Tcp Ip Routing Protocols

TCP/IP & ROUTING PROTOCOLS V.Maheswaran Nair Sub Divisional Engineer BSNL,Trivandrum.

TCP/IP (Transmission Control Protocol/ Internet Protocol)

TCP/IP TCP/IP Protocols provide the ability to connect machines regardless of the underlying network cabling & the Operating Systems in use.

 TCP/IP is a piece of networking software for the Internet and networks worldwide.

TCP/IP protocol suite contain two main things:  Network Applications  HTTP (Hyper Text Transfer Protocol) WWW Services  FTP (File Transfer Protocol) for transferring file  SMTP Simple Mail Transfer Protocol) for E Mail  DNS (Domain Name System) etc… TCP is Used

Networking protocols  Moving packet of data from Source to Destination Internet Protocols (IP) and Routing Protocols are used.

 TCP is responsible for:  Data concurrency  Packet Sequencing  Delivery guarantee  Error Control  Retransmission

Internet address

MAC address

Port address

(Node to Node)

(Application) IP address

16 bit

(Source to

Decimal Notation

Destination)

MAC address 48 bit (6 Bytes) Hexa Decimal Notation

AB

0F Vendor ID

25

B6

C5 Serial No

D2

IP Address

 Each Internet address consists of 4 bytes (32-bits), defining two parts:  Netid  Hostid  These parts are of varying lengths depending upon the class of the address.

Dotted Decimal Notation

 To make the 32-bit address form more compact and easier to read, Internet addresses are usually written in decimal form with decimal points separating the bytes.

10000000 00001011 00000011 00011111

128.11.3.31

Classes

 There are five different IP Address Classes: A, B, C, D & E.  These are designed to cover the needs of different types of organizations.

Each address is a pair (Netid and Hostid) where the Netid identifies a network and the Hostid identifies a host on that network. NIB

Net Id

Host Id

Internet Address Classes Byte 1 A 0

Byte 2

NETID

B 10 C 110

Byte 3

Byte 4

HOSTID

NETID

HOSTID

NETID

D 1110

MULTICAST ADDRESS

E 1111

FOR FUTURE USE

HOSTID

IP Address Classes  Class

A Address : N.H.H.H  Class B Address : N.N.H.H  Class C Address : N.N.N.H

IP address - Classes •IP addressing supports five different address classes - A, B,C,D,E •CLASS A,B,C are available for commercial uses. •The left most bits indicate the network class.

Network Class A 0xxxxxxx

Class B

xxxxxxxx

Host xxxxxxxx

xxxxxxxx

Network 10xxxxxx xxxxxxxx 10

Host xxxxxxxx xxxxxxxx

Network xxxxxxxx

Host xxxxxxxx

Class C 110xxxxx 110

xxxxxxxx

Identifying a class of address Address Identifier

A

0

B

10

C

110

D

1110

E

1111

Network Id

7 bits Network Bits 14 bits Network Bits 21 bits Network Bits

Host Id 24 bits Host Id 16 bits Host Id 8 bits Host Id

Multicast address (224.0.0.0-239.255.255.255) Reserved for future use

IP Address Bit Patterns 8 Bits 8 Bits 8 Bits 8 Bits Class-A:

00000000

01111111

0-127

Class-B:

10000000

10111111

128-191

Class-C:

11000000

11011111

192-223

Class-D:

11100000

11101111

224-239

Class-E:

11110000

11111111

240-255

Address space utilisation 12 12 8 0111111 7 100000 1 00

100%

B-25% 1011111 19 1 1 19 2 110000 00 C-12.5% 1101111 1 22 D-6.25% 3 22 E-6.25%

0 1 0

0

A-50%

1 1

0 1

1110000 4 23 0 0 1110111 9 24 1 0 25 11110000 1111111 5 1

0000000 0 0

Class A A 0

NETID

HOSTID

 Class ‘A’ addresses are The First Octet defines thezero   A lot left-most of addresses bit must are be wasted  Remaining seven bits define designed for organizations that 7  Each network theoretically can netid. Theoretically, There Two special are actually addresses we can 126 have (hostid 2 to in define this class the as class it is as highly ‘A’. different networks. 24 may have a huge number have up to 2 = 16,777,216 = networks all 128 0s and networks. only hostid because all 1s) two areofused of improbable that an organization computers attached to theirfor hosts. the for addresses purposes. are reserved hasspecial so many computers. networks. special purposes.

