- MySQL软件简介
- MHA工作流程
- MHA架构
- MHA注意点
- MHA优点
- MHA工具
- 部署MHA
- 配置MHA
- MHA启动
- MHA日志分析
- MHA集群恢复
- MHA集群恢复脚本
- MHA的VIP漂移（在云主机上，不支持vip漂移）
- 配置MHA识别master_ip_failover脚本
- 解决binlog因为断电断网无法保存
- 1）sersync + rsync
- 2）mysqlbinlog
- MySQL读写分离
- Atlas主要功能
- Atlas相对于官方MySQL-Proxy的优势
- Atlas安装
- Atlas管理接口使用
- 配置版MHA安装脚本
- 免密脚本
- 配置文件
- 执行前提
- 脚本
- read传参版mha安装脚本
MySQL高可用及读写分离
环境准备
| 主机名 | 外网IP | 内网IP | 角色 | 软件 |
|---|---|---|---|---|
| db01 | 10.0.0.51 | 172.16.1.51 | master | mysql, MHA-node |
| db02 | 10.0.0.52 | 172.16.1.52 | slave | mysql, MHA-node |
| db03 | 10.0.0.53 | 172.16.1.53 | slave | mysql, MHA-node |
| db04 | 10.0.0.54 | 172.16.1.54 | slave, MHA-manager | mysql, MHA-node, MHA-manager |
change master to
master_host='172.16.1.51',
master_user='rep',
master_password='123',
master_log_file='mysql-bin.000002',
master_log_pos=445;MySQL软件简介
MHA Master Haproxy
MHA能够在较短的时间内实现自动故障检测和故障转移，通常在10-30秒以内;在复制框架中，MHA能够很好地解决复制过程中的数据一致性问题，由于不需要在现有的replication中添加额外的服务器，仅需要一个manager节点，而一个Manager能管理多套复制，所以能大大地节约服务器的数量;另外，安装简单，无性能损耗，以及不需要修改现有的复制部署也是它的优势之处。
MHA还提供在线主库切换的功能，能够安全地切换当前运行的主库到一个新的主库中(通过将从库提升为主库),大概0.5-2秒内即可完成。
MHA由两部分组成：MHA Manager（管理节点）和MHA Node（数据节点）。MHA Manager可以独立部署在一台独立的机器上管理多个Master-Slave集群，也可以部署在一台Slave上。当Master出现故障时，它可以自动将最新数据的Slave提升为新的Master,然后将所有其他的Slave重新指向新的Master。整个故障转移过程对应用程序是完全透明的。
MHA工作流程
MHA架构
MHA注意点
1.一个MHA可以监控多套MySQL集群
2.MHA可以不安装在MySQL集群中的某一台服务器上
3.MHA尽量不要安装在主库上面
4.MHA通过manager节点监控其它所有node节点
5.所有数据数据库上必须安装node节点
6.MHA是一个C/S架构的服务（CLient/Server）
- 服务端：manager
- 客户端：node
MHA优点
-
Masterfailover and slave promotion can be done very quickly
自动故障转移快
-
Mastercrash does not result in data inconsistency
主库崩溃不存在数据一致性问题
-
Noneed to modify current MySQL settings (MHA works with regular MySQL)
不需要对当前mysql环境做重大修改
-
Noneed to increase lots of servers
不需要添加额外的服务器(仅一台manager就可管理上百个replication)
-
Noperformance penalty
性能优秀，可工作在半同步复制和异步复制，当监控mysql状态时，仅需要每隔N秒向master发送ping包(默认3秒)，所以对性能无影响。你可以理解为MHA的性能和简单的主从复制框架性能一样。
-
Works with any storage engine
只要replication支持的存储引擎，MHA都支持，不会局限于innodb
MHA工具
Masterfailover and slave promotion can be done very quickly
自动故障转移快
Mastercrash does not result in data inconsistency
主库崩溃不存在数据一致性问题
Noneed to modify current MySQL settings (MHA works with regular MySQL)
不需要对当前mysql环境做重大修改
Noneed to increase lots of servers
不需要添加额外的服务器(仅一台manager就可管理上百个replication)
Noperformance penalty
性能优秀，可工作在半同步复制和异步复制，当监控mysql状态时，仅需要每隔N秒向master发送ping包(默认3秒)，所以对性能无影响。你可以理解为MHA的性能和简单的主从复制框架性能一样。
Works with any storage engine
只要replication支持的存储引擎，MHA都支持，不会局限于innodb
## node节点
[root@db01 bin]# ll
save_binary_logs // 保存宕机主库上的所有binlog events
apply_diff_relay_logs // 对比relay log 找到数据最新的从库
filter_mysqlbinlog // 截取binlog，relay log
purge_relay_logs // 删除relay log（我们要关闭relay log 自动删除的功能），定期清理不用的relay log
## manager节点
[root@db01 bin]# ls -1
masterha_check_repl // 检测主从复制，MHA启动的时候，需要对主从进行检测（可以
手动）
masterha_check_ssh // 检测ssh免密，MHA manager一定要跟所有库都免密
masterha_check_status // 检测MHA启动状态的，systemctl status 服务名（手动）
masterha_conf_host // 配置主机，当主库宕机之后，将宕机主库配置，从配置文件中摘除
masterha_manager // 启动MHA（手动）systemctl start MHA
masterha_master_monitor // 检测主库心跳
masterha_master_switch // 做切换
masterha_secondary_check // 建立TCP连接
masterha_stop // 停止MHA systemctl stop MHA部署MHA
## 先决条件
1.主库和从库都要开启binlog
2.主库的server_id和从库不同，从库之间也不能相同
[root@db02 ~]# vim /etc/my.cnf
[mysqld]
log-bin=mysql-bin
server_id=2
[root@db03 ~]# vim /etc/my.cnf
[mysqld]
log-bin=mysql-bin
server_id=3
[root@db04 ~]# vim /etc/my.cnf
[mysqld]
log-bin=mysql-bin
server_id=4
3.主库和从库都要用主从复制用户
# 主库创建：
root@localhost [(none)] >grant replication slave on *.* to rep@'172.16.1.%' identified
by '123';
4.所有数据库之间免密
[root@db01 ~]# ssh-keygen -t dsa -P '' -f ~/.ssh/id_dsa >/dev/null 2>&1
[root@db02 ~]# ssh-keygen -t dsa -P '' -f ~/.ssh/id_dsa >/dev/null 2>&1
[root@db03 ~]# ssh-keygen -t dsa -P '' -f ~/.ssh/id_dsa >/dev/null 2>&1
[root@db04 ~]# ssh-keygen -t dsa -P '' -f ~/.ssh/id_dsa >/dev/null 2>&1
[root@db01 ~]# ssh-copy-id -i ~/.ssh/id_dsa.pub 172.16.1.51
[root@db01 ~]# ssh-copy-id -i ~/.ssh/id_dsa.pub 172.16.1.52
[root@db01 ~]# ssh-copy-id -i ~/.ssh/id_dsa.pub 172.16.1.53
[root@db01 ~]# ssh-copy-id -i ~/.ssh/id_dsa.pub 172.16.1.54
[root@db02 ~]# ssh-copy-id -i ~/.ssh/id_dsa.pub 172.16.1.51
[root@db02 ~]# ssh-copy-id -i ~/.ssh/id_dsa.pub 172.16.1.52
[root@db02 ~]# ssh-copy-id -i ~/.ssh/id_dsa.pub 172.16.1.53
[root@db02 ~]# ssh-copy-id -i ~/.ssh/id_dsa.pub 172.16.1.54
[root@db03 ~]# ssh-copy-id -i ~/.ssh/id_dsa.pub 172.16.1.51
[root@db03 ~]# ssh-copy-id -i ~/.ssh/id_dsa.pub 172.16.1.52
[root@db03 ~]# ssh-copy-id -i ~/.ssh/id_dsa.pub 172.16.1.53
[root@db03 ~]# ssh-copy-id -i ~/.ssh/id_dsa.pub 172.16.1.54
[root@db04 ~]# ssh-copy-id -i ~/.ssh/id_dsa.pub 172.16.1.51
[root@db04 ~]# ssh-copy-id -i ~/.ssh/id_dsa.pub 172.16.1.52
[root@db04 ~]# ssh-copy-id -i ~/.ssh/id_dsa.pub 172.16.1.53
[root@db04 ~]# ssh-copy-id -i ~/.ssh/id_dsa.pub 172.16.1.54
### 检测ssh
[root@db01 ~]# for i in 51 52 53 54;do ssh 172.16.1.$i "echo $i OK";done
51 OK
52 OK
53 OK
54 OK
[root@db02 ~]# for i in 51 52 53 54;do ssh 172.16.1.$i "echo $i OK";done
51 OK
52 OK
53 OK
54 OK
[root@db03 ~]# for i in 51 52 53 54;do ssh 172.16.1.$i "echo $i OK";done
51 OK
52 OK
53 OK
54 OK
[root@db04 ~]# for i in 51 52 53 54;do ssh 172.16.1.$i "echo $i OK";done
51 OK
52 OK
53 OK
54 OK
5.所有库，要禁用自动删除relay log的功能
## 临时生效
root@localhost [(none)] >set global relay_log_purge = 0;
[root@db01 ~]# vim /etc/my.cnf
[mysqld]
relay_log_purge=0
6.从库只读（不能配置在配置文件中）
##只有从库操作
set global read_only=1;
7.命令软链接
软链接mysql和mysqlbinlog
[root@db01 ~]# ln -s /app/mysql/bin/mysql /usr/bin/mysql
[root@db01 ~]# ln -s /app/mysql/bin/mysqlbinlog /usr/bin/mysqlbinlog
[root@db02 ~]# ln -s /app/mysql/bin/mysql /usr/bin/mysql
[root@db02 ~]# ln -s /app/mysql/bin/mysqlbinlog /usr/bin/mysqlbinlog
[root@db03 ~]# ln -s /app/mysql/bin/mysql /usr/bin/mysql
[root@db03 ~]# ln -s /app/mysql/bin/mysqlbinlog /usr/bin/mysqlbinlog
[root@db04 ~]# ln -s /app/mysql/bin/mysql /usr/bin/mysql
[root@db04 ~]# ln -s /app/mysql/bin/mysqlbinlog /usr/bin/mysqlbinlog安装node组件
## 每一台都要装node（安装可能会报错，多安装几次就行）
[root@db01 ~]# yum localinstall -y mha4mysql-node-0.56-0.el6.noarch.rpm
[root@db02 ~]# yum localinstall -y mha4mysql-node-0.56-0.el6.noarch.rpm
[root@db03 ~]# yum localinstall -y mha4mysql-node-0.56-0.el6.noarch.rpm
[root@db04 ~]# yum localinstall -y mha4mysql-node-0.56-0.el6.noarch.rpm安装manager组件
## 安装mha manager节点必须先安装node （安装可能会报错，多安装几次就行）
[root@db04 ~]# yum localinstall -y mha4mysql-manager-0.56-0.el6.noarch.rpm配置MHA
# 1.创建mha配置文件目录
[root@db04 ~]# mkdir /etc/mha
# 2.创建MHA管理用户（所有库都要有，但是只要在主库创建，从库会同步）
root@localhost [(none)] >grant all on *.* to mha@'172.16.1.%' identified by 'mha';
# 3.创建mha配置文件
cat >/etc/mha/app1.cnf<<EOF
[server default]
manager_log=/etc/mha/logs/manager.log
manager_workdir=/etc/mha/app1
master_binlog_dir=/app/mysql/data
user=mha
password=mha
ping_interval=2
repl_password=123
repl_user=rep
ssh_user=root
ssh_port=22
[server1]
hostname=172.16.1.51
port=3306
[server2]
#candidate_master=1
#check_repl_delay=0
hostname=172.16.1.52
port=3306
[server3]
hostname=172.16.1.53
port=3306
[server4]
hostname=172.16.1.54
port=3306
EOF
# 4.检测ssh
[root@db04 ~]# masterha_check_ssh --conf=/etc/mha/app1.cnf
# 5.检测主从复制
[root@db04 ~]# masterha_check_repl --conf=/etc/mha/app1.cnf
# 6.启动mha
[root@db04 ~]# mkdir /etc/mha/logs
[root@db04 ~]# nohup masterha_manager --conf=/etc/mha/app1.cnf --remove_dead_master_conf --ignore_last_failover < /dev/null > /etc/mha/logs/manager.log 2>&1 &
[root@db04 ~]# masterha_check_status --conf=/etc/mha/app1.cnf
app1 (pid:19198) is running(0:PING_OK), master:172.16.1.51MHA启动
# 机制：
1.MHA做完一次切换后，会自动结束进程
2.MHA做完一次切换后，会移除master配置
3.MHA做完一次切换后，会在工作目录下生成一个锁文件(8个小时) .lock .lck
nohup ## 后台运行
masterha_manager ## mha启动命令
--conf=/etc/mha/app1.cnf ## 指定配置文件
--remove_dead_master_conf ## 在配置文件中，移除宕机的master配置
--ignore_last_failover ## 忽略上一次故障转移
< /dev/null > /etc/mha/logs/manager.log 2>&1 &MHA日志分析
# 1.gtid检测
GTID failover mode = 0
Starting Non-GTID based failover.
