当前位置：首页 > 系统 > nginx keepalived双网卡双网段热备配置

nginx keepalived双网卡双网段热备配置

系统2年前 (2021-08-23)

nginx keepalived双网卡双网段热备配置

环境介绍

【web1】

ens33:192.168.3.220  ens37:172.16.1.220

【web2】

ens33:192.168.3.230  ens37:172.16.1.230

【VIP】

vip1:192.168.3.240   vip2:172.16.1.240


修改计算机名

hostnamectl --static set-hostname web1 && su
hostnamectl --static set-hostname web2 && su


关闭selinux和firewalld

setenforce 0 && sed -i 's/^SELINUX=enforcing$/SELINUX=permissive/' /etc/selinux/config
systemctl stop firewalld && systemctl disable firewalld && systemctl status firewalld


更换源

mv /etc/yum.repos.d/* /tmp && curl -o /etc/yum.repos.d/CentOS-Base.repo https://mirrors.aliyun.com/repo/Centos-7.repo && curl -o /etc/yum.repos.d/epel.repo http://mirrors.aliyun.com/repo/epel-7.repo && yum makecache


安装nginx和keepalived

rpm -Uvh http://nginx.org/packages/centos/7/noarch/RPMS/nginx-release-centos-7-0.el7.ngx.noarch.rpm
yum install nginx keepalived -y 
systemctl restart nginx && systemctl enable nginx


生成测试页面

echo 'web1-192.168.3.220' > /usr/share/nginx/html/index.html
echo 'web2-192.168.3.230' > /usr/share/nginx/html/index.html


添加nginx检测脚本

检测nginx进程是否存在，若不存在则重启，若重启失败则直接杀掉keepalived进程触发切换

vim /etc/keepalived/nginx_check.sh
添加
#!/bin/bash
A=`ps -C nginx --no-header |wc -l`
if [ $A -eq 0 ];then
      /usr/local/nginx/sbin/nginx
      if [ `ps -C nginx --no-header |wc -l` -eq 0 ];then
      pkill keep
      fi
fi

执行chmod u+x /etc/keepalived/nginx_check.sh


web1修改keepalived配置

mv /etc/keepalived/keepalived.conf /etc/keepalived/keepalived.conf.bak
vim /etc/keepalived/keepalived.conf
添加
global_defs {
  router_id nginx1
 }

vrrp_script chk_nginx {
        script "/etc/keepalived/nginx_check.sh"
        interval 2
        weight 20
		fall 1
		rise 10
}

vrrp_instance VI_NGINX_1 {
   state MASTER
   interface ens33
   virtual_router_id 51
   priority  100
   advert_int 1
   authentication {
       auth_type PASS
       auth_pass ikingtech
   }

   virtual_ipaddress {
       192.168.3.240
   }
   track_script {
        chk_nginx
    }
}

vrrp_instance VI_NGINX_2 {
   state BACKUP
   interface ens37
   virtual_router_id 52
   priority  90
   advert_int 1
   authentication {
       auth_type PASS
       auth_pass ikingtech
   }
   
   virtual_ipaddress {
       172.16.1.240
   }
   
   track_script {
        chk_nginx
    }
}

#执行systemctl enable keepalived && systemctl start keepalived && systemctl status keepalived

nginx keepalived双网卡双网段热备配置

web2修改keepalived配置

mv /etc/keepalived/keepalived.conf /etc/keepalived/keepalived.conf.bak
vim /etc/keepalived/keepalived.conf
添加
global_defs {
  router_id nginx2
 }

vrrp_script chk_nginx {
  script "/etc/keepalived/nginx_check.sh"
  interval 2
  weight 20
  fall 2
  rise 1
}

vrrp_instance VI_NGINX_1 {
   state BACKUP
   interface ens33
   virtual_router_id 51
   priority  90
   advert_int 1
   authentication {
       auth_type PASS
       auth_pass ikingtech
   }

   virtual_ipaddress {
       192.168.3.240
   }
   
   track_script {
        chk_nginx
    }
}

vrrp_instance VI_NGINX_2 {
   state MASTER
   interface ens37
   virtual_router_id 52
   priority  100
   advert_int 1
   authentication {
       auth_type PASS
       auth_pass ikingtech
   }

   virtual_ipaddress {
       172.16.1.240
   }
   track_script {
        chk_nginx
    }
}

