Chapter -1 INTRODUCTION 1.1 Networking Networking is a practice of linking of two or more computing devices such as PCs, printers, faxes etc., with each other Connection Con nection between two devices is through physical media or logical media to share information, data and resources. Networks are made with the hardware and software.
Cable/media Fig.! Computer network
"here are many different ways to connect your computer to another computer or a network. #sing $indows $indows %&&&, you can connect your computer to! •
'nother computer using a direct cable connection.
•
' private network using a modem or an integrated service digital network ()*+N adapter or a network adopter card.
•
' network using a virtual private network (-PN connection.
•
'nother computer by having another computer call your computer.
"he interconnected collection of autonomous computers is called computer network. "wo computers are said to be interconnected if they are able to exchange information. "he connection need not be via a copper wire fiber optics, microwaves and communication satellites can be used.
1.2 Types of Networking:
Page 1 of 45
•
$ired network
•
$ireless network
ire! networks: $ired networks are almost always faster and less expensive than wireless networks. nce connected, there is little that can disrupt a good0wired connection. $ired networks come in many forms, but the most popular are 1omePN' and 2thernet. 1omePN' uses the existing phone line wires in your home and 2thernet needs special network cabling.
Fig.%! $ired network
ire"ess Networks: 3obile computers, such as notebook computers and personal digital assistants (P+'s are the fastest0 growing segment of the computer industry. 3any of the owner of these computers have desktop machines on 4'Ns and $'Ns back at the office and want to be connected to their home base even when away from home or en route. *ince having a wired connection is impossible in cars and airplanes, there is a lot of interest in wireless networks.
Page 2 of 45
Fig.5! $ireless network
1.# $o!e"s of Networking Networking 3odel means the connectivity of two computers. $e have many types of networking models.
%i&
(i
Client 6 *erver 3odel
(ii
Peer to Peer 3odel ($orkgroup ($orkgroup 3odel
(iii
+omain 3o 3odel
C"ient '(er)er $o!e"
)n a Client server model we have one server and many clients. ' Client can share the resources of server, but a server cannot share the resources on clients. n the point of view of administrator it7s very easy to control the network because we combine with the server also at security point of view. )t is very useful because it uses user level security in which users have to remember only one password to share the resources.
%ii& *eer to *eer $o!e" %orkgro+p $o!e" &
)n Peer to Peer networking model all computers are in e8ual status, that is we cannot manage centrali9at centrali9ation, ion, administrati administration on security. security. )n Peer to Peer networking client client use operating system like $indow :;, $indow
Page # of 45
%iii& Do,ain $o!e"
)t is a mixture of client server and peer0to0peer model. )n this clients can share their resources as peer0to0peer but with the permission of the server as in client server model therefore it is commonly used model because in this security is more as we can put restriction on both server and clients.
1.4 Categories of Network oa" /rea Network %/N& 4'N is a computer network that is used to connect computers and work station to share data and resources such as printers or faxes. 4'N is restricted to a small area such as home, office or college. +evices used in 4'N are! 1#= and switch. 3edia for 4'N is #"P cables.
Fig.>! 4ocal 'rea network
Ca,p+s /rea Network %C/N& Campus 'rea Network is a computer network made up of two or more 4'Ns within a limited area. )t can cover many buildings in an area. "he main feature of C'N is that all of the computers which are connected together have some relationship to each other. )t Page 4 of 45
will help to interconnect academic departments, library and computer laboratories. C'N is larger than 4'N but smaller than $'N. +evices used in C'N are! 1#=, *witch, 4ayer05 switch, 'ccess Point.
$etropo"itan /rea Network %$/N& 3'N is the interconnection of networks in a city. 3'N is not owned by a single organi9ation. 3'N can also be formed by connecting remote 4'Ns through telephone lines or radio links. 3'N supports data and voice transmission. "he best example of 3'N is cable ".- network in a city.
Fig.? 3etropolitan area network
i!e /rea Network Netw ork %/N& %/N& $'N covers a wide geographical area which includes multiple multiple computers or 4'Ns. )t connects computer networks through public networks like, telephone system, microwave, satellite link or leased line. 3ost of the $'Ns $'Ns use leased lines for internet access as they provide faster data transfer. $'N helps an organi9ation to establish network between all its departments and offices Page 5 of 45
locate located d in the same same or diffe differen rentt cities cities.. )t also also enables enables commun communica icatio tion n between between the organi9ation and rest world. +evices used in $'N $'N is is only @outer
Fig.A! $ide area network
Chapter-2 *RO0$ OR$U/TION 2.1 *ro3"e, O)er)iew:
Page of 45
)t is difficult to manage communication between far away branches and absence of networking results in higher cost and low efficiency communication among organi9ation and outside environment.
2.2 O3eti)e of the proet: bBective of proBect is to make communication possible between far away branches, head0offices of any organi9ation with lower cost and higher efficiency. )n this proBect we use routi routing ng protoc protocols ols to have have commun communica icati tion on of an organi organi9at 9ation ion with with it7s it7s far away away branches.
2.# Networking o,ponents: $hen a computer or device ' is re8uesting a resource from another computer or device =, the item ' is is referred to as a client. =ecause all or most items that are part of a network live in association or cooperation, almost any one of them can be referred to as a client. =ased on this, there can be different types of clients. "he most regularly used of them is referred to as a workstation. )f you already have one or more computers that you plan to use as workstations, you can start by checking the hardware parts installed in the computer. "he computer must meet the following re8uirements!
*roessor: 'n )ntel Pentium or Celeron family of processors or an '3+ A/'thlon/+uron family of processors. "he processor should have a 5&& megahert9 clock speed. ' higher speed is recommended.
R/$: "he computer must have a memory of at least A> megabytes (3=. 's memory is not particularly expensive ex pensive nowadays, you should upgrade up grade the computerDs memory to at least ?%3=. Page 6 of 45
7ar! Dri)e: =efore installing 3icrosoft $indows
Network Ca3"es: Cable is used to connect computers. 'lthough we are planning to use as much wireless as possible, you should always have hav e one or more cables ca bles around. )n our network, we will use Category ? cable @0>?. "he ends of the cable appear as follows!
ig+re 2.1: R8 onnetors
Intro!+tion Intro!+tion to Network Distri3+tors: $e can connect one computer to another. "his can be done using their serial ports!
ig+re 2.2: onneting o,p+ters 3y seria" port
7+3: ' hub is rectangular box that is used as the central obBect on which computers and other devices are connected. "o make this possible, a hub is e8uipped with small holes called ports. 1ere is an example of a hub!