Class B B 10

NETID

HOSTID

 The Class ‘B’left-most addresses are  two bits are 10  A lot of addresses are wasted 1 4  Two Octets define the netid We Each can network have 2 theoretically = 16,384 can designed for midsize  The Last next 16-bits 14-bits are used define to define to define the class as ‘B’.  Two special addresses (hostid in this class as it the is hosts highly(or 1also 6 define and two octets networks have up to in 2 class = 65,536 ‘B’. organizations that may have a different the hostid. networks. all 0s and hostid all 1s) are used improbable that an organization hostid. routers). large number of computers for purposes. hasspecial so many computers. attached to their networks.

Class C C 110

NETID

HOSTID

 Class C addresses are  Three Octets define the netid  The three leftmost bits are A class next ‘C’ 21-bits network define can have  Two addresses (all 0s and all designed for small organizations Theoretically, we can have  Eight bits are used to define and one octet define the hostid. 21 110 to define the class as “C’. different 2 =2,097,152 networks. networks. 8 1s) are reserved for special that have a small number of 2the =256 hosts. hostid. addresses. attached to their computers networks.

Class D D 1110

MULTICAST ADDRESS

 Class D address is designed  There is no netid or hostid;  Remaining 28-bits define The first four bits are 1110 to for multicasting. whole address is used for multicast addresses. define the class as ‘D’. multicasting.

Class E E 1111

FOR FUTURE USE

 Class E is reserved by the  The first four bitsuse. arehostid. 1111 to There is no netid or Internet for future define the class as ‘E’.

FROM A B C D E

TO

0.0.0.0

127.255.255.255

128.0.0.0

191.255.255.255

192.0.0.0

223.255.255.255

224.0.0.0

239.255.255.255

240.0.0.0

255.255.255.255

Number of networks and hosts in each Class

Class

No. of networks

No. of hosts

A

27 – 2 = 126

224 – 2 = 16,777,214

B

214 = 16,384

216 – 2 = 65,534

C

221 = 2,097,152

28 – 2 = 254

D

Not Applicable

Not Applicable

E

Not Applicable

Not Applicable

TCP/IP and OSI  OSI

is made of seven layers.  TCP/IP protocol is made of five layers. APPLICATION PRESENTATION

APPLICATION

SESSION

OSI Model

TRANSPORT

TRANSPORT

NETWORK

NETWORK

DATA LINK

DATA LINK

PHYSICAL

PHYSICAL

TCP/IP Model

TCP/IP Protocol Suite A

T N D

P

FTP DHCP SMTP TELNET HTTP

TFTP SNMP DNS TCP

ICMP IGMP

UDP IP

ARP

RARP

Ethernet, Token Ring, FDDI, HDLC, FR, PPP, ATM

Protocols defined by the underlying networks

Data Encapsulation Application

Data

Port add (TCP)

TPT Layer (TCP/UDP)

Data

TCP Segment Port add (UDP)

Data

UDP Message NW Layer (IP)

Data Link (MAC) Physical

Source IP

TCP-UDP

Data

Dest IP

IP Datagram Source MAC

IP Header

TCP-UDP Frame

Bits 10000010101001

Data

Dest MAC

TCP Details 

   

Provides application programs access to the network using a reliable connection-oriented transport layer service TCP sends and receives data reliably using sequence numbers and acknowledgements TCP is a byte oriented protocol i.e. every byte in each packet is assigned a sequence number Data stream handed over to TCP is called an unstructured stream TCP divides this data stream into segments for transmission to remote network

