DHCP服务器只作用于局域网同一网段内，客户端是通过广播消息来获得DHCP服务器响应后才能得到IP地址的，但广播消息不能跨越子网，那么如何让客户端获取到DHCP服务器提供的IP地址呢？这就是DHCP中继服务器的功能了。
实验拓扑图：
实验环境：
DHCP服务器：CentOS Linux release 7.3.1611
DHCP中继代理：CentOS Linux release 7.3.1611
客户端：CentOS 6.9
1、DHCP服务器配置
（1）、配置DHCP服务器的网络参数
[root@host02 ~]# vi /etc/sysconfig/network-scripts/ifcfg-ens33 TYPE="Ethernet"
BOOTPROTO="static"
IPADDR=192.168.8.250
NETMASK=255.255.255.0
GATEWAY=192.168.8.254
DEFROUTE="yes"
PEERDNS="yes"
PEERROUTES="yes"
IPV4_FAILURE_FATAL="no"
IPV6INIT="yes"
IPV6_AUTOCONF="yes"
IPV6_DEFROUTE="yes"
IPV6_PEERDNS="yes"
IPV6_PEERROUTES="yes"
IPV6_FAILURE_FATAL="no"
IPV6_ADDR_GEN_MODE="stable-privacy"
NAME="ens33"
UUID="8225206d-ab64-4101-b343-35316fc76a7c"
DEVICE="ens33"
ONBOOT="yes"
~
~
~
（2）、重启网络服务
[root@host02 ~]# systemctl restart network
查看DHCP服务器IP
[root@host02 ~]# ifconfig
ens33: flags=<UP,BROADCAST,RUNNING,MULTICAST> mtu
inet 192.168.8.250 netmask 255.255.255.0 broadcast 192.168.8.255
inet6 fe80::66d2:4ff4:82d6:a6e4 prefixlen scopeid 0x20<link>
ether :0c:::7e:d0 txqueuelen (Ethernet)
RX packets bytes (18.3 MiB)
RX errors dropped overruns frame
TX packets bytes (5.6 MiB)
TX errors dropped overruns carrier collisions lo: flags=<UP,LOOPBACK,RUNNING> mtu
inet 127.0.0.1 netmask 255.0.0.0
inet6 :: prefixlen scopeid 0x10<host>
loop txqueuelen (Local Loopback)
RX packets bytes (30.4 KiB)
RX errors dropped overruns frame
TX packets bytes (30.4 KiB)
TX errors dropped overruns carrier collisions virbr0: flags=<UP,BROADCAST,MULTICAST> mtu
inet 192.168.122.1 netmask 255.255.255.0 broadcast 192.168.122.255
ether :::f7::1e txqueuelen (Ethernet)
RX packets bytes (432.0 B)
RX errors dropped overruns frame
TX packets bytes (0.0 B)
TX errors dropped overruns carrier collisions
（3） 安装DHCP服务
[root@host02 ~]# yum install dhcp
[root@host02 ~]# rpm -qa dhcp
dhcp-4.2.-.el7.centos.x86_64
（4）编辑DHCP服务器的配置文件
[root@host02 ~]# vi dhcpd.conf #dhcp configuration #整体环境设置 ddns-update-style none;
ignore client-updates;
default-lease-time ;
max-lease-time ;
option routers 192.168.8.254;
option domain-name "centos.me";
option domain-name-servers 202.102.224.68,202.102.227.68; #二个网段IP分配 subnet 192.168.8.0 netmask 255.255.255.0 {
range 192.168.8.200 192.168.8.210;
option subnet-mask 255.255.255.0;
option routers 192.168.8.1;
}
subnet 192.168.100.0 netmask 255.255.255.0 {
range 192.168.100.80 192.168.100.100;
option subnet-mask 255.255.255.0;
option routers 192.168.100.1;
}
将编辑好的DHCP配置文件覆盖初装后的空白配置文件;
[root@host02 ~]# cp dhcpd.conf /etc/dhcp/dhcpd.conf
cp: overwrite ‘/etc/dhcp/dhcpd.conf’? y
（5）启动DHCP服务，并检查状态；
[root@host02 ~]# systemctl start dhcpd
[root@host02 ~]# systemctl status dhcpd
● dhcpd.service - DHCPv4 Server Daemon
Loaded: loaded (/usr/lib/systemd/system/dhcpd.service; disabled; vendor preset: disabled)
Active: active (running) since Thu -- :: CST; 6s ago
Docs: man:dhcpd()
man:dhcpd.conf()
Main PID: (dhcpd)
Status: "Dispatching packets..."
