完整实验命令解析：从集群搭建到负载均衡配置（2）
一、环境准备与基础网络配置
1.1 节点角色与网络规划
| 节点角色 | 主机名 | 所属网段 | IP 地址 | 网关 | 核心功能 |
|---|---|---|---|---|---|
| Web 服务器 | web1 | 10.1.8.0/24 | 10.1.8.11 | 10.1.8.10（后期调整为 10.1.8.20） | 部署 Nginx/HTTPD，提供 Web 服务 |
| Web 服务器 | web2 | 10.1.8.0/24 | 10.1.8.12 | 10.1.8.10（后期调整为 10.1.8.20） | 部署 Nginx/HTTPD，提供 Web 服务 |
| Web 服务器 | web3 | 10.1.8.0/24 | 10.1.8.13 | 10.1.8.10（后期调整为 10.1.8.20） | 部署 Nginx/HTTPD，提供 Web 服务 |
| 负载均衡器（初期） | lb | 10.1.1.0/24 | 10.1.1.10 | 无需配置 | 部署 LVS，实现负载均衡 |
| 客户端 | client1 | 10.1.8.0/24 | 10.1.8.21 | 10.1.8.10（后期调整为 10.1.8.20） | 测试 Web 服务访问 |
| 客户端 | client2 | 10.1.1.0/24 | 10.1.1.21 | 10.1.1.10（后期调整为 10.1.1.20） | 测试负载均衡与高可用 |
| 路由器 | router | 多网段（10.1.8.0/24、10.1.1.0/24、10.1.2.0/24） | 10.1.8.20、10.1.1.20、10.1.2.20 | - | 实现多网段互通 |
| NFS 服务器 | nfs | 10.1.2.0/24 | 10.1.2.100 | 10.1.2.20 | 提供共享存储，存储 Web 静态资源 |
| 高可用负载均衡器 | ha1 | 10.1.8.0/24 | 10.1.8.14 | 10.1.8.20 | 部署 HAProxy+Keepalived，主负载均衡节点 |
| 高可用负载均衡器 | ha2 | 10.1.8.0/24 |
1.2 基础网络配置（全节点）
1.2.1 10.1.8.0/24 网段节点（web1、web2、web3、client1）
web 服务器
[root@web1-3 ~]#yum install -y nginxsystemctl enable nginx --nowecho Welcome to $(hostname) > /usr/share/nginx/html/index.html
lvs 服务器
[root@lvs ~]# yum install -y ipvsadm# systemctl enable ipvsadm# 等ipvs规则配置完成后再启动ipvsadm服务
client服务器
# 10.1.8.0/24 网段网关为10.1.8.10 #web123 client1nmcli connection modify ens33 ipv4.gateway 10.1.8.10nmcli connection up ens33 # 10.1.1.0/24 网段网关为10.1.1.10 #client2 lb不用配置nmcli connection modify ens33 ipv4.gateway 10.1.1.10nmcli connection up ens33
二、Web 服务部署（web1、web2、web3 节点）
配置 web
[root@web1-3 ~]## 部署 webyum install -y nginxecho Welcome to $(hostname) > /usr/share/nginx/html/index.html systemctl enable nginx.service --now#验证 Web 服务可用性[root@client1 ~ 10:56:20]# curl 10.1.8.11Welcome to web1.laoma.cloud[root@client1 ~ 10:57:49]# curl 10.1.8.12Welcome to web2.laoma.cloud[root@client1 ~ 10:57:53]# curl 10.1.8.13Welcome to web3.laoma.cloud
三、LVS 负载均衡部署（lb 节点）
配置 LVS
[root@lb ~ 10:55:16]# echo "net.ipv4.ip_forward=1" >> /etc/sysctl.conf[root@lb ~ 10:59:15]# sysctl -pnet.ipv4.ip_forward = 1[root@lb ~ 10:59:21]# yum install -y ipvsadm[root@lb ~ 10:59:38]# touch /etc/sysconfig/ipvsadm[root@lb ~ 10:59:38]# systemctl enable ipvsadm --now# 创建轮询负载[root@lb ~ 10:59:38]# ipvsadm -A -t 10.1.1.10:80 -s rr[root@lb ~ 11:00:03]# ipvsadm -a -t 10.1.1.10:80 -r 10.1.8.11 -m[root@lb ~ 11:00:03]# ipvsadm -a -t 10.1.1.10:80 -r 10.1.8.12 -m[root@lb ~ 11:00:03]# ipvsadm -a -t 10.1.1.10:80 -r 10.1.8.13 -m[root@lb ~ 11:00:03]# ipvsadm-save -n > /etc/sysconfig/ipvsadm# 核实配置是否生效[root@lb ~ 11:00:04]# ipvsadm -Ln[root@lb ~ 11:00:59]# for i in {1..90};do curl -s 10.1.1.10 ;done|sort|uniq -c30 Welcome to web1.laoma.cloud30 Welcome to web2.laoma.cloud30 Welcome to web3.laoma.cloud负载均衡模式更改为加权轮询
