Centos7编译安装Nginx+keepalived

1、安装环境、主机信息及软件版本

Nginx：1.12.2
keepalived:2.0.12
时间同步(同步后确认各服务器时间是否一致，不一致须要修改一下时区)
关闭防火墙

2、编译安装Nginx

 1.编译安装Nginx

[root@k8s-node-207 ~]# wget http://nginx.org/download/nginx-1.12.2.tar.gz
[root@k8s-node-207 ~]# yum -y install gcc gcc-c++ automake pcre pcre-devel zlib zlib-devel open openssl-devel #安装相关依赖包 [root@k8s-node-207 ~]# tar xf nginx-1.12.2.tar.gz [root@k8s-node-207 ~]# cd nginx-1.12.2/ [root@k8s-node-207 nginx-1.12.2]# ./configure  --prefix=/usr/local/nginx --with-http_ssl_module --with-http_stub_status_module --with-http_realip_module --with-threads  --with-pcre  --with-http_gzip_static_module #设置编译参数，特别是--prefix [root@k8s-node-207 nginx-1.12.2]# make && make install [root@k8s-node-207 nginx-1.12.2]# scp -r /usr/local/nginx  172.16.155.208:/usr/local/  #拷贝到另一台Nginx服务器

2.配置Nginx为系统服务

[root@k8s-node-207 nginx-1.12.2]# vim /lib/systemd/system/nginx.service  #建立Nginx服务系统启动文件 [Unit] Description=nginx After=network.target [Service] Type=forking ExecStart=/usr/local/nginx/sbin/nginx  #注意要替换为本身编译安装的路径 ExecReload=/usr/local/nginx/sbin/nginx -s reload ExecStop=/usr/local/nginx/sbin/nginx -s quit PrivateTmp=true [Install] WantedBy=multi-user.target
 [root@k8s-node-207 nginx-1.12.2]# systemctl start nginx #测试脚本，启动 [root@k8s-node-207 nginx-1.12.2]# netstat -tnlp|grep :80 #检查端口 tcp 0      0 0.0.0.0:80              0.0.0.0:*               LISTEN      7834/nginx: master [root@k8s-node-207 nginx-1.12.2]# ps aux|grep nginx #检查进程 root 7834  0.0  0.0  45956  1124 ?        Ss   16:24   0:00 nginx: master process /usr/local/nginx/sbin/nginx nobody 7835  0.0  0.0  48484  1976 ?        S    16:24   0:00 nginx: worker process root 7944  0.0  0.0 112720   968 pts/0    S+   16:25   0:00 grep -E --color=auto nginx [root@k8s-node-207 nginx-1.12.2]# systemctl enable nginx  #设置开机自启 [root@k8s-node-207 nginx-1.12.2]# scp /lib/systemd/system/nginx.service 172.16.155.208:/lib/systemd/system/nginx.service  #拷贝到另外一台机器
#在另外一台机器上启动Nginx [root@k8s-node-208 ~]# systemctl start nginx [root@k8s-node-208 ~]# systemctl status nginx [root@k8s-node-208 ~]# systemctl enable nginx

3、编译安装配置keepalived

1.编译安装keepalived

[root@k8s-node-207 ~]# wget https://www.keepalived.org/software/keepalived-2.0.12.tar.gz #下载源码
[root@k8s-node-207 ~]# yum install -y openssl openssl-devel libnl libnl-devel  #安装依赖文件 [root@k8s-node-207 ~]# tar xf keepalived-2.0.12.tar.gz [root@k8s-node-207 ~]# cd keepalived-2.0.12/ [root@k8s-node-207 keepalived-2.0.12]# ./configure --prefix=/usr/local/keepalived #只有编译安装的目录 [root@k8s-node-207 keepalived-2.0.12]# make && make install

