MHA，把已经故障的master主机，重新加入

第一次正常实验，70为主，80为备主，90为从

部署 MHA 高可用集群
192.168.0.70 centos-70 master  mha4mysql-node
192.168.0.80 centos-80 slave1,Candidate master  mha4mysql-node
192.168.0.90 centos-90 slave2  mha4mysql-node mha4mysql-manage

[root@centos-70 ~]# ssh-keygen
[root@centos-80 ~]# ssh-keygen
[root@centos-80 ~]# ssh-keygen

[root@centos-70 ~]# ssh-copy-id 192.168.0.70
[root@centos-70 ~]# ssh-copy-id 192.168.0.80
[root@centos-70 ~]# ssh-copy-id 192.168.0.90

[root@centos-80 ~]# ssh-copy-id 192.168.0.70
[root@centos-80 ~]# ssh-copy-id 192.168.0.80
[root@centos-80 ~]# ssh-copy-id 192.168.0.90

[root@centos-90 ~]# ssh-copy-id 192.168.0.70
[root@centos-90 ~]# ssh-copy-id 192.168.0.80
[root@centos-90 ~]# ssh-copy-id 192.168.0.90

70-80-90执行以下命令
wget -O /etc/yum.repos.d/epel.repo http://mirrors.aliyun.com/repo/epel-7.repo
yum clean all
yum makecache
yum -y install perl-DBD-MySQL perl-Config-Tiny perl-Log-Dispatch perl-Parallel-ForkManager libappstream-glib deltarpm patchutils
rpm -ivh mha4mysql-node-0.57-0.el7.noarch.rpm

90执行以下命令
yum -y install perl-DBD-MySQL perl-Config-Tiny perl-Log-Dispatch perl-Parallel-ForkManager perl-Time-HiRes perl-ExtUtils-CBuilder perl-ExtUtils-MakeMaker perl-CPAN
rpm -ivh mha4mysql-manager-0.57-0.el7.noarch.rpm
 

70-80-90执行以下命令
systemctl start mysqld
vim /etc/my.cnf 
validate-password=OFF
systemctl restart mysqld
grep 'password' /var/log/mysqld.log
mysql -u root -p'buL.UJp!T2Od'  
set password for root@localhost = password('123456');
flush privileges;

70执行以下命令
mysql -u root -p123456
create database ha;
use ha;
create table test(id int,name varchar(20));
insert into test values (1,"jack");
exit
vim /etc/my.cnf
log-bin=mysql-bin-master
server-id=1
binlog-do-db=ha
binlog-ignore-db=mysql
systemctl restart mysqld
mysql -u root -p123456
grant replication slave on *.* to repl@'192.168.0.%' identified by '123456';
flush privileges;
show master status;
exit
mysqldump -u root -p123456 -B ha >ha.sql
scp ha.sql root@192.168.0.80:/root
scp ha.sql root@192.168.0.90:/root

80执行以下命令
mysql -u root -p123456 <ha.sql
vim /etc/my.cnf
log-bin=mysql-slave1
server-id=2
binlog-do-db=ha
binlog-ignore-db=mysql
log_slave_updates=1
systemctl restart mysqld
mysql -u root -p123456
grant replication slave on *.* to 'repl'@'192.168.0.%' identified by '123456';
flush privileges;
stop slave;
change master to master_host='192.168.0.70',master_user='repl',master_password='123456';
start slave;
show slave status\G
exit

90执行以下命令
mysql -u root -p123456 <ha.sql
vim /etc/my.cnf
log-bin=mysql-slave2
server-id=3
binlog-do-db=ha
binlog-ignore-db=mysql
log_slave_updates=1
systemctl restart mysqld
mysql -u root -p123456
grant replication slave on *.* to 'repl'@'192.168.0.%' identified by '123456';
flush privileges;
stop slave;
change master to master_host='192.168.0.70',master_user='repl',master_password='123456';
start slave;
show slave status\G
exit

测试 mysql 主从数据同步
70执行以下命令
mysql -uroot -p123456
use ha;
insert into test values(2,"man");
80执行以下命令
mysql -uroot -p123456
use ha;
select * from test;

