项目：构建高可用、负载均衡的高效Web服务器

hotslow2024-09-15 9:14

一、项目需求

实现高可用，防止单点故障，实现负载均衡

负载均衡：将流量分摊给一个或多个后端服务器，提高公司业务的可用性、方便业务后期的水平动态扩展。

二、项目描述

本项目旨在构建一个集高可用性、负载均衡、高效Web服务及缓存技术于一体的综合解决方案， 以显著提升Web应用的可靠性、性能和响应速度。通过精细规划和实施，我们成功搭建了一个能够应对高并 发访问、确保业务连续性的Web应用架构。

三、设备要求

准备四台主机

1.准备两台主机实现keepalived+haproxy

第一台主机：

第二台主机：

2.准备两台主机作为后端服务器（realserver）

第一台主机：

第二台主机：

四、搭建

4.1 基础设置

注：四台主机都需同样的操作，也可以准备好一台主机，再克隆，修改IP，hostname即可

1.配置网络

ka1：

bash 复制代码 
[root@localhost ~]# nmcli connection modify ens160 ipv4.addresses 172.25.254.30/24
[root@localhost ~]# nmcli connection modify ens160 ipv4.gateway 172.25.254.2
[root@localhost ~]# nmcli connection modify ens160 ipv4.dns 114.114.114.114
[root@localhost ~]# nmcli connection modify ens160 ipv4.method manual 
[root@localhost ~]# nmcli connection modify ens160 connection.autoconnect yes
[root@localhost ~]# nmcli connection up ens160 
Connection successfully activated (D-Bus active path: /org/freedesktop/NetworkManager/ActiveConnection/12)

ka2：

bash 复制代码 
[root@localhost ~]# vim /etc/NetworkManager/system-connections/ens160.nmconnection 
connection]
id=ens160
type=ethernet
interface-name=ens160

[ethernet]

[ipv4]
address1=172.25.254.40/24,172.25.254.2
dns=114.114.114.114;
method=manual

[ipv6]
addr-gen-mode=eui64
method=auto

[root@localhost ~]# nmcli connection reload 
[root@localhost ~]# nmcli connection up ens160

注意：分别使用两种方式更改IP

realserver1：172.25.254.101

realserver2：172.25.254.102

使用上面的方式修改IP

2.配置软件仓库

bash 复制代码 
[root@localhost ~]# mkdir /rhel9
[root@localhost ~]# echo /dev/sr0 /rhel9 >> /etc/rc.d/rc.local 
[root@localhost ~]# chmod +x /etc/rc.d/rc.local 
[root@localhost ~]# mount /dev/sr0 /rhel9
mount: /rhel9: WARNING: source write-protected, mounted read-only.

3.设置主机名和本地解析

bash 复制代码 
[root@ka1 ~]# hostnamectl set-hostname ka1.lm.org
[root@ka2 ~]# hostnamectl set-hostname ka2.lm.org

4.2 搭建高可用负载均衡集群

1.安装keepalived，haproxy

ka1，ka2主机都需此操作

bash 复制代码 
[root@ka1 ~]# hostnamectl set-hostname ka1
[root@ka1 ~]# dnf install -y keepalived
[root@ka1 ~]# dnf install -y haproxy
# 开机自启
[root@ka1 ~]# systemctl enable --now haproxy
Created symlink /etc/systemd/system/multi-user.target.wants/haproxy.service → /usr/lib/systemd/system/haproxy.service.
[root@ka1 ~]# systemctl enable --now keepalived
Created symlink /etc/systemd/system/multi-user.target.wants/keepalived.service → /usr/lib/systemd/system/keepalived.service.

2.修改keepalived和haproxy配置文件

ka1

bash 复制代码 
[root@ka1 ~]# vim /etc/keepalived/keepalived.conf 
global_defs {
   notification_email {
        3041609716@qq.com
   }
   notification_email_from keepalived@lm.org
   smtp_server 127.0.0.1
   smtp_connect_timeout 30
   router_id ka1.lm.org
   vrrp_skip_check_adv_addr
   #vrrp_strict
   vrrp_garp_interval 0
   vrrp_gna_interval 0
}
vrrp_script check_haproxy {
    script "/etc/keepalived/scripts/haproxy.sh"
    interval 1
    weight -30
    fall 2
    rise 2
    timeout 2
}
vrrp_instance web1 {
    state MASTER
    interface ens160
    virtual_router_id 100
    priority 100
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass 1111
    }
    virtual_ipaddress {
        172.25.254.100 dev ens160 label ens160:1
    }
    track_script {
        check_haproxy
    }
}
vrrp_instance web2 {
    state SLAVE
    interface ens160
    virtual_router_id 200
    priority 80
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass 1111
    }
    virtual_ipaddress {
        172.25.254.200 dev ens160 label ens160:2
    }

    track_script {
        check_haproxy
    }
}

[root@ka1 ~]# vim /etc/haproxy/haproxy.cfg 
listen webcluster1
    bind 172.25.254.100:80
    server web1 172.25.254.101:80 check
    server web2 172.25.254.102:80 check
listen webcluster2
    bind 172.25.254.200:80
    server web1 172.25.254.101:80 check
    server web2 172.25.254.102:80 check

