Introduction to DHCP service

Posted by agret on Mon, 24 Jan 2022 13:13:29 +0100

1. Introduction to DHCP

DHCP: Dynamic Host Configuration Protocol, which is a local area network protocol working in the application layer. UDP is used for data transmission. It works with unreliable transmission protocol. It is usually used in large-scale local area network environment. Its main function is to centrally manage and allocate network resources;

In the network environment, not all devices can dynamically obtain ip addresses, gateways, etc,

If DHCP is assigned to the client, the speed limit can be unified and the device can be managed;

In fact, the operation is very simple, that is, turn on the device – > automatically obtain the IP address,

2. How DHCP works

Lease Trilogy

The client is connected to the network environment

Then it will issue DHCP discover, which will find whether there is a dhcp server;
Simple understanding: the client is shouting: can anyone assign an ip address
And it uses 0.0.0.0 as a bit source address; Use 255.255.255.255 as the destination address
Who is the DHCP server? You can give me an ip address
I'm a client. Use MAC address or computer name to confirm yourself;

Server side:

With this DHCP service, it found that someone asked if there was a DHCP device
So they will send him a DHCP Offer and ask him if this is OK, accept it or not, and so on
When the DHCP server receives the ip address information requested by the client,
He will find out whether there is a legal IP address in his IP resource pool,
He will record the ip address and package it into the offer package
- offer package: the mac address of the customer;
- There are legal IP, subnet, mask, default gateway, etc
- Also have their own ip address, and mac address, tell the server;
- It will use UDP 68 to broadcast these messages;

client:

After receiving the offer, he just sends another request to the server
The client finally received the offer information from the server;
- Then the IP address will be selected,
- After you have selected it, you will reply to a requested message,
The client sends the requested information content
- Including the ip he wants to select. He wants to tell the dhcp server that it wants to use this ip
- However, the client still does not use this ip address, and it will continue to use 0.0.0.0 as the source address; Then it is sent to the dhcp server to broadcast the request information;

The server:

Received the request from the client. OK, I'll send you an ack confirmation message;
After the server receives the request information from the client, it will judge whether this address is used
- If someone uses it, they will send NACK reply to reject it
- If no one uses it, all right, here you are

Renewal

The client and server have reached an agreement before, and it is already using this dhcpip

Because dhcp ip has a lease term;

If the client is 50% past the lease term;

The client will send the message packet of DHCP request to the server in time;

If DHCP does not reply, the longest lease will start

client

When the lease reaches 50%, it sends a request to the server
If you receive a reply from the server, it will update the lease term; So continue to use
If no reply is received from the server
- Then the client will continue to use it because it still has 50% of the time
- Until 87.5%, it will still send request information to the dhcp server
- What if you haven't received a reply? When it reaches 100%, it will directly discard the ip
- Then broadcast the request and try every 5 minutes;

3. DHCP service setup

Preparation environment:

Two servers:

The network mode is vmnet mode
192.168.75.40: HDCP server
192.168.75.41: acting as a client
Turn off firewall and selinux
Be sure to turn off the dhcp function of vmware;

[root@master ~]# iptables -F
[root@master ~]# service iptables stop
iptables: Set chain as policy ACCEPT: filter                    [determine]
iptables: Clear firewall rules:                                 [determine]
iptables: Uninstalling module:                                   [determine]
[root@master ~]# getenforce 
Disabled
[root@slave41 ~]# setenforce 0
setenforce: SELinux is disabled

DHCP server configuration

DHCP information

Software name:
- dhcp: dhcp Server package
- dhcp common: installation package of dhcp related commands (installed by default)
service name
- dhcp PD: dhcp service name
- dhcp relay: dhcp intermediate service name
Port number:
- udp: 67 – > the target port of the client, which accepts the request of the client and DHCP request
- udp: 68 – > can be the source port of the server, which is used to recover packets from the client
Profile:
- /etc/dhcp/dhcpd.conf: the default is an empty configuration file
- /usr/share/doc/dhcp-4.1.1/dhcpd6.conf.sample: the default template file for the configuration file
- /etc/sysconfig/dhcrelay: the configuration file used in the relay experiment

/etc/dhcp/dhcpd. Explanation of conf configuration file

When you see conf.sample, it is basically the template file of the configuration file

Note: at least one subnet segment is consistent with the local ip segment, otherwise an error will be reported

There are too many semicolons at the end of each statement; End, or you will report an error

#Check it out
[root@master ~]# vim /etc/dhcp/dhcpd.conf 
# DHCP Server Configuration file.
#   see /usr/share/doc/dhcp*/dhcpd.conf.sample		#Here you are prompted to take a look at the configuration template file
#   see 'man 5 dhcpd.conf'						#The methods are all given to you. You can use the man 5 configuration file