80-90执行以下命令
mysql -u root -p123456 -e 'set global read_only=1'

70-80-90执行以下命令
mysql -uroot -p123456
grant all privileges on *.* to 'mha'@'localhost' identified by '123456';
grant all privileges on *.* to 'mha'@'192.168.0.%' identified by '123456';
flush privileges;
exit

90执行以下命令
mkdir -p /etc/masterha
mkdir -p /var/log/masterha/app1
cd /etc/masterha/
rz 上传app1.cnf
vim /etc/masterha/app1.cnf

[server default]
manager_workdir=/var/log/masterha/app1
manager_log=/var/log/masterha/app1/manager.log
master_binlog_dir=/var/lib/mysql
master_ip_failover_script=/usr/local/bin/master_ip_failover
password=123456
user=mha
ping_interval=1
remote_workdir=/tmp
repl_password=123456
repl_user=repl
ssh_user=root

[server1]
hostname=192.168.0.70
port=3306

[server2]
hostname=192.168.0.80
port=3306
candidate_master=1
check_repl_delay=0

[server3]
hostname=192.168.0.90
port=3306

rz 上传 master_ip_failover
mv master_ip_failover /usr/local/bin/
chmod +x /usr/local/bin/master_ip_failover
vim /usr/local/bin/master_ip_failover

80-90执行以下命令
mysql -uroot -p123456 -e 'set global relay_log_purge=0'

90执行以下命令
masterha_check_ssh --conf=/etc/masterha/app1.cnf
masterha_check_repl --conf=/etc/masterha/app1.cnf
nohup masterha_manager --conf=/etc/masterha/app1.cnf --remove_dead_master_conf --ignore_last_failover < /dev/null > /var/log/masterha/app1/manager.log 2>&1 &

90新开ssh执行以下命令
tailf /var/log/masterha/app1/manager.log

测试故障转移
70执行以下命令
systemctl stop mysqld
90执行以下命令
mysql -u root -p123456
show slave status\G
查看主库是否变更为80
80执行以下命令
insert into ha.test values(3,'test');
90执行以下命令
select * from ha.test;
查看是否同步80的数据

故障恢复后重新加入 mysql 主从集群
70执行以下命令
systemctl start mysqld
mysql -uroot -p123456 -e 'set global read_only=1'
mysql -uroot -p123456 -e ' set global relay_log_purge=0'
mysql -uroot -p123456
select * from ha.test;
查看下当前数据
stop slave;
change master to master_host='192.168.0.80',master_user='repl',master_password='123456';
flush privileges;
start slave;
show slave status\G
select * from ha.test;
查看70宕机后，80新增的数据是否同步过来，数据是否一致。
exit

90执行以下命令
vim /etc/masterha/app1.cnf

[server default]
manager_workdir=/var/log/masterha/app1
manager_log=/var/log/masterha/app1/manager.log
master_binlog_dir=/var/lib/mysql
master_ip_failover_script=/usr/local/bin/master_ip_failover
password=123456
user=mha
ping_interval=1
remote_workdir=/tmp
repl_password=123456
repl_user=repl
ssh_user=root

[server1]
candidate_master=1
check_repl_delay=0
hostname=192.168.0.70
port=3306

[server2]
hostname=192.168.0.80
port=3306

[server3]
hostname=192.168.0.90
port=3306

masterha_check_repl --conf=/etc/masterha/app1.cnf
nohup masterha_manager --conf=/etc/masterha/app1.cnf --remove_dead_master_conf --ignore_last_failover < /dev/null > /var/log/masterha/app1/manager.log 2>&1 &

80执行以下命令
systemctl stop mysqld
 

90查看日志，是否master切换回70

70执行以下命令
select * from ha.test; 查看数据和90的数据是否一致。
insert into ha.test values (4,'abc');
exit
ip a
查看虚拟ip是否为192.168.0.66

90执行以下命令
select * from ha.test; 查看数据是否和70一致。

整个流程走完，数据一致，没有问题，虚拟ip，也会跟随master漂移。

 

本文地址：https://blog.csdn.net/workey/article/details/108755672