https://boke.wsfnk.com/archives/537.html
https://www.aliyun.com/jiaocheng/132307.html?spm=5176.100033.2.6.51e94d54AD4NcE
一、配置主從同步
1.1 本例中主從ip及端口
Master：192.168.20.50:3306
Slave1：192.168.31.3:3306 （候選master）
Slave2：192.168.31.2:3306
1.2、安裝mysql
wget -i -c http://dev.mysql.com/get/mysql57-community-release-el7-10.noarch.rpm
yum -y install mysql57-community-release-el7-10.noarch.rpm
yum -y install mysql-community-server
systemctl start mysqld.service
grep "password" /var/log/mysqld.log
mysql -uroot -p #登錄
ALTER USER 'root'@'localhost' IDENTIFIED BY 'My2019..';
#因為安裝了Yum Repository，以后每次yum操作都會自動更新，需要把這個卸載掉
yum -y remove mysql57-community-release-el7-10.noarch
1.3.Master配置參數
[mysqld] user=mysql pid_file = /var/lib/mysql/mysqld.pid socket = /var/lib/mysql/mysql.sock port = 3306 basedir = /usr datadir = /var/lib/mysql tmpdir = /tmp log-bin = master-bin log-bin-index = master-bin.index server_id = 1 innodb_log_file_size = 256M expire-logs-days = 1
validate_password=off #取消密碼驗證
#mysql5.6已上的特性，開啟gtid，必須主從全開
gtid_mode = on
enforce_gtid_consistency = 1
log_slave_updates = 1
#開啟半同步復制 否則自動切換主從的時候會報主鍵錯誤
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 = 5000
1.4.在master創建一個復制用戶
主節點：
mysql> create user repl_user; 所有主機上授權： mysql> grant replication slave on *.* to repl_user identified by 'tt123456'; mysql> flush privileges; mysql>grant all on *.* to root identified by 'my123456'; #很重要
1.5. Slave1 配置參數
[mysqld]
pid_file = /var/lib/mysql/mysqld.pid
socket = /var/lib/mysql/mysql.sock
basedir=/usr
port=3306
user=mysql
tmpdir= /tmp
server_id= 2
relay_log_index = slave_relay_bin.index
relay_log= slave_relay_bin
innodb_log_file_size= 256M
expire-logs-days = 1
log-bin = mysql-bin （候選需要配）
log-bin-index = mysql-bin.index （候選需要配）
read_only=1
relay_log_purge=0 #（一主一叢不需要此項，兩從及以上建議開次參數，防止切換為成主庫的從庫自動刪除中繼日志后，無法給其他從庫應用這部分日志）
validate_password=off #取消密碼驗證
gtid_mode = on
enforce_gtid_consistency = 1
log_slave_updates = 1
#開啟半同步復制 否則自動切換主從的時候會報主鍵錯誤
plugin_load = "rpl_semi_sync_master=semisync_master.so;rpl_semi_sync_slave=semisync_slave.so"（候選需要配）
loose_rpl_semi_sync_master_enabled = 1（候選需要配）
loose_rpl_semi_sync_slave_enabled = 1（候選需要配）
loose_rpl_semi_sync_master_timeout = 5000（候選需要配）
1.6.連接Master和Slave
參考：https://blog.csdn.net/lichangzai/article/details/50423906
mysql> CHANGE MASTER TO
MASTER_HOST='192.168.31.3',
MASTER_PORT=3306,
MASTER_AUTO_POSITION=1,
MASTER_USER='repl_user',
MASTER_PASSWORD='tt123456';
#master_log_file='master-bin.000007',#5.6后不需要指定
#master_log_pos=194;
mysql> start slave;
mysql> show slave status\G;
1.7查看半同步是否開啟
master： mysql> show status like 'Rpl_semi_sync_master_status'; +-----------------------------+-------+ | Variable_name | Value | +-----------------------------+-------+ | Rpl_semi_sync_master_status | ON | +-----------------------------+-------+ slave： mysql> show status like 'Rpl_semi_sync_slave_status'; +----------------------------+-------+ | Variable_name | Value | +----------------------------+-------+ | Rpl_semi_sync_slave_status | ON | +----------------------------+-------+
二、配置ssh公鑰互信
2.1. 本例中manager節點和node節點ip
manager：192.168.20.50
node1：192.168.20.50
node2：192.168.31.3
node3：192.168.31.2
注：manager節點可以安裝獨立的服務器上，本例為了節省機器，manager安裝在了主庫（192.168.20.50）上.