ka2

bash 复制代码 
# 40
[root@ka2 ~]# vim /etc/keepalived/keepalived.conf 
global_defs {
   notification_email {
        3041609716@qq.com
   }
   notification_email_from keepalived@lm.org
   smtp_server 127.0.0.1
   smtp_connect_timeout 30
   router_id ka1.lm.org
   vrrp_skip_check_adv_addr
   #vrrp_strict
   vrrp_garp_interval 0
   vrrp_gna_interval 0
}
vrrp_script check_haproxy {
    script "/etc/keepalived/scripts/haproxy.sh"
    interval 1
    weight -30
    fall 2
    rise 2
    timeout 2
}
vrrp_instance web1 {
    state SLAVE
    interface ens160
    virtual_router_id 100
    priority 80
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass 1111
    }
    virtual_ipaddress {
        172.25.254.100/24 dev ens160 label ens160:1
    }
    track_script {
        check_haproxy
    }
}
vrrp_instance web2 {
    state MASTER
    interface ens160
    virtual_router_id 200
    priority 100
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass 1111
    }
    virtual_ipaddress {
        172.25.254.200 dev ens160 label ens160:2
    }
    track_script {
        check_haproxy
    }
}
[root@ka1 ~]# vim /etc/haproxy/haproxy.cfg 
listen webcluster1
    bind 172.25.254.100:80
    server web1 172.25.254.101:80 check
    server web2 172.25.254.102:80 check
listen webcluster2
    bind 172.25.254.200:80
    server web1 172.25.254.101:80 check
    server web2 172.25.254.102:80 check

3.修改内核参数

bash 复制代码 
[root@ka1 ~]# vim /etc/sysctl.conf 
net.ipv4.ip_nonlocal_bind = 1
[root@ka1 ~]# sysctl -p
[root@ka2 ~]# vim /etc/sysctl.conf 
net.ipv4.ip_nonlocal_bind = 1
[root@ka2 ~]# sysctl -p
bash 复制代码 
[root@localhost ~]# hostnamectl set-hostname realserver1
[root@realserver1 ~]# dnf install -y nginx
[root@realserver1 ~]# systemctl enable --now nginx
Created symlink /etc/systemd/system/multi-user.target.wants/nginx.service → /usr/lib/systemd/system/nginx.service.
[root@realserver1 ~]# echo 172.25.254.101 > /usr/share/nginx/html/index.html
[root@realserver1 ~]# curl 172.25.254.101
172.25.254.101

4.3 后端服务器设置

realserver1，realserver2都是几乎相同的操作

bash 复制代码 
[root@localhost ~]# hostnamectl set-hostname realserver1
[root@realserver1 ~]# dnf install -y nginx
[root@realserver1 ~]# echo 172.25.254.101 > /usr/share/nginx/html/index.html
[root@realserver1 ~]# curl 172.25.254.101
172.25.254.101


[root@localhost ~]# hostnamectl set-hostname realserver2
[root@realserver2 ~]# dnf install -y nginx
[root@realserver2 ~]# echo 172.25.254.102 > /usr/share/nginx/html/index.html
[root@realserver2 ~]# curl 172.25.254.102
172.25.254.102

4.4 测试

bash 复制代码 
[root@ka1 ~]# ifconfig
ens160: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 172.25.254.30  netmask 255.255.255.0  broadcast 172.25.254.255
        inet6 fe80::20c:29ff:fef1:9875  prefixlen 64  scopeid 0x20<link>
        ether 00:0c:29:f1:98:75  txqueuelen 1000  (Ethernet)
        RX packets 3109  bytes 211759 (206.7 KiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 3775  bytes 227206 (221.8 KiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

ens160:1: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 172.25.254.100  netmask 255.255.255.255  broadcast 0.0.0.0
        ether 00:0c:29:f1:98:75  txqueuelen 1000  (Ethernet)
[root@ka1 ~]# curl 172.25.254.100
172.25.254.102
[root@ka1 ~]# curl 172.25.254.100
172.25.254.101
[root@ka1 ~]# curl 172.25.254.100
172.25.254.102
[root@ka1 ~]# curl 172.25.254.200
172.25.254.101
[root@ka1 ~]# curl 172.25.254.200
172.25.254.102
[root@ka1 ~]# curl 172.25.254.200
172.25.254.101
[root@ka1 ~]# curl 172.25.254.200
172.25.254.102
bash 复制代码 
[root@ka2 ~]# ifconfig
ens160: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 172.25.254.40  netmask 255.255.255.0  broadcast 172.25.254.255
        inet6 fe80::20c:29ff:fed5:289d  prefixlen 64  scopeid 0x20<link>
        ether 00:0c:29:d5:28:9d  txqueuelen 1000  (Ethernet)
        RX packets 306952  bytes 22903523 (21.8 MiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 560789  bytes 37705879 (35.9 MiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