Page 9 of 45
ig+re 2.#: 7+3
Ro+ters: @outers @outers are networ networkin king g device devicess that that forwar forward d data data packets packets betwee between n networ networks ks using using headers and forwarding tables to determine the best path to forward the packets. @outers work at the network layer of the "CP/)P model or layer 5 of the *) model. @outers also provide interconnectivity between like and unlike media. 1ere is an example of a router!
ig+re 2.4: Ro+ter
Network Car!s: )n order to connect to a network, a computer must be e8uipped with a device called a network card. ' network card, or a network adapter, also called a network interface card, or N)C, allows a computer to connect to the exterior. )f you buy a computer from one of those popular stores or big companies on the )nternet, most of their computers have a network card tested and already. Gou can reliably use it. )f you go to a store that sells or manufactures computers, you can ask them to install or make sure that the computer has a network card. $hen it comes to their installation, there are roughly two categories of network cards! internal and external. 'n internal network card looks like a printed circuit board with some obBects HattachedH or HgluedH to it and it appears as follows!
(with:
Page of 45
' netw networ ork k
swit witch or switc witchi hing ng
hub hub is
a com comput puter
net network workiing
devi device ce that that
connects network segments. ' network swith is a small hardware device that Boins multiple computers together within one local area network (4'N. "echnically, network switches operate at layer two (+ata 4ink 4ayer o f the *) model. Network switches appear nearly identical to network hubs, but a switch generally contains more intelligence (and a slightly higher price tag than a hub. #nlike hubs, network switches are capable of inspecting data packets as they are received, determining the source and destination device of each packet, and forwarding them appropriately. =y delivering delive ring messages only to the connected device intended, a networ network k switc switch h conser conserves ves network bandwidth and offers generally better performance than a hub.
ig+re 2.5: (with
(er)er: ' network server is a computer designed to process re8uests and deliver data to other (client computers over a local network or the )nternet. 2xamples include $eb servers, proxy servers, and F"P servers. Not only should you learn about servers on the )nternet, private network servers for business and personal use are also becoming more common.
Page 1; of 45
ig+re 2.: (er)er
/ess *oint: )n computer networking, a wire"ess aess point ($'P is a device that allows wireless devices to connect to a wired network using $i0Fi, =luetooth or related standards. "he $'P usually connects to a router (via a wired network, and can relay data between the wireless devices (such as computers or printers and wired devices on the network.
ig+re 2.6: /ess *oint
Network (oftware: Operating (yste,s : ' workstation is a computer that is a member of a network. 't homes and small businesses, the most regular operating system, at the time of this writing, is probably 3icrosoft $indows
3icrosoft $indows %&&& Professional. n this site, we will mostly cover 3icrosoft $indows
*aket Traer: *aket *aket Traer is a Cisco router simulator that can be utili9ed in training and education, but also in research for simple computer network simulations. "he tool is created by Cisco *ystems and provided for free distribution to faculty, students, and alumni who are or have participated in the Cisco Networking 'cademy. "he purpose of Packet "racer is to offer students and teachers a tool to learn the principles of networking as well as develop Cisco technolog y specific skills. "he current version of Packet "racer supports an array of simulated 'pplication 4ayer protocols, as well as basic routing with @)P, *PF, and 2)I@P, to the extent re8uired by the current CCN' curriculum. $hile Packet "racer aims to provide a realistic simulation of functional networks, the application itself utili9es only a small number of features found within the actual hardware running a current Cisco )* version.
Chapter # Page 12 of 45
*RO8CT TC7NI
decimal values, each representing ; bits, in the range & to %?? (known as octets separated by decimal points. "his is known as Hdotted decimalH notation. 2xample! >&.J:.%%&.%&& )t is some someti time mess usef useful ul to view view the the value valuess in thei theirr bina binary ry form form.. >& >& .J: .J: .%%& .%%& .%&& .%&& &&&&&.& &&&&&.&&& &&. .& &&&. &&.&&& &&&&& && 2very )P address address consists consists of two parts, parts, one identifying the network and one identifying the node. "he Class of the address and the subnet mask determine which part belongs to the network address and which part belongs to the node address. "he four numbers in an )P address are called octets, because they each have eight positions when viewed in binary form. )f you add all the positions together, you get 5%, which is why )P addresses are considered 5%0bit numbers. *ince each of the eight positions can have two different states ( or & the total number of possible combinations per octet is %; or %?A. *o each octet can contain any value between & and %??. Combine the four octets and you get %5% or a possible >,%:>,:AJ,%:A uni8ue values. ut of the almost >.5 billion possible combinations, certain values are restricted from use as typical )P addresses. For example, the )P address &.&.&.& is reserved for the default default network network and the address address %??.%??.%??.%?? %??.%??.%??.%?? is used for broadcasts broadcasts.. #nderstandi #nderstanding ng )P 'ddresses 'n )P address is an address used in order to uni8uely identify a device on an )P network. "he address is made up of 5% binary bits, which can be divisible into a network portion and host portion with the help of a subnet mask. "he 5% binary bits are broken into four octets ( octet K ; bits. 2ach octet is converted to decimal and separated by a period (dot. For this reason, an )P address is said to be expressed in dotted decimal format (for example, J%.A.;.&&. "he value in each octet ranges from & to %?? decimal, or &&&&&&&& 0 binary. 1ere is how binary octets convert to decimal! "he right most bit, or least significant bit, of an octet holds a value of %&. "he bit Bust to the left of that holds a value of %. "his continues until the left0most bit, or most significant bit, which holds a value of %J. *o if all binary bits are a one, the decimal e8uivalent would be %?? as shown here! Page 1# of 45
%; A> 5% A ; > % (%;LA>L5%LAL;L>L%LK%?? 1ere is a sample octet conversion when not all of the bits are set to . & & & & & & & A> & & & & & (&LA>L&L&L&L&L&LKA? 'nd this is sample shows an )P address represented in both binary and decimal. &. . %5. : (decimal &&&&&&.&&&&&&&.&&&&.&&&&& (binary "here are five )P classes plus certain special addresses!
