Keepalived两个集群实验

LVS+Keepalived高可用高并发集群

• 两台Keepalived作为Web服务入口

• 在Keepalived中配置LVS

实验：

LVS1	192.168.221.10 ，VIP：192.168.221.100
LVS2	192.168.221.20， VIP：192.168.221.100
RS1	192.168.221.30
RS2	192.168.221.40

实验目标

1、使用LVS DR 模式直接路由模式）+ Keepalived实现高可用调度器。

2、VIP 由 Keepalived 在 LVS1/LVS2 之间漂移。

3、Client 访问 VIP（192.168.221.100），请求被转发到 RS1/RS2。

4、模拟 LVS1 故障时，LVS2 自动接管 VIP。

实验步骤：

（1）在RS1、RS2配置web服务
# 安装httpd/nginx
# sudo apt install -y nginx(Ubuntu)
yum install -y httpd(CentOS)
# 修改首页文件
echo "This is RS1 (192.168.221.30)" > /var/www/html/index.html
echo "This is RS2 (192.168.221.40)" > /var/www/html/index.html
# 启动服务
systemctl start httpd（2）在RS1、RS2上绑定VIP到lo:0
# 在 RS1 / RS2 都执行
ip addr add 192.168.221.100/32 dev lo
ip link set lo up
# 永久关闭 ARP
echo "net.ipv4.conf.all.arp_ignore = 1" | sudo tee -a /etc/sysctl.conf
echo "net.ipv4.conf.all.arp_announce = 2" | sudo tee -a /etc/sysctl.conf
echo "net.ipv4.conf.lo.arp_ignore = 1" | sudo tee -a /etc/sysctl.conf
echo "net.ipv4.conf.lo.arp_announce = 2" | sudo tee -a /etc/sysctl.conf
sysctl -p
# 验证：
ip addr show lo
# 应该能看到 192.168.192.168.221.100/32（3）在 LVS1、LVS2 上安装 Keepalived 和 ipvsadm
yum install -y keepalived ipvsadm
​
（4）LVS1 配置（192.168.221.30——Master）
vim /etc/keepalived/keepalived.conf
++++++++++++++++++++++++++++++++++++++++++++++++++++++++
! Configuration File for keepalived
​
global_defs {router_id lvs_1
}
​
vrrp_instance VI_1 {state MASTERinterface ens33virtual_router_id 51priority 100advert_int 1authentication {auth_type PASSauth_pass 1111}
​virtual_ipaddress {192.168.221.100}
}
​
#下面可选
virtual_server 192.168.221.100 80 {delay_loop 6lb_algo rr #算法轮询lb_kind DR #LVS-DR模式persistence_timeout 50protocol TCP
​real_server 192.168.221.30 80 { #主节点的IPweight 1TCP_CHECK {connect_timeout 3nb_get_retry 3delay_before_retry 3}}
​real_server 192.168.221.40 80 { #从节点的IPweight 1TCP_CHECK {connect_timeout 3nb_get_retry 3delay_before_retry 3}}
}
++++++++++++++++++++++++++++++++++++++++++++++++++++++++
​
（5）LVS2 配置（192.168.221.40——Backup）
vim /etc/keepalived/keepalived.conf
++++++++++++++++++++++++++++++++++++++++++++++++++++++++
! Configuration File for keepalived
​
global_defs {router_id lvs_1
}
​
vrrp_instance VI_1 {state BACKUPinterface ens33virtual_router_id 51priority 90advert_int 1authentication {auth_type PASSauth_pass 1111}
​virtual_ipaddress {192.168.221.100}
}
​
#下面可选
virtual_server 192.168.221.100 80 {delay_loop 6lb_algo rrlb_kind DRpersistence_timeout 50protocol TCP
​real_server 192.168.221.30 80 {weight 1TCP_CHECK {connect_timeout 3nb_get_retry 3delay_before_retry 3}}real_server 192.168.221.40 80 {weight 1TCP_CHECK {connect_timeout 3nb_get_retry 3delay_before_retry 3}}
}
++++++++++++++++++++++++++++++++++++++++++++++++++++++++（6）启动 LVS + Keepalived
systemctl enable --now keepalived
# 查看 LVS 规则：
ipvsadm -Ln
# 应该能看到：
TCP  192.168.221.100:80 rr
-> 192.168.221.30:80
-> 192.168.221.40:80
（7）测试高可用与负载均衡
浏览器输入

测试故障切换

停止 LVS1：
systemctl stop keepalived
- VIP 自动漂移到 LVS2（`ip addr show` 验证）。
- Client 依旧访问 `192.168.221.100`，负载均衡继续生效。

---

搭建Nginx+Keepalived高可用集群

一、环境规划（以 4 台服务器为例）

角色	主机名 / IP	软件	说明
分发层（主）	192.168.221.10	Nginx + Keepalived	承载虚拟 IP，反向代理到后端
分发层（备）	192.168.221.20	Nginx + Keepalived	主节点故障时自动接管 VIP
后端 Web1	192.168.221.30	Nginx（提供 Web 服务）	真实业务节点 1
后端 Web2	192.168.221.40	Nginx（提供 Web 服务）	真实业务节点 2
虚拟 IP（VIP）	192.168.221.100	-	对外提供统一访问入口