2.配置keepalived master节点

[root@k8s-node-207 keepalived-2.0.12]# cd /usr/local/keepalived/etc/keepalived/  #进入keepalived配置目录 [root@k8s-node-207 keepalived]# cp keepalived.conf keepalived.conf-$(date +%F-%H:%M:%S)  #拷贝默认的配置 ! Configuration File for keepalived global_defs { #默认该字段是配置发送邮件通知，因为我使用微信进行通知因此忽略 } vrrp_script chk_http_port { #配置服务的健康检查   script "/mnt/chk_nginx.sh" #检查是使用的脚本路径 interval 2 #监控间隔   weight -5 #每检测失败一次，若是weight大于0则当前节点的priority增长该配置的值，不然减小 fall 2 #执行几回才会认为是失败 rise 1 #执行多少次才会认为是成功   } vrrp_instance VI_1 { state MASTER  #指定当前节点的初始状态 interface eth0 #vrrp实例绑定的网卡接口 用于发送vrrp包  
 nopreempt #设置为非抢占模式，优先级高的设置 解决优先级高的恢复后再次抢断 测试时能够先不配置 virtual_router_id 51 #指定vrrp实例的ID 范围是0-255 主备节点必须一致  priority 101 #指定当前节点的优先级 优先级高的为MASTER  advert_int 1 #指定发送vrrp间隔时间 主备必须一致  authentication {  #主备必须一致   auth_type PASS #指定认证方式 这里使用简单密码认证   auth_pass 1111 #指定认证使用的密码 最大为8位 } virtual_ipaddress { 172.16.155.209 #指定VIP地址 } notify_master "/root/script_dir/wechat.py master test keepalived状态发送改变,master切换至172.16.155.207" #设置通知脚本路径及通知信息 此处根据本身实际状况自定义，此处配置不影响启动，若是没有告警通知能够稍后建立 notify_backup "/root/script_dir/wechat.py backup test keepalived状态发送改变,backup切换至172.16.168.207"  notify_fault "/root/script_dir/wechat.py fault test keepalived发送故障,故障主机为：172.16.168.207" #以上配置为：当节点成为master时执行的操做 成为backup是执行的操做 当发生故障时执行的操做 track_script { #监控脚本执行的状态  chk_http_port } } [root@k8s-node-207 keepalived]# cd /usr/local/keepalived/etc/sysconfig/  #因为启动命令默认会去/etc/keepalived/下读取keepalived.conf，而且默认日志会写入/var/log/messages文件中，因此咱们须要修改相关配置 [root@k8s-node-207 sysconfig]# cat keepalived KEEPALIVED_OPTIONS="-f /usr/local/keepalived/etc/keepalived/keepalived.conf -S 0 -D" #-f 指定配置文件路径 -S 指定日志路径 0 表示local0.*  [root@k8s-node-207 sysconfig]# cat /etc/rsyslog.conf #修改rsyslog配置文件 #Save keepalived log local0.*                                                /var/log/keepalived.log  #指定日志文件路径
#暂时不拷贝至另外一台服务器

 3.配置keepalived backup节点

#207上拷贝相关文件到208上
[root@k8s-node-207 sysconfig]# scp -r /usr/local/keepalived/ 172.16.155.208:/usr/local/   [root@k8s-node-207 sysconfig]# scp /etc/rsyslog.conf 172.16.155.208:/etc/rsyslog.conf 
[root@k8s-node-207 sysconfig]# scp /lib/systemd/system/keepalived.service 172.16.155.208:/lib/systemd/system/keepalived.service #默认编译keepalived时会自动生成系统服务配置文件
[root@k8s-node-207 sysconfig]# systemctl restart rsyslog.service 

#208上修改配置文件为backup [root@k8s-node-208 ~]# cd /usr/local/keepalived/etc/keepalived/ [root@k8s-node-208 keepalived]# vim keepalived.conf ! Configuration File for keepalived global_defs { #可自定义 } vrrp_script chk_http_port {  #可自定义 script "/mnt/chk_nginx.sh" interval 2 weight -5 fall 2 rise 1 } vrrp_instance VI_1 { state backup  #指定当前节点状态 interface eth0 nopreempt virtual_router_id 51 #确保与master保持一致 priority 100 #确保小于master设置的值 advert_int 1 #确保与master保持一致 authentication { #确保与master保持一致 auth_type PASS auth_pass 1111 } virtual_ipaddress { #确保与master保持一致 172.16.155.209 } notify_master "/root/script_dir/wechat.py master test keepalived状态发送改变,master切换至172.16.155.208"  #可自定义 notify_backup "/root/script_dir/wechat.py backup test keepalived状态发送改变,backup切换至172.16.168.208" notify_fault "/root/script_dir/wechat.py fault test keepalived发送故障,故障主机为：172.16.168.208" track_script { #可自定义 chk_http_port } }

默认编译后会自动生成系统服务配置文件，若是没有参考如下配置建立

[root@k8s-node-207 sysconfig]# vim /lib/systemd/system/keepalived.service [Unit] Description=LVS and VRRP High Availability Monitor After= network-online.target syslog.target Wants=network-online.target [Service] Type=forking PIDFile=/var/run/keepalived.pid KillMode=process EnvironmentFile=-/usr/local/keepalived/etc/sysconfig/keepalived ExecStart=/usr/local/keepalived/sbin/keepalived $KEEPALIVED_OPTIONS ExecReload=/bin/kill -HUP $MAINPID [Install] WantedBy=multi-user.target