# 2.还没有设置vip切换脚本
master_ip_failover_script is not set. Skipping invalidating dead master IP address.
save_binary_logs \
--command=save \
--start_file=mysql-bin.000002 \
--start_pos=1024 \
--binlog_dir=/app/mysql/data \
--
output_file=/var/tmp/saved_master_binlog_from_172.16.1.51_3306_20230809184858.binlog \
--handle_raw_binlog=1 \
--disable_log_bin=0 \
--manager_version=0.56
# 3.保存binlog
先保存在宕机主库的/var/tmp下
然后scp到manager所在机器的mha工作目录下/etc/mha/app1
再从MHA机器上，发送到新的主库上/var/tmp
# 4.检测配置文件有没有 candidate_master=1
Candidate masters from the configuration file:
Non-candidate masters:
# 5.所有其他从库想要加进集群
CHANGE MASTER TO MASTER_HOST='172.16.1.52', MASTER_PORT=3306, MASTER_LOG_FILE='mysqlbin.000002', MASTER_LOG_POS=154, MASTER_USER='rep', MASTER_PASSWORD='123';
## MHA切换优先级
1.candidate_master
2.数据量最新
3.配置文件中server标签值，越小，优先级越高MHA集群恢复
# 1.修复宕机主库
systemctl start mysqld
# 2.找到change master语句
[root@db04 ~]# grep -i 'change master to' /etc/mha/logs/manager.log |awk -F: '{print $4}'|sed "s#xxx#123#g"
# 3.在宕机主库中执行
# 4.开启主从复制
start slave;
# 5.修复 MHA 配置文件
# 6.启动MHA
[root@db04 ~]# nohup masterha_manager --conf=/etc/mha/app1.cnf --remove_dead_master_conf --ignore_last_failover < /dev/null > /etc/mha/logs/manager.log 2>&1 &MHA集群恢复脚本
#!/bin/bash
. /etc/init.d/functions
mha_log="/etc/mha/logs/manager.log"
pass="123"
down_master=$(sed -nr 's#^M.* (.*)\(.*\).*!$#\1#gp' ${mha_log})
# 2.找到change master语句
change=$(grep -i 'change master to' ${mha_log} |awk -F: '{print $4}'|sed "s#xxx#$pass#g")
mha_user=mha
mha_pass=mha
# 1.修复宕机主库
ssh $down_master "systemctl start mysqld"
while true;do
mysql_count=`ps -ef|grep -c [m]ysqld`
if [ $mysql_count -gt 0 ];then
action "$down_master MySQL Server" /bin/true
# 3.在宕机主库中执行，开启主从复制
mysql -u${mha_user} -p${mha_pass} -h$down_master -e "${change};start slave;" &>/dev/null
break
fi
done
# 4.修复 MHA 配置文件
cat > /etc/mha/app1.cnf <<EOF
[server default]
manager_log=${mha_log}
manager_workdir=/etc/mha/app1
master_binlog_dir=/app/mysql/data
master_ip_failover_script=/etc/mha/app1/master_ip_failover
password=${mha_user}
ping_interval=2
repl_password=123
repl_user=rep
ssh_port=22
ssh_user=root
user=${mha_pass}
[server1]
hostname=172.16.1.51
port=3306
[server2]
hostname=172.16.1.52
port=3306
[server3]
hostname=172.16.1.53
port=3306
[server4]
hostname=172.16.1.54
port=3306
[binlog1]
no_master=1
master_binlog_dir=/data/mysql/binlog
hostname=172.16.1.54
EOFMHA的VIP漂移（在云主机上，不支持vip漂移）
不能添加ip地址，不能自定义路由
1）MHA + keepalived
缺陷：
- keepalived指定两台设备漂移，另一台必须提升为主库 （candidate_master）
- 如何确定另一台BACKUP数据一定最新 （半同步）
- 半同步会阻塞主库写入，影响主库性能，随时恢复到异步
- （不对外提供服务）
- （提升半同步复制机器的配置）
2）MHA自带vip脚本
- 手动将vip绑定到主库上
[root@db02 ~]# ifconfig eth1:1 172.16.1.55/24配置MHA识别master_ip_failover脚本
# 1.修改配置文件
[root@db04 ~]# vim /etc/mha/app1.cnf
[server default]
master_ip_failover_script=/etc/mha/app1/master_ip_failover
# 2.授权
[root@db04 ~]# chmod +x /etc/mha/app1/master_ip_failover
# 4.手动在主库上绑定VIP
[root@db02 ~]# ifconfig eth1:1 172.16.1.55/24
## 导致MHA起不来
1）脚本不在指定目录
2）没有执行权
3）格式问题 127:0
[root@db04 ~]# dos2unix /etc/mha/app1/master_ip_failover
dos2unix: converting file /etc/mha/app1/master_ip_failover to Unix format ...
# 脚本内容
[root@db04 bin]# cat /etc/mha/app1/master_ip_failover
#!/usr/bin/env perl
use strict;
use warnings FATAL => 'all';
use Getopt::Long;
my (
$command, $ssh_user, $orig_master_host, $orig_master_ip,
$orig_master_port, $new_master_host, $new_master_ip, $new_master_port
);
my $vip = '172.16.1.55/24';
my $key = '1';
my $ssh_start_vip = "/sbin/ifconfig eth1:$key $vip";
my $ssh_stop_vip = "/sbin/ifconfig eth1:$key down";
GetOptions(
'command=s' => \$command,
'ssh_user=s' => \$ssh_user,
'orig_master_host=s' => \$orig_master_host,
'orig_master_ip=s' => \$orig_master_ip,
'orig_master_port=i' => \$orig_master_port,
'new_master_host=s' => \$new_master_host,
'new_master_ip=s' => \$new_master_ip,
'new_master_port=i' => \$new_master_port,
);
exit &main();
sub main {
print "\n\nIN SCRIPT TEST====$ssh_stop_vip==$ssh_start_vip===\n\n";
if ( $command eq "stop" || $command eq "stopssh" ) {
my $exit_code = 1;
eval {
print "Disabling the VIP on old master: $orig_master_host \n";
&stop_vip();
$exit_code = 0;
};
if ($@) {
warn "Got Error: $@\n";
exit $exit_code;
}
exit $exit_code;
}
elsif ( $command eq "start" ) {
my $exit_code = 10;
eval {
print "Enabling the VIP - $vip on the new master - $new_master_host \n";
&start_vip();
$exit_code = 0;
};
if ($@) {
warn $@;
exit $exit_code;
}
exit $exit_code;
}
elsif ( $command eq "status" ) {
print "Checking the Status of the script.. OK \n";
exit 0;
}
else {
&usage();
exit 1;
}
}
sub start_vip() {
`ssh $ssh_user\@$new_master_host \" $ssh_start_vip \"`;
}
sub stop_vip() {
return 0 unless ($ssh_user);
`ssh $ssh_user\@$orig_master_host \" $ssh_stop_vip \"`;
}
sub usage {
print
"Usage: master_ip_failover --command=start|stop|stopssh|status --orig_master_host=host --orig_master_ip=ip --orig_master_port=port --new_master_host=host --new_master_ip=ip --new_master_port=port\n";
}
解决binlog因为断电断网无法保存
# 1.修改配置文件
[root@db04 ~]# vim /etc/mha/app1.cnf
[server default]
master_ip_failover_script=/etc/mha/app1/master_ip_failover
# 2.授权
[root@db04 ~]# chmod +x /etc/mha/app1/master_ip_failover
# 4.手动在主库上绑定VIP
[root@db02 ~]# ifconfig eth1:1 172.16.1.55/24
## 导致MHA起不来
1）脚本不在指定目录
2）没有执行权
3）格式问题 127:0
[root@db04 ~]# dos2unix /etc/mha/app1/master_ip_failover
dos2unix: converting file /etc/mha/app1/master_ip_failover to Unix format ...