2.2. 配置manager和node各節點間的root用戶的ssh公鑰互信
在三個 mysql 節點分別執行如下操作：（三個都有，包括自己ssh自己） ssh-keygen -t rsa ssh-copy-id -i /root/.ssh/id_rsa.pub root@192.168.20.50 ssh-copy-id -i /root/.ssh/id_rsa.pub root@192.168.31.2 ssh-copy-id -i /root/.ssh/id_rsa.pub root@192.168.31.3
三、安裝 MHA 包
3.1. MHA安裝
manager和node節點安裝：
#先安裝依賴
wget http://dl.fedoraproject.org/pub/epel/epel-release-latest-7.noarch.rpm
rpm -ivh epel-release-latest-7.noarch.rpm
yum install -y perl-DBD-MySQL perl-Config-Tiny perl-Log-Dispatch perl-Parallel-ForkManager
wget https://qiniu.wsfnk.com/mha4mysql-node-0.58-0.el7.centos.noarch.rpm
rpm -ivh mha4mysql-node-0.58-0.el7.centos.noarch.rpm
manager上安裝：
wget https://qiniu.wsfnk.com/mha4mysql-manager-0.58-0.el7.centos.noarch.rpm
rpm -ivh mha4mysql-manager-0.58-0.el7.centos.noarch.rpm
3.2、MHA Manager 端配置
配置主配置文件
vi /etc/mha/app1.cnf
[server default] manager_workdir=/var/log/mha/app1 manager_log=/var/log/mha/app1/manager.log
user=root #mysql用戶
password=my123456 #mysql密碼
ssh_user=root
repl_user=repl_user
repl_password=tt123456
ping_interval=1
#master_binlog_dir= /var/lib/mysql,/var/log/mysql
#secondary_check_script=masterha_secondary_check -s 192.168.20.50 -s 192.168.31.3 -s 192.168.31.2
#master_ip_failover_script="/etc/mha/scripts/master_ip_failover"
#master_ip_online_change_script="/etc/mha/scripts/master_ip_online_change"
#report_script="/etc/mha/scripts/send_report"
[server1]#這里模塊不注釋，ssh測試會報錯，不過沒影響 hostname=192.168.20.50 port=3306 master_binlog_dir=/var/lib/mysql candidate_master=1 check_repl_delay=0 [server2] hostname=192.168.31.3 port=3306 master_binlog_dir=/var/lib/mysql candidate_master=1 #如果候選master有延遲的話，relay日志超過100m，failover切換不能成功，加上此參數后會忽略延遲日志大小。 check_repl_delay=0 [server3] hostname=192.168.31.2 port=3306 master_binlog_dir=/var/lib/mysql ignore_fail=1 #如果這個節點掛了，mha將不可用，加上這個參數，slave掛了一樣可以用 no_master=1 #從不將這台主機轉換為master
PS:以上注釋不能寫在配置文件里，不然會報錯
四、運行
4.1. 檢查SSH配置
masterha_check_ssh --conf=/etc/masterha/app1/app1.cnf
4.2檢查MHA當前配置
masterha_check_repl --conf=/etc/masterha/app1/app1.cnf
4.3啟動mha
nohup masterha_manager --conf=/etc/masterha/app1/app1.cnf --remove_dead_master_conf --ignore_last_failover < /dev/null > /var/log/mha/app1/manager.log 2>&1 &
4.4查看狀態
masterha_check_status --conf=/etc/masterha/app1/app1.cnf
4.5關閉mha
masterha_stop --conf=/etc/masterha/app1/app1.cnf
4.6查看日志
tail -f /etc/masterha/app1/mha_manager.log
4.7從庫從新加入新主
grep "CHANGE MASTER TO MASTER" /var/log/mha/app1/manager.log | tail -1
五、Failover應用場景測試
自動failover測試
應用場景1：master dead后，MHA當時已經開啟，候選Master庫（Slave）會自動failover為Master.