[root@lb ~ 11:01:12]# ipvsadm -E -t 10.1.1.10:80 -s wrr[root@lb ~ 11:01:33]# ipvsadm -e -t 10.1.1.10:80 -r 10.1.8.12 -m -w 2[root@lb ~ 11:01:33]# ipvsadm -e -t 10.1.1.10:80 -r 10.1.8.13 -m -w 3[root@lb ~ 11:01:34]# ipvsadm -Ln访问验证[root@lb ~ 11:01:46]# for i in {1..90};do curl -s 10.1.1.10 ;done|sort|uniq -c15 Welcome to web1.laoma.cloud30 Welcome to web2.laoma.cloud45 Welcome to web3.laoma.cloud思考
此时client1是否可以通过10.1.1.10访问后端服务器？
答：不能访问
client1 发出去数据包经过10.1.1.10的ipvs模块处理，而后端web收到数据包后根据来源地址10.1.8.21进行回复，也就是直接返回client1 导致数据包没有返回LVS处理
如果不能，需要如何配置才能实现访问?
[root@web1 ~ 10:57:32]# nmcli connection modify ens33 ipv4.routes '10.1.8.21 255.255.255.255 10.1.8.10'[root@web1 ~ 11:02:32]# nmcli connection up ens33[root@web2 ~ 10:57:32]# nmcli connection modify ens33 ipv4.routes '10.1.8.21 255.255.255.255 10.1.8.10'[root@web2 ~ 11:02:32]# nmcli connection up ens33[root@web3 ~ 10:57:32]# nmcli connection modify ens33 ipv4.routes '10.1.8.21 255.255.255.255 10.1.8.10'[root@web3 ~ 11:02:32]# nmcli connection up ens33
#清空之前实验[root@lb ~ 11:02:09]# > /etc/sysconfig/ipvsadm[root@lb ~ 11:16:16]# systemctl restart ipvsadm.service 进入虚拟机把lb的第二张网卡拔掉#web1 2 3 client1 执行# 10.1.8.0/24 网段网关为10.1.8.20nmcli connection modify ens33 ipv4.gateway 10.1.8.20nmcli connection up ens33#client2 执行# 10.1.1.0/24 网段网关为10.1.1.20nmcli connection modify ens33 ipv4.gateway 10.1.1.20nmcli connection up ens33
四、LVS-DR 模式部署（全节点适配）
4.1 虚拟网卡（Dummy）配置
Web 节点（web1、web2、web3）配置
# 增加虚拟网卡，子网掩码一定要32位 [root@client2 ~ 11:20:31]# nmcli connection add type dummy ifname dummy con-name dummy ipv4.method manual ipv4.addresses 10.1.8.100/32 连接 "dummy" (cafa29cd-6424-4356-9dc0-edc6b044be44) 已成功添加。 [root@client2 ~ 11:32:10]# nmcli connection up dummy[root@web1 ~ 11:20:31]# nmcli connection add type dummy ifname dummy con-name dummy ipv4.method manual ipv4.addresses 10.1.8.100/32 连接 "dummy" (c1d840b5-f6f9-4aa9-9688-2318a45628e1) 已成功添加。 [root@web1 ~ 11:34:09]# nmcli connection up dummy[root@web2 ~ 11:20:31]# nmcli connection add type dummy ifname dummy con-name dummy ipv4.method manual ipv4.addresses 10.1.8.100/32 连接 "dummy" (c1d840b5-f6f9-4aa9-9688-2318a45628e1) 已成功添加。 [root@web2 ~ 11:34:09]# nmcli connection up dummy[root@web3 ~ 11:20:31]# nmcli connection add type dummy ifname dummy con-name dummy ipv4.method manual ipv4.addresses 10.1.8.100/32 连接 "dummy" (c1d840b5-f6f9-4aa9-9688-2318a45628e1) 已成功添加。 [root@web3 ~ 11:34:09]# nmcli connection up dummy[root@client2 ~ 11:32:11]# nmcli connection delete dummy 成功删除连接 "dummy" (cafa29cd-6424-4356-9dc0-edc6b044be44)。 [root@client2 ~ 11:35:56]# nmcli c NAME UUID TYPE DEVICE ens33 555eece5-af4c-45ae-bab9-c07e68d0e649 ethernet ens33 # LVS 节点（lb）配置 [root@web1 ~ 11:33:40]# ip -br a lo UNKNOWN 127.0.0.1/8 ::1/128 ens33 UP 10.1.8.11/24 fe80::20c:29ff:feb2:fcae/64 dummy0 DOWN dummy UNKNOWN 10.1.8.100/32 fe80::6f18:c0cb:74d0:ea0d/64 [root@web2 ~ 11:34:04]# ip -br a lo UNKNOWN 127.0.0.1/8 ::1/128 ens33 UP 10.1.8.12/24 fe80::20c:29ff:fefe:b2c/64 dummy0 DOWN dummy UNKNOWN 10.1.8.100/32 fe80::f4aa:4d23:ac32:7129/64[root@web3 ~ 11:36:41]# ip -br a lo UNKNOWN 127.0.0.1/8 ::1/128 ens33 UP 10.1.8.13/24 fe80::20c:29ff:fe1d:4a9c/64 dummy0 DOWN dummy UNKNOWN 10.1.8.100/32 fe80::30ca:ef21:2f8:fe5c/64
4.2 LVS-DR 模式规则配置（lb 节点）
[root@lb ~ 11:36:58]# nmcli connection add type dummy ifname dummy con-name dummy ipv4.method manual ipv4.addresses 10.1.8.100/32
连接 "dummy" (8cdb619b-460e-4b83-afe3-5f855a601d4d) 已成功添加。
[root@lb ~ 11:40:46]# nmcli connection up dummy
连接已成功激活（D-Bus 活动路径：/org/freedesktop/NetworkManager/ActiveConnection/4）
[root@lb ~ 11:40:48]# ip -br a
lo UNKNOWN 127.0.0.1/8 ::1/128
ens33 UP 10.1.8.10/24 fe80::20c:29ff:fe55:d621/64
dummy0 DOWN
dummy UNKNOWN 10.1.8.100/32 fe80::d941:efae:a684:17ea/64 [root@lb ~ 11:40:59]# ipvsadm -A -t 10.1.8.100:80 -s rr
[root@lb ~ 11:42:26]# ipvsadm -a -t 10.1.8.100:80 -r 10.1.8.11:80
[root@lb ~ 11:42:26]# ipvsadm -a -t 10.1.8.100:80 -r 10.1.8.12:80
[root@lb ~ 11:42:26]# ipvsadm -a -t 10.1.8.100:80 -r 10.1.8.13:80
[root@lb ~ 11:42:26]# ipvsadm-save -n > /etc/sysconfig/ipvsadm
[root@lb ~ 11:42:27]# ipvsadm -Ln
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags-> RemoteAddress:Port Forward Weight ActiveConn InActConn
TCP 10.1.1.10:80 wrr-> 10.1.8.11:80 Masq 1 0 0 -> 10.1.8.12:80 Masq 2 0 0 -> 10.1.8.13:80 Masq 3 0 0
TCP 10.1.8.100:80 rr-> 10.1.8.11:80 Route 1 0 0 -> 10.1.8.12:80 Route 1 0 0 -> 10.1.8.13:80 Route 1 0 0
[root@lb ~ 11:42:59]# ipvsadm -D -t 10.1.1.10:80
[root@lb ~ 11:43:20]# ipvsadm -Ln
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags-> RemoteAddress:Port Forward Weight ActiveConn InActConn
TCP 10.1.8.100:80 rr-> 10.1.8.11:80 Route 1 0 0 -> 10.1.8.12:80 Route 1 0 0 -> 10.1.8.13:80 Route 1 0 0 #DR 模式可用性测试（client1、client2 节点）
[root@client2 ~ 11:36:58]# curl http://10.1.8.100
Welcome to web3.laoma.cloud
[root@client2 ~ 11:44:04]# curl http://10.1.8.100
Welcome to web2.laoma.cloud
[root@client2 ~ 11:44:06]# curl http://10.1.8.100
Welcome to web1.laoma.cloud[root@client1 ~ 11:36:58]# curl http://10.1.8.100
Welcome to web3.laoma.cloud
[root@client1 ~ 11:45:11]# curl http://10.1.8.100
Welcome to web2.laoma.cloud
[root@client1 ~ 11:45:12]# curl http://10.1.8.100
Welcome to web1.laoma.cloud[root@client1 ~ 11:45:12]# for i in {1..90};do curl -s 10.1.8.100 ;done|sort|uniq
Welcome to web1.laoma.cloud
Welcome to web2.laoma.cloud