# 脚本内容
[root@db04 bin]# cat /etc/mha/app1/master_ip_failover
#!/usr/bin/env perl
use strict;
use warnings FATAL => 'all';
use Getopt::Long;
my (
$command, $ssh_user, $orig_master_host, $orig_master_ip,
$orig_master_port, $new_master_host, $new_master_ip, $new_master_port
);
my $vip = '172.16.1.55/24';
my $key = '1';
my $ssh_start_vip = "/sbin/ifconfig eth1:$key $vip";
my $ssh_stop_vip = "/sbin/ifconfig eth1:$key down";
GetOptions(
'command=s' => \$command,
'ssh_user=s' => \$ssh_user,
'orig_master_host=s' => \$orig_master_host,
'orig_master_ip=s' => \$orig_master_ip,
'orig_master_port=i' => \$orig_master_port,
'new_master_host=s' => \$new_master_host,
'new_master_ip=s' => \$new_master_ip,
'new_master_port=i' => \$new_master_port,
);
exit &main();
sub main {
print "\n\nIN SCRIPT TEST====$ssh_stop_vip==$ssh_start_vip===\n\n";
if ( $command eq "stop" || $command eq "stopssh" ) {
my $exit_code = 1;
eval {
print "Disabling the VIP on old master: $orig_master_host \n";
&stop_vip();
$exit_code = 0;
};
if ($@) {
warn "Got Error: $@\n";
exit $exit_code;
}
exit $exit_code;
}
elsif ( $command eq "start" ) {
my $exit_code = 10;
eval {
print "Enabling the VIP - $vip on the new master - $new_master_host \n";
&start_vip();
$exit_code = 0;
};
if ($@) {
warn $@;
exit $exit_code;
}
exit $exit_code;
}
elsif ( $command eq "status" ) {
print "Checking the Status of the script.. OK \n";
exit 0;
}
else {
&usage();
exit 1;
}
}
sub start_vip() {
`ssh $ssh_user\@$new_master_host \" $ssh_start_vip \"`;
}
sub stop_vip() {
return 0 unless ($ssh_user);
`ssh $ssh_user\@$orig_master_host \" $ssh_stop_vip \"`;
}
sub usage {
print
"Usage: master_ip_failover --command=start|stop|stopssh|status --orig_master_host=host --orig_master_ip=ip --orig_master_port=port --new_master_host=host --new_master_ip=ip --new_master_port=port\n";
}MHA在主库故障的时候首先会备份binlog
binlog Server
## binlog server是MySQL集群内的某一台机器
[server1]
hostname=172.16.1.51
port=3306
[server2]
hostname=172.16.1.52
port=3306
[server3]
hostname=172.16.1.53
port=3306
[server4]
hostname=172.16.1.54
port=3306
[binlog1]
no_master=1
master_binlog_dir=/data/mysql/binlog
hostname=172.16.1.54
## binlog server是MySQL集群外的某一台机器
[server1]
hostname=172.16.1.51
port=3306
[server2]
hostname=172.16.1.52
port=3306
[server3]
hostname=172.16.1.53
port=3306
[server4]
hostname=172.16.1.54
port=3306
[binlog1]
master_binlog_dir=/data/mysql/binlog
hostname=172.16.1.58
## 创建binlog存放目录
mkdir -p /data/mysql/binlog
[root@db04 binlog]# mysqlbinlog -R --host=172.16.1.55 --user=mha --password=mha --raw --stop-never mysql-bin.000001 &
2）mysqlbinlog
- 查看binlog --base64-output=decode-rows
- 截取binlog --start-position --stop-position
- 同步binlog1）sersync + rsync
问题：MHA怎么知道，你binlog在哪？
[binlog1]
master_binlog_dir=/data/mysql/binlog
hostname=172.16.1.542）mysqlbinlog
– 查看binlog –base64-output=decode-rows
– 截取binlog –start-position –stop-position
- 同步binlog
MySQL读写分离
- mysql-proxy 服务自带：Lua
- atlas 360：C语言 Alibaba：Java语言
- mycat 重量级Atlas主要功能
1.读写分离
2.从库负载均衡
3.IP过滤
4.自动分表（不用）
5.DBA可平滑上下线DB
6.自动摘除宕机的DB
分表功能
-
垂直拆分
-
水平拆分
Atlas相对于官方MySQL-Proxy的优势
1.将主流程中所有Lua代码用C重写，Lua仅用于管理接口
2.重写网络模型、线程模型
3.实现了真正意义上的连接池
4.优化了锁机制，性能提高数十倍
Atlas安装
## 安装
[root@db01 ~]# rpm -ivh Atlas-2.2.1.el6.x86_64.rpm
## 修改配置文件
[root@db01 ~]# vim /usr/local/mysql-proxy/conf/test.cnf
# 对所用数据库账户的密码进行加密
[root@db04 bin]# cd /usr/local/mysql-proxy/bin/
[root@db04 bin]# ll
total 44
-rwxr-xr-x 1 root root 9696 Dec 17 2014 encrypt
-rwxr-xr-x 1 root root 23564 Dec 17 2014 mysql-proxy
-rwxr-xr-x 1 root root 1552 Dec 17 2014 mysql-proxyd
-rw-r--r-- 1 root root 6 Dec 17 2014 VERSION
[root@db01 bin]# ./encrypt 123
3yb5jEku5h4=
[root@db01 bin]# ./encrypt mha
O2jBXONX098=
## 管理接口
#管理接口的用户名
admin-username = user
#管理接口的密码
admin-password = pwd
#Atlas监听的管理接口IP和端口
admin-address = 0.0.0.0:2345
# Atlas配置文件
[root@db01 conf]# cat /usr/local/mysql-proxy/conf/lol.cnf
[mysql-proxy]
#带#号的为非必需的配置项目
#管理接口的用户名
admin-username = user
#管理接口的密码
admin-password = pwd
#Atlas后端连接的MySQL主库的IP和端口，可设置多项，用逗号分隔
proxy-backend-addresses = 172.16.1.55:3306
#Atlas后端连接的MySQL从库的IP和端口，@后面的数字代表权重，用来作负载均衡，若省略则默认为1，可设
置多项，用逗号分隔
proxy-read-only-backend-addresses = 172.16.1.51:3306,172.16.1.53:3306,172.16.1.54:3306
#用户名与其对应的加密过的MySQL密码，密码使用PREFIX/bin目录下的加密程序encrypt加密，下行的user1>和user2为示例，将其替换为你的MySQL的用户名和加密密码！
pwds = rep:3yb5jEku5h4=, mha:O2jBXONX098=
#设置Atlas的运行方式，设为true时为守护进程方式，设为false时为前台方式，一般开发调试时设为false>，线上运行时设为true,true后面不能有空格。
daemon = true
#设置Atlas的运行方式，设为true时Atlas会启动两个进程，一个为monitor，一个为worker，monitor在worker意外退出后会自动将其重启，设为false时只有worker，没有monitor，一般开发调试时设为false，线上运行时设为true,true后面不能有空格。
keepalive = true
#工作线程数，对Atlas的性能有很大影响，可根据情况适当设置,CPU核心数
event-threads = 8
#日志级别，分为message、warning、critical、error、debug五个级别
log-level = error
#日志存放的路径
log-path = /usr/local/mysql-proxy/log
#SQL日志的开关，可设置为OFF、ON、REALTIME，OFF代表不记录SQL日志，ON代表记录SQL日志，REALTIME代>表记录SQL日志且实时写入磁盘，默认为OFF
sql-log = ON
#慢日志输出设置。当设置了该参数时，则日志只输出执行时间超过sql-log-slow（单位：ms)的日志记录。>不设置该参数则输出全部日志。
sql-log-slow = 10
#实例名称，用于同一台机器上多个Atlas实例间的区分
#instance = test
#Atlas监听的工作接口IP和端口
proxy-address = 0.0.0.0:3307
#Atlas监听的管理接口IP和端口
admin-address = 0.0.0.0:2345
#分表设置，此例中person为库名，mt为表名，id为分表字段，3为子表数量，可设置多项，以逗号分隔，若>不分表则不需要设置该项
#tables = person.mt.id.3
#默认字符集，设置该项后客户端不再需要执行SET NAMES语句
charset = utf8
#允许连接Atlas的客户端的IP，可以是精确IP，也可以是IP段，以逗号分隔，若不设置该项则允许所有IP连>接，否则只允许列表中的IP连接
client-ips = 127.0.0.1
#Atlas前面挂接的LVS的物理网卡的IP(注意不是虚IP)，若有LVS且设置了client-ips则此项必须设置，否则>可以不设置
#lvs-ips = 192.168.1.1
# 启动
[root@db04 bin]# /usr/local/mysql-proxy/bin/mysql-proxyd test startAtlas管理接口使用
## 管理接口
#管理接口的用户名
admin-username = user
#管理接口的密码
admin-password = pwd
#Atlas监听的管理接口IP和端口
admin-address = 0.0.0.0:2345
[root@db01 ~]# mysql -uuser -ppwd -h127.0.0.1 -P2345
user@127.0.0.1 [(none)] >SELECT * FROM help;
+----------------------------+---------------------------------------------------------+
| command | description |
+----------------------------+---------------------------------------------------------+
| SELECT * FROM help | 查看帮助 |
| SELECT * FROM backends | 查看后端状态 |
| SET OFFLINE $backend_id | set offline 3; 下线指定的数据库 |
| SET ONLINE $backend_id | set online 3; 上线指定的数据库 |
| ADD MASTER $backend | add master 127.0.0.1:3306 添加一个主库 |
| ADD SLAVE $backend | add slave 172.6.1.59:3306 添加一个从库 |
| REMOVE BACKEND $backend_id | 摘除后端指定的数据库 |
| SELECT * FROM clients | 查看后端可以连接的客户端 |
| ADD CLIENT $client | example: "add client 192.168.1.2", ... |
| REMOVE CLIENT $client | example: "remove client 192.168.1.2", ... |
| SELECT * FROM pwds | 查看atlas允许连接的用户 |
| ADD PWD $pwd | 添加一个用户，密码写明文即可 |
| ADD ENPWD $pwd | 添加一个用户，密码必须先加密再添加 |
| REMOVE PWD $pwd | 删除一个用户 |
| SAVE CONFIG | 保存配置文件 |
| SELECT VERSION | 查看版本 |
+----------------------------+---------------------------------------------------------+
mysql> select * from pwds;
+----------+--------------+
| username | password |
+----------+--------------+
| user1 | +jKsgB3YAG8= |
| user2 | GS+tr4TPgqc= |
+----------+--------------+
mysql> select * from backends;
+-------------+------------------+-------+------+
| backend_ndx | address | state | type |
+-------------+------------------+-------+------+
| 1 | 172.16.1.55:3306 | up | rw |
| 2 | 172.16.1.52:3306 | up | ro |
| 3 | 172.16.1.53:3306 | up | ro |