CGroup: /system.slice/dhcpd.service
└─ /usr/sbin/dhcpd -f -cf /etc/dhcp/dhcpd.conf -user dhcpd -group dhcpd --no-pid Aug :: host02.linux.com dhcpd[]: No subnet declaration for virbr0 (192.168.122.1).
Aug :: host02.linux.com dhcpd[]: ** Ignoring requests on virbr0. If this is not what
Aug :: host02.linux.com dhcpd[]: you want, please write a subnet declaration
Aug :: host02.linux.com dhcpd[]: in your dhcpd.conf file for the network segment
Aug :: host02.linux.com dhcpd[]: to which interface virbr0 is attached. **
Aug :: host02.linux.com dhcpd[]:
Aug :: host02.linux.com dhcpd[]: Listening on LPF/ens33/:0c:::7e:d0/192.168.8.0/
Aug :: host02.linux.com dhcpd[]: Sending on LPF/ens33/:0c:::7e:d0/192.168.8.0/
Aug :: host02.linux.com dhcpd[]: Sending on Socket/fallback/fallback-net
Aug :: host02.linux.com systemd[]: Started DHCPv4 Server Daemon.
[root@host02 ~]# netstat -tunlp |grep dhcp
udp 0.0.0.0: 0.0.0.0:* /dhcpd
udp 0.0.0.0: 0.0.0.0:* /dhcpd
udp6 ::: :::* /dhcpd
DHCP服务器一切正常！至此DHCP服务器的配置已经完成；下一步配置DHCP中继代理服务器。
2、配置DHCP中继服务器
（1）、做为中继服务器首先要确保有两张网卡，否则没法中继；
本实验我已经在中继服务器上虚拟出两个网卡；
[root@host-a ~]# lspci |grep ethernet
[root@host-a ~]# lspci |grep Ethernet
:01.0 Ethernet controller: Intel Corporation 82545EM Gigabit Ethernet Controller (Copper) (rev )
:05.0 Ethernet controller: Intel Corporation 82545EM Gigabit Ethernet Controller (Copper) (rev )
（2）查看网卡信息
[root@host-a ~]# ifconfig
ens33: flags=<UP,BROADCAST,RUNNING,MULTICAST> mtu
ether :0c::a9::dd txqueuelen (Ethernet)
RX packets bytes (561.4 KiB)
RX errors dropped overruns frame
TX packets bytes (78.5 KiB)
TX errors dropped overruns carrier collisions ens37: flags=<UP,BROADCAST,RUNNING,MULTICAST> mtu
ether :0c::a9::e7 txqueuelen (Ethernet)
RX packets bytes (2.8 KiB)
RX errors dropped overruns frame
TX packets bytes (68.6 KiB)
TX errors dropped overruns carrier collisions lo: flags=<UP,LOOPBACK,RUNNING> mtu
inet 127.0.0.1 netmask 255.0.0.0
inet6 :: prefixlen scopeid 0x10<host>
loop txqueuelen (Local Loopback)
RX packets bytes (12.5 KiB)
RX errors dropped overruns frame
TX packets bytes (12.5 KiB)
TX errors dropped overruns carrier collisions virbr0: flags=<UP,BROADCAST,MULTICAST> mtu
inet 192.168.122.1 netmask 255.255.255.0 broadcast 192.168.122.255
ether :::7a:2c:f4 txqueuelen (Ethernet)
RX packets bytes (0.0 B)
RX errors dropped overruns frame
TX packets bytes (0.0 B)
TX errors dropped overruns carrier collisions
（2）接下来分别配置这两个网卡的网络参数
网卡ens33与DHCP服务器相连，属于192.168.8.0/24网段，注意网卡ens33的IP地址是DHCP服务器的网关地址，
配置如下：
[root@host-a ~]# vi /etc/sysconfig/network-scripts/ifcfg-ens33