Welcome to web3.laoma.cloud[root@client2 ~ 11:46:31]# for i in {1..90};do curl -s 10.1.8.100 ;done|sort|uniq -c30 Welcome to web1.laoma.cloud30 Welcome to web2.laoma.cloud30 Welcome to web3.laoma.cloud
五、Keepalived 高可用部署（web1、web2 节点，后期迁移至 ha1、ha2）
5.1 初期 Web 节点高可用配置（web1 为主、web2 为备）
[root@web1 ~ 14:02:06]# nmcli connection delete dummy 成功删除连接 "dummy" (6a249f96-28ab-41c5-8f22-e9f0f3e395bc)。[root@lb ~ 14:05:28]#nmcli connection delete dummy reboot
配置 web2，作为备节点
[root@web2 ~]#
yum install -y keepalived
cp /etc/keepalived/keepalived.conf{,.ori}
vim /etc/keepalived/keepalived.conf! Configuration File for keepalivedglobal_defs {router_id web2
}
#vim
vrrp_instance nginx {state BACKUPinterface ens33virtual_router_id 51priority 100advert_int 1authentication {auth_type PASSauth_pass lyk@123}virtual_ipaddress {10.1.8.100/24}
}[root@web2 ~ 14:17:54]# systemctl enable keepalived.service --now[root@web2 ~ 14:18:09]# systemctl restart keepalived.service [root@web2 ~ 14:19:28]# ip -br a
lo UNKNOWN 127.0.0.1/8 ::1/128
ens33 UP 10.1.8.12/24 10.1.8.100/24 fe80::20c:29ff:fefe:b2c/64 [root@client1 ~ 14:05:28]# while true;do curl -s http://10.1.8.100/;sleep 1;done
Welcome to web2.laoma.cloud
Welcome to web2.laoma.cloud
......[root@client2 ~ 14:05:27]# while true;do curl -s http://10.1.8.100/;sleep 1;done
Welcome to web2.laoma.cloud
Welcome to web2.laoma.cloud
Welcome to web2.laoma.cloud
Welcome to web2.laoma.cloud
.....5.1.2 主节点配置（web1 节点）
[root@web2 ~ 14:19:38]# scp /etc/keepalived/keepalived.conf web1:/etc/keepalived/keepalived.conf
Warning: Permanently added 'web1,10.1.8.11' (ECDSA) to the list of known hosts.
keepalived.conf 100% 320 998.8KB/s 00:00 # 双 VIP 高可用配置（web1、web2 节点）
# web1 节点配置（主 10.1.8.100，备 10.1.8.200）
[root@web1 ~ 14:16:44]# vim /etc/keepalived/keepalived.conf
[root@web1 ~ 14:36:08]# cat /etc/keepalived/keepalived.conf
! Configuration File for keepalivedglobal_defs {router_id web1
}vrrp_instance web {state MASTERinterface ens33virtual_router_id 51priority 200advert_int 1authentication {auth_type PASSauth_pass lyk@123}virtual_ipaddress {10.1.8.100/24}
}[root@web1 ~ 14:36:19]# systemctl restart keepalived.service
#client链接显示
.......
Welcome to web1.laoma.cloud
Welcome to web1.laoma.cloud
Welcome to web1.laoma.cloud
Welcome to web1.laoma.cloud
Welcome to web1.laoma.cloud
Welcome to web1.laoma.cloud
......
[root@web2 ~ 14:34:11]# ip -br a
lo UNKNOWN 127.0.0.1/8 ::1/128
ens33 UP 10.1.8.12/24 fe80::20c:29ff:fefe:b2c/64 [root@web1 ~ 14:38:42]# systemctl stop keepalived.service #client链接显示
.......
Welcome to web2.laoma.cloud
Welcome to web2.laoma.cloud
Welcome to web2.laoma.cloud
Welcome to web2.laoma.cloud
Welcome to web2.laoma.cloud
Welcome to web2.laoma.cloud
......