Defa+"t Network !0 "he )P address of &.&.&.& is used for the default network.
C"ass / !0 "his class is for very large networks, such as a maBor international company might have. )P addresses with a first octet from to %A are part of this class. "he other three octets are used to identify each host. "his means that there are %A Class ' networks each with A,JJJ,%> (%%> 0% possible hosts for a total of %,>J,>;5,A>; (%5 uni8ue )P addresses. Class ' networks account for half of the total available )P addresses. )n Class ' networks, the high order bit value (the very first binary number in the first octet is always &.
oop3ak:"he )P address %J.&.&. is used as the loopba ck address. "his means that it is used by the host computer to send a message back to itself. )t is commonly used for troubleshooting and network testing.
C"ass 0:0 Class = is used for medium0si9ed networks. ' good example is a large college campus. )P addresses with a first octet from %; to : are part of this class. Class = addresses also includes the second octet as part of the Net identifier. "he other two octets are used to identify each host. "his means that there are A,5;> (%> Class = networks each with A?,?5> (%A 0% possible hosts for a total of ,&J5,J>,;%> (%5& uni8ue )P addresses. Class = networks make up a 8uarter of the total available )P addresses. Class = networks have a first bit value of and a second bit value of & in the first octet.
Page 14 of 45
6 C"ass C: 6 Class C addresses are commonly used for small to mid0si9e businesses. )P addresses with a first octet from :% to %%5 are part of this class. Class C addresses also include the second and third octets as part of the Net identifier. "he last octet is used to identify each host. "his means that there are %,&:J,?% (%% Class C networks networks each with %?> (%; 0% possible hosts for a total of ?5A,;J&,:% (%%: uni8ue )P addresses. Class C networks make up an eighth of the total available )P addresses. Class C networks have a first bit value of , second bit value of and a third bit value of & in the first octet.
C"ass D: 6 6 #sed for multicasts, Class + is slightly different from the first three classes. )t has a first bit value of , second bit value of , third bit value of and fourth bit value of &. "he other other %; bits bits are used to identi identify fy the group of comput computers ers the multic multicast ast message message is intended for. Class + accounts for /Ath (%A;,>5?,>?A or %%; of the available )P addresses.
6 C"ass : 6 Class 2 is used for experimental purposes only. 4ike Class +, it is different from the first three classes. )t has a first bit value of , second bit value of , third bit value of and fourth bit value of . "he other %; bits are used to identify the group of computers the multicast message is intended for. Class 2 accounts for /Ath (%A;,>5?,>?A or %%; of the available )P addresses.
Page 15 of 45
Fig 5. )P address
0roa!ast !0 3essages that are intended for all computers on a network are sent as broadcasts. "hese messages always use the )P address %??.%??.%??.%??.
/!!ress!0 "he uni8ue number )+ assigned to one host or interface in a network.
(+3net !0 ' portion of a network sharing a particular subnet address.
(+3net ,ask !0 !0 ' 5%0bit combination used to describe which portion of an address refers to the subnet and which part refers to the host.
Page 1 of 45
I* a!!ressing: Iiven an )P address, its class can be determined from the three high0order bits. Figure shows the significance in the three high order bits and the range of addresses that fall into each class. For informational purposes, Class + and Class 2 addresses are also shown.
ig+re #.2: I* /!!ress /!!ress
$ain feat+res Of I* are
*aketi=ation: +ata from an upper layer protocol is encapsulated inside one or
more packets/datagrams (the terms are basically synonymous in )P. No circuit setup is needed before a host h ost tries to send packets to a host it has ha s previously not communicated with (this is the point of a packet0switched network, thus )P ()nternet protocol is a connectionless protocol. Page 16 of 45
I* *aket or,at: 'n )P packet contains several types of information.
>ersion- )ndicates the version of )P currently used.
I* 7ea!er ength %I7&- )ndicates the datagram header length in 5%0bit words
Type-of-(er)ie0 *pecifies how an upper0layer protocol would like a current
datagram to be handled, and assigns datagrams various levels of importance.
Tota" ength *pecifies the length, in bytes, of the entire )P packet, including the
data and header.
I!entifiation 0 Contains an integer that identifies the current datagram. "his field
is used to help piece together datagram fragments.
"ags- Consists of a 50bit field of which the two low0order (least0significant (least0significant bits
control fragmentation. "he low0order bit specifies whether the packet can be fragmented. "he middle bit specifies whether the packet is the last fragment in a series of fragmented packets. "he third or high0order bit is not used.
rag,ent Offset- )ndicates the position of the fragment7s data relative to the
beginning of the data in the original datagram, which allows the destination )P process to properly reconstruct the original datagram.
Ti,e-to-i)e- 3aintains a counter that gradually decrements down to 9ero, at
which point the datagram is discarded. "his keeps packets from looping endlessly.
*rotoo"- )ndicates which upper0layer protocol receives incoming packets after
)P processing is complete.
Page 19 of 45
integrity. 7ea!er Cheks+,- 1elps ensure )P header integrity.
(o+re /!!ress- *pecifies the sending node.
Destination /!!ress- *pecifies the receiving node.
security. Options- 'llows )P to support various options, such as security.
#.2 (+3netting: =y extending the mask to be %??.%??.%??.%%>, you have taken three bits (indicated by HsubH from the original host portion of the address and used them to make subnets. $ith these three bits, it is p ossible to create eight subnets. $ith the remaining five host )+ bits, each subnet can have up to 5% host addresses, 5& of which can actually be assigned to a device since host ids of all 9eros or all ones are not allowed (it is very important to rememb remember er this. this. *o, with with this this in mind, mind, these these subnet subnetss have been been create created. d. %&>.J. %&>.J.?.& ?.& %??.%??.%??.%%> host address range to 5& %&>.J.?.5% %??.%??.%??.%%> host address range 55 to A% %&>.J.?.A> %??.%??.%??.%%> host address range A? to :> %&>.J.?.:A%??.%??.%??.%%> host address range :J to %A %&>.J.?.%; %??.%??.%??.%%> host address range %: to ?; %&>.J.?.A&%??.%??.%??.%%> host address range A to :& %&>.J.?.:% %??.%??.%??.%%> host address range :5 to %%% %&>.J.?.%%> %??.%??.%??.%%> host address range %%? to %?>
Types of (+3netting: •
i?e! ength (+3net $ask %($&
•
>aria3"e ength (+3net $ask %>($&
Page 1 of 45
($: (teps of (+3netting for ($
For )P address :%.A;.&.& (Class C •
)dentify the total no. of subnets %Mn K no.of subnets
•
$here n are the no.s and borrowed borrowed bytes from host )+ portion. 4et we are given that we have to make > subnets. "herefore %Mn K> i.e nK%
•
"o idettify the total no. of the valid hosts for each subnet.