TCP Header.. Octet +0

Octet +1

Octet +2

Octet +3

0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7

DESTINATION PORT

SOURCE PORT

SEQUENCE NUMBER ACKNOWLEDGEMENT NUMBER HEAD LEN

U A P R S FI R C S S Y N G K H T N CHECKSUM

WINDOW SIZE URGENT POINTER

OPTIONS AND PADDING

TCP Header… 

Source & Destination Port (16 Bits)    

 

Port numbers are used to identify a unique application in a machine 65536 (0-65535) port numbers can be defined Theoretically it is possible to run 65535 simultaneous applications in a host The first 1024 ports, port numbers 0-1023 known as well known port numbers, are assigned and are reserved for standard applications and are controlled by IANA The remaining ports, 1024-65535, are dynamic and can be used freely by applications Source port is randomly generated by the source machine

Well known port numbers PORT

DESCRIPTION

20

File Transfer-Data

21

File Transfer-Control

23

Telnet

25

SMTP

53

Domain Name Server

69

Trivial File Transfer

80

WWW

123

Network Time Protocol

179

Border Gateway Protocol

TCP Header… 

Sequence Number (32 Bits)    



Helps in establishing TCP connections, along with SYN bit, called as Three Way Handshake Helps in maintaining account of amount of data being transferred Identifies where the encapsulated data fits within a data stream from the sender Sequence number is incremented, in the system, every 4 microsecond

Acknowledgement Number (32 Bits)  

Helps in maintaining account of amount of data being transferred Identifies the sequence number expected from the other end of data transmission unit

Seq/Ack numbers relation  During

TCP Connection Establishment/ Three way handshake 

Acknowledgement Number Sent = Sequence Number Received+1

 During 

Data Transfer

Acknowledgement Number Sent = Sequence Number Received + Data Received in Bytes

Three-Way-Handshake Sender

0

Receiver

1

SN-95426

2

AN-00000

1 0

000B

SN-16780

3

AN- 95427

1 1

000B

SN-95427

4

AN-16781

0 1

000B

0-Closed; 1-Listen; 2-SYN-Sent; 3-SYN-Received; 4-Established

Data Transfer 0

Sender

2

SN-95426 AN-00000

1 0

000B

1 AN- 95427 1 1

4

Receiver

SN-16780

3

000B

SN-95427

AN-16781

0 1

000B

SN-16781

5

AN- 95428 0 1

100B

SN-95428

5

AN-16881

0 1

200B

SN-16881 AN- 95628 0 1

5

150B

SN-95628

5

AN-17031

0 1

250B

SN-17031

AN- 95878 0 1

5

300B

0-Closed; 1-Listen; 2-SYN-Sent; 3-SYN-Received; 4-Established; 5-Data Transfer

Closing a TCP Connection SN - 95880

Sender

6

AN -17334 0 1 1

Receiver

0B

SN - 17334 AN - 95881 0 1 0

0B

WAIT SN - 17334 AN - 95881 0 1 1

SN - 95881

6

0B

AN -17334

0

0 1 0

0B

0 6-Finish; 0- Closed

TCP Header…. 

Header Length (4 Bits)    



Sometimes called Data Offset Indicates the length of header in 32-bit words Identifies the beginning of data Typical value is 5 unless there are options

Flags (6 Bits)      

Urgent (URG) Acknowledgement (ACK) Push (PSH) Reset (RST) Synchronisation (SYN) Finish (FIN)

TCP Header…..  Window  

 

Size (16 Bits)

Indicates the size of the sliding window Specifies the number of octets, starting with the octet indicated by the acknowledgement number, that the sender of the segment will accept from its peer at the other end of the connection before the peer must stop transmitting and wait for an acknowledgement A default window size is 4096 bytes Used for flow control by using Sliding window mechanism

Flow Control 



Sender retains a copy of transmitted data until it receives an acknowledgment from the remote network. If no acknowledgment is received, within a specified time, the data is retransmitted by using adaptive retransmission algorithm.   



TCP records the time of the transmission and sequence number of the segment. TCP again records the time of the acknowledgement received. Using this delta, TCP builds a sample round-trip delay time and uses this to build an average time for a packet to be sent and to receive an acknowledgement

TCP will time out after a number of unsuccessful retransmissions

Sliding Window-Flow Control

Moves to right when ack is received.