#web2 节点配置（备 10.1.8.100，主 10.1.8.200）
[root@web1 ~ 14:49:14]# vim /etc/keepalived/keepalived.conf
[root@web1 ~ 14:49:18]# cat /etc/keepalived/keepalived.conf
! Configuration File for keepalivedglobal_defs {router_id web1
}vrrp_instance web_1 {state MASTERinterface ens33virtual_router_id 51priority 200advert_int 1authentication {auth_type PASSauth_pass lyk@123}virtual_ipaddress {10.1.8.100/24}
}
vrrp_instance web_2 {state BACKUPinterface ens33virtual_router_id 52priority 100advert_int 1authentication {auth_type PASSauth_pass lyk@123}virtual_ipaddress {10.1.8.200/24}
}#web2
[root@web2 ~ 14:37:35]# vim /etc/keepalived/keepalived.conf
[root@web2 ~ 14:50:03]# cat /etc/keepalived/keepalived.conf
! Configuration File for keepalivedglobal_defs {router_id web2
}vrrp_instance web_1 {state BACKUPinterface ens33virtual_router_id 52priority 100advert_int 1authentication {auth_type PASSauth_pass lyk@123}virtual_ipaddress {10.1.8.100/24}
}vrrp_instance web_2 {state MASTERinterface ens33virtual_router_id 51priority 200advert_int 1authentication {auth_type PASSauth_pass lyk@123}virtual_ipaddress {10.1.8.200/24}
}[root@web1 ~ 14:51:38]# systemctl restart keepalived.service
[root@web1 ~ 14:52:10]# ip -br a
lo UNKNOWN 127.0.0.1/8 ::1/128
ens33 UP 10.1.8.11/24 10.1.8.100/24 fe80::20c:29ff:feb2:fcae/64 [root@web2 ~ 14:51:12]# systemctl restart keepalived.service
[root@web2 ~ 14:52:16]# ip -br a
lo UNKNOWN 127.0.0.1/8 ::1/128
ens33 UP 10.1.8.12/24 10.1.8.200/24 fe80::20c:29ff:fefe:b2c/64 #验证 高可用切换测试（client1 节点）
[root@client1 ~ 14:40:09]# while true;do curl -s http://10.1.8.100/;sleep 1;done
Welcome to web1.laoma.cloud
Welcome to web1.laoma.cloud
Welcome to web1.laoma.cloud
^C
[root@client1 ~ 14:53:26]# while true;do curl -s http://10.1.8.200/;sleep 1;done
Welcome to web2.laoma.cloud
Welcome to web2.laoma.cloud
Welcome to web2.laoma.cloud
Welcome to web2.laoma.cloud
^C#断开web1，查看web2
[root@web1 ~ 14:52:26]# systemctl stop keepalived.service
[root@web2 ~ 14:55:19]# ip -br a
lo UNKNOWN 127.0.0.1/8 ::1/128
ens33 UP 10.1.8.12/24 10.1.8.200/24 10.1.8.100/24 fe80::20c:29ff:fefe:b2c/64
[root@web1 ~ 14:55:07]# nmcli c NAME UUID TYPE DEVICE 有线连接 1 57b39c8d-d270-3ce9-95e4-48b5823381a6 ethernet ens36 ens33 555eece5-af4c-45ae-bab9-c07e68d0e649 ethernet ens33 [root@web1 ~ 15:05:28]# nmcli connection modify 有线连接\ 1 ipv4.method manual ipv4.addresses 10.1.2.11/24 connection.id ens36;nmcli connection up ens36[root@web2 ~ 15:07:10]# nmcli connection modify 有线连接\ 1 ipv4.method manual ipv4.addresses 10.1.2.12/24 connection.id ens36;nmcli connection up ens36[root@web3 ~ 15:07:31]# nmcli connection modify 有线连接\ 1 ipv4.method manual ipv4.addresses 10.1.2.13/24 connection.id ens36;nmcli connection up ens36[root@web1 ~ 15:07:04]# systemctl stop keepalived [root@web2 ~ 15:07:04]# systemctl stop keepalived [root@web3 ~ 15:07:04]# systemctl stop keepalived [root@lb ~ 14:05:28]# init 0[root@web 1 2 3 ~ 15:13:37]# ip -br a
用cluster-tpl 克隆三台虚拟机nfs（修改网卡为vm （仅主机2）） ha1 ha2（不变）#nfs （2.100）为2.0网络段 router 主机添加网卡（一共三个）添加网卡（设置vm （2仅主机））