ens160:2: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 172.25.254.200  netmask 255.255.255.255  broadcast 0.0.0.0
        ether 00:0c:29:d5:28:9d  txqueuelen 1000  (Ethernet)
[root@ka2 ~]# curl 172.25.254.100
172.25.254.101
[root@ka2 ~]# curl 172.25.254.100
172.25.254.102
[root@ka2 ~]# curl 172.25.254.100
172.25.254.101
[root@ka2 ~]# curl 172.25.254.100
172.25.254.102
[root@ka2 ~]# curl 172.25.254.100
172.25.254.101
[root@ka2 ~]# curl 172.25.254.100
172.25.254.102
[root@ka2 ~]# curl 172.25.254.100
172.25.254.101
[root@ka2 ~]# curl 172.25.254.200
172.25.254.101
[root@ka2 ~]# curl 172.25.254.200
172.25.254.102
[root@ka2 ~]# curl 172.25.254.200
172.25.254.101
[root@ka2 ~]# curl 172.25.254.200
172.25.254.102

模拟故障：

关闭ka1的keepalived

复制代码 
[root@ka1 ~]# systemctl stop keepalived
[root@ka1 ~]# ifconfig
ens160: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 172.25.254.30  netmask 255.255.255.0  broadcast 172.25.254.255
        inet6 fe80::20c:29ff:fef1:9875  prefixlen 64  scopeid 0x20<link>
        ether 00:0c:29:f1:98:75  txqueuelen 1000  (Ethernet)
        RX packets 927274  bytes 72325472 (68.9 MiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 1854187  bytes 129793327 (123.7 MiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0



[root@ka2 ~]# ifconfig
ens160: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 172.25.254.40  netmask 255.255.255.0  broadcast 172.25.254.255
        inet6 fe80::20c:29ff:fed5:289d  prefixlen 64  scopeid 0x20<link>
        ether 00:0c:29:d5:28:9d  txqueuelen 1000  (Ethernet)
        RX packets 683628  bytes 53310970 (50.8 MiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 1365919  bytes 95613730 (91.1 MiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

ens160:1: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 172.25.254.100  netmask 255.255.255.0  broadcast 0.0.0.0
        ether 00:0c:29:d5:28:9d  txqueuelen 1000  (Ethernet)

ens160:2: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 172.25.254.200  netmask 255.255.255.255  broadcast 0.0.0.0
        ether 00:0c:29:d5:28:9d  txqueuelen 1000  (Ethernet)


# 即使发生故障，依然可以成功访问
[root@ka1 ~]# curl 172.25.254.100
172.25.254.102
[root@ka1 ~]# curl 172.25.254.100
172.25.254.101
[root@ka1 ~]# curl 172.25.254.100
172.25.254.102
[root@ka1 ~]# curl 172.25.254.100
172.25.254.101
[root@ka1 ~]# curl 172.25.254.200
172.25.254.102
[root@ka1 ~]# curl 172.25.254.200
172.25.254.101
[root@ka1 ~]# curl 172.25.254.200
172.25.254.102
[root@ka1 ~]# curl 172.25.254.200
172.25.254.101
相关推荐
荣--2 天前
一键部署不是为了省时间 —— 它是把"买来的 PaaS"变成"自己的平台"的拐点
运维·zabbix·工程化·一键部署·平台化·边界设计
江华森2 天前
动手实战学 Docker — 从零到集群编排完全指南
运维
Avan_菜菜3 天前
FRP 内网穿透完整实战：从 HTTP 映射到 HTTPS 自签代理
运维·nginx·https
SelectDB4 天前
Litefuse 开源并推出单进程轻量模式，25 秒就能跑起来的 Agent 可观测与评估平台
运维·后端·自动化运维
XIAOHEZIcode5 天前
Linux系统鼠标偏移常见原因以及修复方案
linux·运维·游戏
用户0328472220706 天前
如何搭建本地yum源(上)
运维
ping某7 天前
为什么 Nginx 明明监听了 80，转发后端时却用了 4xxxx 端口？
后端·nginx
大树889 天前
金刚石散热越强，管路越先见顶
大数据·运维·服务器·人工智能·ai
摇滚侠9 天前
Linux CentOS7 rpm 安装 MySQL 5.7
linux·运维·mysql
霸道流氓气质9 天前
领域驱动设计（DDD）在 Spring Boot 微服务中的实践指南
运维·spring boot·微服务
热门推荐
012026年6月AI大模型全景报告：GPT-5.6、Claude Opus 4.8、Gemini 3.5，中美AI三足鼎立谁主沉浮？022026年6月AI行业全景：从百模大战到Agent元年，这30天发生了什么？032026 年 AI 编程工具终极横评：Cursor vs Claude Code vs Copilot vs Windsurf04【AI】2026 年具身智能模型和世界模型总结05Claude Code、Codex、Cursor三分天下：2026年AI编程Agent生态全景剖析06飞书长连接_事件订阅（接收消息，审批任务状态变更）07GitHub 镜像站点08【AI总结】2026年6月 主流国内外大模型总结09Trae国际版与国内版深度测评：AI原生IDE的双生花10AI科技热点日报 | 2026年6月1日