二、具体操作

## 基础环境配置（四台虚拟机）
# 关闭防火墙（或放行80/443端口）
systemctl stop firewalld && systemctl disable firewalld
# 关闭SELinux（临时+永久）
setenforce 0
sed -i 's/SELINUX=enforcing/SELINUX=disabled/g' /etc/selinux/config
# 安装依赖（如epel-release）
yum install -y epel-release
# 安装Nginx
yum -y install make zlib zlib-devel gcc-c++ libtool  openssl openssl-devel
# nginx依赖pcre库
wget  http://192.168.57.200/Software/pcre-8.45.tar.bz2
tar -xvf pcre-8.45.tar.bz2
cd pcre-8.45/
./configure
make && make install
# 检验是否安装
pcre-config --version
# 下载编译nginx
cd ..
wget http://192.168.57.200/Software/nginx-1.21.4.tar.gz
tar -xvf nginx-1.21.4.tar.gz
cd nginx-1.21.4
./configure --with-http_stub_status_module --with-http_ssl_module --with-pcre=../pcre-8.45
make && make install
# 安装成功后直接执行nginx启动
/usr/local/nginx/sbin/nginx
#软链接（是为了不用写路径直接执行Nginx命令）
ln -s /usr/local/nginx/sbin/nginx /usr/sbin/nginx
​
# 1. 安装Keepalived（192.168.221.10和20）
yum install -y keepalived
# 编写Nginx健康检查脚本，用于检测Nginx是否存活
vim /etc/keepalived/check_nginx.sh
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
#!/bin/bash
if ! pgrep -x "nginx" > /dev/null; thenexit 1  # Nginx未运行，返回非0表示故障
fi
exit 0    # Nginx运行正常
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
# 赋予执行权限
chmod +x /etc/keepalived/check_nginx.sh
​
# 2. 主节点（192.168.221.10）Keepalived 配置
# 编辑/etc/keepalived/keepalived.conf：
vim /etc/keepalived/keepalived.conf
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
! Configuration File for keepalived
global_defs {router_id nginx_master  # 唯一标识，备节点需不同
}
​
# 定义Nginx健康检查脚本
vrrp_script chk_nginx {script "/etc/keepalived/check_nginx.sh"interval 2  # 检查间隔（秒）weight -50  # 检查失败时优先级减少值fall 2      # 连续2次失败视为故障rise 1      # 1次成功视为恢复
}
​
vrrp_instance VI_1 {state MASTER            # 主节点角色interface ens33         # 网卡名（用`ip addr`确认）virtual_router_id 51    # 虚拟路由ID（主备需相同）priority 100            # 优先级（主节点需高于备节点）advert_int 1            # 心跳间隔（秒）# 认证（主备需相同）authentication {auth_type PASSauth_pass 1111}# 关联健康检查脚本track_script {chk_nginx}# 虚拟IP（VIP）virtual_ipaddress {192.168.221.100/24 dev ens33}
}
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
​
# 3. 备节点（192.168.221.20）Keepalived 配置
vim /etc/keepalived/keepalived.conf
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
! Configuration File for keepalived
global_defs {router_id nginx_backup  # 唯一标识，与主节点不同
}
​
# 定义Nginx健康检查脚本（与主节点一致）
vrrp_script chk_nginx {script "/etc/keepalived/check_nginx.sh"interval 2weight -50fall 2rise 1
}
​
vrrp_instance VI_1 {state BACKUP            # 备节点角色interface ens33         # 网卡名（与主节点一致）virtual_router_id 51    # 虚拟路由ID（与主节点一致）priority 90             # 优先级（低于主节点）advert_int 1            # 心跳间隔（与主节点一致）# 认证（与主节点一致）authentication {auth_type PASSauth_pass 1111}# 关联健康检查脚本track_script {chk_nginx}# 虚拟IP（与主节点一致）virtual_ipaddress {192.168.221.100/24 dev ens33}
}
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
​
# 4. 分发层 Nginx 反向代理配置（主/备节点）
vim /usr/local/nginx/conf/nginx.conf
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
#在第34行左右直接添加
upstream web_cluster {server 192.168.221.30:80 weight=1;  # Web1节点server 192.168.221.40:80 weight=1;  # Web2节点
}
#在第41行左右
server {listen       80;#server_name  localhost;
​#charset koi8-r;#access_log  logs/host.access.log  main;
​location / {##     root   html;##     index  index.html index.htm;proxy_pass http://web_cluster;proxy_set_header Host $host;proxy_set_header X-Real-IP $remote_addr;}
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
# 重启Nginx生效
nginx -s stop
nginx
​
# 五、服务启动与测
## 配置web服务（192.168.221.30和40）
# Web1节点：创建首页以区分节点
echo "<h1>Web1 - 192.168.221.30</h1>" > /usr/local/nginx/html/index.html
# Web2节点：创建首页以区分节点
echo "<h1>Web2 - 192.168.221.40</h1>" > /usr/local/nginx/html/index.html
# 重启httpd生效
nginx -s stop
nginx
​
六、测试
# 启动Keepalived并设置开机自启
systemctl start keepalived && systemctl enable keepalived

验证高可用功能（停掉一个测试另一个）

步骤 1：检查VIP绑定
在主节点执行 `ip addr`，确认 `192.168.221.100` 已绑定到网卡（如`ens33`）；备节点暂未绑定。

步骤 2：访问 VIP 测试负载均衡
浏览器访问 `http://192.168.221.100`，刷新页面应交替显示 “Web1” 和 “Web2”（若未配置`ip_hash`）。

步骤 3：故障切换测试
- 模拟主节点 Nginx 故障：在主节点执行 `systemctl stop nginx`，等待约 5 秒后，在备节点执行 `ip addr`，确认`192.168.221.100`已漂移到备节点。
- 模拟主节点宕机：关闭主节点电源，备节点应自动接管 VIP，浏览器访问`http://192.168.221.100`仍正常。