| 4 | 172.16.1.54:3306 | up | ro |
+-------------+------------------+-------+------+
mysql> set offline 3;
+-------------+------------------+---------+------+
| backend_ndx | address | state | type |
+-------------+------------------+---------+------+
| 3 | 172.16.1.53:3306 | offline | ro |
+-------------+------------------+---------+------+
mysql> select * from backends;
+-------------+------------------+---------+------+
| backend_ndx | address | state | type |
+-------------+------------------+---------+------+
| 1 | 172.16.1.55:3306 | up | rw |
| 2 | 172.16.1.52:3306 | up | ro |
| 3 | 172.16.1.53:3306 | offline | ro |
| 4 | 172.16.1.54:3306 | up | ro |
+-------------+------------------+---------+------+
mysql> set online 3;
+-------------+------------------+---------+------+
| backend_ndx | address | state | type |
+-------------+------------------+---------+------+
| 3 | 172.16.1.53:3306 | unknown | ro |
+-------------+------------------+---------+------+
mysql> select * from backends;
+-------------+------------------+-------+------+
| backend_ndx | address | state | type |
+-------------+------------------+-------+------+
| 1 | 172.16.1.55:3306 | up | rw |
| 2 | 172.16.1.52:3306 | up | ro |
| 3 | 172.16.1.53:3306 | up | ro |
| 4 | 172.16.1.54:3306 | up | ro |
+-------------+------------------+-------+------+
mysql> add master 172.16.1.58:3306;
Empty set (0.00 sec)
mysql> add master 172.16.1.51:3306;
Empty set (0.00 sec)
mysql> select * from backends;
+-------------+------------------+-------+------+
| backend_ndx | address | state | type |
+-------------+------------------+-------+------+
| 1 | 172.16.1.55:3306 | up | rw |
| 2 | 172.16.1.58:3306 | down | rw |
| 3 | 172.16.1.51:3306 | up | rw |
| 4 | 172.16.1.54:3306 | up | ro |
| 5 | 172.16.1.52:3306 | up | ro |
| 6 | 172.16.1.53:3306 | up | ro |
+-------------+------------------+-------+------+
6 rows in set (0.00 sec)
[root@db01 ~]# mysql -uuser -ppwd -h127.0.0.1 -P2345 -e "add slave 172.16.1.59:3306"
配置版MHA安装脚本
免密脚本
vim mianmi.sh
#!/bin/bash
source /etc/init.d/functions
function push_public_key(){
yum install -y sshpass &>/dev/null
if [ ! -f ~/.ssh/id_dsa.pub ];then
ssh-keygen -t dsa -P '' -f ~/.ssh/id_dsa >/dev/null 2>&1
fi
for ip in $@;do
ping -W1 -c1 $ip &>/dev/null
if [ $? -eq 0 ];then
sshpass -p '1' ssh-copy-id -o StrictHostKeyChecking=no -i ~/.ssh/id_dsa.pub root@$ip &>/dev/null
action "主机：$ip" /bin/true
else
action "主机：$ip" /bin/false
fi
done
}
function hhh(){
for ipa in $@;do
scp /root/ssh.sh $ipa:/root/ssh.sh
ssh $ipa "bash /root/ssh.sh ${slave_ip[*]}"
done
}
read -p "请输入slave的IP地址（多台使用空格开）：" -a slave_ip
push_public_key ${slave_ip[*]}
hhh ${slave_ip[*]}
exit
vim ssh.sh
#!/bin/bash
source /etc/init.d/functions
yum install -y sshpass &>/dev/null
if [ ! -f ~/.ssh/id_dsa.pub ];then
ssh-keygen -t dsa -P '' -f ~/.ssh/id_dsa >/dev/null 2>&1
fi
for ip in $@;do
ping -W1 -c1 $ip &>/dev/null
if [ $? -eq 0 ];then
sshpass -p '1' ssh-copy-id -o StrictHostKeyChecking=no -i ~/.ssh/id_dsa.pub root@$ip &>/dev/null
action "主机：$ip" /bin/true
else
action "主机：$ip" /bin/false
fi
done
配置文件
vim /root/mha_peizhi.cnf
# 设置主机列表
hosts=("172.16.1.51" "172.16.1.52" "172.16.1.53" "172.16.1.54")
slave_hosts=("172.16.1.52" "172.16.1.53" "172.16.1.54")
# 主从的变量设置
zhu_user="rep"
zhu_password="123"
mha_user="mha"
mha_password="mha"
primary_host="172.16.1.51"
primary_user="root"
primary_password="123"
slave_user="root"
slave_password="123"
slave1_host="172.16.1.52"
slave2_host="172.16.1.53"
slave3_host="172.16.1.54"
# mha参数变量
mha_log="/etc/mha/logs/manager.log"
pass="123"
mha_user=mha
mha_pass=mha
执行前提
# 执行前先创建可以互相访问的root用户
mysql> grant all on *.* to root@'172.16.1.%' identified by '123';
# 上传node包与manager包至脚本主机上
172.16.1.51（脚本主机）
# 上传VIP漂移脚本至脚本主机上
172.16.1.51（脚本主机）
# 创建mha_peizhi.cnf的配置脚本
# 创建mianmi.sh的免密传输脚本
# 创建ssh.sh的免密脚本
# 创建mha.sh的脚本
脚本
#!/bin/bash
. /root/mha_peizhi.cnf
chongqi(){
for host in "${hosts[@]}";do
ssh "$host" "/etc/init.d/mysqld restart"
done
}
# 获取用户输入的复制方式（GTID或传统）
read -p "请选择复制方式（1. GTID复制，2. 传统主从复制）: " replication_type
# 配置主库的复制方式
if [ $replication_type -eq 1 ]; then
# GTID复制
echo "GTID复制方式已选择。"
else
# 传统主从复制
echo "传统主从复制方式已选择。"
fi
# 配置从库的复制方式
if [ $replication_type -eq 1 ]; then
# 配置GTID主从配置文件
for host in "${hosts[@]}"; do
ssh "${host}" "cat > /etc/my.cnf <<EOF
[mysqld]
basedir=/app/mysql
datadir=/app/mysql/data
server_id=$(expr ${host: -2} + 0)
log-bin=mysql-bin
gtid_mode=on
enforce_gtid_consistency
relay_log_purge=0
EOF"
done
chongqi
echo "GTID配置文件已完成"
# 配置主库的主从复制用户
mysql -u${primary_user} -p${primary_password} -e "GRANT REPLICATION SLAVE ON *.* TO ${zhu_user}@'172.16.1.%' IDENTIFIED BY '${zhu_password}';"
echo "主从复制用户已完成"
# 配置从库的主从复制
for slave_host in "${slave_hosts[@]}"; do
mysql -u${slave_user} -p${slave_password} -h${slave_host} -e "CHANGE MASTER TO MASTER_HOST=\"${primary_host}\", MASTER_USER=\"${zhu_user}\", MASTER_PASSWORD=\"${zhu_password}\",master_auto_position='1',master_port=3306;"
done
echo "GTID主从复制已完成"
else
# 配置主从配置文件
for host in "${hosts[@]}"; do
ssh "${host}" "cat > /etc/my.cnf <<EOF
[mysqld]
basedir=/app/mysql
datadir=/app/mysql/data
server_id=$(expr ${host: -2} + 0)
log-bin=mysql-bin
relay_log_purge=0
EOF"
done
chongqi
echo "主从配置已完成"
# 配置主库的主从复制用户
mysql -u${primary_user} -p${primary_password} -e "GRANT REPLICATION SLAVE ON *.* TO ${zhu_user}@'172.16.1.%' IDENTIFIED BY '${zhu_password}';"
echo "主从复制用户已完成"
# 获取主库的二进制日志文件和位置信息
master_log_file=$(mysql -u${primary_user} -p${primary_password} -e "SHOW MASTER STATUS" | awk 'NR==2{print $1}')
master_log_pos=$(mysql -u${primary_user} -p${primary_password} -e "SHOW MASTER STATUS" | awk 'NR==2{print $2}')
# 配置从库的主从复制
for slave_host in "${slave_hosts[@]}"; do
mysql -u${slave_user} -p${slave_password} -h${slave_host} -e "CHANGE MASTER TO MASTER_HOST=\"${primary_host}\", MASTER_USER=\"${zhu_user}\", MASTER_PASSWORD=\"${zhu_password}\", MASTER_LOG_FILE=\"${master_log_file}\", MASTER_LOG_POS=${master_log_pos},master_port=3306;"
done
echo "从库复制已完成"
fi
# 启动从库的复制进程
for slave_host in "${slave_hosts[@]}"; do
mysql -u${slave_user} -p${slave_password} -h${slave_host} -e 'stop slave;'
mysql -u${slave_user} -p${slave_password} -h${slave_host} -e 'start slave;'
done
echo "启动从库"
# 安装MHA Node组件
for host in "${hosts[@]}"; do
scp mha4mysql-node-0.56-0.el6.noarch.rpm "$host":/root/
ssh "$host" "yum localinstall -y mha4mysql-node-0.56-0.el6.noarch.rpm"
done
echo "安装Node组件"
# 在从库中安装manager组件
scp mha4mysql-manager-0.56-0.el6.noarch.rpm "$slave3_host":/root/
ssh "$slave3_user@$slave3_host" "yum localinstall -y mha4mysql-manager-0.56-0.el6.noarch.rpm"
echo "安装manager组件"
# 创建主从用户及高可用用户
mysql -u${primary_user} -p${primary_password} -e "GRANT REPLICATION SLAVE ON *.* TO ${zhu_user}@'172.16.1.%' IDENTIFIED BY '${zhu_password}';"
mysql -u${primary_user} -p${primary_password} -e "GRANT ALL ON *.* TO ${mha_user}@'172.16.1.%' IDENTIFIED BY '${mha_password}';"
echo "创建主从用户与高可用用户"
# 设置从库只读
for slave_host in "${slave_hosts[@]}"; do
mysql -u${slave_user} -p${slave_password} -h${slave_host} -e 'set global read_only=1;'
done
echo "设置从库只读已完成"
# 命令软链接
for host in "${hosts[@]}"; do
ssh "$host" "ln -s /app/mysql/bin/mysql /usr/bin/mysql"