# **nfs**虚拟机界面执行 hostnamectl set-hostname nfs.lyk.cloud nmcli c nmcli connection modify ens33 ipv4.addresses 10.1.2.100/24 ipv4.gateway 10.1.2.20 nmcli connection up ens33 bash ping 1.1.1.1#Xshell执行 [root@router ~ 15:26:51]# nmcli c NAME UUID TYPE DEVICE ens33 555eece5-af4c-45ae-bab9-c07e68d0e649 ethernet ens33 ens36 c4a81250-34ce-3a67-a3ea-bacfb0289b97 ethernet ens36 有线连接 1 e7758ba8-0110-3e86-883c-8aa1bbdf1f2b ethernet -- [root@router ~ 15:27:05]# nmcli connection modify 有线连接\ 1 ipv4.method manual ipv4.addresses 10.1.2.20/24 connection.id ens37 [root@router ~ 15:29:39]# nmcli connection up ens37 连接已成功激活（D-Bus 活动路径：/org/freedesktop/NetworkManager/ActiveConnection/11）[root@router ~ 15:31:07]# ip -br a lo UNKNOWN 127.0.0.1/8 ::1/128 ens33 UP 10.1.8.20/24 fe80::20c:29ff:fe62:b97a/64 ens36 UP 10.1.1.20/24 fe80::f529:7e26:4c51:56cc/64 ens37 UP 10.1.2.20/24 fe80::1334:fcc3:8f69:5e5a/64 [root@nfs ~ 15:28:08]# ping 1.1.1.1 PING 1.1.1.1 (1.1.1.1) 56(84) bytes of data. 64 bytes from 1.1.1.1: icmp_seq=2 ttl=127 time=144 ms 64 bytes from 1.1.1.1: icmp_seq=3 ttl=127 time=186 ms 64 bytes from 1.1.1.1: icmp_seq=4 ttl=127 time=390 ms 64 bytes from 1.1.1.1: icmp_seq=5 ttl=127 time=145 ms ......
hostnamectl set-hostname ha1.lyk.cloud nmcli connection modify ens33 ipv4.addresses 10.1.8.14 ipv4.gateway 10.1.8.20 nmcli connection up ens33 bashhostnamectl set-hostname ha2.lyk.cloud nmcli connection modify ens33 ipv4.addresses 10.1.8.15 ipv4.gateway 10.1.8.20 nmcli connection up ens33 bash
[root@client2 ~ 15:52:26]# vim /etc/hosts
[root@client2 ~ 15:52:35]# cat /etc/hosts
127.0.0.1 localhost localhost.localdomain localhost4 localhost4.localdomain4
::1 localhost localhost.localdomain localhost6 localhost6.localdomain6########### cluster ################
10.1.8.100 www.laoma.cloud www
10.1.8.10 lb.laoma.cloud lb
10.1.8.11 web1.laoma.cloud web1
10.1.8.12 web2.laoma.cloud web2
10.1.8.13 web3.laoma.cloud web3
10.1.8.14 ha1.laoma.cloud ha1
10.1.8.15 ha2.laoma.cloud ha2
10.1.8.20 router.laoma.cloud router
10.1.8.21 client1.laoma.cloud client1
10.1.1.21 client2.laoma.cloud client2#client2推送到其他节点
[root@client2 ~ 15:59:47]# for host in 10.1.8.1{1..5} 10.1.8.20 10.1.8.21 ; do scp /etc/hosts $host:/etc/hosts; done
六、NFS 共享存储部署（nfs 节点与 web 节点）
6.1 NFS 服务器配置（nfs 节点）
6.1.1 安装 NFS 服务并创建共享目录
web三个加网卡，vm （仅主机2）
# 安装软件 web1 2 3 nfs yum install -y nfs-utils#再次添加一行nfs 再次推送 [root@client2 ~ 16:01:05]# vim /etc/hosts [root@client2 ~ 16:00:37]# cat /etc/hosts 127.0.0.1 localhost localhost.localdomain localhost4 localhost4.localdomain4 ::1 localhost localhost.localdomain localhost6 localhost6.localdomain6########### cluster ################ 10.1.8.100 www.laoma.cloud www 10.1.8.10 lb.laoma.cloud lb 10.1.8.11 web1.laoma.cloud web1 10.1.8.12 web2.laoma.cloud web2 10.1.8.13 web3.laoma.cloud web3 10.1.8.14 ha1.laoma.cloud ha1 10.1.8.15 ha2.laoma.cloud ha2 10.1.8.20 router.laoma.cloud router 10.1.8.21 client1.laoma.cloud client1 10.1.1.21 client2.laoma.cloud client2 10.1.2.100 nfs.laoma.cloud nfs
安装 NFS 服务并创建共享目录
