linux之mysl57高可用mha4mysql
1 环境准备 master （主机） salve1（从机） slave2（从机） mha（mha管理） 都安装了mysql 正常
2 ssh免密登录 master （主机） salve1（从机） slave2（从机） mha（mha管理）分别安装
ssh-keygen -t rsa 三回车
ssh-copy-id -i ~/.ssh/id_rsa.pub 192.168.106.128
ssh-copy-id -i ~/.ssh/id_rsa.pub 192.168.106.129
ssh-copy-id -i ~/.ssh/id_rsa.pub 192.168.106.130
ssh-copy-id -i ~/.ssh/id_rsa.pub 192.168.106.131
添加完 用 ping 都测试下
3 master （主机） salve1（从机） slave2（从机） mha（mha管理）分别安装
1)添加依赖
wget http://dl.fedoraproject.org/pub/epel/epel-release-latest-7.noarch.rpm
yum install perl-DBD-MySQL -y
yum install perl-Config-Tiny -y
yum install epel-release -y
yum install perl-Log-Dispatch -y
yum install perl-Parallel-ForkManager -y
yum install perl-Time-HiRes -y
yum install perl-CPAN -y
2)安装节点
wget https://github.com/yoshinorim/mha4mysql-node/releases/download/v0.58/mha4mysql-node-0.58-0.el7.centos.noarch.rpm
rpm -ivh mha4mysql-node-0.58-0.el7.centos.noarch.rpm
4 mha（mha管理）安装 注意：centos8 不支持 mha4mysql-manager
wget https://github.com/yoshinorim/mha4mysql-manager/releases/download/v0.58/mha4mysql-manager-0.58-0.el7.centos.noarch.rpm
rpm -ivh mha4mysql-manager-0.58-0.el7.centos.noarch.rpm
以上 4不都正常安装 接下来就开始分别配置 测试
5 关闭所有库的防火墙
systemctl stop firewalld
6 配置 master （主机）
#bin_log配置
log_bin=mysql-bin
server-id=1
sync-binlog=1
binlog-ignore-db=information_schema
binlog-ignore-db=mysql
binlog-ignore-db=performation_schema
binlog-ignore-db=sys
#relay_log配置
relay_log=mysql-relay-bin
log_slave_updates=1
relay_log_purge=0
#gtid配置
#开启gtid
gtid_mode=on
enforce_gtid_consistency=1
#半同步复制配置
#自动加载半同步插件
plugin_load="rpl_semi_sync_master=semisync_master.so;rpl_semi_sync_slave=semisync_slave.so"
loose_rpl_semi_sync_master_enabled=1
loose_rpl_semi_sync_slave_enabled=1
loose_rpl_semi_sync_master_timeout=1000
7 重启mysql
systemctl restart mysqld
mysql -uroot -p
8 进行授权
grant replication slave on *.* to 'root'@'%' identified by 'Yumeko213@';
flush privileges;
grant all privileges on *.* to 'root'@'%' identified by 'Yumeko213@';
flush privileges;
9 配置 salve1（从机） slave2（从机）
#bin_log配置
log_bin=mysql-bin
#服务器ID,从库1是2,从库2是3
server-id=2
sync-binlog=1
binlog-ignore-db=information_schema
binlog-ignore-db=mysql
binlog-ignore-db=performation_schema
binlog-ignore-db=sys
#relay_log配置
relay_log=mysql-relay-bin
log_slave_updates=1
relay_log_purge=0
read_only=1
#gtid配置
#开启gtid
gtid_mode=on
enforce_gtid_consistency=1
#半同步复制配置
#自动加载半同步插件
plugin_load="rpl_semi_sync_master=semisync_master.so;rpl_semi_sync_slave=semisync_slave.so"
loose_rpl_semi_sync_master_enabled=1
loose_rpl_semi_sync_slave_enabled=1
loose_rpl_semi_sync_master_timeout=1000
10 重启mysql
systemctl restart mysqld
mysql -uroot -p
11进行授权
grant replication slave on *.* to 'root'@'%' identified by 'Yumeko213@';
flush privileges;
grant all privileges on *.* to 'root'@'%' identified by 'Yumeko213@';
flush privileges;
12 设置从库复制的主库地址
change master to master_host='192.168.106.128',master_port=3306,master_user='root',master_password='Yumeko213@',master_auto_position=1;
13 测试以上无问题
主库 show master status;
从库 start slave; show slave status \G;
14 mha配置
1)创建目录 和 文件
mkdir -p /etc/mha/
touch /var/log/mha/manager.log
mkdir -p /data/log/mha
2)公共配置vim /etc/masterha_default.cnf
[server default]
user=root
password=Yumeko213@
repl_user=root
repl_password=Yumeko213@
ssh_user=root
master_binlog_dir=/var/lib/mysql
remote_workdir=/data/log/mha
ping_interval=1
secondary_check_script=masterha_secondary_check -s 192.168.106.128 -s 192.168.106.129 -s 192.168.106.130
3)主配置vim /etc/mha/app1.cnf
[server default]
manager_workdir=/var/log/mha/manager
manager_log=/var/log/mha/manager.log
[server1]
hostname=192.168.106.128
port=3306
candidate_master=1
[server2]
hostname=192.168.106.129
port=3306
candidate_master=1
[server3]
hostname=192.168.106.130
port=3306
15 测试管理
1)检查SSH连接
masterha_check_ssh --conf=/etc/mha/app1.cnf
出现All SSH connection tests passed successfully.
2)检查复制配置
masterha_check_repl --conf=/etc/mha/app1.cnf
出现MySQL Replication Health is OK!
3)后台启动
nohup masterha_manager --conf=/etc/mha/app1.cnf < /dev/null > /var/log/mha/manager.log 2>&1 &
出现app1 (pid:5057) is running(0:PING_OK), master:192.168.106.128
4)检查启动状态
masterha_check_status --conf=/etc/mha/app1.cnf
出现app1 (pid:6312) is running(0:PING_OK), master:192.168.106.128
5)停止
masterha_stop --conf=/etc/mha/app1.cnf
出现 Stopped app1 successfully.
16 测试主故障转移
killall -9 mysqld
cat /var/log/mha/manager.log
17撤销主从关系
stop slave；
reset slave all;
show slave status; 就为空了，并且删除了本地内存重点
删除master.info和relay-log.info文件；
删除所有的relay log（包括还没有应用完的日志），创建一个新的relay log文件