TYPE="Ethernet"
BOOTPROTO="static"
IPADDR=192.168.8.254
NETMASK=255.255.255.0
DEFROUTE="yes"
PEERDNS="yes"
PEERROUTES="yes"
IPV4_FAILURE_FATAL="no"
IPV6INIT="yes"
IPV6_AUTOCONF="yes"
IPV6_DEFROUTE="yes"
IPV6_PEERDNS="yes"
IPV6_PEERROUTES="yes"
IPV6_FAILURE_FATAL="no"
IPV6_ADDR_GEN_MODE="stable-privacy"
NAME="ens33"
UUID="065ff1ad-9117-4cf4-a716-c61d5fdc0c00"
DEVICE="ens33"
ONBOOT="yes"
网卡ens37与192.168.100.0/24网段相连，注意网卡ens37的IP地址是192.168.100.0/24网段的网关地址，配置如下：
[root@host-a ~]# cat /etc/sysconfig/network-scripts/ifcfg-ens37
TYPE=Ethernet
BOOTPROTO=static
IPADDR=192.168.100.1
PREFIX=
DEFROUTE=yes
IPV4_FAILURE_FATAL=no
IPV6INIT=yes
IPV6_AUTOCONF=yes
IPV6_DEFROUTE=yes
IPV6_PEERDNS=yes
IPV6_PEERROUTES=yes
IPV6_FAILURE_FATAL=no
IPV6_ADDR_GEN_MODE=stable-privacy
NAME="ens37"
UUID=e3854409-f832-4d4c-b7ae-271d0aa5ecc5
ONBOOT=yes
（3）配置好后，重起网络服务使设置生效
[root@host-a ~]# systemctl restart network
[root@host-a ~]# ifconfig
ens33: flags=<UP,BROADCAST,RUNNING,MULTICAST> mtu
inet 192.168.8.254 netmask 255.255.255.0 broadcast 192.168.8.255
inet6 fe80::f31:c380:a398:1df6 prefixlen scopeid 0x20<link>
ether :0c::a9::dd txqueuelen (Ethernet)
RX packets bytes (562.4 KiB)
RX errors dropped overruns frame
TX packets bytes (91.6 KiB)
TX errors dropped overruns carrier collisions ens37: flags=<UP,BROADCAST,RUNNING,MULTICAST> mtu
inet 192.168.100.1 netmask 255.255.255.0 broadcast 192.168.100.255
inet6 fe80:::4bde:b57c:957d prefixlen scopeid 0x20<link>
ether :0c::a9::e7 txqueuelen (Ethernet)
RX packets bytes (2.8 KiB)
RX errors dropped overruns frame
TX packets bytes (76.6 KiB)
TX errors dropped overruns carrier collisions lo: flags=<UP,LOOPBACK,RUNNING> mtu
inet 127.0.0.1 netmask 255.0.0.0
inet6 :: prefixlen scopeid 0x10<host>
loop txqueuelen (Local Loopback)
RX packets bytes (12.5 KiB)
RX errors dropped overruns frame
TX packets bytes (12.5 KiB)
TX errors dropped overruns carrier collisions virbr0: flags=<UP,BROADCAST,MULTICAST> mtu
inet 192.168.122.1 netmask 255.255.255.0 broadcast 192.168.122.255
ether :::7a:2c:f4 txqueuelen (Ethernet)
RX packets bytes (0.0 B)
RX errors dropped overruns frame
TX packets bytes (0.0 B)
TX errors dropped overruns carrier collisions
（4）安装DHCP
#安装
[root@host-a network-scripts]# yum install dhcp
#安装成功
[root@host-a network-scripts]# rpm -qa dhcp
dhcp-4.2.-.el7.centos.x86_64
#注意dhcp包含dhcp中继服务
[root@host-a ~]# rpm -ql dhcp |grep dhcrelay
/usr/lib/systemd/system/dhcrelay.service
/usr/sbin/dhcrelay
/usr/share/man/man8/dhcrelay..gz
（5）打开DHCP中继服务器路由转发功能
编辑/etc/sysctl.conf 文件在文件添加一行内容为：
“net.ipv4.ip_forward=1”
