码迷,mamicode.com
首页 > 数据库 > 详细

mysql复制+keepalived+haproxy配置(负载均衡)

时间:2020-09-24 22:10:47      阅读:75      评论:0      收藏:0      [点我收藏+]

标签:tco   name   绑定   tcp   address   后端   realm   isp   master   

双主+keepalived+haproxy配置(负载均衡)

实验系统:CentOS 6.5_x86_64
实验前提:防火墙和selinux都关闭
实验软件:keepalived-1.2.13  haproxy-1.8.13  mysql—5.7.21
主1 ip:192.168.226.134
主2 ip:192.168.226.135
vip 192.168.226.150

一、安装mysql

获取mysql安装包:wget https://dev.mysql.com/get/Downloads/MySQL-5.7/mysql-5.7.21-linux-glibc2.12-x86_64.tar.gz
安装mysql过程省略

二、配置主主复制

1 命令 service iptables stop 
2 检查:service iptables status
3 关闭vi /etc/selinux/config 
4 SELINUX=disabled

 

两台服务器上都要执行:
创建复制用户

1 grant replication slave on *.* to ‘repl‘@‘%‘ identified by ‘123‘;

进入从服务器mysql
命令: # mysql -uroot -p
关闭slave
命令:stop slave;
开始配置:
输入下面代码即可:

1 CHANGE MASTER TO MASTER_HOST=‘192.168.226.135‘, MASTER_USER=‘repl‘, MASTER_PASSWORD=‘123‘, MASTER_LOG_FILE=‘mysql-bin.000002‘, MASTER_LOG_POS=681;


先在从服务器配置完成,启动从服务器:
命令: start slave;
反向再配置一次

三、安装haproxy(两台节点都要安装)

1 tar -zxvxf haproxy-1.8.13.tar.gz 
2 cd haproxy-1.8.13
3 make TARGET=linux2628 //根据自己主机设定
4 make install

 

提供启动脚本

 
  1 vi /etc/init.d/haproxy
  2 
  3 #!/bin/sh
  4 #
  5 # haproxy
  6 #
  7 # chkconfig: - 85 15
  8 # description: HAProxy is a free, very fast and reliable solution   9 # offering high availability, load balancing, and  10 # proxying for TCP and HTTP-based applications
 11 # processname: haproxy
 12 # config: /etc/haproxy/haproxy.cfg
 13 # pidfile: /var/run/haproxy.pid
 14 
 15 # Source function library.
 16 . /etc/rc.d/init.d/functions
 17 
 18 # Source networking configuration.
 19 . /etc/sysconfig/network
 20 
 21 # Check that networking is up.
 22 [ "$NETWORKING" = "no" ] && exit 0
 23 
 24 exec="/usr/local/sbin/haproxy"
 25 prog=$(basename $exec)
 26 
 27 [ -e /etc/sysconfig/$prog ] && . /etc/sysconfig/$prog
 28 
 29 cfgfile=/etc/haproxy/haproxy.cfg
 30 pidfile=/var/run/haproxy.pid
 31 lockfile=/var/lock/subsys/haproxy
 32 
 33 check() {
 34 $exec -c -V -f $cfgfile $OPTIONS
 35 }
 36 
 37 start() {
 38 $exec -c -q -f $cfgfile $OPTIONS
 39 if [ $? -ne 0 ]; then
 40 echo "Errors in configuration file, check with $prog check."
 41 return 1
 42 fi
 43 
 44 echo -n $"Starting $prog: "
 45 # start it up here, usually something like "daemon $exec"
 46 daemon $exec -D -f $cfgfile -p $pidfile $OPTIONS
 47 retval=$?
 48 echo
 49 [ $retval -eq 0 ] && touch $lockfile
 50 return $retval
 51 }
 52 
 53 stop() {
 54 echo -n $"Stopping $prog: "
 55 # stop it here, often "killproc $prog"
 56 killproc $prog
 57 retval=$?
 58 echo
 59 [ $retval -eq 0 ] && rm -f $lockfile
 60 return $retval
 61 }
 62 
 63 restart() {
 64 $exec -c -q -f $cfgfile $OPTIONS
 65 if [ $? -ne 0 ]; then
 66 echo "Errors in configuration file, check with $prog check."
 67 return 1
 68 fi
 69 stop
 70 start
 71 }
 72 
 73 reload() {
 74 $exec -c -q -f $cfgfile $OPTIONS
 75 if [ $? -ne 0 ]; then
 76 echo "Errors in configuration file, check with $prog check."
 77 return 1
 78 fi
 79 echo -n $"Reloading $prog: "
 80 $exec -D -f $cfgfile -p $pidfile $OPTIONS -sf $(cat $pidfile)
 81 retval=$?
 82 echo
 83 return $retval
 84 }
 85 
 86 force_reload() {
 87 restart
 88 }
 89 
 90 fdr_status() {
 91 status $prog
 92 }
 93 
 94 case "$1" in
 95 start|stop|restart|reload)
 96 $1
 97 ;;
 98 force-reload)
 99 force_reload
100 ;;
101 check)
102 check
103 ;;
104 status)
105 fdr_status
106 ;;
107 condrestart|try-restart)
108 [ ! -f $lockfile ] || restart
109 ;;
110 *)
111 echo $"Usage: $0 {start|stop|status|restart|try-restart|reload|force-reload}"
112 exit 2
113 esac
 