#执行systemctl enable keepalived && systemctl start keepalived && systemctl status keepalived

nginx keepalived双网卡双网段热备配置


测试

搭建好后，正常状态应该为web1负责vip1，web2负责vip2

nginx keepalived双网卡双网段热备配置   nginx keepalived双网卡双网段热备配置


现在将web1关闭web1上的keepalived，触发切换

pkill keep
ps -ef|grep keep

nginx keepalived双网卡双网段热备配置

web1已经将keepailved进程杀死，检查ip已经看不到任何vip


测试访问vip1和vip2，发现web2已经全部接管vip

nginx keepalived双网卡双网段热备配置   nginx keepalived双网卡双网段热备配置


如果nginx为docker模式

[IP规划]nginx1为master nginx2为backup

业务网

nginx1:192.168.0.20
nginx2:192.168.0.21
nginx vip:192.168.0.30

局域网

nginx1:10.211.131.20
nginx2:10.211.131.21
nginx vip:10.211.131.30



[解决双网卡双网关问题]

vim /etc/rc.d/rc.local
route add -net 10.211.131.0 netmask 255.255.255.0 dev ens224
route add -net 10.211.131.0 netmask 255.255.255.0 gw 10.211.131.254
chmod +x /etc/rc.d/rc.local


[换源]

mv /etc/yum.repos.d/* /tmp && curl -o /etc/yum.repos.d/CentOS-Base.repo https://mirrors.aliyun.com/repo/Centos-7.repo && curl -o /etc/yum.repos.d/epel.repo http://mirrors.aliyun.com/repo/epel-7.repo && yum makecache
yum install open-vm-tools wget vim tree -y


[系统配置]

节点1
hostnamectl --static set-hostname nginx1 && su
节点2
hostnamectl --static set-hostname nginx1 && su

systemctl start firewalld && systemctl status firewalld

节点1
firewall-cmd --permanent --add-rich-rule="rule family="ipv4" source address="192.168.0.21" protocol value="vrrp" accept"
firewall-cmd --permanent --add-rich-rule="rule family="ipv4" source address="10.211.131.21" protocol value="vrrp" accept"
firewall-cmd --reload
节点2
firewall-cmd --permanent --add-rich-rule="rule family="ipv4" source address="192.168.0.20" protocol value="vrrp" accept"
firewall-cmd --permanent --add-rich-rule="rule family="ipv4" source address="10.211.131.20" protocol value="vrrp" accept"
firewall-cmd --reload


[nginx配置]

docker pull nginx
docker run --name nginx-container -d nginx:latest
mkdir -p /home/docker/nginx/conf
docker cp nginx-container:/etc/nginx/nginx.conf /home/docker/nginx/conf
docker rm -f nginx-container

vim /root/docker_nginx.sh
添加
#!/bin/bash
docker run --name nginx --restart=always -p 80:80 \
    -v /home/docker/nginx/conf/nginx.conf:/etc/nginx/nginx.conf:ro \
    -v /home/docker/nginx/conf/conf.d:/etc/nginx/conf.d \
    -v /home/docker/nginx/html:/usr/share/nginx/html \
    -v /home/docker/nginx/logs:/var/log/nginx \
    -d nginx:latest

执行chmod +x docker_nginx.sh    
systemctl restart docker && sh /root/docker_nginx.sh
#如果挂载错误，删除容器后重新执行运行
#docker container rm nginx
#docker container ls -all

节点1
echo 'web1-192.168.0.20' > /home/docker/nginx/html/index.html
节点2
echo 'web2-192.168.0.21' > /home/docker/nginx/html/index.html

vim /home/docker/nginx/conf/nginx.conf
添加
server {
        listen       80;
        server_name  localhosttest.com;
        root         /usr/share/nginx/html;
        #charset koi8-r;
        #access_log  logs/host.access.log  main;
        location / {
           # root   html;
            index  index.html index.htm;
        }
        error_page  404              /404.html;
        error_page   500 502 503 504  /50x.html;
        location = /50x.html {
            #root   html;
        }
        }
执行docker restart nginx
#docker exec -it nginx bash