ssh "$host" "ln -s /app/mysql/bin/mysqlbinlog /usr/bin/mysqlbinlog"
done
echo "命令软链接"
# 创建mha配置文件目录并编写配置文件
ssh "$slave3_host" "mkdir /etc/mha"
ssh "$slave3_host" "cat >/etc/mha/app1.cnf<<EOF
[server default]
manager_log=${mha_log}
manager_workdir=/etc/mha/app1
master_binlog_dir=/app/mysql/data
master_ip_failover_script=/etc/mha/app1/master_ip_failover
password=${mha_user}
ping_interval=2
repl_password=123
repl_user=rep
ssh_port=22
ssh_user=root
user=${mha_pass}
[server1]
hostname=172.16.1.51
port=3306
[server2]
hostname=172.16.1.52
port=3306
[server3]
hostname=172.16.1.53
port=3306
[server4]
hostname=172.16.1.54
port=3306
[binlog1]
no_master=1
master_binlog_dir=/data/mysql/binlog
hostname=172.16.1.54
EOF"
echo "创建mha配置文件目录并编写配置文件已完成"
# 传输跳转VIP脚本
scp master_ip_failover ${slave3_host}:/etc/mha/app1/
# 创建日志目录
ssh "$slave3_host" "mkdir /etc/mha/logs"
echo "创建日志目录已完成"
# 启动 MHA Manager
ssh "$slave3_host" "nohup masterha_manager --conf=/etc/mha/app1.cnf --remove_dead_master_conf --ignore_last_failover < /dev/null > /etc/mha/logs/manager.log 2>&1 &"
echo "启动 MHA Manager已完成"
# 设置虚拟IP
ifconfig eth1:1 172.16.1.55/24
read传参版mha安装脚本
[root@db01 MHA]# cat install_mha.sh
#!/bin/bash
. /etc/init.d/functions
#-------------- 变量 ------------------
check_ip="[1-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}"
mha_worker_dir=/etc/mha
mha_log_dir=/etc/mha/logs
binlog_dir=/app/mysql/data
mha_manager_user=mha
mha_manager_password=mha
repl_user=rep
repl_password=123
#------------ 函数 -------------------
function push_public_key(){
yum install -y sshpass &>/dev/null
if [ ! -f ~/.ssh/id_dsa.pub ];then
ssh-keygen -t dsa -P '' -f ~/.ssh/id_dsa >/dev/null 2>&1
fi
for ip in $@;do
ping -W1 -c1 $ip &>/dev/null
if [ $? -eq 0 ];then
sshpass -p '1' ssh-copy-id -o StrictHostKeyChecking=no -i ~/.ssh/id_dsa.pub root@$ip &>/dev/null
action "主机：$ip" /bin/true
else
action "主机：$ip" /bin/false
fi
done
}
function check_ipaddress(){
if [[ $1 =~ $check_ip ]];then
echo "$1 ok" &>/dev/null
else
echo "Usage：请输入正确的IP地址。"
exit 2
fi
}
#------------------- 程序 -----------------------
read -p "请输入集群个数：" num
expr 1 + $num &>/dev/null
if [ $? -ne 0 ];then
echo "Usage：集群个数必须是一个整形"
exit 1
fi
read -p "请输入master的IP地址：" master_ip
check_ipaddress $master_ip
read -p "请输入slave的IP地址（多台使用空格开）：" -a slave_ip
check_ipaddress ${slave_ip[0]}
push_public_key $master_ip ${slave_ip[*]}
for id in ${slave_ip[*]};do
echo ${id##*.}
done
read -p "请输入manager的IP地址：" manager_ip
check_ipaddress $manager_ip
if [ $manager_ip == $master_ip ];then
echo -e "\e[1;4;5;31mFBI Warning： 尽量不要把mha装在主库上\e[0m"
fi
cat > app1.cnf <<EOF
[server default]
manager_log=${mha_log_dir}/manager.log
manager_workdir=${mha_worker_dir}/app1
master_binlog_dir=$binlog_dir
user=$mha_manager_user
password=$mha_manager_password
ping_interval=2
repl_password=$repl_password
repl_user=$repl_user
ssh_user=root
ssh_port=22
[server100]
hostname=$master_ip
port=3306
EOF
for((n=1;n<=$num-1;n++));do
cat >> app1.cnf <<EOF
[server$n]
hostname=${slave_ip[(($n-1))]}
port=3306
EOF
done
ssh $manager_ip "mkdir -p $mha_worker_dir $mha_log_dir"
scp app1.cnf $manger_ip:$mha_worker_dir
## 管理接口
#管理接口的用户名
admin-username = user
#管理接口的密码
admin-password = pwd
#Atlas监听的管理接口IP和端口
admin-address = 0.0.0.0:2345
[root@db01 ~]# mysql -uuser -ppwd -h127.0.0.1 -P2345
user@127.0.0.1 [(none)] >SELECT * FROM help;
+----------------------------+---------------------------------------------------------+
| command | description |
+----------------------------+---------------------------------------------------------+
| SELECT * FROM help | 查看帮助 |
| SELECT * FROM backends | 查看后端状态 |
| SET OFFLINE $backend_id | set offline 3; 下线指定的数据库 |
| SET ONLINE $backend_id | set online 3; 上线指定的数据库 |
| ADD MASTER $backend | add master 127.0.0.1:3306 添加一个主库 |
| ADD SLAVE $backend | add slave 172.6.1.59:3306 添加一个从库 |
| REMOVE BACKEND $backend_id | 摘除后端指定的数据库 |
| SELECT * FROM clients | 查看后端可以连接的客户端 |
| ADD CLIENT $client | example: "add client 192.168.1.2", ... |
| REMOVE CLIENT $client | example: "remove client 192.168.1.2", ... |
| SELECT * FROM pwds | 查看atlas允许连接的用户 |
| ADD PWD $pwd | 添加一个用户，密码写明文即可 |
| ADD ENPWD $pwd | 添加一个用户，密码必须先加密再添加 |
| REMOVE PWD $pwd | 删除一个用户 |
| SAVE CONFIG | 保存配置文件 |
| SELECT VERSION | 查看版本 |
+----------------------------+---------------------------------------------------------+
mysql> select * from pwds;
+----------+--------------+
| username | password |
+----------+--------------+
| user1 | +jKsgB3YAG8= |
| user2 | GS+tr4TPgqc= |
+----------+--------------+
mysql> select * from backends;
+-------------+------------------+-------+------+
| backend_ndx | address | state | type |
+-------------+------------------+-------+------+
| 1 | 172.16.1.55:3306 | up | rw |
| 2 | 172.16.1.52:3306 | up | ro |
| 3 | 172.16.1.53:3306 | up | ro |
| 4 | 172.16.1.54:3306 | up | ro |
+-------------+------------------+-------+------+
mysql> set offline 3;
+-------------+------------------+---------+------+
| backend_ndx | address | state | type |
+-------------+------------------+---------+------+
| 3 | 172.16.1.53:3306 | offline | ro |
+-------------+------------------+---------+------+
mysql> select * from backends;
+-------------+------------------+---------+------+
| backend_ndx | address | state | type |
+-------------+------------------+---------+------+
| 1 | 172.16.1.55:3306 | up | rw |
| 2 | 172.16.1.52:3306 | up | ro |
| 3 | 172.16.1.53:3306 | offline | ro |
| 4 | 172.16.1.54:3306 | up | ro |
+-------------+------------------+---------+------+
mysql> set online 3;
+-------------+------------------+---------+------+
| backend_ndx | address | state | type |
+-------------+------------------+---------+------+
| 3 | 172.16.1.53:3306 | unknown | ro |
+-------------+------------------+---------+------+
mysql> select * from backends;
+-------------+------------------+-------+------+
| backend_ndx | address | state | type |
+-------------+------------------+-------+------+
| 1 | 172.16.1.55:3306 | up | rw |
| 2 | 172.16.1.52:3306 | up | ro |
| 3 | 172.16.1.53:3306 | up | ro |
| 4 | 172.16.1.54:3306 | up | ro |
+-------------+------------------+-------+------+
mysql> add master 172.16.1.58:3306;
Empty set (0.00 sec)
mysql> add master 172.16.1.51:3306;
Empty set (0.00 sec)
mysql> select * from backends;
+-------------+------------------+-------+------+
| backend_ndx | address | state | type |
+-------------+------------------+-------+------+
| 1 | 172.16.1.55:3306 | up | rw |
| 2 | 172.16.1.58:3306 | down | rw |
| 3 | 172.16.1.51:3306 | up | rw |
| 4 | 172.16.1.54:3306 | up | ro |
| 5 | 172.16.1.52:3306 | up | ro |
| 6 | 172.16.1.53:3306 | up | ro |
+-------------+------------------+-------+------+
6 rows in set (0.00 sec)
[root@db01 ~]# mysql -uuser -ppwd -h127.0.0.1 -P2345 -e "add slave 172.16.1.59:3306"vim mianmi.sh
#!/bin/bash
source /etc/init.d/functions
function push_public_key(){
yum install -y sshpass &>/dev/null
if [ ! -f ~/.ssh/id_dsa.pub ];then
ssh-keygen -t dsa -P '' -f ~/.ssh/id_dsa >/dev/null 2>&1
fi
for ip in $@;do
ping -W1 -c1 $ip &>/dev/null
if [ $? -eq 0 ];then
sshpass -p '1' ssh-copy-id -o StrictHostKeyChecking=no -i ~/.ssh/id_dsa.pub root@$ip &>/dev/null