•
%Mm0%K no.of valid hosts. $here m are the remaining no. of bits in host )+ %MA0 %KA%
•
Calculate the subnet mask and range
•
*ubnet
mask
for
n/w
:%.A;.&.&/%A
is
. . . . .&&&&& .&&&&&&&& &&& ie %??.%??.%? %??.% ??.%??.:% ?.:% •
rangeK %?A0:%KA>
•
)dentify the total no of subnets, no. of valid hosts and the broadcast address.
>($ )n -4*3 to allocate )P addresses to subnets depending upon the no. of hosts. "he network having more no of hosts is given priority and the one having least no of host comes at last and for each network the subnet is assigned separately.
ig #.#: )aria3"e s+3net ,ask
Page 2; of 45
>($ ?a,p"e: Iiven the same network and re8uirements as in develop a sub0netting scheme using -4*3, given! net'! must support > hosts net=! must support %; hosts netC! must support % hosts net+! must support J hosts net2! must support %; host +etermine what mask allows the re8uired number of hosts. net'! re8uires a /%; (%??.%??.%??.%>& mask to support > hosts net=! re8uires a /%J (%??.%??.%??.%%> mask to support %; hosts netC! re8uires a /5& (%??.%??.%??.%?% mask to support % hosts net+O! re8uires a /%; (%??.%??.%??.%>& (%??.%??.%??.%> & mask to support J hosts net2! re8uires a /%J (%??.%??.%??.%%> mask to support %; hosts ' /%: (%??.%??.%??.%>; would only allow A usable host addresses "herefore net+ re8uires a /%; mask. "he easiest way to assign the subnets is to assign the largest first. For example, you can assign in this manner! net=! %&>.?.?.&/%J host address range to 5& net2! %&>.?.?.5%/%J host address range 55 to A% net'! %&>.?.?.A>/%; host address range A? to J; net+! %&>.?.?.;&/%; host address range ; to :> netC! %&>.?.?.:A/5& host address range :J to :;
#.# ra,e Re"ay: Frame @elay is still one of the most popular $'N services deployed over the past decade, and there7s a good reason for thiscost. =y default, Frame @elay is classified as a non0broadcast multi0access (N=3' network, meaning it doesn7t send any broadcasts like @)P updates across the network. Frame @elay has at its roots a technology called <.%?, and it essentially incorporates the components of <.%? that are still relevant to today7s reliable reliable and relatively QcleanR telecommunications networks while leaving out the no0longer0needed error0correction components. )t7s substantially more complex than the simple leased0line networks you learned about when ) discussed the 1+4C and PPP protocols, but is still relevant when looking at event the most commonly used networks Page 21 of 45
from providers such as o%, or other similar companies. "he leased0line networks are easy to conceptuali9e 0 but not so much when it comes to Frame @elay. )t can be significantly more more comple complex x and and vers versat atil ile, e, whic which h is why why it7 it7s ofte often n repr repres esen ente ted d as a Qclo QcloudR udR in networking graphics. Gou won7t won7t be usin using g the the encap encapsu sula lati tion on 1+4C 1+4C or enca encaps psul ulat atio ion n PPP PPP comma command nd to configure it.Frame @elay doesn7t work like a point0to0point leased line (although it can be made to look and act like one.Frame @elay is usually less expensive than leased lines are, but there are some sacrifices to make to get that savings. )f, for example, you had to add seven remote sites to the corporate office and had only one free serial port on your routerit7s Frame @elay to the rescueS f course, ) should probably mention that you now also have one single point of failure, which is not so good. =ut Frame @elay is used to save money, not to make a network more resilient. "ake "ake a look at Fig. >5 to get an idea of what a network looked like before and after Frame @elay.
ig #.4: ra,e Re"ay
#.4 >/N: 's networks have grown in si9e and complexity, many companies have turned to virtual local area networks networks (-4'Ns to provide provide some way of structuri structuring ng this growth logically logically. =asically, a -4'N is a collection of nodes that are grouped together in a single broadcast domain that is based on something other than physical location. 1ere are some common reasons why a company might have -4'Ns!
Page 22 of 45
•
(e+ (e+rit rity y 0 *epara *eparatin ting g syste systems ms that that have sensitiv sensitivee data data from from the rest rest of the
network decreases the chances that people will gain access to information they are not authori9ed to see. •
*roets@(peia" app"iations 0 3anaging a proBect or working with a speciali9ed
application can be simplified by the use of a -4'N that brings all of the re8uired nodes together. •
*erfor, *erfor,an ane@0a e@0an!wi n!wi!th !th 0 Care Carefu full moni monito tori ring ng of netw networ ork k use use allo allows ws the the
network administrator to create -4'Ns that reduce the n umber of router hops and increase the apparent bandwidth for network users. •
0roa!asts@Traffi f"ow 0 *ince a principle element of a -4'N is the fact that it
does does not not pass pass broa broadca dcast st traf traffi ficc to node nodess that that are are not part of the the -4'N -4'N,, it automatically reduces broadcasts. 'ccess lists provide the network administrator with a way to control who sees what network traffic. 'n access list is a table the networ network k admini administr strato atorr create createss that that lists lists which which addres addresses ses have access access to that that network. •
Depart Depart,ent ,ents@(p s@(pei eifi fi o3 types types 0 Comp Compani anies es may may want want -4'N -4'Nss set set up for for
departments that are heavy network users (such as multimedia or engineering, or a -4'N across departments that is dedicated to specific types of employees (such as managers or sales people.
