Chapter 4 Introduction to TCP/IP Protocols

Introduction to TCP/IP Protocols

Protocol (as described by Webster): Code prescribing strict adherence to correct etiquette and precedence (as in diplomatic exchange and in the military services).

TCP/IP Protocols were developed by the Department of Defense–ARPANET (1960s). Its purpose is to share data/talk to other computes and only share/talk to computers that we trust and want to allow access.

TCP/IP (Transmission Control Protocol/ Internet Protocol) is not just one protocol but a collection of specialized protocols/subprotocols. TCP/IP has become the protocol of choice and recently Microsoft and NetWare/Novell use it as their default protocol.

Certain subprotocals of the TCP/IP collection are known as the core protocols and operate in the Network layers of the OSI model. TCP and IP being the most significant in the TCP/IP collection.

TCP (Transmission Control Protocol) operates in the Transport layer of the OSI model and is a reliable data delivery service. It is connection oriented; meaning it must establish communication nodes prior to transmitting data.

UDP (User Datagram Protocol) also operates in the Transport layer of the ISO model. However, it is lightweight and has little overhead. It is connectionless; meaning that packets have no assurance that they will be received in the correct sequence. UDP is used where data must be transferred quickly such as live audio or video.

IP (Internet Protocol) operates at the Network layer of the OSI, providing how and where the data should be delivered. It enables TCP/IP to internetwork more than one LAN segment and more than one type of network through a router. Like UDP, it is connectionless, although it does contain a reliability component, the Header checksum, which verifies the integrity of the routing information in its header.

ICMP (Internet Control Message Protocol) is a Network layer protocol that can detect when part of a network is congested and reports on data delivery success/failure announcing failure to the sender. These announcements provide critical information for troubleshooting it can’t correct errors and the data will be discarded when TTL (Time to Live) expires.

IGMP (Internet Group Management Protocol) operates at the Network layer and manages multicasting (allowing one node to send data to defined group of nodes) used for Internet teleconferencing, videoconferencing, routers, network nodes

ARP (Address Resolution Protocol) operates at the Network layer protocol and is a method for finding a host's (node) MAC address when only its Internet Layer (IP) or some other Network Layer address is known. It then creates a database ARP table (ARP cache) that maps the MAC address to the host’s IP (Logical address). RARP (Reverse Address Resolution Protocol) is used when you do not know the (IP) address. RARP was made obsolete by BOOTP, which has been superseded by the Dynamic Host Configuration Protocol (DHCP).

IPv4 Addressing Networks recognize two addresses: Logical (Network layer) IP address assigned dynamically or manually, and the Physical (MAC, hardware) address is static, assigned by the manufacture.

IP protocol handles logical addressing. The IP address contains specific parameters (unique 32-bit number, divided into four octets, and separated by periods). Example: 192.168.1.99. The addresses are then further divided in classes. The first octet specifies the network class you are using on your network.

Classes

A 1-126 00000001 - 01111111 Subnet Mask 255.0.0.0

B 128-191 10000000 - 10111111 Subnet Mask 255.255.0.0

C 192-223 11000000- 11011111 Subnet Mask 255.255.255.0

D 224-230 11100000-11101111 Reserved for Multi Casting Video Conferencing

E 240-254 11110000-11110111 Testing and Research

IPv4 has 2 to the power of 32 with total number of 4,294,000,000 addresses.

IP Address Break down: 192.168.1.37

192.168 Network Part First two sets of number

1.37 Host Part second set of numbers

Reserved Addresses

127.0.0.1 local host HOME There is no place like home.

APIPA (Automatic Private IP Addressing) 169.254.0.1-169.254.255.254

Can’t use 169.254.0.0 & 169.254.255.255 it is part of a block reserved by IANA.

0.0.0.0 is network 00000000

255.0.0.0 Broadcast 11111111

I need an address.

For the node/workstation to have a unique IP address it can be statically setup or dynamically setup using the following methods:


BOOTP (Bootstrap Protocol) an Application layer protocol which sends a broadcast message asking to be assigned an address. The broadcast will include the MAC address. The BOOTP server then responds with a client IP address, Server IP address, server host name and the IP address of the default router. This has been surpassed by the following protocol. However, you may still encounter these most likely in diskless workstations.

DHCP (Dynamic Host Configuration Protocol) operates at the application layer protocol. It is widely used to assign the network device’s unique IP address automatically (BOOTP replacement). This saves time and reduces headaches due to not having to document and resolve address duplication on workstations/networks.

DHCP leasing is a process where the device borrows/leases IP addresses. Again like any lease car, house, building, it is for a limited time. Once the lease is over, another lease is negotiated or terminated.

APIPA (Automatic Private IP Addressing) if a client cannot communicate without a valid IP address, APIPA will assign it. ***** IANA (Internet Assigned Numbers Authority) reserved predefined pool of addresses 169.254.0.0 through 169.254.255.255

IPv6 Addressing is the next generation of IP addressing that is gradually replacing IPv4. Most new development of applications, servers, and network devices support it. Advantages are a more efficient header, better security, better prioritization provisions, automatic IP address configuration, and it also adds billions of additional IP addresses.

Note: IPv6 2 to the power of 128 with the total number of 340,282,366,920,938,000,000,000,000,000,000,000,000 addresses.

Sockets and Ports
Processes assigned unique port numbers ranging from 0- 65535. A port number is similar to a building number “80” and the IP address “192.168.1.34” may be the suite number within the building. Together the IP and port number make up a socket “192.168.1.34:80“

Example:
Telnet is port: 23
Host IP: 10.43.3.87
Socket: 10.43.3.87:23

Host Names and DNS (Domain Name System) TCP/IP addressing is made up of numbers which are not easily remembered by most humans. Therefore a naming system was established. Which takes the host/device or domain IP and gives it a common name such as Host Name “SERVER” or Domain Name “procomps.com”. The DNS (Domain Name System) relies on many computers across the globe. These computers are related in hierarchical and distributed manners, with 13 computers, known as root servers, so DNS will not fail if there are a handful of the servers experiencing errors. Thus, when changing your DNS for a mail server, it can take up to 24 hours for all DNS servers in the system to update.

Application Layer Protocols

Telnet: Terminal emulation protocol often used to connect two dissimilar systems.

FTP (File Transfer Protocol): sending and receiving files via TCP/IP with no special client software required.

TFTP (Trivial File Transfer Protocol) also enables file transfers between computers and is simpler (more trivial) than FTP. It is connectionless and useful in loading data and programs on diskless workstation

NTP (Network Time Protocol) Synchronizes network computer clocks.

PING (Packet Internet Groper) is a utility used to assist in verifying TCP/IP installed bound to NIC, configured correctly, and communicating with network.

Ping the IP address or host name
Ping loopback address: 127.0.0.1