action "主机：$ip" /bin/true
else
action "主机：$ip" /bin/false
fi
done
}
function hhh(){
for ipa in $@;do
scp /root/ssh.sh $ipa:/root/ssh.sh
ssh $ipa "bash /root/ssh.sh ${slave_ip[*]}"
done
}
read -p "请输入slave的IP地址（多台使用空格开）：" -a slave_ip
push_public_key ${slave_ip[*]}
hhh ${slave_ip[*]}
exit
vim ssh.sh
#!/bin/bash
source /etc/init.d/functions
yum install -y sshpass &>/dev/null
if [ ! -f ~/.ssh/id_dsa.pub ];then
ssh-keygen -t dsa -P '' -f ~/.ssh/id_dsa >/dev/null 2>&1
fi
for ip in $@;do
ping -W1 -c1 $ip &>/dev/null
if [ $? -eq 0 ];then
sshpass -p '1' ssh-copy-id -o StrictHostKeyChecking=no -i ~/.ssh/id_dsa.pub root@$ip &>/dev/null
action "主机：$ip" /bin/true
else
action "主机：$ip" /bin/false
fi
done配置文件
vim /root/mha_peizhi.cnf
# 设置主机列表
hosts=("172.16.1.51" "172.16.1.52" "172.16.1.53" "172.16.1.54")
slave_hosts=("172.16.1.52" "172.16.1.53" "172.16.1.54")
# 主从的变量设置
zhu_user="rep"
zhu_password="123"
mha_user="mha"
mha_password="mha"
primary_host="172.16.1.51"
primary_user="root"
primary_password="123"
slave_user="root"
slave_password="123"
slave1_host="172.16.1.52"
slave2_host="172.16.1.53"
slave3_host="172.16.1.54"
# mha参数变量
mha_log="/etc/mha/logs/manager.log"
pass="123"
mha_user=mha
mha_pass=mha
执行前提
# 执行前先创建可以互相访问的root用户
mysql> grant all on *.* to root@'172.16.1.%' identified by '123';
# 上传node包与manager包至脚本主机上
172.16.1.51（脚本主机）
# 上传VIP漂移脚本至脚本主机上
172.16.1.51（脚本主机）
# 创建mha_peizhi.cnf的配置脚本
# 创建mianmi.sh的免密传输脚本
# 创建ssh.sh的免密脚本
# 创建mha.sh的脚本
脚本
#!/bin/bash
. /root/mha_peizhi.cnf
chongqi(){
for host in "${hosts[@]}";do
ssh "$host" "/etc/init.d/mysqld restart"
done
}
# 获取用户输入的复制方式（GTID或传统）
read -p "请选择复制方式（1. GTID复制，2. 传统主从复制）: " replication_type
# 配置主库的复制方式
if [ $replication_type -eq 1 ]; then
# GTID复制
echo "GTID复制方式已选择。"
else
# 传统主从复制
echo "传统主从复制方式已选择。"
fi
# 配置从库的复制方式
if [ $replication_type -eq 1 ]; then
# 配置GTID主从配置文件
for host in "${hosts[@]}"; do
ssh "${host}" "cat > /etc/my.cnf <<EOF
[mysqld]
basedir=/app/mysql
datadir=/app/mysql/data
server_id=$(expr ${host: -2} + 0)
log-bin=mysql-bin
gtid_mode=on
enforce_gtid_consistency
relay_log_purge=0
EOF"
done
chongqi
echo "GTID配置文件已完成"
# 配置主库的主从复制用户
mysql -u${primary_user} -p${primary_password} -e "GRANT REPLICATION SLAVE ON *.* TO ${zhu_user}@'172.16.1.%' IDENTIFIED BY '${zhu_password}';"
echo "主从复制用户已完成"
# 配置从库的主从复制
for slave_host in "${slave_hosts[@]}"; do
mysql -u${slave_user} -p${slave_password} -h${slave_host} -e "CHANGE MASTER TO MASTER_HOST=\"${primary_host}\", MASTER_USER=\"${zhu_user}\", MASTER_PASSWORD=\"${zhu_password}\",master_auto_position='1',master_port=3306;"
done
echo "GTID主从复制已完成"
else
# 配置主从配置文件
for host in "${hosts[@]}"; do
ssh "${host}" "cat > /etc/my.cnf <<EOF
[mysqld]
basedir=/app/mysql
datadir=/app/mysql/data
server_id=$(expr ${host: -2} + 0)
log-bin=mysql-bin
relay_log_purge=0
EOF"
done
chongqi
echo "主从配置已完成"
# 配置主库的主从复制用户
mysql -u${primary_user} -p${primary_password} -e "GRANT REPLICATION SLAVE ON *.* TO ${zhu_user}@'172.16.1.%' IDENTIFIED BY '${zhu_password}';"
echo "主从复制用户已完成"
# 获取主库的二进制日志文件和位置信息
master_log_file=$(mysql -u${primary_user} -p${primary_password} -e "SHOW MASTER STATUS" | awk 'NR==2{print $1}')
master_log_pos=$(mysql -u${primary_user} -p${primary_password} -e "SHOW MASTER STATUS" | awk 'NR==2{print $2}')
# 配置从库的主从复制
for slave_host in "${slave_hosts[@]}"; do
mysql -u${slave_user} -p${slave_password} -h${slave_host} -e "CHANGE MASTER TO MASTER_HOST=\"${primary_host}\", MASTER_USER=\"${zhu_user}\", MASTER_PASSWORD=\"${zhu_password}\", MASTER_LOG_FILE=\"${master_log_file}\", MASTER_LOG_POS=${master_log_pos},master_port=3306;"
done
echo "从库复制已完成"
fi
# 启动从库的复制进程
for slave_host in "${slave_hosts[@]}"; do
mysql -u${slave_user} -p${slave_password} -h${slave_host} -e 'stop slave;'
mysql -u${slave_user} -p${slave_password} -h${slave_host} -e 'start slave;'
done
echo "启动从库"
# 安装MHA Node组件
for host in "${hosts[@]}"; do
scp mha4mysql-node-0.56-0.el6.noarch.rpm "$host":/root/
ssh "$host" "yum localinstall -y mha4mysql-node-0.56-0.el6.noarch.rpm"
done
echo "安装Node组件"
# 在从库中安装manager组件
scp mha4mysql-manager-0.56-0.el6.noarch.rpm "$slave3_host":/root/
ssh "$slave3_user@$slave3_host" "yum localinstall -y mha4mysql-manager-0.56-0.el6.noarch.rpm"
echo "安装manager组件"
# 创建主从用户及高可用用户
mysql -u${primary_user} -p${primary_password} -e "GRANT REPLICATION SLAVE ON *.* TO ${zhu_user}@'172.16.1.%' IDENTIFIED BY '${zhu_password}';"
mysql -u${primary_user} -p${primary_password} -e "GRANT ALL ON *.* TO ${mha_user}@'172.16.1.%' IDENTIFIED BY '${mha_password}';"
echo "创建主从用户与高可用用户"
# 设置从库只读
for slave_host in "${slave_hosts[@]}"; do
mysql -u${slave_user} -p${slave_password} -h${slave_host} -e 'set global read_only=1;'
done
echo "设置从库只读已完成"
# 命令软链接
for host in "${hosts[@]}"; do
ssh "$host" "ln -s /app/mysql/bin/mysql /usr/bin/mysql"
ssh "$host" "ln -s /app/mysql/bin/mysqlbinlog /usr/bin/mysqlbinlog"
done
echo "命令软链接"
# 创建mha配置文件目录并编写配置文件
ssh "$slave3_host" "mkdir /etc/mha"
ssh "$slave3_host" "cat >/etc/mha/app1.cnf<<EOF
[server default]
manager_log=${mha_log}
manager_workdir=/etc/mha/app1
master_binlog_dir=/app/mysql/data
master_ip_failover_script=/etc/mha/app1/master_ip_failover
password=${mha_user}
ping_interval=2
repl_password=123
repl_user=rep
ssh_port=22
ssh_user=root
user=${mha_pass}
[server1]
hostname=172.16.1.51
port=3306
[server2]
hostname=172.16.1.52
port=3306
[server3]
hostname=172.16.1.53
port=3306
[server4]
hostname=172.16.1.54
port=3306
[binlog1]
no_master=1
master_binlog_dir=/data/mysql/binlog
hostname=172.16.1.54
EOF"
echo "创建mha配置文件目录并编写配置文件已完成"
# 传输跳转VIP脚本
scp master_ip_failover ${slave3_host}:/etc/mha/app1/
# 创建日志目录
ssh "$slave3_host" "mkdir /etc/mha/logs"
echo "创建日志目录已完成"
# 启动 MHA Manager
ssh "$slave3_host" "nohup masterha_manager --conf=/etc/mha/app1.cnf --remove_dead_master_conf --ignore_last_failover < /dev/null > /etc/mha/logs/manager.log 2>&1 &"
echo "启动 MHA Manager已完成"
# 设置虚拟IP
ifconfig eth1:1 172.16.1.55/24
read传参版mha安装脚本
[root@db01 MHA]# cat install_mha.sh
#!/bin/bash
. /etc/init.d/functions
#-------------- 变量 ------------------
check_ip="[1-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}"
mha_worker_dir=/etc/mha
mha_log_dir=/etc/mha/logs
binlog_dir=/app/mysql/data
mha_manager_user=mha
mha_manager_password=mha
repl_user=rep
repl_password=123
#------------ 函数 -------------------
function push_public_key(){
yum install -y sshpass &>/dev/null
if [ ! -f ~/.ssh/id_dsa.pub ];then
ssh-keygen -t dsa -P '' -f ~/.ssh/id_dsa >/dev/null 2>&1
fi
for ip in $@;do
ping -W1 -c1 $ip &>/dev/null
if [ $? -eq 0 ];then
sshpass -p '1' ssh-copy-id -o StrictHostKeyChecking=no -i ~/.ssh/id_dsa.pub root@$ip &>/dev/null
action "主机：$ip" /bin/true
else
action "主机：$ip" /bin/false
fi
done
}
function check_ipaddress(){
if [[ $1 =~ $check_ip ]];then
echo "$1 ok" &>/dev/null
else
echo "Usage：请输入正确的IP地址。"
exit 2
fi
}
#------------------- 程序 -----------------------
read -p "请输入集群个数：" num
expr 1 + $num &>/dev/null
if [ $? -ne 0 ];then
echo "Usage：集群个数必须是一个整形"
exit 1
fi
read -p "请输入master的IP地址：" master_ip
check_ipaddress $master_ip
read -p "请输入slave的IP地址（多台使用空格开）：" -a slave_ip
check_ipaddress ${slave_ip[0]}
push_public_key $master_ip ${slave_ip[*]}
for id in ${slave_ip[*]};do