#.5 (panning Tree *rotoo" %(T*& ' robust network design not only includes efficient transfer of packets or frames, but also considers how to recover 8uickly from faults in the network. )n a 4ayer 5 environment, env ironment, the routing protocols in use keep track of redundant paths to a destination network so that a secondary path can be used 8uickly if the primary path fails. 4ayer 5 routing allows many paths to a destination to remain up and active, and allows load sharing across multiple paths.
Page 2# of 45
)n a 4ayer % environment (switching or bridging, however, no routing protocols are used, and active redundant paths are neither allowed nor desirable. )nstead, some form of bridging provides data transport between networks or switch ports. "he *panning "ree Protocol (*"P provides network link redundancy so that a 4ayer % switched network can recover from failures without intervention in a timely manner. "he *"P is defined in the )222 ;&%.+ standard.
*re)enting oops with (panning Tree *rotoo"
=ridging loops form because parallel switches (or bridges are unaware of each other. *"P was developed to overcome the possibility of bridging loops so that redundant switches and switch paths could be used for their benefits. =asically, the protocol enables switches to become aware of each other so they can negotiate a loop0free path through the network.
4oops are discovered before they are made available for use, and redundant links are effect shut down to prevent the loops from forming. )n the case of redundant links, switches can be made aware that a link shut down for loop prevention should be brought up 8uickly in case of a link failure.
*"P is communicated among all connected switches on a network. 2ach switch executes the spanning0tr spanning0tree ee algorithm algorithm based on informati information on received received from other neighbourin neighbouring g switches. "he algorithm chooses a reference point in the network and calculates all the reduct paths to that reference point. $hen redund ant paths are found, the spanning0tree algorithm picks one path by which to forward frames and disables, or blocks, forwarding on the other redundant paths.
's its name implies, *"P computes a tree structure that spans all switches in a subnet or network. @edundant paths are placed in a =locking or *tandby state to prevent frame forwarding. Page 24 of 45
"he switched network is then in a loop0free condition. 1owever, if a forwarding port fails or becomes disconnected, the spanning0tree algorithm re computes the spanning tree topology so that the appropriate blocked links can be reactivated.
7ow (T* orksA
"eting a Root 0ri!ge
For all switches in a network to agree on a loop0free topology, a common frame of reference must exist to use as a guide. "his reference point is called the root bridge. bridge. ("he term term brid bridge ge continues to be used even in a switched environment because *"P was developed for use in bridges. "herefore, when you see bridge, bridge, think switch think switch. . 'n election process among all connected switches chooses the root bridge. 2ach switch has a uni8ue bridge ID that identifies it to other switches. "he bridge )+ is an ;0byte value consisting of the following fields!
"he priority or weight of a switch in relation to all other 0ri!ge *riority %2 3ytes&B "he switches. "he Priority field can have a value of & to A?,?5? and defaults to 5%,JA; (or &x;&&& on every Catalyst switch.
$/C /!!ress % 3ytes&B "he "he 3'C address used by a switch can come from the
*upervisor module, the backplane, or a pool of ,&%> addresses that are assigned assigned to every supervisor or backplane, depending on the switch model. )n any event, this address is hard0coded and uni8ue, and the user cannot change it.
's an example, consider the small network shown in Figure. For simplicity, assume that each Catalyst switch has a 3'C address of all &s, with the last hex digit e8ual to the switch label.
Page 25 of 45
ig #.5: $/C a!!ress
)n this network, each switch has the default bridge priority of 5%,JA;. "he switches are interconnected Fast 2thernet links. 'll three switches try to elect themselves as the root, but all of them have e8ual =ridge Priority values. "he election outcome produces the root bridge, determined by the lowest 3'C addressthat of Catalyst '.
"eting Root *orts
Now that a reference point has been nominated and elected for the entire switched network, each non root switch must figure out where it is in relation to the root bridge. "his action can be performed by selecting only one root port on each non root switch. "he root port always points toward the current root bridge. *"P uses the concept of cost to determine many things. *electing a root port involves evaluating evaluating the root path cost . "his value is the cumulative cost of all the links leading to the root bridge. ' particular switch link also has a cost associated with it, called the path cost . "o understand the difference between these values, remember that only the root path cost is carried inside the =P+#. 's the root path cost travels along, other switches can modify its value to make it cumulative. "he path cost, however, is not contained in the
Page 2 of 45
=P+#. )t is known only to the local switch where the port (or QpathR to a neighboring switch resides. "he original )222 ;&%.+ standard defined path cost as &&& 3bps divided by the link bandwidth in megabits per second. "hese values are shown in the center column of the table. 3odern networks commonly use Iigabit 2thernet and C0>; '"3, '"3, which are both either too close to or greater than the maximum scale of &&& 3bps. "he )222 now use sa nonlinear scale for path cost "he root path cost value is determined in the following manner! 1. "he root bridge sends out a =P+# with a root path cost value of & because its ports
sit directly on the root bridge. 2. $hen the next0closest neighbor receives the =P+#, it adds the path cost of its own
port where the =P+# arrived. ("his is done as the =P+# is received . . #. "he neighbor sends out =P+#s with this new cumulative value as the root path cost. 4. "he root path cost is incremented by the ingress port path cost as the =P+# is
received at each switch down the line. 5. Notice the emphasis on incrementing the root path cost as =P+#s are received .
$hen computing the spanning0tree algorithm manually, remember to compute a newroot path cost as =P+#s come in to a switch port, not as they go out.
ig #.: "eting Root 0ri!ge
Page 26 of 45
"eting Designate! *orts
' starting or reference point has been identified, and each switch QconnectsR itself toward the reference point with w ith the single link that has ha s the best path. ' tree structure is beginning to emerge, but links have only been identified at this point. 'll links still are connected and could be active, leaving bridging loops. "o remove remove the possibility possibility of bridging bridging loops, *"P makes a final computation computation to identify identify one designated designated port on each network segment. *uppose that two or more switches have ports connected to a single common network segment. )f a frame appears on that segment, all the bridges attempt to forward it to its destination. )n each determination process discussed so far, two or more links might have identical root path costs. "his results in a tie condition, unless other factors are co nsidered. 'll 'll tie *"P decisions are based on the following se8uence of four conditions!