Sent and ack

Moves to right when data is sent. Moves to right or left to fix the size of the window.

Window Size

Sent but not ack Can be sent Can’t be sent

TCP Header….. 

Checksum(16 Bits)  



Urgent Pointer(16 Bits)  



Used for error detection Covers both header and the encapsulated data Used only when urgent flag is set Points to the last octet of urgent data

Options 

One of the important options is MSS (Maximum Segment Size)  Informs the receiver of the largest segment the sender is willing to accept, without causing fragmentation

TCP Header……  Padding 

Consists of 1-3 octets, each equal to zero, to force the length of TCP header to be in multiples of four octets.

User Datagram Protocol  Provides

unreliable connectionless service  Transfers data without establishing a session  Used for services that have an inbuilt reliability  Does not use end to end error checking and correction  Does not order the packets; may loose or duplicate a packet  Runs faster than TCP due to less overheads

UDP Header..

Octet +0

Octet +1

Octet +2

Octet +3

0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 SOURCE PORT

DESTINATION PORT

MESSAGE LENGTH

CHECKSUM

UDP Header... 

Source Port (16 Bits) 



Destination Port (16 Bits)  



Identifies the receiving process. Some fixed, pre-assigned port numbers used for services on the Internet.  7 for UDP; 69 for TFTP

Message length (16 Bits)  



Identifies the sending process.

Indicates the size of the UDP header and its data in bytes. Minimum size is 8, if carries no data.

Checksum (16 Bits)  

Covers the UDP header and UDP data. Optional; If not used, set to all zeros.

Internet Protocol.      

Provides best-effort or connectionless delivery service. No error checking or tracking If reliability is important, IP must be paired with a reliable protocol like TCP Transmits blocks of data called datagrams each of which is transported separately Responsible for IP addressing Datagrams may travel along different routes and may arrive out of sequence or duplicated.

IP Header.. Octet +0

Octet +1

Octet +2

Octet +3

0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7

VER

HLEN

TOS

IDENTIFICATION TIME TO LIVE

PROTOCOL

TOTAL LENGTH DM F F

FRAGMENT OFFSET HEADER CHECKSUM

SOURCE ADDRESS OF HOST DESTINATION ADDRESS OF HOST OPTIONS

PADDING

IP Header… 

Version (4 Bits) 



Header Length (4 Bits)   



Identifies the IP version to which the packet belongs Indicates the length of IP header in 32 bit words. Minimum length is 20 octets. Options may increase the size up to a maximum of 24 octets.

Type of Service (8 Bits)  

Used for specifying special handling of packet. Has two sub-fields:  Precedence  TOS

IP Header…. P

P

P

D

T

Precedence 000-Routine 001-Priority 010-Immediate 011-Flash 100-Flash Override 101-CRITIC/ECP 110-Internetwork Control 111-Network Control

0

0

0

0

R

C

0

Delay 0-Normal 1-Minimise

Reliability 0-Normal 1-Maximise

Throughput 0-Normal 1-Maximise

0

0

0

Reserved: Always set to ‘0’

0

Cost 0-Normal 1-Minimise

= No TOS

IP Header….. 

Total Length (16 Bits) 







Specifies total length of the packet, including header, in octets Largest decimal number =216= 65535, the maximum possible size of an IP packet is 65535 octets Total length - header length = Packet’s data payload

Identification (16 Bits) 



Each datagram is identified by a identification number set by the source. Normally incremented by 1 for each datagram sent.