echo ${id##*.}
done
read -p "请输入manager的IP地址：" manager_ip
check_ipaddress $manager_ip
if [ $manager_ip == $master_ip ];then
echo -e "\e[1;4;5;31mFBI Warning： 尽量不要把mha装在主库上\e[0m"
fi
cat > app1.cnf <<EOF
[server default]
manager_log=${mha_log_dir}/manager.log
manager_workdir=${mha_worker_dir}/app1
master_binlog_dir=$binlog_dir
user=$mha_manager_user
password=$mha_manager_password
ping_interval=2
repl_password=$repl_password
repl_user=$repl_user
ssh_user=root
ssh_port=22
[server100]
hostname=$master_ip
port=3306
EOF
for((n=1;n<=$num-1;n++));do
cat >> app1.cnf <<EOF
[server$n]
hostname=${slave_ip[(($n-1))]}
port=3306
EOF
done
ssh $manager_ip "mkdir -p $mha_worker_dir $mha_log_dir"
scp app1.cnf $manger_ip:$mha_worker_dir
vim /root/mha_peizhi.cnf
# 设置主机列表
hosts=("172.16.1.51" "172.16.1.52" "172.16.1.53" "172.16.1.54")
slave_hosts=("172.16.1.52" "172.16.1.53" "172.16.1.54")
# 主从的变量设置
zhu_user="rep"
zhu_password="123"
mha_user="mha"
mha_password="mha"
primary_host="172.16.1.51"
primary_user="root"
primary_password="123"
slave_user="root"
slave_password="123"
slave1_host="172.16.1.52"
slave2_host="172.16.1.53"
slave3_host="172.16.1.54"
# mha参数变量
mha_log="/etc/mha/logs/manager.log"
pass="123"
mha_user=mha
mha_pass=mha# 执行前先创建可以互相访问的root用户
mysql> grant all on *.* to root@'172.16.1.%' identified by '123';
# 上传node包与manager包至脚本主机上
172.16.1.51（脚本主机）
# 上传VIP漂移脚本至脚本主机上
172.16.1.51（脚本主机）
# 创建mha_peizhi.cnf的配置脚本
# 创建mianmi.sh的免密传输脚本
# 创建ssh.sh的免密脚本
# 创建mha.sh的脚本脚本
#!/bin/bash
. /root/mha_peizhi.cnf
chongqi(){
for host in "${hosts[@]}";do
ssh "$host" "/etc/init.d/mysqld restart"
done
}
# 获取用户输入的复制方式（GTID或传统）
read -p "请选择复制方式（1. GTID复制，2. 传统主从复制）: " replication_type
# 配置主库的复制方式
if [ $replication_type -eq 1 ]; then
# GTID复制
echo "GTID复制方式已选择。"
else
# 传统主从复制
echo "传统主从复制方式已选择。"
fi
# 配置从库的复制方式
if [ $replication_type -eq 1 ]; then
# 配置GTID主从配置文件
for host in "${hosts[@]}"; do
ssh "${host}" "cat > /etc/my.cnf <<EOF
[mysqld]
basedir=/app/mysql
datadir=/app/mysql/data
server_id=$(expr ${host: -2} + 0)
log-bin=mysql-bin
gtid_mode=on
enforce_gtid_consistency
relay_log_purge=0
EOF"
done
chongqi
echo "GTID配置文件已完成"
# 配置主库的主从复制用户
mysql -u${primary_user} -p${primary_password} -e "GRANT REPLICATION SLAVE ON *.* TO ${zhu_user}@'172.16.1.%' IDENTIFIED BY '${zhu_password}';"
echo "主从复制用户已完成"
# 配置从库的主从复制
for slave_host in "${slave_hosts[@]}"; do
mysql -u${slave_user} -p${slave_password} -h${slave_host} -e "CHANGE MASTER TO MASTER_HOST=\"${primary_host}\", MASTER_USER=\"${zhu_user}\", MASTER_PASSWORD=\"${zhu_password}\",master_auto_position='1',master_port=3306;"
done
echo "GTID主从复制已完成"
else
# 配置主从配置文件
for host in "${hosts[@]}"; do
ssh "${host}" "cat > /etc/my.cnf <<EOF
[mysqld]
basedir=/app/mysql
datadir=/app/mysql/data
server_id=$(expr ${host: -2} + 0)
log-bin=mysql-bin
relay_log_purge=0
EOF"
done
chongqi
echo "主从配置已完成"
# 配置主库的主从复制用户
mysql -u${primary_user} -p${primary_password} -e "GRANT REPLICATION SLAVE ON *.* TO ${zhu_user}@'172.16.1.%' IDENTIFIED BY '${zhu_password}';"
echo "主从复制用户已完成"
# 获取主库的二进制日志文件和位置信息
master_log_file=$(mysql -u${primary_user} -p${primary_password} -e "SHOW MASTER STATUS" | awk 'NR==2{print $1}')
master_log_pos=$(mysql -u${primary_user} -p${primary_password} -e "SHOW MASTER STATUS" | awk 'NR==2{print $2}')
# 配置从库的主从复制
for slave_host in "${slave_hosts[@]}"; do
mysql -u${slave_user} -p${slave_password} -h${slave_host} -e "CHANGE MASTER TO MASTER_HOST=\"${primary_host}\", MASTER_USER=\"${zhu_user}\", MASTER_PASSWORD=\"${zhu_password}\", MASTER_LOG_FILE=\"${master_log_file}\", MASTER_LOG_POS=${master_log_pos},master_port=3306;"
done
echo "从库复制已完成"
fi
# 启动从库的复制进程
for slave_host in "${slave_hosts[@]}"; do
mysql -u${slave_user} -p${slave_password} -h${slave_host} -e 'stop slave;'
mysql -u${slave_user} -p${slave_password} -h${slave_host} -e 'start slave;'
done
echo "启动从库"
# 安装MHA Node组件
for host in "${hosts[@]}"; do
scp mha4mysql-node-0.56-0.el6.noarch.rpm "$host":/root/
ssh "$host" "yum localinstall -y mha4mysql-node-0.56-0.el6.noarch.rpm"
done
echo "安装Node组件"
# 在从库中安装manager组件
scp mha4mysql-manager-0.56-0.el6.noarch.rpm "$slave3_host":/root/
ssh "$slave3_user@$slave3_host" "yum localinstall -y mha4mysql-manager-0.56-0.el6.noarch.rpm"
echo "安装manager组件"
# 创建主从用户及高可用用户
mysql -u${primary_user} -p${primary_password} -e "GRANT REPLICATION SLAVE ON *.* TO ${zhu_user}@'172.16.1.%' IDENTIFIED BY '${zhu_password}';"
mysql -u${primary_user} -p${primary_password} -e "GRANT ALL ON *.* TO ${mha_user}@'172.16.1.%' IDENTIFIED BY '${mha_password}';"
echo "创建主从用户与高可用用户"
# 设置从库只读
for slave_host in "${slave_hosts[@]}"; do
mysql -u${slave_user} -p${slave_password} -h${slave_host} -e 'set global read_only=1;'
done
echo "设置从库只读已完成"
# 命令软链接
for host in "${hosts[@]}"; do
ssh "$host" "ln -s /app/mysql/bin/mysql /usr/bin/mysql"
ssh "$host" "ln -s /app/mysql/bin/mysqlbinlog /usr/bin/mysqlbinlog"
done
echo "命令软链接"
# 创建mha配置文件目录并编写配置文件
ssh "$slave3_host" "mkdir /etc/mha"
ssh "$slave3_host" "cat >/etc/mha/app1.cnf<<EOF
[server default]
manager_log=${mha_log}
manager_workdir=/etc/mha/app1
master_binlog_dir=/app/mysql/data
master_ip_failover_script=/etc/mha/app1/master_ip_failover
password=${mha_user}
ping_interval=2
repl_password=123
repl_user=rep
ssh_port=22
ssh_user=root
user=${mha_pass}
[server1]
hostname=172.16.1.51
port=3306
[server2]
hostname=172.16.1.52
port=3306
[server3]
hostname=172.16.1.53
port=3306
[server4]
hostname=172.16.1.54
port=3306
[binlog1]
no_master=1
master_binlog_dir=/data/mysql/binlog
hostname=172.16.1.54
EOF"
echo "创建mha配置文件目录并编写配置文件已完成"
# 传输跳转VIP脚本
scp master_ip_failover ${slave3_host}:/etc/mha/app1/
# 创建日志目录
ssh "$slave3_host" "mkdir /etc/mha/logs"
echo "创建日志目录已完成"
# 启动 MHA Manager
ssh "$slave3_host" "nohup masterha_manager --conf=/etc/mha/app1.cnf --remove_dead_master_conf --ignore_last_failover < /dev/null > /etc/mha/logs/manager.log 2>&1 &"
echo "启动 MHA Manager已完成"
# 设置虚拟IP
ifconfig eth1:1 172.16.1.55/24
read传参版mha安装脚本
[root@db01 MHA]# cat install_mha.sh
#!/bin/bash
. /etc/init.d/functions
#-------------- 变量 ------------------
check_ip="[1-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}"
mha_worker_dir=/etc/mha
mha_log_dir=/etc/mha/logs
binlog_dir=/app/mysql/data
mha_manager_user=mha
mha_manager_password=mha
repl_user=rep
repl_password=123
#------------ 函数 -------------------
function push_public_key(){
yum install -y sshpass &>/dev/null
if [ ! -f ~/.ssh/id_dsa.pub ];then
ssh-keygen -t dsa -P '' -f ~/.ssh/id_dsa >/dev/null 2>&1
fi
for ip in $@;do
ping -W1 -c1 $ip &>/dev/null
if [ $? -eq 0 ];then
sshpass -p '1' ssh-copy-id -o StrictHostKeyChecking=no -i ~/.ssh/id_dsa.pub root@$ip &>/dev/null
action "主机：$ip" /bin/true
else
action "主机：$ip" /bin/false
fi
done
}
function check_ipaddress(){
if [[ $1 =~ $check_ip ]];then
echo "$1 ok" &>/dev/null
else
echo "Usage：请输入正确的IP地址。"
exit 2
fi
}
#------------------- 程序 -----------------------
read -p "请输入集群个数：" num
expr 1 + $num &>/dev/null
if [ $? -ne 0 ];then
echo "Usage：集群个数必须是一个整形"
exit 1
fi
read -p "请输入master的IP地址：" master_ip
check_ipaddress $master_ip
read -p "请输入slave的IP地址（多台使用空格开）：" -a slave_ip
check_ipaddress ${slave_ip[0]}
push_public_key $master_ip ${slave_ip[*]}
