5.7
mha4mysql-manager-0.58
存放 /data/software/
部署：
一、 mysql5.7 一主二从
mysql安装略
1、 三 个 mysql 节点做主从复制授权
mysql> grant replication slave on . to ‘mysql’@’192.168.X.%’ identified by ‘passwd’;
mysql> grant all on . to ‘root’@’192.168.X.%’ identified by ‘passwd’;
2、 在两个salve节点上执行，只读限制
3、 主master
4、 两个slave节点执行
mysql> start slave;
二、 MHA环境和安装
1、 配置三台机器的ssh互信
ssh-keygen -t rsa
ssh-copy-id -i /root/.ssh/id_rsa.pub root@192.168.2.50
ssh-copy-id -i /root/.ssh/id_rsa.pub root@192.168.2.51
ssh-copy-id -i /root/.ssh/id_rsa.pub root@192.168.2.52
2、 MHA安装
三节点安装依赖
wget http://dl.fedoraproject.org/pub/epel/epel-release-latest-7.noarch.rpm
rpm -ivh epel-release-latest-7.noarch.rpm
三节点部署node
rpm -ivh mha4mysql-node-0.58-0.el7.centos.noarch.rpm
master节点部署manager
rpm -ivh mha4mysql-manager-0.58-0.el7.centos.noarch.rpm
全局配置文件
/etc/masterha_default.cnf
scp 192.168.8.59:/data/software/masterha_default.cnf /etc/
修改账号密码ip
主配置文件
/etc/mha/app1.cnf
修改账号密码ip
配置VIP
/etc/mha/master_ip_failover
scp 192.168.8.59:/data/software/master_ip_failover /etc/mha/
定义VIP变量
my $vip = ‘192.168.2.49/24’;
my $key = ‘1’;
my $ssh_stop_vip = “/sbin/ifconfig em1:$key down”;
配置编写VIP脚本
/etc/mha/master_ip_online_change
服务的网卡识别名一样
stop_vip=echo "ssh root@$orig_master_host /usr/sbin/ifconfig em1:$key down" ##em1 网卡名
start_vip=echo "ssh root@$new_master_host /usr/sbin/ifconfig em1:$key $vip"
chmod +x /etc/mha/master_ip_failover chmod +x /etc/mha/master_ip_online_change
配置报警邮件（非必须）
yum install mailx
/etc/mha/mail.rc ## 8.59:/data/software/
/etc/mha/send_report
chmod +x /etc/mha/send_report
验证 mysql 主从复制是否成功
masterha_check_repl —conf=/etc/mha/app1.cnf （成功：MySQL Replication Health is OK.）
三、 启动MHA（注意：MHA监控脚本切换一次就会退出，需要再次启动）
ifconfig em1:1 192.168.2.49/24
nohup masterha_manager —conf=/etc/mha/app1.cnf —remove_dead_master_conf —ignore_last_failover < /dev/null > /opt/log/mha/app1/manager.log 2>&1 &
tail -f /opt/log/mha/app1/manager.log
启动成功 [info] Ping(SELECT) succeeded, waiting until MySQL doesn’t respond..
检查集群状态
masterha_check_status —conf=/etc/mha/app1.cnf
四、 切换测试
停MHA脚本
masterha_stop —conf=/etc/mha/app1.cnf
切换 master
masterha_master_switch —conf=/etc/mha/app1.cnf —master_state=alive —new_master_host=192.168.2.51 —orig_master_is_new_slave —running_updates_limit=10000 —interactive=0
故障master重新加入集群
修复故障老的 master,设置为 slave （新master 2.51）
mysql> change master to master_host=’192.168.2.51’,master_user=’mysql’,master_password=’passwd’,master_log_file=’mysql-bin.0000X’,master_log_pos=XXX;
mysql> start slave;
在新manger 节点上重新启动监控进程
nohup masterha_manager —conf=/etc/mha/app1.cnf —remove_dead_master_conf —ignore_last_failover < /dev/null > /opt/log/mha/app1/manager.log 2>&1 &
五、 从1和从2之间的切换
192.168.2.50 主
192.168.2.51 从1 (自用读库）
六、从1和从2之间，使用keepalive+vip， 如从2业务读库宕机后，vip 自动漂移 从1 自用读库
安装keepalived
wget http://www.keepalived.org/software/keepalived-1.2.13.tar.gz
yum -y install popt-*
./configure —prefix=/usr/local/keepalived —enable-snmp
make && make install
cp /usr/local/keepalived/etc/rc.d/init.d/keepalived /etc/rc.d/init.d/
cp /usr/local/keepalived/etc/sysconfig/keepalived /etc/sysconfig/
chmod +x /etc/rc.d/init.d/keepalived
chkconfig keepalived on
mkdir /etc/keepalived
编辑 /etc/keepalived/keepalived.conf
创建check_mysql.sh ,master_sh
模拟关闭（vip）上mysql,vip会进行飘移