IP Header…… 

Flags (3 Bits)   





First bit is not used. Second bit is Don’t Fragment (DF) bit Third bit if More Fragment (MF) bit

Maximum Transmit Unit (MTU) is the size of the largest packet, including IP Header, that can be transmitted or received through a data link Default MTU is 576 bytes, which can be handled by any network without fragmentation

IP Header……  Fragment 

 



Offset (13 Bits)

The fragmentation occurs at the routers, if the original packet length exceeds the MTU of a data link Used only in the cases when a datagram is fragmented on its way Specifies the offset, in units of eight octets, from the beginning of header to the beginning of the fragment Each fragment is marked, by router, with the same identifier number

Fragmentation.. MTU-1500

MTU-1500

MTU-576

172.16.2.0

172.16.3.0

1500 B IP

TCP

Data

512 B IP

IP

TCP

Data

DF=0; MF=1; Offset=0 IP

TCP IP

IP

512 B Data

DF=0; MF=1; Offset=64 Data TCP

Data Data

IP

476 B Data

DF=0; MF=0; Offset=128 Data

Fragmentation  





Only the receiver host reassembles the datagram The destination machine starts a reassembly timer for about 60-120 seconds. If not all fragments were received, then hosts discard the packets and sends a time exceeded ICMP message to the source machine If a single fragment is lost during a transmission, the entire packet must be resent

IP Header……  Time 

to live-TTL (8 Bits)

Assigns a life to an IP datagram

 Protocol  

Specifies the protocol that runs on the top of IP. TCP-6; EGP-8; UDP-17; OSPF-89

 Header  

(8 Bits)

Checksum (16 Bits)

Error detection field for IP header As each router decrements the TTL, checksum is calculated by each router

the

IP Header…….  Source 

Address of Host (32 Bits)

IP Address of the Originating Machine

 Destination 

Address of Host (32 Bits)

IP Address of the Destination Machine

 Options 

Security: 



Strict Source Routing(SSR): 



Specifies how secret the datagram is Gives the complete path to be followed

Loose Source Routing(LSR): 

Gives the list of routers not to be missed

IP Header…….. 

Record Route: 



Makes each router to append its IP address.

Time Stamp: 

Makes each router to append its IP address and time stamp.

 Padding 

Ensures that the header ends on a 32 bit boundary by adding zeros after the option field.

Well known port numbers PORT

DESCRIPTION

20

File Transfer-Data

21

File Transfer-Control

23

Telnet

25

SMTP

53

Domain Name Server

69

Trivial File Transfer

80

WWW

123

Network Time Protocol

179

Border Gateway Protocol

Domain Name System (DNS) DNS Server What is the IP Address of www.Yahoo.com

What is the IP Address of www.Yahoo.com DNS Server

Internet User traffic

www.yahoo.com , IP address is 210.212.90.15

www.yahoo.com , IP address is 210.212.90.15

yahoo.com

ROUTING PROTOCOLS

Routing Protocol  It

is a language a router speaks with other routers Functions of RP  Forwarding,  Sharing  Updating

information about the reachability and status of the network

Static Routing  Routes

to destinations are set up manually  Route may be up or down but static routes will remain in the routing tables and traffic would still be sent towards the route  Not suitable for large networks  Also known as Non-adaptive routing

Dynamic Routing  Routes

are learnt via an internal or external routing protocols  Network reachability is dependent on the existence and state of the network  Routing decisions change to reflect the changes in topology  Also known as Adaptive routing

Routing Table A

Data base to be maintained by each router.  Created by using algorithms.  It contains Network address Interface address for reaching the next router (Hope) Metric

Types of R P Routes Static Distance Vector Protocols Dynamic

RIP,IGRP

Link State Protocols OSPF,IS-IS

Path Determination Router-A Network Router

Next Hop A

B 192.168.1. 0

192.168.7. 192.168.1.0 0 192.168.2.0 Direct 192.168.3.0 Direct 192.168.3. C 192.168.6. 0 0 192.168.4.0 B,C 192.168.4. 192.168.5. 192.168.5.0 B,C 0 0 192.168.6.0 B,C 192.168.7.0 B,C •Networks192.168.4.0 to 192.168.7.0 can be reached via either router B or C, which path is preferable? •Metrics are needed to rank the alternatives. 192.168.2. Direct 0

Routing Protocols contd..  Distance

Vector Routing Protocols

eg. RIP V1 (Routing Information protocol) RIP V2  Link

State Routing Protocols

eg. OSPF (Open Shortest Path First) IS-IS (Intermediate System-Intermediate System)