for id in ${slave_ip[*]};do
echo ${id##*.}
done
read -p "请输入manager的IP地址：" manager_ip
check_ipaddress $manager_ip
if [ $manager_ip == $master_ip ];then
echo -e "\e[1;4;5;31mFBI Warning： 尽量不要把mha装在主库上\e[0m"
fi
cat > app1.cnf <<EOF
[server default]
manager_log=${mha_log_dir}/manager.log
manager_workdir=${mha_worker_dir}/app1
master_binlog_dir=$binlog_dir
user=$mha_manager_user
password=$mha_manager_password
ping_interval=2
repl_password=$repl_password
repl_user=$repl_user
ssh_user=root
ssh_port=22
[server100]
hostname=$master_ip
port=3306
EOF
for((n=1;n<=$num-1;n++));do
cat >> app1.cnf <<EOF
[server$n]
hostname=${slave_ip[(($n-1))]}
port=3306
EOF
done
ssh $manager_ip "mkdir -p $mha_worker_dir $mha_log_dir"
scp app1.cnf $manger_ip:$mha_worker_dir
#!/bin/bash
. /root/mha_peizhi.cnf
chongqi(){
for host in "${hosts[@]}";do
ssh "$host" "/etc/init.d/mysqld restart"
done
}
# 获取用户输入的复制方式（GTID或传统）
read -p "请选择复制方式（1. GTID复制，2. 传统主从复制）: " replication_type
# 配置主库的复制方式
if [ $replication_type -eq 1 ]; then
# GTID复制
echo "GTID复制方式已选择。"
else
# 传统主从复制
echo "传统主从复制方式已选择。"
fi
# 配置从库的复制方式
if [ $replication_type -eq 1 ]; then
# 配置GTID主从配置文件
for host in "${hosts[@]}"; do
ssh "${host}" "cat > /etc/my.cnf <<EOF
[mysqld]
basedir=/app/mysql
datadir=/app/mysql/data
server_id=$(expr ${host: -2} + 0)
log-bin=mysql-bin
gtid_mode=on
enforce_gtid_consistency
relay_log_purge=0
EOF"
done
chongqi
echo "GTID配置文件已完成"
# 配置主库的主从复制用户
mysql -u${primary_user} -p${primary_password} -e "GRANT REPLICATION SLAVE ON *.* TO ${zhu_user}@'172.16.1.%' IDENTIFIED BY '${zhu_password}';"
echo "主从复制用户已完成"
# 配置从库的主从复制
for slave_host in "${slave_hosts[@]}"; do
mysql -u${slave_user} -p${slave_password} -h${slave_host} -e "CHANGE MASTER TO MASTER_HOST=\"${primary_host}\", MASTER_USER=\"${zhu_user}\", MASTER_PASSWORD=\"${zhu_password}\",master_auto_position='1',master_port=3306;"
done
echo "GTID主从复制已完成"
else
# 配置主从配置文件
for host in "${hosts[@]}"; do
ssh "${host}" "cat > /etc/my.cnf <<EOF
[mysqld]
basedir=/app/mysql
datadir=/app/mysql/data
server_id=$(expr ${host: -2} + 0)
log-bin=mysql-bin
relay_log_purge=0
EOF"
done
chongqi
echo "主从配置已完成"
# 配置主库的主从复制用户
mysql -u${primary_user} -p${primary_password} -e "GRANT REPLICATION SLAVE ON *.* TO ${zhu_user}@'172.16.1.%' IDENTIFIED BY '${zhu_password}';"
echo "主从复制用户已完成"
# 获取主库的二进制日志文件和位置信息
master_log_file=$(mysql -u${primary_user} -p${primary_password} -e "SHOW MASTER STATUS" | awk 'NR==2{print $1}')
master_log_pos=$(mysql -u${primary_user} -p${primary_password} -e "SHOW MASTER STATUS" | awk 'NR==2{print $2}')
# 配置从库的主从复制
for slave_host in "${slave_hosts[@]}"; do
mysql -u${slave_user} -p${slave_password} -h${slave_host} -e "CHANGE MASTER TO MASTER_HOST=\"${primary_host}\", MASTER_USER=\"${zhu_user}\", MASTER_PASSWORD=\"${zhu_password}\", MASTER_LOG_FILE=\"${master_log_file}\", MASTER_LOG_POS=${master_log_pos},master_port=3306;"
done
echo "从库复制已完成"
fi
# 启动从库的复制进程
for slave_host in "${slave_hosts[@]}"; do
mysql -u${slave_user} -p${slave_password} -h${slave_host} -e 'stop slave;'
mysql -u${slave_user} -p${slave_password} -h${slave_host} -e 'start slave;'
done
echo "启动从库"
# 安装MHA Node组件
for host in "${hosts[@]}"; do
scp mha4mysql-node-0.56-0.el6.noarch.rpm "$host":/root/
ssh "$host" "yum localinstall -y mha4mysql-node-0.56-0.el6.noarch.rpm"
done
echo "安装Node组件"
# 在从库中安装manager组件
scp mha4mysql-manager-0.56-0.el6.noarch.rpm "$slave3_host":/root/
ssh "$slave3_user@$slave3_host" "yum localinstall -y mha4mysql-manager-0.56-0.el6.noarch.rpm"
echo "安装manager组件"
# 创建主从用户及高可用用户
mysql -u${primary_user} -p${primary_password} -e "GRANT REPLICATION SLAVE ON *.* TO ${zhu_user}@'172.16.1.%' IDENTIFIED BY '${zhu_password}';"
mysql -u${primary_user} -p${primary_password} -e "GRANT ALL ON *.* TO ${mha_user}@'172.16.1.%' IDENTIFIED BY '${mha_password}';"
echo "创建主从用户与高可用用户"
# 设置从库只读
for slave_host in "${slave_hosts[@]}"; do
mysql -u${slave_user} -p${slave_password} -h${slave_host} -e 'set global read_only=1;'
done
echo "设置从库只读已完成"
# 命令软链接
for host in "${hosts[@]}"; do
ssh "$host" "ln -s /app/mysql/bin/mysql /usr/bin/mysql"
ssh "$host" "ln -s /app/mysql/bin/mysqlbinlog /usr/bin/mysqlbinlog"
done
echo "命令软链接"
# 创建mha配置文件目录并编写配置文件
ssh "$slave3_host" "mkdir /etc/mha"
ssh "$slave3_host" "cat >/etc/mha/app1.cnf<<EOF
[server default]
manager_log=${mha_log}
manager_workdir=/etc/mha/app1
master_binlog_dir=/app/mysql/data
master_ip_failover_script=/etc/mha/app1/master_ip_failover
password=${mha_user}
ping_interval=2
repl_password=123
repl_user=rep
ssh_port=22
ssh_user=root
user=${mha_pass}
[server1]
hostname=172.16.1.51
port=3306
[server2]
hostname=172.16.1.52
port=3306
[server3]
hostname=172.16.1.53
port=3306
[server4]
hostname=172.16.1.54
port=3306
[binlog1]
no_master=1
master_binlog_dir=/data/mysql/binlog
hostname=172.16.1.54
EOF"
echo "创建mha配置文件目录并编写配置文件已完成"
# 传输跳转VIP脚本
scp master_ip_failover ${slave3_host}:/etc/mha/app1/
# 创建日志目录
ssh "$slave3_host" "mkdir /etc/mha/logs"
echo "创建日志目录已完成"
# 启动 MHA Manager
ssh "$slave3_host" "nohup masterha_manager --conf=/etc/mha/app1.cnf --remove_dead_master_conf --ignore_last_failover < /dev/null > /etc/mha/logs/manager.log 2>&1 &"
echo "启动 MHA Manager已完成"
# 设置虚拟IP
ifconfig eth1:1 172.16.1.55/24[root@db01 MHA]# cat install_mha.sh
#!/bin/bash
. /etc/init.d/functions
#-------------- 变量 ------------------
check_ip="[1-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}"
mha_worker_dir=/etc/mha
mha_log_dir=/etc/mha/logs
binlog_dir=/app/mysql/data
mha_manager_user=mha
mha_manager_password=mha
repl_user=rep
repl_password=123
#------------ 函数 -------------------
function push_public_key(){
yum install -y sshpass &>/dev/null
if [ ! -f ~/.ssh/id_dsa.pub ];then
ssh-keygen -t dsa -P '' -f ~/.ssh/id_dsa >/dev/null 2>&1
fi
for ip in $@;do
ping -W1 -c1 $ip &>/dev/null
if [ $? -eq 0 ];then
sshpass -p '1' ssh-copy-id -o StrictHostKeyChecking=no -i ~/.ssh/id_dsa.pub root@$ip &>/dev/null
action "主机：$ip" /bin/true
else
action "主机：$ip" /bin/false
fi
done
}
function check_ipaddress(){
if [[ $1 =~ $check_ip ]];then
echo "$1 ok" &>/dev/null
else
echo "Usage：请输入正确的IP地址。"
exit 2
fi
}
#------------------- 程序 -----------------------
read -p "请输入集群个数：" num
expr 1 + $num &>/dev/null
if [ $? -ne 0 ];then
echo "Usage：集群个数必须是一个整形"
exit 1
fi
read -p "请输入master的IP地址：" master_ip
check_ipaddress $master_ip
read -p "请输入slave的IP地址（多台使用空格开）：" -a slave_ip
check_ipaddress ${slave_ip[0]}
push_public_key $master_ip ${slave_ip[*]}
for id in ${slave_ip[*]};do
echo ${id##*.}
done
read -p "请输入manager的IP地址：" manager_ip
check_ipaddress $manager_ip
if [ $manager_ip == $master_ip ];then
echo -e "\e[1;4;5;31mFBI Warning： 尽量不要把mha装在主库上\e[0m"
fi
cat > app1.cnf <<EOF
[server default]
manager_log=${mha_log_dir}/manager.log
manager_workdir=${mha_worker_dir}/app1
master_binlog_dir=$binlog_dir
user=$mha_manager_user
password=$mha_manager_password
ping_interval=2
repl_password=$repl_password
repl_user=$repl_user
ssh_user=root
ssh_port=22
[server100]
hostname=$master_ip
port=3306
EOF
for((n=1;n<=$num-1;n++));do
cat >> app1.cnf <<EOF
[server$n]
hostname=${slave_ip[(($n-1))]}
port=3306
EOF
done
ssh $manager_ip "mkdir -p $mha_worker_dir $mha_log_dir"
scp app1.cnf $manger_ip:$mha_worker_dir