建立Nginx状态检查脚本

[root@k8s-node-207 sysconfig]# cat /mnt/chk_nginx.sh #!/bin/bash ngix_status=`ps -C nginx --no-header |wc -l` if [[ ${ngix_status} -eq 0 ]];then /usr/local/nginx/sbin/nginx sleep 2 new_nginx_status=$(ps -C nginx --no-header |wc -l) if [[ ${new_nginx_status} -eq 0 ]];then killall keepalived fi fi [root@k8s-node-207 sysconfig]# scp /mnt/chk_nginx.sh 172.16.155.208:/mnt/chk_nginx.sh [root@k8s-node-207 sysconfig]# chmod +x /mnt/chk_nginx.sh [root@k8s-node-207 sysconfig]# ssh 172.16.155.208 "chmod +x /mnt/chk_nginx.sh"

4.启动服务并检查当前状态

[root@k8s-node-207 sysconfig]# systemctl restart rsyslog.service #重启rsyslog服务，使keepalived的日志配置生效 [root@k8s-node-207 sysconfig]# echo "k8s-node-207" > /usr/local/nginx/html/index.html  #修改Nginx index文件 稍后测试使用 [root@k8s-node-207 sysconfig]# systemctl start keepalived.service #启动keepalived服务 #208上 同207操做 [root@k8s-node-208 keepalived]# systemctl restart rsyslog.service [root@k8s-node-208 keepalived]#  echo "k8s-node-208" > /usr/local/nginx/html/index.html [root@k8s-node-208 keepalived]# systemctl start keepalived.service 
 [root@k8s-node-207 sysconfig]# ip a|grep 172.16  #207上检查当前的IP地址 inet 172.16.155.207/24 brd 172.16.155.255 scope global eth0 inet 172.16.155.209/32 scope global eth0  #VIP地址当前绑定在当前主机eth0网卡上 [root@k8s-node-208 keepalived]# ip a|grep 172.16  #208上检查当前用有的IP地址 主要检查是否发生脑裂问题 inet 172.16.155.208/24 brd 172.16.155.255 scope global eth0 [root@k8s-node-207 sysconfig]# curl http://172.16.155.209:80  #使用crul命令范围VIP查看访问到哪一个节点 显示为207节点
k8s-node-207

 5.测试keepalived主从是否能够切换

[root@k8s-node-207 sysconfig]# systemctl stop keepalived.service #在207上关闭keepalived服务 [root@k8s-node-207 sysconfig]# ip a|grep 172.16 #查看207上的拥有的IP地址 VIP已经没有了 inet 172.16.155.207/24 brd 172.16.155.255 scope global eth0 [root@k8s-node-208 keepalived]# ip a|grep 172.16 #查看208上的拥有的IP地址 VIP已经漂移到208上了 inet 172.16.155.208/24 brd 172.16.155.255 scope global eth0 inet 172.16.155.209/32 scope global eth0 [root@k8s-node-207 sysconfig]# curl http://172.16.155.209:80 #访问一下VIP 显示为208节点
k8s-node-208 [root@k8s-node-207 sysconfig]# systemctl start keepalived.service #207上从新启动keepalived [root@k8s-node-207 sysconfig]# ip a|grep 172.16 #VIP没有漂移回来，符合预期，若是想在master恢复后VIP漂移回来则能够删除nopreempt配置 inet 172.16.155.207/24 brd 172.16.155.255 scope global eth0

5.测试Nginx服务故障时keepalived是否能够切换

[root@k8s-node-208 mnt]# ip a|grep 172.16  #当前VIP在208上 inet 172.16.155.208/24 brd 172.16.155.255 scope global eth0 inet 172.16.155.209/32 scope global eth0 [root@k8s-node-208 mnt]# systemctl stop nginx #关闭208上Nginx服务 稍等几秒钟 [root@k8s-node-208 mnt]# ip a|grep 172.16 #再次检查208上的IP，VIP已经没有了 inet 172.16.155.208/24 brd 172.16.155.255 scope global eth0
 [root@k8s-node-207 mnt]# ip a|grep 172.16  #207上查看，VIP已经漂移到207上了 inet 172.16.155.207/24 brd 172.16.155.255 scope global eth0 inet 172.16.155.209/32 scope global eth0