Metrics  Hop

 Cost

Count -- Distance Vector

(BW) – Link State

Routing Updates 20.0.0.10 20.0.0.0

A

21.0.0.0 .1

.2

Routing Table-A NW 20.0.0.0 21.0.0.0

VIA HOP ---------0 ---------0

22.0.0.0 21.0.0.2

1

23.0.0.0 21.0.0.2

2

 After

B

.2

C

22.0.0.0 .1

.2

Routing Table-B NW 21.0.0.0 22.0.0.0

VIA HOP ---------0 ---------0

20.0.0.0 21.0.0.1 23.0.0.0 22.0.0.2

1 1

23.0.0.15 23.0.0.0 .1

Routing Table-C NW 22.0.0.0 23.0.0.0

VIA HOP ---------0 ---------0

21.0.0.0 22.0.0.1

1

20.0.0.0 22.0.0.1

2

exchanging 2 periodic updates, the network is converged.

Routing Updates 20.0.0.0 .2

Routing Table-A

A

.1

NW VIA HOP 20.0.0.0 D 0 21.0.0.0 D 0 22.0.0.0 21.0.0.2 1 23.0.0.0 21.0.0.2 UR 2



B

21.0.0.0

.1

.2

Routing Table-B

NW 20.0.0.0 21.0.0.0 22.0.0.0 23.0.0.0

C

22.0.0.0 .2

VIA HOP 21.0.0.1 1 D 0 D 0 22.0.0.2 UR 1

23.0.0.0 .1

Routing Table-C

NW 20.0.0.0 21.0.0.0 22.0.0.0 23.0.0.0

VIA HOP 22.0.0.1 2 22.0.0.1 1 D 0 D 0 C UR

Router-C in its next scheduled update, flags the network as unreachable and passes the information along.

Routing Updates 20.0.0.0 .2

Routing Table-A

A

.1

NW VIA HOP 20.0.0.0 D 0 21.0.0.0 D 0 22.0.0.0 21.0.0.2 1 23.0.0.0 21.0.0.2 2

 

B

21.0.0.0

.1

.2

Routing Table-B

NW 20.0.0.0 21.0.0.0 22.0.0.0 23.0.0.0

C

22.0.0.0 .2

VIA HOP 21.0.0.1 1 D 0 D 0 22.0.0.2 1

23.0.0.0 .1

Routing Table-C

NW 20.0.0.0 21.0.0.0 22.0.0.0 23.0.0.0

VIA HOP 22.0.0.1 2 22.0.0.1 1 D 0 D 0

Routers-A & B still have entries in the route table about 23.0.0.0. The information is no longer valid but there is no router to inform them of this fact, thus creating a black-hole in the network.

Route Invalidation Timer 20.0.0.0

A

.2

Routing Table-A

NW VIA HOP TIME 20.0.0.0 C 0 RIT 21.0.0.0 C 0 RIT 22.0.0.0 21.0.0.2 1 RIT 2 RIT 0 23.0.0.0 21.0.0.2 UR

 

21.0.0.0 .1

B

.2

Routing Table-B

C

22.0.0.0 .1

.2

NW VIA HOP TIME 20.0.0.0 21.0.0.1 1 RIT 21.0.0.0 C 0 RIT 22.0.0.0 C 0 RIT 1 RIT 23.0.0.0 22.0.0.2 UR 0

23.0.0.0 .1

Routing Table-C

NW VIA HOP TIME 20.0.0.0 22.0.0.1 2 RIT 21.0.0.0 22.0.0.1 1 RIT 22.0.0.0 C 0 RIT 23.0.0.0 C 0 RIT

Another Timer, Garbage Collection or Flush Timer, 60 Seconds longer than RIT, is set. On the expiry of which the route entry will be flushed from the routing table.