1. 4owest root bridge )+ 2. 4owest root path cost to root bridge #. 4owest sender bridge )+ 4. 4owest sender port )+
ig #.6: "eting Designate! *orts
Page 29 of 45
"he three switches have chosen their designated ports (+P for the following reasons!
Cata"yst /
=ecause =ecause this switch switch is the root bridge bridge,, all its active active ports are design designate ated d ports, ports, by definition. 't the root bridge, the root path cost of each port is &.
Cata"yst 0
Catalyst ' port / is the +P for the *egment '6= because it has the lowest root path cost (&. Catalyst = port /% is the +P for segment =6C. "he root path cost for each end of this segment is :, determined from the incoming =P+# on port /. =ecause the root path cost is e8ual on both ports of the segment, the +P must be chosen by the next criteria criteria the lowest sender bridge )+. $hen Catalyst = sends a =P+# to Catalyst C, it has the lowest 3'C address in the bridge )+. Catalyst C also sends a =P+# to Catalyst =, but its sender bridge )+ is higher. "herefore, Catalyst = port /% is selected as the segment7s +P.
Cata"yst C
Catalyst ' port /% is the +P for *egment '6C because it has the lowest root path cost (&. Catalyst = port /% is the +P for *egment =6C. "herefore, Catalyst C port /% will be neither a root port nor a designated port. 's discussed in the next section, any port that is not elected to either position enters the =locking state.
(T* (tates
"o participate in *"P, each port of a switch must progress through several states. ' port begins its life in a +isabled state, moving through several passive states and, finally, finally, into an active state if allowed to forward traffic. "he *"P port states are as follows!
Disa3"e!B Ports Ports that are administratively shut down by the network administrator, or by
the system because of a fault condition, are in the +isabled state. "his state is special and is not part of the normal *"P progression for a port.
Page 2 of 45
'fter a port initiali9es, it begins in the =locking state so that no bridging 0"okingB 'fter loops can form. )n the =locking state, a port cannot receive or transmit data and cannot add 3'C addresses to its address table. )nstead, a port is allowed to receive only =P+#s so that the switch can hear from other neighboring switches. )n addition, ports that are put into standby mode to remove a bridging loop enter the =locking state.
isteningB ' port is moved from =locking to 4istening if the switch thinks that the port
can be selected as a root port or designated port. )n other words, the port is on its way to begin forwarding traffic.
)n the 4istening state, the port still cannot send or receive data frames. 1owever, the port is allowed to receive and send =P+#s so that it can actively participate in the *panning "ree "ree topol topology ogy proc proces ess. s. 1ere, 1ere, the the port port final finally ly is allo allowed wed to beco become me a root root port port or designated port because the switch can advertise the port by sending =P+#s to other switches. )f the port loses its root port or designated port status, it returns to the =locking state.
'fter a period of time called the Forward the Forward Delay in the 4istening state, the earningB 'fter port is allowed to move into the 4earning state. "he port still sends and receives =P+#s as before. )n addition, the switch now can learn new 3'C addresses to add to its address table. "his gives the port an extra period of silent participation and allows the switch to assemble at least some address information. "he port cannot yet send any data frames, however.
'fter another Forward +elay period of time in the 4earning state, the port orwar!ingB 'fter is allowed to move into the Forwarding state. "he port now can send and receive data frames, collect 3'C addresses in its address table, and send an d receive =P+#s. "he port is now a fully functioning switch port within the spanning0tree topology. @emember that a switch port is allowed into the Forwarding state only if no redundant links (or loops are detected and if the port has the best path to the root bridge as the root port or designated port. Page #; of 45
#. Ro+ting: @outing is the process of directing packets from a source node to a destination node on a different network. )t is of two types!
(tati ro+ting: "he process of manually adding routes in each routerDs routing table. "he administrator configures the destination network, next hop, and appropriate metrics. "he route doesnDt change until the network administrator ad ministrator changes it.
/!)antages:•
No overhead on router CP#.
•
No bandwidth usage between links.
•
*ecurity (only administrator adds routes.
Disa!)antages:•
'dmini 'dministr strato atorr must must really really underst understand and intern internetw etwork ork and how each each router router is connected.
•
)f a new network is added, administrator must update all routers. Not practical on large networks as it is time intensive.
Dyna,i ro+ting: +ynamic routes adBust to changes within the internetwork environment automatically. $hen network changes occur, routers begin to converge by recalculating routes and distributing route updates. "he route update messages spread through the network, which causes other routers to recalculate their routes. "he process continues until all routes have converged. #ses protocols to find and update routes on a routing table. )t uses CP# time and consumes bandwidth between links. "he routing protocol defines the rules used by the routers when they communicate with each other. "here are two types of routing protocols on internetworks, )nterior Iateway Protocol ()IP and 2xterior Iateway Page #1 of 45
Protocol (2IP. )IP is used in networks in the same administrative domain. 2IPs are used to communicate between the domains.
#.6 Ro+te! protoo"s: @outed protocols are nothing more than data being transported across the networks. @outed protocols include! •
)nternet Protocol
•
"elnet
•
@emote Procedure Call (@PC
•
*N3P
•
*3"P
•
Novell )P<
•
pen *tandards )nstitute networking protocol
•
+2Cnet
•
'ppletalk
•
=anyan -ines -ines
•
#.9 Ro+ting protoo"s: protoo"s: @outing Protocols are the software that allow routers to dynamically advertise and learn routes, determine which routes are available and which are the most efficient routes to a destination. @outing protocols used by the )nternet Protocol suite include! •
@outing )nformation Protocol (@)P and @)P ))
•
pen *hortest Path First (*PF Page #2 of 45
•
)ntermediate *ystem to )ntermediate *ystem ()*0)*
•
)nterior Iateway @outing Protocol ()I@P
•
2nhanced )nterior Iateway @outing Protocol (2)I@P
RI* %Ro+ting Infor,ation *rotoo"&: @)P (@outi (@outing ng )nform )nformati ation on Protoc Protocol ol is a widely widely0use 0used d protoc protocol ol for managi managing ng router router inform informati ation on within within a self0c self0cont ontain ained ed networ network k such such as a corpor corporate ate local local area area networ network k (4'N (4'N or an interc interconne onnecte cted d group group of such such 4'Ns. 4'Ns. @)P is classi classifi fied ed by the )ntern )nternet et 2ngineering "ask Force ()2"F as one of several internal gateway protocols ()nterior Iateway Protocol. #sing @)P, a gateway host (with a router sends its entire routing table (which lists all the other hosts it knows about to its closest neighbor host every 5& seconds. "he neighbor host in turn will pass the information on to its next neighbor and so on until all hosts within the network have the same knowledge of routing paths, a state known as network convergence. @)P uses a hop count as a way to determine network distance. (ther protocols use more sophisticated algorithms that include timing as well. 2ach host with a router in the network uses the routing table information to determine the next host to route a packet to for a specified destination. @)P is considered an effective solution for small homogeneous networks. For larger, more complicated networks, @)PDs transmission of the entire routing table every 5& seconds may put a heavy amount a mount of extra traffic in the network. "he maBor alternative to @)P is the pen *hortest Path First Protocol (*PF.