#Copy the template file to the real configuration file
[root@master dhcp-4.1.1]# cp -a dhcpd.conf.sample /etc/dhcp/dhcpd.conf 
cp: Overwrite"/etc/dhcp/dhcpd.conf"？ y


#Real configuration file explanation
[root@master ~]# vim /etc/dhcp/dhcpd.conf 
	#Global configuration
default-lease-time 600;			#The default lease is s
max-lease-time 7200;			#Maximum lease by s
log-facility local7;			#The log storage device is actually under / var/log/message



............
	#The declared resource network segment and subnet mask. At least one of the network segments declared by dhcp is the same as the network segment used by the local machine;
subnet 192.168.75.0 netmask 255.255.255.0 {		
	##Declare the available IP address pool and the range that can be given to the client
		#This network segment should be noted that broadcast address and gateway address cannot be assigned
  range 192.168.75.100 192.168.75.110;
  	#Here is the dns address specified
  option domain-name-servers 114.114.114.114;
  	#dns domain name is represented here
  option domain-name "liangjiawei.net";
  	#Gateway address, which specifies the gateway address assigned to the client
  option routers 192.168.75.2;
  	#This is the broadcast address
  option broadcast-address 192.168.75.255;
  	#Default lease
  default-lease-time 600;
  	#Maximum lease time
  max-lease-time 7200;
}





	#Assign a fixed ip to a host for ip address binding;
host fantasia {
  hardware ethernet 08:00:07:26:c0:a5;
  fixed-address fantasia.fugue.com;
}




	#Superscope
class "foo" {
  match if substring (option vendor-class-identifier, 0, 4) = "SUNW";
}

shared-network 224-29 {
  subnet 10.17.224.0 netmask 255.255.255.0 {
    option routers rtr-224.example.org;
  }
  subnet 10.0.29.0 netmask 255.255.255.0 {
    option routers rtr-29.example.org;
  }
  pool {
    allow members of "foo";
    range 10.17.224.10 10.17.224.250;
  }
  pool {
    deny members of "foo";
    range 10.0.29.10 10.0.29.230;
  }
}

DHCP basic function test

dhcp server configuration

#Install software
[root@master ~]# yum -y install dhcp dhcp-common

#Prepare profile
[root@master ~]# cp -a dhcpd.conf.sample /etc/dhcp/dhcpd.conf 

#Modify profile
[root@master ~]# cp -a /etc/dhcp/dhcpd.conf  /etc/dhcp/dhcpd.conf.bak
[root@master ~]# vim /etc/dhcp/dhcpd.conf 
	#Put all the previous notes
subnet 192.168.75.0 netmask 255.255.255.0 {
  range 192.168.75.100 192.168.75.110;
  #option domain-name-servers 114.114.114.114;
  #option domain-name "liangjiawei.net";
  option routers 192.168.75.2;
  #option broadcast-address 192.168.75.255;
  default-lease-time 600;
  max-lease-time 7200;
}


#start-up
[root@master ~]# service dhcpd start
 Starting dhcpd:                                            [determine]


#Check whether port 67 is enabled
[root@master ~]# netstat -ntulp 
Active Internet connections (only servers)
Proto Recv-Q Send-Q Local Address               Foreign Address             State       PID/Program name   
tcp        0      0 0.0.0.0:22                  0.0.0.0:*                   LISTEN      1118/sshd           
tcp        0      0 127.0.0.1:25                0.0.0.0:*                   LISTEN      1197/master         
tcp        0      0 :::22                       :::*                        LISTEN      1118/sshd           
tcp        0      0 ::1:25                      :::*                        LISTEN      1197/master         
udp        0      0 0.0.0.0:67                  0.0.0.0:*                               1422/dhcpd

Check the log file on the server side

[root@master ~]# tail -10 /var/log/messages 
..........
May  5 10:28:45 master dhcpd: Sending on   Socket/fallback/fallback-net
	#The logs are recorded here. dhcp finds that there are requests from this mac address
May  5 10:31:05 master dhcpd: DHCPDISCOVER from 00:0c:29:57:c1:ff via eth0		
	#dhcp gives an offer
May  5 10:31:06 master dhcpd: DHCPOFFER on 192.168.75.100 to 00:0c:29:57:c1:ff (slave41) via eth0  
	#There's something here. I got a request
May  5 10:31:06 master dhcpd: DHCPREQUEST for 192.168.75.100 (192.168.75.40) from 00:0c:29:57:c1:ff (slave41) via eth0
	#ack is restored here
May  5 10:31:06 master dhcpd: DHCPACK on 192.168.75.100 to 00:0c:29:57:c1:ff (slave41) via eth0