[root@nfs ~ 16:19:02]# mkdir /var/www/html/ -p [root@nfs ~ 16:19:55]# echo Welcome to www.lyk.cloud > /var/www/html/index.html [root@nfs ~ 16:20:01]# echo '/var/www 10.1.2.0/24(rw,sync)' >> /etc/exports [root@nfs ~ 16:27:46]# systemctl restart nfs-server [root@nfs ~ 16:20:06]# systemctl enable nfs-server.service --now [root@nfs ~ 16:21:12]# systemctl status nfs[root@web1 ~ 16:22:14]# systemctl disable nginx.service --now Removed symlink /etc/systemd/system/multi-user.target.wants/nginx.service. [root@web1 ~ 16:26:11]# systemctl start httpd[root@web1 ~ 16:22:14]# systemctl disable nginx.service --now Removed symlink /etc/systemd/system/multi-user.target.wants/nginx.service. [root@web1 ~ 16:26:11]# systemctl start httpd[root@web2 ~ 16:26:32]# yum install -y httpd [root@web2 ~ 16:26:54]# echo Welcome to $(hostname) > /var/www/html/index.html [root@web2 ~ 16:26:54]# systemctl enable httpd.service --now Created symlink from /etc/systemd/system/multi-user.target.wants/httpd.service to /usr/lib/systemd/system/httpd.service.[root@web3 ~ 16:27:03]# systemctl disable nginx --now Removed symlink /etc/systemd/system/multi-user.target.wants/nginx.service. [root@web3 ~ 16:27:09]# systemctl enable httpd.service --now#验证 NFS 共享（web1 节点） [root@web1 ~ 16:28:00]# showmount -e nfs Export list for nfs: /var/www 10.1.2.0/24# Web 节点挂载 NFS（web1、web2、web3 节点） # 配置永久挂载（/etc/fstab） [root@web1 ~ 16:35:19]# vim /etc/fstab #最后一行添加 【】echo 'nfs.laoma.cloud:/var/www /var/www/ nfs defaults 0 0' >> /etc/fstab #或者 [root@web1 ~ 17:01:10]# cat /etc/fstab # # /etc/fstab # Created by anaconda on Fri Aug 1 15:45:32 2025 # # Accessible filesystems, by reference, are maintained under '/dev/disk' # See man pages fstab(5), findfs(8), mount(8) and/or blkid(8) for more info # /dev/mapper/centos-root / xfs defaults 0 0 UUID=b54b3764-2b2b-4a76-a0ec-83e308071ae5 /boot xfs defaults 0 0 /dev/mapper/centos-swap swap swap defaults 0 0 nfs.laoma.cloud:/var/www /var/www/ nfs defaults 0 0[root@web1 ~ 17:01:10]# mount /var/www/ [root@web1 ~ 17:02:20]# df -h /var/www/ 文件系统 容量 已用 可用 已用% 挂载点 nfs.laoma.cloud:/var/www 50G 1.6G 49G 4% /var/www#web2 3 [root@web2 ~ 16:26:54]# echo 'nfs.laoma.cloud:/var/www /var/www/ nfs defaults 0 0' >> /etc/fstab [root@web2 ~ 16:42:10]# mount /var/www/ [root@web2 ~ 17:10:11]# df -h /var/www/ 文件系统 容量 已用 可用 已用% 挂载点 nfs.laoma.cloud:/var/www 50G 1.6G 49G 4% /var/www[root@web3 ~ 16:26:54]# echo 'nfs.laoma.cloud:/var/www /var/www/ nfs defaults 0 0' >> /etc/fstab [root@web3 ~ 16:42:10]# mount /var/www/ [root@web3 ~ 17:10:11]# df -h /var/www/ 文件系统 容量 已用 可用 已用% 挂载点 nfs.laoma.cloud:/var/www 50G 1.6G 49G 4% /var/www#验证 [root@client2 ~ 16:58:02]# curl 10.1.8.13 Welcome to www.lyk.cloud [root@client2 ~ 17:12:12]# curl 10.1.8.12 Welcome to www.lyk.cloud [root@client2 ~ 17:12:15]# curl 10.1.8.11 Welcome to www.lyk.cloud
七、HAProxy 负载均衡部署（ha1、ha2 节点）
7.1 HAProxy 基础配置（ha1、ha2 节点）
[root@web1 ~ 17:09:54]# umount /var/www [root@web2 ~ 17:09:54]# umount /var/www [root@web3 ~ 17:09:54]# umount /var/www #验证 [root@client2 ~ 17:12:17]# curl 10.1.8.11 Welcome to web1.laoma.cloud [root@client2 ~ 17:14:47]# curl 10.1.8.12 Welcome to web2.laoma.cloud [root@client2 ~ 17:14:50]# curl 10.1.8.13 Welcome to web3.laoma.cloud
7.1.1 安装 HAProxy 并备份配置
[root@ha1-2 ~]# yum install -y haproxy
