集群概述、keepalived介绍、用keepalived配置高可用集群

2019独角兽企业重金招聘Python工程师标准>>>

根据功能划分为两大类&＃xff1a;高可用和负载均衡高可用集群通常为两台服务器&＃xff0c;一台工作&＃xff0c;另外一台作为冗余&＃xff0c;当提供服务的机器宕机&＃xff0c;冗余将接替继续提供服务实现高可用的开源软件有&＃xff1a;heartbeat、keepalived 负载均衡集群&＃xff0c;需要有一台服务器作为分发器&＃xff0c;它负责把用户的请求分发给后端的服务器处理&＃xff0c;在这个集群里&＃xff0c;除了分发器外&＃xff0c;就是给用户提供服务的服务器了&＃xff0c;这些服务器数量至少为2 实现负载均衡的开源软件有LVS、keepalived、haproxy、nginx&＃xff0c;商业的有F5、Netscaler

Keepalive介绍&＃xff08;HA集群&＃xff09;HA即高可用&＃xff0c;又叫做双机热备&＃xff0c;用于关键性的业务&＃xff0c;简单理解就是有两台机器A和B&＃xff0c;正常是A提供服务&＃xff0c;B待命闲置&＃xff0c;当A宕机或者服务宕掉&＃xff0c;会切换至B机器继续提供服务。在这里我们使用keepalived来实现高可用集群&＃xff0c;因为heartbeat在centos6上有一些问题&＃xff0c;影响实验效果 keepalived通过VRRP&＃xff08;Virtual Router Redundancy Protocl&＃xff09;来实现高可用。在这个协议里会将多台功能相同的路由器组成一个小组&＃xff0c;这个小组里会有1个master角色和N&＃xff08;N>&＃61;1&＃xff09;个backup角色。 master会通过组播的形式向各个backup发送VRRP协议的数据包&＃xff0c;当backup收不到master发来的VRRP数据包时&＃xff0c;就会认为master宕机了。此时就需要根据各个backup的优先级来决定谁成为新的mater。 Keepalived要有三个模块&＃xff0c;分别是core、check和vrrp。其中core模块为keepalived的核心&＃xff0c;负责主进程的启动、维护以及全局配置文件的加载和解析&＃xff0c;check模块负责健康检查&＃xff0c;vrrp模块是来实现VRRP协议的。

用keepalived配置高可用集群

keepalived下需要有一个服务来实现高可用&＃xff0c;用nginx来代替实现服务

准备两台机器130和132&＃xff0c;130作为master&＃xff0c;132作为backup 两台机器都执行yum install -y keepalived 两台机器都安装nginx&＃xff0c;其中130上已经编译安装过nginx&＃xff0c;132上需要yum安装nginx: yum install -y nginx

设定vip为100

#vim /etc/keepalived/keepalived.conf 编辑130上keepalived配置文件&＃xff0c;内容从https://coding.net/u/aminglinux/p/aminglinux-book/git/blob/master/D21Z/master_keepalived.conf获取先清空原先配置文件的内容&＃xff0c;用快捷键>!$&＃xff0c;然后再将复制的文件放到#/etc/keepalived/keepalived.conf下

global_defs { notification_email { //可以不用定义 aming&＃64;aminglinux.com } notification_email_from root&＃64;aminglinux.com smtp_server 127.0.0.1 smtp_connect_timeout 30 router_id LVS_DEVEL } vrrp_script chk_nginx { //检测服务是否正常 script "/usr/local/sbin/check_ng.sh" //定义脚本路径 interval 3 } vrrp_instance VI_1 { state MASTER interface ens33 virtual_router_id 51 //主从id一致 priority 100 //权重 advert_int 1 authentication { auth_type PASS auth_pass aminglinux>com } virtual_ipaddress { 192.168.37.100 } track_script { chk_nginx } }

#vim /usr/local/sbin/check_ng.sh 130编辑监控脚本&＃xff0c;内容从https://coding.net/u/aminglinux/p/aminglinux-book/git/blob/master/D21Z/master_check_ng.sh获取

#!/bin/bash #时间变量&＃xff0c;用于记录日志 d&＃61;&＃96;date --date today &＃43;%Y%m%d_%H:%M:%S&＃96; #计算nginx进程数量 n&＃61;&＃96;ps -C nginx --no-heading|wc -l&＃96; #如果进程为0&＃xff0c;则启动nginx&＃xff0c;并且再次检测nginx进程数量&＃xff0c; #如果还为0&＃xff0c;说明nginx无法启动&＃xff0c;此时需要关闭keepalived if [ $n -eq "0" ]; then /etc/init.d/nginx start n2&＃61;&＃96;ps -C nginx --no-heading|wc -l&＃96; if [ $n2 -eq "0" ]; then echo "$d nginx down,keepalived will stop" >> /var/log/check_ng.log systemctl stop keepalived fi fi

#chmod 755 /usr/local/sbin/check_ng.sh //给脚本755权限

#systemctl start keepalived //master上启动keepalived服务

132上#vim /etc/keepalived/keepalived.conf 编辑配置文件&＃xff0c;内容从https://coding.net/u/aminglinux/p/aminglinux-book/git/blob/master/D21Z/backup_keepalived.conf获取

global_defs { notification_email { aming&＃64;aminglinux.com } notification_email_from root&＃64;aminglinux.com smtp_server 127.0.0.1 smtp_connect_timeout 30 router_id LVS_DEVEL } vrrp_script chk_nginx { script "/usr/local/sbin/check_ng.sh" interval 3 } vrrp_instance VI_1 { state BACKUP interface ens33 virtual_router_id 51 priority 90 //权重 advert_int 1 authentication { auth_type PASS auth_pass aminglinux>com } virtual_ipaddress { 192.168.37.100 } track_script { chk_nginx } }

132上编辑监控脚本&＃xff0c;内容从https://coding.net/u/aminglinux/p/aminglinux-book/git/blob/master/D21Z/backup_check_ng.sh获取

#时间变量&＃xff0c;用于记录日志 d&＃61;&＃96;date --date today &＃43;%Y%m%d_%H:%M:%S&＃96; #计算nginx进程数量 n&＃61;&＃96;ps -C nginx --no-heading|wc -l&＃96; #如果进程为0&＃xff0c;则启动nginx&＃xff0c;并且再次检测nginx进程数量&＃xff0c; #如果还为0&＃xff0c;说明nginx无法启动&＃xff0c;此时需要关闭keepalived if [ $n -eq "0" ]; then systemctl start nginx n2&＃61;&＃96;ps -C nginx --no-heading|wc -l&＃96; if [ $n2 -eq "0" ]; then echo "$d nginx down,keepalived will stop" >> /var/log/check_ng.log systemctl stop keepalived fi fi

#chmod 755 /usr/local/sbin/check_ng.sh //给脚本755权限

132上也启动服务 #systemctl start keepalived

模拟之前先要关闭防火墙和selinux&＃xff0c; 模拟生产环境中宕机的情况&＃xff0c;测试高可用&＃xff1a; 最好的手段是关闭主上的keepalived服务来测试高可用