实验目的:
使用Keepalive来保证Nginx的高可用,进而实现网站的高可用!
实验拓扑:
实验环境:
如上拓扑图所示,master与slave是mysql主从,且有discuz论坛,且已经做高可用。(可参加《构建高可用服务器之三Keepalive冗余Mysql》)。现在是刚搭建两台Nginx,均已安装keepalive,IP地址分别为192.168.1.2和192.168.1.5(可参加《构建Nginx服务器之一安装及虚拟主机配置》和《构建高可用服务器之一Keepalive介绍及安装》)。
一、配置主机Nginx(192.168.1.2)
nginx服务的配置文件如下:
[root@nginx ~]# cat/usr/local/nginx/conf/nginx.conf worker_processes 1; events { worker_connections 1024; } http { include mime.types; default_type application/octet-stream; sendfile on; keepalive_timeout 65; ############################################################################### upstream discuz { server 192.168.1.3:80 weight=1max_fails=2 fail_timeout=30s; server 192.168.1.4:80 weight=1max_fails=2 fail_timeout=30s; } server { listen 80; server_name localhost; location/ { root html; index index.html index.htm; proxy_set_header Host $host; proxy_set_header X-Real-IP$remote_addr; proxy_set_headerX-Forwarded-For $proxy_add_x_forwarded_for; proxy_pass http://discuz; } ################################动静分离####################################### location~ .*\.(php|jsp|cgi|shtml)?$ { proxy_set_header Host $host; proxy_set_header X-Real-IP$remote_addr; proxy_set_headerX-Forwarded-For $proxy_add_x_forwarded_for; proxy_pass http://discuz; } location ~ .*\.(html|htm|gif|jpg|jpeg|bmp|png|ico|txt|js|css)$ { root/usr/local/nginx/html; expires 30d; } ####################################################################### error_page 500 502 503 504 /50x.html; location = /50x.html { root html; } } } [root@nginx ~]#
keepaive配置文件如下:
[root@nginx ~]# cat/etc/keepalived/keepalived.conf ! Configuration File for keepalived global_defs { notification_email { [email protected] } notification_email_from [email protected] smtp_server 127.0.0.1 smtp_connect_timeout 30 router_id LVS_DEVEL } vrrp_script chk_nginx { script"/data/sh/check_nginx.sh" interval2 weight2 } vrrp_instance VI_1 { state BACKUP interface eth0 virtual_router_id 251 priority 100 mcast_src_ip 192.168.1.2 advert_int 1 nopreempt authentication { auth_type PASS auth_pass 1111 } virtual_ipaddress { 192.168.1.88 } track_script{ chk_nginx } }
主机nginx上keepalive脚本文件:
[root@nginx ~]# mkdir -p /data/sh [root@nginx ~]# touch check_nginx.sh [root@nginx ~]# chmod +x check_nginx.sh [root@nginx ~]# vim check_nginx.sh [root@nginx ~]# mv check_nginx.sh /data/sh/ [root@nginx ~]# cat /data/sh/check_nginx.sh #!/bin/bash #Target: If Nginx Down, Stop keepalive! # killall -0 nginx if [[ ! $? -eq 0 ]];then /etc/init.d/keepalived stop fi [root@nginx ~]#
加载模块
[root@nginx ~]# modprobe ip_vs [root@nginx ~]# modprobe ip_vs_rr [root@nginx ~]# modprobe ip_vs_wrr
启动服务
[root@nginx ~]# /etc/init.d/keepalivedrestart Stopping keepalived: [FAILED] Starting keepalived: [ OK ] [root@nginx ~]# /usr/local/nginx/sbin/nginx [root@nginx ~]#
二、配置主机nginx2(192.168.1.5)
关于nginx服务配置文件:
与nginx主机(192.168.1.2完全一致)
关于keepalive
只需要把配置文件keepalived.conf中的mcast_src_ip 192.168.1.2改为mcast_src_ip192.168.1.5
关于keepalived的脚步配置文件
与nginx主机(192.168.1.2完全一致)
启动nginx与keepalive服务
三、测试
可以通过Nginx上的VIP 192.168.1.88访问
停止nginx主机上的nginx服务,VIP自动漂移到nginx2上,论坛已经可以访问;
[root@nginx ~]# ps -ef | grep nginx root 2067 1 0 21:31 ? 00:00:00 nginx: master process/usr/local/nginx/sbin/nginx nobody 2070 2067 0 21:31 ? 00:00:00 nginx: worker process root 6302 1946 0 22:24 pts/0 00:00:00 grep nginx [root@nginx ~]# /etc/init.d/keepalivedstatus keepalived (pid 4430) is running... [root@nginx ~]# pkill nginx [root@nginx ~]# /etc/init.d/keepalivedstatus keepalived is stopped [root@nginx ~]# tail /var/log/messages May 21 22:18:25 nginx NetworkManager[1517]:<info> prefix 24 (255.255.255.0) May 21 22:18:25 nginx NetworkManager[1517]:<info> gateway 192.168.33.2 May 21 22:18:25 nginx NetworkManager[1517]:<info> nameserver '192.168.33.2' May 21 22:18:25 nginx NetworkManager[1517]:<info> domain name 'localdomain' May 21 22:24:20 nginx Keepalived:Terminating on signal May 21 22:24:20 nginx Keepalived_vrrp:Terminating VRRP child process on signal May 21 22:24:20 nginx Keepalived_vrrp:VRRP_Instance(VI_1) removing protocol VIPs. May 21 22:24:20 nginxKeepalived_healthcheckers: Netlink reflector reports IP 192.168.1.88 removed May 21 22:24:20 nginxKeepalived_healthcheckers: Terminating Healthchecker child process on signal May 21 22:24:20 nginx Keepalived: StoppingKeepalived v1.1.15 (05/21,2015) [root@nginx ~]#
此时VIP在nginx2主机上(192.168.1.5)
论坛依旧可以访问,Nginx高可用实现成功。
四、总结
1、 selinux与iptables注意要配置或关闭及清空。
2、 注意加载ip_vs、ip_vs_rr、ip_vs_wrr模块.(根据需要加载即可)。
3、 Nginx做的有动静分离,那么两台nginx上的网站发布目录要和后台apache网站发布目录内容完全一致。
4、编译安装keepalive的时候,指定的内核一定要与本机内核一致。
5、同一组keepalive的virtual_router_id必须一致,同一局域网内有多组keepalive的时候,virtual_router_id不能一致。
(这个拓扑图中,Mysql那一组keepalive的virtual_router_id一致均为51,Nginx那一组keepalive的virtual_router_id一致均为251,因为mysql与nginx在同一局域网内,所以这两组keepalive的virtual_router_id一定不能一样。)
6、 当keepalive中开启nopreempt的时候,两台keepalive的state均为BACKUP
(nopreempt这个选项为非抢占,忽略优先级的存在,当nginx主机宕机的时候,VIP会飘到nginx2主机上,并且即使nginx主机恢复正常,VIP依旧指向nginx2,这样可以减少短暂服务中断的次数,2次变1次)