O(* %Open (hortest *ath irst&:
pen *hortest Path First is a true link state protocol developed as an open standard for routing )P across large multi0vendor networks. ' link state protocol will send link state adverti adve rtisem sements ents to all con connec nected ted nei neighbo ghbors rs of the sam samee are areaa to com commun municat icatee rou route te information. 2ach *PF enabled router, when started, will send hello packets to all directly connected *PF routers. "he hello packets contain information such as router
Page ## of 45
timers, router )+ and subnet mask. )f the routers agree on the information they become *PF * PF ne neig ighb hbor ors. s. n nce ce ro rout uter erss be becom comee ne neig ighb hbor orss th they ey es esta tabl blis ish h ad adBa Bacen cenci cies es by exchanging link state databases. @outers on point0to0point and point0to0multipoint links (as specified with the *PF interface typesetting automatically establish adBacencies.
IR* %nhane! %nhane! Interior ateway Ro+ting *rotoo" &:
2nhanced 2nhance d )nter )nterior ior Iateway @outing Protocol is a hybri hybrid d routi routing ng protoc protocol ol developed by Cisco systems for routing many protocols across an enterprise Cisco network. )t has characteristics of both distance vector routing protocols and link state routing protocols. )t is proprietary which re8uires that you use Cisco routers. 2)I@P will route the same protocols that )I@P routes ()P ()P,, )P< and use the same composite metrics as )I@P to select a best path destination. 's well there is the option to load balance traffic across e8uall or une e8ua une8ual 8ual metric metric cos costt pat paths. hs. *um *ummar mari9a i9atio tion n is aut automa omatic tic at a net networ work k cla class ss address addr ess how howeve everr it can be con config figure ured d to sum summar mari9e i9e at sub subnet net boun boundar daries ies as wel well. l. @edistribution between )I@P and 2)I@P is automatic as well. "here is support for a hop count of %?? and variable length subnet masks.
#. TNT! "elnet is a protocol which is part of the "CP/)P suite. )t is 8uite similar to the #N)< rlogin program. "elnet allows you to control a remote computer from your own one. )t is terminal emulation software. )n the old days hard drives were humongous and expensive and there were no personal computers. "o make use of existing computers you had to lease hard rive space and use terminals to operate the system. For developers this was great because computing became beca me lots cheaper. Gou Gou needed a server and many connections connec tions could be made. $ith telnet u can emulate this type of distributed computing and for example operate a supercomputer from a distance.
#.1; D7C* %Dyna,i 7ost Config+ration *rotoo"&: +1CP (+ynamic 1ost Configuration Protocol is a communications protocol that lets networ network k admini administr strato ators rs central centrally ly manage manage and automa automate te the assign assignmen mentt of )ntern )nternet et Protocol ()P addresses in an organi9ationDs network. #sing the )nternet Protocol, each Page #4 of 45
machine that can connect to the )nternet needs a uni8ue )P address, which is assigned when an )nternet connection is created for a specific computer. $ithout +1CP, the )P address must be entered manually at each computer in an organi9ation and a new )P address must be entered each time a computer moves to a new location on the network. +1CP lets a network administrator supervise and distribute )P addresses from a central point and automatically sends a new )P address when a computer is plugged into a different place in the network.
Page #5 of 45
Chapter no. 4 /0OUT OUR *RO8CT 4.1 Intro!+tion: "he proBect is a communicati communication on model which represents represents a mesh of networking networking elements including routers, switches, servers (+1CP and +N* in this model, frame relay, access points, computers and different types of cables cab les to connect them. "he proBect is based on the technol technology ogy of 1ardwar 1ardwaree and Networ Networkin king, g, typic typicall ally y on CCN' CCN' (Cisco (Cisco Certi Certifie fied d Network 'ssociate. 'ssociate. "he essence of this proBect lies in the configuration of these network elements so that they can communicate with each other as re8uired by the network administrator. "he kind of access rights given to each network element and the services each computer can use depends on the configuration done by the network administrator. "he proBect typically shows the communication model for an organi9ation with its two branches and one headoffice. "he obBective of planning such a network model is to make easy the task of actual set up of a network. "he communication model prepared (as in this proBect acts as a guide while connecting the real computers and other network devices like routers, switches, and different types of servers. "he detailed obBectives are given as! •
asy to set +p at+a" network:
)t becomes so easy to set up an network that is prepared in the model. )t acts in similar way as a map of a building to be built. 's it is so diff diffic icul ultt Bob Bob to cons constr truc uctt a hous housee whose whose map map is not not avail availabl able, e, simi simila larl rly y it is very very cumbersome Bob to start connecting a lot of networking devices available in absence of a model. •
I* a!!ressing:
'ssigning the )P addresses to the network is the first and the most important task. )P addresses are actually uni8ue addresses to each network element. )t is
Page # of 45
the uni8ue code that identified the network element in the network. )n the network model, we have all the elements visible to us at a time, so we can assign them )P addresses easily, but the same Bob will be difficult to do on a group of computers, at different different locations. •
asy to ,ake hanges an! e?ten! the network:
)t becomes easy to make further changes in the network if is network model is available. $e can have a look on the settings and the implementations already done on it and so can modify it. the same Bob is really hectic to be done on actual network and can result the introduction of errors in it. •
asy to +n!erstan!:
' network can be easily understood for its structure, characteristics and configuration from the network model. )n the absence of this, each network element will need to be visited at different locations and checked for its configuration. •
sti,ation of the o,p+ters an! har!ware re+ire!:
'ny organi9ation first prepares a network model, before actual set up. "he main things like number of departments, ad number of computers in each department etc. are taken care of while preparing the network model. 'fter the network model is ready, the organi9ation can easily estimate the cost re8uired to have such network, the type of network elements re8uired and the number of these, thereby avoiding the wastage. •
Ca3"ing:
+ifferent network elements needed to be connected by different types of cables. )n actual, mistake can be done while connecting different elements with wrong type of cables and on the wrong interfaces as well. =ut a network model provides the types of cables and the detail of interfaces on which they should be connected, which helps a lot while its actual implementation.