[安装keepalived]

yum install keepalived -y
mv /etc/keepalived/keepalived.conf /etc/keepalived/keepalived.conf.bak
vim /etc/keepalived/nginx_pid.sh
添加
#!/bin/bash
A=`docker inspect --format '{{.State.Running}}' nginx`
if [ "${A}" != "true" ]; then
     docker restart nginx
      sleep 3
            if [ `docker inspect --format '{{.State.Running}}' nginx` != "true" ]; then
                  systemctl stop keepalived
fi 
fi
执行chmod +x /etc/keepalived/nginx_pid.sh

节点1
vim /etc/keepalived/keepalived.conf
改为
global_defs {
   router_id nginx1
}

vrrp_script chk_nginx {
    script "/etc/keepalived/nginx_pid.sh"
    interval 2
    weight 20
}

vrrp_instance VI_NGINX_1 {
    state MASTER
    interface ens192
    virtual_router_id 51
    priority 100
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass ikingtech
    }
    virtual_ipaddress {
        192.168.0.30
    }
    track_script {
        chk_nginx
    }
}

vrrp_instance VI_NGINX_2 {
    state MASTER
    interface ens224
    virtual_router_id 52
    priority 100
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass ikingtech
    }
    virtual_ipaddress {
        10.221.131.30
    }
    track_script {
        chk_nginx
    }
}

节点2
global_defs {
   router_id nginx2
   }

vrrp_script chk_nginx {
    script "/etc/keepalived/nginx_pid.sh"
    interval 2
    weight 20
}

vrrp_instance VI_NGINX_1 {
    state BACKUP
    interface ens192
    virtual_router_id 51
    priority 90
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass ikingtech
    }
    virtual_ipaddress {
        192.168.0.30
    }
    track_script {
        chk_nginx
    }
}

vrrp_instance VI_NGINX_2 {
    state BACKUP
    interface ens224
    virtual_router_id 52
    priority 90
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass ikingtech
    }
    virtual_ipaddress {
        10.221.131.30
    }
    track_script {
        chk_nginx
    }
}

执行systemctl enable keepalived && systemctl start keepalived && systemctl status keepalived


[测试]

docker stop nginx


本站所有文章均可随意转载，转载时请保留原文链接及作者。

本文链接：https://www.vos.cn/os/503.html

返回列表

上一篇：删除回收站右键菜单的固定到"开始"屏幕

下一篇：彻底解决oracle安装流程无法x11弹出窗口问题

相关文章

台式机使用u盘安装chromeos第三方cloudready

台式机使用u盘安装chromeos第三方cloudready

 首先准备好一个8G或者16G的u盘https://www.neverware.com/freedownload下载32位或者64位的系统将下载好的文件解压出bin格式的文件本地需要安装ch...

CentOS7 Samba配置

CentOS7 Samba配置

在RAID的基础上搭建samba服务提供文件共享参数介绍[global] workgroup = 工作组名称 server string = 服务...

CentOS7 NFS配置

CentOS7 NFS配置

网络文件系统(Network File System)是一种将远程主机上的分区(目录)经网络挂载到本地的一种机制，通过对网络文件系统的支持，用户可以在本地系统上像操作本地分区一样来对远程主机的共享分区...

Redhat 6.5网卡聚合绑定

Redhat 6.5网卡聚合绑定

linux下网卡bonding配置是通过多张网卡绑定为一个逻辑网卡，实现本地网卡的冗余，带宽扩容和负载均衡，在生产场景中是一种常用的技术，常用聚合模式为0，1，60(balance-rr)轮循(RR)...

Red Hat Enterprise Linux 6.5安装和分区

Red Hat Enterprise Linux 6.5安装和分区

系统镜像名称为rhel-server-6.5-x86_64-dvd.iso安装Install or upgrade an existing system跳过安装源测试检查Next语言English键盘...

EVE-NG配置静态固定地址

EVE-NG配置静态固定地址

sudo vi /etc/network/interfaces# The primary network interfaceiface eth0 inet manual ...

陕ICP备19010041号