Client authentication

#Modify profile
[root@slave41 ~]# vim /etc/sysconfig/network-scripts/ifcfg-eth0 
DEVICE=eth0
TYPE=Ethernet
ONBOOT=yes
BOOTPROTO=dhcp				#This is modified to dhcp
#IPADDR=192.168.75.41		#Note original
#PREFIX=24
#GATEWAY=192.168.75.2
#DNS1=114.114.114.114
NAME="System eth0"

#Then restart the network check
	#The original has changed, so the line is broken directly;
[c:\~]$ ssh root@192.168.75.100


Connecting to 192.168.75.100:22...
Connection established.
To escape to local shell, press 'Ctrl+Alt+]'.

Last login: Wed May  5 10:31:16 2021
[root@slave41 ~]# ip a 
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN 
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 00:0c:29:57:c1:ff brd ff:ff:ff:ff:ff:ff
    inet 192.168.75.100/24 brd 192.168.75.255 scope global eth0
    inet6 fe80::20c:29ff:fe57:c1ff/64 scope link 
       valid_lft forever preferred_lft forever

DHCP (fixed address assignment test)

Scenario:

Some servers need to be bound with fixed ip,

This is the requirement of the client;

Every time you want to get dhcp, you will get a fixed ip address

The server keeps a fixed address for the client;

Server configuration

#View the mac address of the client
[root@slave41 ~]# arp -a
? (192.168.75.1) at 00:50:56:c0:00:08 [ether] on eth0
? (192.168.75.40) at 00:0c:29:fd:e7:9f [ether] on eth0

[root@slave41 ~]# ifconfig 
eth0      Link encap:Ethernet  HWaddr 00:0C:29:57:C1:FF  
          inet addr:192.168.75.100  Bcast:192.168.75.255  Mask:255.255.255.0
          inet6 addr: fe80::20c:29ff:fe57:c1ff/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:618 errors:0 dropped:0 overruns:0 frame:0
          TX packets:417 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000 
          RX bytes:55303 (54.0 KiB)  TX bytes:48062 (46.9 KiB)


#Modify configuration file directly
[root@master ~]# vim /etc/dhcp/dhcpd.conf
subnet 192.168.75.0 netmask 255.255.255.0 {
  range 192.168.75.100 192.168.75.110;
  #option domain-name-servers 114.114.114.114;
  #option domain-name "liangjiawei.net";
  option routers 192.168.75.2;
  #option broadcast-address 192.168.75.255;
  default-lease-time 600;
  max-lease-time 7200;
}

................
host fantasia {
  hardware ethernet 00:0C:29:57:C1:FF;
  fixed-address 192.168.75.120;					#This can be on an ip other than the address
}


#First restart the dhcp service
[root@master ~]# service dhcpd restart
 close dhcpd:                                                [determine]
Starting dhcpd:                                            [determine]

Server log information

[root@master ~]# tail -7 /var/log/messages 
May  5 10:43:06 master dhcpd: Sending on   Socket/fallback/fallback-net
May  5 10:43:26 master dhcpd: DHCPREQUEST for 192.168.75.100 from 00:0c:29:57:c1:ff via eth0: lease 192.168.75.100 unavailable.
May  5 10:43:26 master dhcpd: DHCPNAK on 192.168.75.100 to 00:0c:29:57:c1:ff via eth0
May  5 10:43:26 master dhcpd: DHCPDISCOVER from 00:0c:29:57:c1:ff via eth0
May  5 10:43:26 master dhcpd: DHCPOFFER on 192.168.75.120 to 00:0c:29:57:c1:ff via eth0
May  5 10:43:26 master dhcpd: DHCPREQUEST for 192.168.75.120 (192.168.75.40) from 00:0c:29:57:c1:ff via eth0
May  5 10:43:26 master dhcpd: DHCPACK on 192.168.75.120 to 00:0c:29:57:c1:ff via eth0

	#It can also be seen from the log that the experiment was successful

DHCP superscope experiment (same network segment)

What is superscope?

The dhcp server can provide multiple scope renewal addresses for clients on a single physical device;

There is one situation:

If dhcp ip is not enough and new devices come in, how to solve this problem?
Multiple addresses can be provided for the LAN,
Simply put, the client can not only allocate ip, but also communicate
So what?