•
Config+ration of network:
' lot of computers, routers and switches connected together can7t be called a network. Configuration is to be done on each network element that decides the
Page #6 of 45
working of network. 'll that configuration is already done in the network model. )t is easy to do the configuration on the model itself than to do the same on the actual network first time. "his will cause a lot of time waste and errors as well. $hen we have a network model, we can easily see the configuration done in the model and can implement the same on the actual network. n etwork. "his reduces the errors and saves the time. •
Ro+ting protoo"s:
$henever a group of elements are connected in a network, a routing protocol must be used in order to tell each network element the way or path to use for transmitting a packet from a particular source to the destination. "he routing protocols are also implemented in a communication model much easily. "he software used for the development of the proBect is Qcisco packet tracerR whose opening window is shown below in the figure. "his work area is used to prepare the network model. 1ere we can sele select ct the the nece necess ssar ary y hard hardwa ware re neede needed d to prep prepar aree the the model model and and also also can can alte alterr its its properties such as, we can add interfaces to the routers, wireless 4'N cards to the computers.
ig 4.1: Ciso *aket Traer
=efore coming to the proBect, here are some main points that demonstrate the features of packet tracer, which will be re8uired later to operate the proBect. $henever we place the cursor on the terminal, the packet tracer shows its )P address allocated, gateway, gateway, and all other properties which are assigned to it when it works within a network. Page #9 of 45
*ame is the case, when the cursor is pointed on the router, which is also a networking element, its various interfaces, active interfaces, )P address, 3'C address its hostname etc. are shown to enhance the understandability 't the bottom of the packet tracer screen, various devices are available for constructing a commun communica icatio tion n model. model. $hen $hen any device device is select selected, ed, its corres correspond ponding ing models models are available. 2xample, when a router is selected, its models in different series eg. %?&&, %A&& are displayed. *ame in the case of terminals, different kinds of computers like desktops, laptops, telephones that can be used as data terminal devices in a network are available.
Traer ig 4.2: The en! !e)ies a)ai"a3"e in *aket Traer
two similar kind of devices, eg both +"2 or +C2, then a cross cable (shown with dotted lines will be used. 'nd when different types of devices are being connected, like one +"2 and other +C2, the n serial cables will be used (with an exception ex ception of routers.
ig+re 4.#: The onnetions a)ai"a3"e in the *aket Traer Traer
4.2 *roet Detai"s: "he network model which we are designing will be consisting of routers, switches, computers, servers, hubs. 'll the above elements together represent an organi9ation. )n
Page # of 45
proBect, different technology of networking is implemented. "hese technologies are like named below! •
@outing protocol ! *PF (pen *hortest Path First Protocol
•
-4'N (-irtual 4ocal 'rea Networks
•
'C4('ccess Control 4ist both standard and an d extended 'C4 implementation
•
+N* (+omain Name *pace
•
1""P
•
+N*
•
+1CP server
•
Configuration of routers, switches, servers, access points and PC7s.
1ere in the figure, the complete model is shown which has been constructed in the proBect. 2ach part of the organi9ation has been given a different background color and according to the configuration done on it.
Page 4; of 45
ig 4.4: *roet O+t"ay
$hen ever any computer in the communication model is selected, packet tracer shows the window, shown in the figure below. "his window basically shows the options that any computer computer have. 2.g. command command prompt, option to allocate )P address address etc. we can use any of the service to ensure that the terminal connected is working correctly in the network.
Page 41 of 45
ig 4.5: The options a)ai"a3"e for a aptop@ Ter,ina"
=elow is the example given to check if one computer is communicating to another. "his is done by using QpingR command in the command prompt. "yping the keyword QpingR and then the )P address shows the result. "he reply is shown from the address to which we wanted wanted to commun communica icate, te, if they they are connected connected in right right manner manner,, or not blocked blocked explicitly, otherwise, failure is shown.
Page 42 of 45
ig 4.: Using *ing Co,,an!
Page 4# of 45
Chapter 5 Con"+sion an! +t+re (ope
Computer Networking is a very vast proBect in the present developing era of electronics and and comm commun unic icat atio ion. n. Now Now day days, comp comput uter erss are are used used in a wide widerr rang range. e. 'll the the organi9ations are using multiple computers within their departments to perform their day to day work. Computer network allows the user to share data, share folders and files with other users connected in a network. Computer Networking has bound the world in a very small area with it wide networking processes like 4'N, 3'N, $'N. Networking $'N. Networking inside your organi9ation is valuable also. )n larger companies, many people never meet others in the organi9ation that can facilitate facilitate solving problems or getting resources. "his proBect is forward compatible as we can add more branches at low cost and high efficiency with effective communication between head office and various branches of an organi9ation.
Page 44 of 45
Referenes "he following web links are visited for the theory reference! http!//www.cisco.com/web/learning/netacad/courseTcatalog/Packet"racer http!//www.cisco.com/web/learning/netacad/courseTcatalog/Packet"racer.html .html http!//www.cisco.com/web/learning/netacad/index.html http!//www.cisco.com/en/#*/docs/internetworking/technology/handbook/itoTdoc.html http!//netcert.tripod.com/ccna/routers/routeprotocols.html http!//www.livinginternet.com/i/iwTroute.htm http!//en.wikipedia.org/wiki/+ynamicT1ostTConfigurationTProtocol http!//www.isc.org/software/dhcp http!//www.cisco.com/web/)N/products/routers/index.html http!//www.webopedia.com/"2@3/@/router. http!//www.webopedia.com/"2@3/@/router.html html
Page 45 of 45