开启路由转发功能
[root@host-a ~]# vi /etc/sysctl.conf
# sysctl settings are defined through files in
# /usr/lib/sysctl.d/, /run/sysctl.d/, and /etc/sysctl.d/.
#
# Vendors settings live in /usr/lib/sysctl.d/.
# To override a whole file, create a new file with the same in
# /etc/sysctl.d/ and put new settings there. To override
# only specific settings, add a file with a lexically later
# name in /etc/sysctl.d/ and put new settings there.
#
# For more information, see sysctl.conf() and sysctl.d().
net.ipv4.ip_forward =
使设置立即生效，执行以下命令：
[root@host-a ~]# sysctl -p
net.ipv4.ip_forward = 1
查看内核参数设置，确认生效；
[root@host-a ~]# sysctl -a |grep ip_forward
net.ipv4.ip_forward = 1
net.ipv4.ip_forward_use_pmtu = 0
[root@host-a ~]#
（6）开启DHCP中继服务
[root@host-a ~]# dhcrelay 192.168.8.250
Dropped all unnecessary capabilities.
Internet Systems Consortium DHCP Relay Agent 4.2.
Copyright - Internet Systems Consortium.
All rights reserved.
For info, please visit https://www.isc.org/software/dhcp/
Listening on LPF/virbr0/:::7a:2c:f4
Sending on LPF/virbr0/:::7a:2c:f4
Listening on LPF/ens37/:0c::a9::e7
Sending on LPF/ens37/:0c::a9::e7
Listening on LPF/ens33/:0c::a9::dd
Sending on LPF/ens33/:0c::a9::dd
Sending on Socket/fallback
[root@host-a ~]# [root@host-a ~]# netstat -tunlp |grep dhcrelay
udp 0.0.0.0: 0.0.0.0:* /dhcrelay
udp 0.0.0.0: 0.0.0.0:* /dhcrelay
udp6 ::: :::* /dhcrelay
3、客户端测试
（1）配置客户端网络为自动获取IP
（2）重启网络服务
（3）查看IP
（4）查看客户端在重启网络时都做了什么
（5）查看DHCP服务器日志文件
[root@host02 ~]# tail -n /var/log/messages
Aug :: host02 systemd: Stopping user-.slice.
Aug :: host02 dhcpd: DHCPREQUEST for 192.168.6.6 from :0c:::4c:ea via 192.168.100.1: ignored (not authoritative).
Aug :: host02 dhcpd: DHCPREQUEST for 192.168.6.6 from :0c:::4c:ea via 192.168.100.1: ignored (not authoritative).
Aug :: host02 dhcpd: DHCPDISCOVER from :0c:::4c:ea via 192.168.100.1
Aug :: host02 dhcpd: DHCPOFFER on 192.168.100.80 to :0c:::4c:ea (host01) via 192.168.100.1
Aug :: host02 dhcpd: DHCPREQUEST for 192.168.100.80 (192.168.8.250) from :0c:::4c:ea (host01) via 192.168.100.1
Aug :: host02 dhcpd: DHCPACK on 192.168.100.80 to :0c:::4c:ea (host01) via 192.168.100.1
Aug :: host02 dhcpd: DHCPREQUEST for 192.168.100.80 (192.168.8.250) from :0c:::4c:ea (host01) via 192.168.100.1
Aug :: host02 dhcpd: DHCPACK on 192.168.100.80 to :0c:::4c:ea (host01) via 192.168.100.1
(6)检查客户端获取的DNS是否正确
DNS与DHCP服务器设置的一致！
可以看到客户端已经可以正常获取DHCP服务器设置的网络参数了！！