# 备份 haproxy 配置文件
cp /etc/haproxy/haproxy.cfg{,.ori}# 修改 haproxy 配置文件，最后添加以下内容
echo '
########### web 代理 ###########
frontend http_frontbind *:80use_backend http_back
backend http_backbalance roundrobinserver node1 10.1.8.11:80 checkserver node2 10.1.8.12:80 checkserver node3 10.1.8.13:80 check
' >> /etc/haproxy/haproxy.cfg# 启用并启动服务
systemctl enable haproxy.service --now#验证 HAProxy 负载效果（client2 节点）
[root@client2 ~ 17:14:51]# curl 10.1.8.14
Welcome to web1.laoma.cloud
[root@client2 ~ 17:24:28]# curl 10.1.8.14
Welcome to web2.laoma.cloud
[root@client2 ~ 17:24:30]# curl 10.1.8.14
Welcome to web3.laoma.cloud
7.2 HAProxy+Keepalived 高可用配置（ha1 为主、ha2 为备）
7.2.1 主节点配置（ha1 节点）
[root@ha1 ~ 17:25:35]# cp /etc/keepalived/keepalived.conf{,.bak}
[root@ha1 ~ 17:25:35]# vim /etc/keepalived/keepalived.conf! Configuration File for keepalivedglobal_defs {router_id ha1
}vrrp_instance nginx {state MASTERinterface ens33virtual_router_id 51priority 110advert_int 1authentication {auth_type PASSauth_pass lyk@123}virtual_ipaddress {10.1.8.100/24}
}[root@ha1 ~ 17:26:07]# systemctl enable keepalived.service --now
Created symlink from /etc/systemd/system/multi-user.target.wants/keepalived.service to /usr/lib/systemd/system/keepalived.service.7.2.2 备节点配置（ha2 节点）
yum install -y keepalived ipvsadm
cp /etc/keepalived/keepalived.conf{,.bak}
vim /etc/keepalived/keepalived.conf! Configuration File for keepalivedglobal_defs {router_id ha2
}vrrp_instance nginx {state BACKUPinterface ens33virtual_router_id 51priority 100advert_int 1authentication {auth_type PASSauth_pass lyk@123}virtual_ipaddress {10.1.8.100/24}
}systemctl enable keepalived.service --now八、集群功能与高可用测试
8.1 功能性测试（client2 节点）
[root@client1 ~ 17:30:02]# while true ;do curl -s www.laoma.cloud;sleep 1;done 1 2 3轮流 #好像没效果 [root@ha1 ~ 17:26:12]# systemctl stop keepalived.service#停web3 [root@web3 ~ 17:20:45]# systemctl stop httpd.service Welcome to web1.laoma.cloud Welcome to web2.laoma.cloud...... 没有web3了#web1 挂载 [root@web1 ~ 17:19:57]# mount -a Welcome to web2.laoma.cloud Welcome to www.lyk.cloud Welcome to web2.laoma.cloud .... web1变成www
#恢复web1 3 [root@web1 ~ 17:35:32]# umount /var/www [root@web3 ~ 17:34:32]# systemctl restart httpd.service
8.2 高可用切换测试
8.2.1 负载均衡主节点故障测试
[root@client2 ~ 17:29:31]# while true ;do curl -s http://10.1.8.100;sleep 1;done Welcome to web3.laoma.cloud Welcome to web2.laoma.cloud Welcome to web1.laoma.cloud Welcome to web3.laoma.cloud ......
测试1：停止 ha1 上 keepalived 服务。
[root@ha1 ~ 17:32:39]# systemctl stop keepalived.service[root@client2 ~ 17:29:31]# while true ;do curl -s http://10.1.8.100;sleep 1;done Welcome to web3.laoma.cloud Welcome to web2.laoma.cloud Welcome to web1.laoma.cloud Welcome to web3.laoma.cloud ...... 无影响
结果：客户端无感知故障，正常访问集群。
测试2：恢复 ha1 上 keepalived 服务。
[root@ha1 ~ 17:46:04]# systemctl start keepalived.service **结果**：客户端无感知故障，正常访问集群。
负载均衡测试
测试1：停止 web2 上 httpd.service ，监控客户端访问情况。
[root@web2 ~ 17:48:27]# systemctl stop httpd.service
结果：大概 15 秒，LVS 将 web2 从后端虚拟主机中剔除
测试2：启动 web2 上 httpd.service，监控客户端访问情况。
[root@web2 ~]# systemctl start httpd.service
结果：大概 5 秒，LVS将web2加入后端虚拟主机中。