 

1 chkconfig --add haproxy
2 chkconfig haproxy on
3 chmod +x /etc/init.d/haproxy

 

提供配置文件

1 mkdir /etc/haproxy
2 mkdir /var/lib/haproxy
3 useradd -r haproxy

 

 
 1 vi /etc/haproxy/haproxy.cfg
 2 
 3 global
 4 
 5 log 127.0.0.1 local2
 6 
 7 chroot /var/lib/haproxy
 8 pidfile /var/run/haproxy.pid
 9 maxconn 4000
10 user haproxy
11 group haproxy
12 daemon
13 
14 stats socket /var/lib/haproxy/stats
15 
16 defaults
17 mode tcp //haproxy运行模式
18 log global
19 option dontlognull
20 option redispatch
21 retries 3
22 timeout http-request 10s
23 timeout queue 1m
24 timeout connect 10s
25 timeout client 1m
26 timeout server 1m
27 timeout http-keep-alive 10s
28 timeout check 10s
29 maxconn 600 //最大连接数
30 
31 listen stats //配置haproxy状态页
32 mode http
33 bind :6677 //找一个比较特殊的端口
34 stats enable
35 stats hide-version //隐藏haproxy版本号
36 stats uri /haproxyadmin?stats //一会用于打开状态页的uri
37 stats realm Haproxy\ Statistics //输入账户密码时的提示文字
38 stats auth admin:admin //用户名:密码
39 stats admin if TRUE //开启状态页的管理功能
40 
41 frontend main 
42 bind *:80 
43 default_backend mysql //后端服务器组名
44 
45 backend mysql
46 balance leastconn //使用最少连接方式调度
47 server m1 192.168.226.134:80 check port 80 maxconn 300
48 server m2 192.168.226.135:80 check port 80 maxconn 300
 



启动日志:

 
1 vi /etc/rsyslog.conf
2 
3 # Provides UDP syslog reception //去掉下面两行注释,开启UDP监听
4 $ModLoad imudp
5 $UDPServerRun 514
6 
7 local2.* /var/log/haproxy.log //添加此行
8 
9 service rsyslog restart
 

 

启动测试haproxy:

1 service haproxy start
2 netstat -tnlp

 

四、安装keepalived (两台服务器都要执行)

1 yum install -y keepalived
2 
3 chkconfig --add keepalived
4 chkconfig keepalived on
5 
6 mv /etc/keepalived/keepalived.conf /etc/keepalived/keepalived.conf.bak
7 vi /etc/keepalived/keepalived.conf

 

粘贴如下内容

 
 1 ! Configuration File for keepalived
 2 global_defs {
 3 router_id Mysql_ha
 4 }
 5 
 6 vrrp_script chk_mysql {
 7 script "/etc/keepalived/check_mysql.sh"
 8 interval 2
 9 weight 5
10 }
11 
12 vrrp_script chk_haproxy {
13 script "/etc/keepalived/chk.sh"
14 interval 2
15 weight 5
16 }
17 
18 vrrp_instance mysql-instance {
19 state MASTER #另一台为BACKUP
20 interface enp4s0 #与网卡名称对应
21 virtual_router_id 11 #每一个IP唯一,另一台绑定相同IP要与整个ID一致
22 priority 10 #另一台为9
23 advert_int 1
24 authentication {
25 auth_type PASS
26 auth_pass password321
27 }
28 track_script {
29 chk_mysql
30 }
31 virtual_ipaddress {
32 192.168.226.150/24
33 }
34 }
35 
36 
37 vrrp_instance mysql-ha {
38 state MASTER
39 interface enp4s0
40 virtual_router_id 13
41 priority 10
42 advert_int 1
43 # nopreempt
44 authentication {
45 auth_type PASS
46 auth_pass password321
47 }
48 track_script {
49 chk_haproxy
50 }
51 virtual_ipaddress {
52 192.168.226.14/24
53 }
54 notify_backup "/etc/init.d/haproxy restart"
55 notify_fault "/etc/init.d/haproxy stop"
56 }
 



在两台机器上创建chk.sh文件:

 
 1 vi /etc/keepalived/chk.sh
 2 
 3 #!/bin/bash
 4 #
 5 if [ $(ps -C haproxy --no-header | wc -l) -eq 0 ]; then
 6 /etc/init.d/keepalived stop
 7 fi
 8 
 9 chmod +x /etc/keepalived/chk.sh
10 
11 vi /etc/keepalived/check_mysql.sh
12 #!/bin/bash
13 MYSQL_PORT=:3306
14 alive=$(netstat -apn |grep $MYSQL_PORT|grep LISTEN)
15 if [ -z "$alive" ] 
16 then
17 exit 1
18 else
19 exit 0
20 fi

 

五、在两台机器上测试
.在浏览器打开http://192.168.226.150:6677/haproxyadmin?stats,打开haproxy状态页:

技术图片

 

技术图片

 

mysql复制+keepalived+haproxy配置(负载均衡)

标签:tco   name   绑定   tcp   address   后端   realm   isp   master   

原文地址:https://www.cnblogs.com/ataoxz/p/13725089.html

(0)
(0)
   
举报
评论 一句话评论(0
登录后才能评论!
© 2014 mamicode.com 版权所有  联系我们:gaon5@hotmail.com
迷上了代码!