后面我的主庫改為192.168.31.3，在這機器執行：systemctl stop mysqld.service
然后192.168.20.50執行：mysql> show master status\G;
從庫機器192.168.31.2執行：mysql> show slave status\G;
最后把原主庫192.168.31.3修復成一個新的slave：
#查看具體修復語句
# grep "CHANGE MASTER TO MASTER" /etc/masterha/app1/manager.log | tail -1 CHANGE MASTER TO MASTER_HOST='192.168.20.50',MASTER_PORT=3306, MASTER_LOG_FILE='master-bin.000009', MASTER_AUTO_POSITION=1,MASTER_USER='repl_user', MASTER_PASSWORD='xxx';
mysql> CHANGE MASTER TO MASTER_HOST='192.168.20.50',MASTER_PORT=3306, MASTER_LOG_FILE='master-bin.000009', MASTER_AUTO_POSITION=1,MASTER_USER='repl_user', MASTER_PASSWORD='tt123456';
mysql>start slave;
mysql>show slave status\G;
驗證。
報錯
1.replicates is not defined in the configuration file!
masterha_check_repl --conf=/etc/masterha/app1/app1.cnf Thu Jan 31 11:36:27 2019 - [warning] Global configuration file /etc/masterha_default.cnf not found. Skipping. Thu Jan 31 11:36:27 2019 - [info] Reading application default configuration from /etc/masterha/app1/app1.cnf.. Thu Jan 31 11:36:27 2019 - [info] Reading server configuration from /etc/masterha/app1/app1.cnf.. Thu Jan 31 11:36:27 2019 - [info] MHA::MasterMonitor version 0.58. Thu Jan 31 11:36:28 2019 - [error][/usr/share/perl5/vendor_perl/MHA/ServerManager.pm, ln671] Master 192.168.20.50:3306 from w hich slave 192.168.31.2(192.168.31.2:3306) replicates is not defined in the configuration file! Thu Jan 31 11:36:28 2019 - [error][/usr/share/perl5/vendor_perl/MHA/MasterMonitor.pm, ln427] Error happened on checking confi gurations. at /usr/share/perl5/vendor_perl/MHA/MasterMonitor.pm line 329. Thu Jan 31 11:36:28 2019 - [error][/usr/share/perl5/vendor_perl/MHA/MasterMonitor.pm, ln525] Error happened on monitoring ser vers. Thu Jan 31 11:36:28 2019 - [info] Got exit code 1 (Not master dead).
解決：配置文件里面沒有manager的主機信息，添加上去。
[server1]
hostname=192.168.20.50 port=3306 master_binlog_dir=/var/lib/mysql candidate_master=1 check_repl_delay=0
2、[error][/usr/share/perl5/vendor_perl/MHA/MasterMonitor.pm
[error][/usr/share/perl5/vendor_perl/MHA/MasterMonitor.pm, ln427] Error happened on checking confi gurations. Argument "1 #M-dM-;M-^NM-dM-8M-\rM-eM-0M-^FM-hM-?M-^YM-eM-^OM-0M-d..." isn't numeric in numeric ge (>=) at /usr/s hare/perl5/vendor_perl/MHA/ServerManager.pm line 1157. Thu Jan 31 14:39:52 2019 - [error][/usr/share/perl5/vendor_perl/MHA/MasterMonitor.pm, ln525] Error happened on monitoring ser vers.
解決：
配置文件不能有中文注釋。。