RIP Timers  Update 

30 Seconds

 Route 

Timer

Invalidation Timer

180 Seconds (6 Times the Update Timer)

 Garbage 

Collection Timer

240 seconds (60 Seconds longer than RIT)

LINK STATE ROUTING (OSPF)  Sharing knowledge about the neighbourhood.  Sharing with every other router in the area.  Sharing when there is a change.

OSPF operation….  OSPF-

Routers send Hello packets out OSPF-enabled interfaces

 Two

routers sharing a common link, after exchanging Hello packets, become neighbors

OSPF operation… 





Link State Advertisements (LSAs) i.e. router’s links and their state, are exchanged between adjacent routers Each router receiving an LSA from a neighbor records the LSA in Link State Database and sends a copy of the LSA to all of its other neighbors LSAs are exchanged, until all the routers build identical Link State Databases i.e. the link state databases have been synchronized

OSPF operation…. 

 

Each router uses SPF algorithm to calculate a shortest path to every known destination, with itself as root Each router builds its router table from its SPF Tree After this, in a stable internetwork, all activities stop except 



Hello packets are exchanged, after regular intervals of 10 seconds (Hello Interval) between neighbors, as keepalives LSAs are exchanged every 30 minutes

Metrics Speed

Cost

>= 100Mbps

1

Ethernet/802.3

10

E1(2.048Mbps)

48

64Kbps

1562

Metric=108/Interface Speed in bits per sec. e.g. 100000000/2048000=48.828125

Fully adjacent router network RA-RB-4 RA-RC-20 RA-RD-5 RB-RA-4 RB-RD-3 RC-RA-20 RC-RD-2 RD-RA-5 RD-RB-3 RD-RC-2

A

C RB-RA-4 RB-RD-3 RD-RA-5 RD-RB-3 RD-RC-2 RA-RB-4 RA-RC-20 RA-RD-5 RC-RA-20 RC-RD-2

4

A

C

2 5 0 2 0

4

5 2

2

tn+2 tn+1 tn

RB-RA-4 RB-RD-3 RA-RB-4 RA-RC-20 B RA-RD-5 RD-RA-5 3 RD-RB-3 3 RD-RC-2 D RC-RA-20 RC-RD-2

B

D RC-RA-20 RC-RD-2 RB-RA-4 RB-RD-3 RA-RB-4 RA-RC-20 RA-RD-5 RD-RA-5 RD-RB-3 RD-RC-2

Shortest Path Tree of Router-A 4

A

4

20 5

C

B

A

3 5

2 0 2

2

4

B

5

3

D

C

2

D

PATH VECTOR ROUTING Path vector routing is similar to distance vector routing. There is at least one node, called the speaker node, in each AS that creates a routing table and advertises it to speaker nodes in the neighboring ASs..

Boarder Gateway Protocol Between Two AS

Autonomous System  Definition

An

autonomous system is a network under a common administration

Autonomous systems

Autonomous systems

AS-0

BG P AS-1

IGP

IGP

BG P

BG P

AS-2

IGP

Note: In distance vector routing, each node shares the knowledge about the entire AS with its immediate neighbors periodically .

Initial routing tables in path vector routing

Popular routing protocols

Stabilized tables for four autonomous systems

Administrative Distances  Diversity

of metrics poses problems in routers running more than one routing protocol.  Router may learn a route to the same destination from each of the protocols  Administrative distances are the route sources to determine most preferred source  Administrative distance is a measure of believability

Administrative Distances  The

administrative distance of various protocols is as below:  Connected

Interface - 0  Static Route -1  OSPF - 110  IS-IS - 115  RIP - 120  Unknown - 255  The lower the administrative distance, the more believable the protocol

Flow Chart of a Packet Packet Received NO If route available, search MAC in ARP cache

NO

Header & Checksum Valid YES

NO

Decrement TTL; TTL>=0

Send ICMP error message

YES

Send ARP request and wait for a response

Discard original Packet NO

YES

YES

YES

Route table lookup on Dest. Add.

Route Found

MAC Address Found

NO

Received ARP Reply

NO

Default route available YES

Build new packet with MAC address and route through port found in routing table

Received ARP reply, insert MAC and IP address into ARP table