The referenced technology is superscope;

There are two network segments involved. Normally, it is impossible to communicate normally,
The function of supporting routing and forwarding is required;
For example, 192.168.75.0 and 192.168.72.0 cannot communicate;

Experimental planning

DHCP server:

You need to set up a single arm and another network card with sub interface;
The server needs to have routing and forwarding function

Client: two

You need to set dhcp to obtain IP

Server configuration

#Prepare subinterface network card
[root@master network-scripts]# cp -a ifcfg-eth0 ifcfg-eth0:0
	#Modify sub profile
[root@master network-scripts]# vim ifcfg-eth0:0
DEVICE=eth0:0
TYPE=Ethernet
ONBOOT=yes
BOOTPROTO=static
IPADDR=192.168.72.40
PREFIX=24
GATEWAY=192.168.75.2
DNS1=114.114.114.114
NAME="System eth0:0"

#Restart the network card
[root@master network-scripts]# service network restart
 Closing interface eth0:                                         [determine]
Close loopback interface:                                             [determine]
Pop up loopback interface:                                             [determine]
Pop up interface eth0:  Determining if ip address 192.168.75.40 is already in use for device eth0...
Determining if ip address 192.168.72.40 is already in use for device eth0...
                                                           [determine]

#Check whether the configuration is successful
[root@master network-scripts]# ifconfig 
eth0      Link encap:Ethernet  HWaddr 00:0C:29:FD:E7:9F  
          inet addr:192.168.75.40  Bcast:192.168.75.255  Mask:255.255.255.0
          inet6 addr: fe80::20c:29ff:fefd:e79f/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:4381 errors:0 dropped:0 overruns:0 frame:0
          TX packets:2886 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000 
          RX bytes:384017 (375.0 KiB)  TX bytes:432768 (422.6 KiB)

eth0:0    Link encap:Ethernet  HWaddr 00:0C:29:FD:E7:9F  
          inet addr:192.168.72.40  Bcast:192.168.72.255  Mask:255.255.255.0
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1


#Turn on routing forwarding
[root@master network-scripts]# vim /etc/sysctl.conf 
.........
net.ipv4.ip_forward = 1

#Use the command to reread the load file
[root@master network-scripts]# sysctl -p
net.ipv4.ip_forward = 1					#It's good to see this
net.ipv4.conf.default.rp_filter = 1
net.ipv4.conf.default.accept_source_route = 0
kernel.sysrq = 0
kernel.core_uses_pid = 1
net.ipv4.tcp_syncookies = 1
kernel.msgmnb = 65536
kernel.msgmax = 65536
kernel.shmmax = 68719476736
kernel.shmall = 4294967296


#The reference of superscope has been modified
#Modify dhcp configuration file

[root@master ~]# vim /etc/dhcp/dhcpd.conf
	#Just keep the content
class "foo" {
  match if substring (option vendor-class-identifier, 0, 4) = "SUNW";
}
	
shared-network 75-72 {				#The name of the superscope is specified here,
  subnet 192.168.75.0 netmask 255.255.255.0 {	#Here is the network segment of 75.0
    option routers 192.168.75.40;				#Note that the gateway specifies the ip address of the local machine
    range 192.168.75.90 192.168.75.90;			#The specially specified range is a
  }
  subnet 192.168.72.0 netmask 255.255.255.0 {	#This is the network segment of 72.0
   option routers 192.168.72.40;				#This is also the local ip gateway
   range 192.168.72.80 192.168.72.80;
  }



#Restart dhcp service
[root@master ~]# service dhcpd restart
 close dhcpd:                                                [determine]
Starting dhcpd:                                            [determine]

test

Change the client to dhcp and restart the network view

After direct restart, it is found that a device obtains 75.90
The other one got 72.80

ping the two devices to each other. ping 192.168.75.90 or 192.168.72.80 can absorb each other,

The experiment is successful ~!

DHCP relay experiment

What is dhcp relay?

dhcp relay (dhcp PR) is a small program,
It can process and forward dhcp information between different subnets and physical network segments
Breaking through the physical layer, dhcp can allocate dhcp ip for different network segments

Understanding experiment

dhcp server:

DHCP relay: consistent with DHCP server network segment

Client 2: different from dhcp server network segment,

Simple understanding:

DHCP server: the network segment used is 75.0,
Client: the network segment used is 72.0;
- The client cannot communicate with the dhcp server, so it cannot automatically dhcp to ip
dhcp relay:
- This server can communicate with the dhcp server through the eth0 network card
- You can also communicate with the network segment of client 72.0 – > eth1 is used
- Then it can accept the request from the client and tell the dhcp server
- Let the dhcp server assign ip to the client

The general understanding is like this;

Experimental process:

plan:

Prepare three servers
- dhcp server – > 192.168.75.0 network segment is used;
- dhcp relay server:
  - eth0: the network segment of net card – > 192 is used to communicate with DHCP server
  - eth1: host only mode – > 1100 network segment, used to communicate with clients
- client:
  - eth0: host mode only – > 100 network segments. The mode is dhcp;