Copyright 2005-2009 Kenneth M. Chipps Ph.D. Routing and Routing Protocols Last Update 2009.07.17...

Copyright 2005-2009 Kenneth M. Chipps Ph.D. www.chipps.com

Routing and Routing Protocols Last Update 2009.07.17

1.5.0

1


2

Objectives


3

Routing Tables Population

• For the router to be able to handle arriving frames, entries must be made into the routing table

• Entries in a routing table can be generated in three ways– Directly connected routes– Static routes– Dynamic routes

Directly Connected Routes

• When a data line exists• When a cable is connected between the

demarc of that data line and an interface of the router

• When the data line is active• When the interface on the router is

activated• A directly connected route is added to the

routing tableCopyright 2005-2009 Kenneth M. Chipps Ph.D.

www.chipps.com4

Directly Connected Routes


5

Static Routes

• When the administrator uses the ip route command to add a route to the routing table

• Static routes are used when– The internetwork is small, may seldom

change, or has no redundant links– The routers need to use dial backup to

dynamically call another router when a leased line fails


6

Static Routes

– An enterprise internetwork has many small branch offices, each with only one possible path to reach the rest of the internetwork

– An enterprise wants to forward packets to hosts in the Internet, not to hosts in the enterprise network


7

Result of Using a Static Route


8

Result of Using a Static Route


9

Dynamic Routes

• When a routing protocol is activated on the router

• When other routers running the same routing protocol talk to each other

• Then the routes know by the other routers are added to the routing table as dynamic routes


10

Routing Table Entries

• The command show ip route shows the routing table

• Each of these methods of entering routes in the routing table has an indicator associated with it– C for directly connected networks– S for static routes– R for routes learned through the RIP routing

protocol as an example of a dynamic routeCopyright 2005-2009 Kenneth M. Chipps Ph.D.

www.chipps.com11


12

Two Ways to Look at Protocols

• Routing Protocols• Routed Protocols


13

What is a Routing Protocol

• What is a routing protocol• These are network layer protocols that are

responsible for path determination and traffic switching

• These have to do with the actual routes the packets take and how that path is calculated

• These protocols include RIP, EIGRP, OSPF, IS-IS, and BGP


14

What is a Routing Protocol


15

What is a Routed Protocol

• What is a routed protocol• These protocols are routed by the routing

protocols• They are concerned with the construction

and transport of the data itself regardless of how it arrives at its destination

• When the OSI model talks about encapsulation, this is what it is referring to


16


• These cover all 7 layers of the OSI model• These protocols contain enough

information in the fields in their headers that allow the packet to be routed from one network to another by the routing protocol


17



18

Static v Dynamic Routing

• Static routes, as we will see, are entered from the keyboard and do not require routing protocols

• Dynamic routes are created by routing protocols


19


• Static routing consists of entries made into the routing table in the router by the network administrator prior to the beginning of routing

• These entries do not change unless the network administrator alters them


20


• This method works well in environments where network traffic is relatively predictable and where network design is relatively simple

• Because static routing systems cannot react to network changes, they generally are considered unsuitable for today's large, changing networks


21


• Whereas dynamic routing protocols can adjust to changing network circumstances by analyzing incoming routing update messages

• If the message indicates that a network change has occurred, the routing software recalculates routes and sends out new routing update messages


22


• These messages permeate the network, stimulating routers to rerun their algorithms and change their routing tables accordingly


23




24


25

Dynamic Routing Protocols

• Dynamic routing protocols usually have one or more of the following design goals– Optimality – Low overhead – Robustness– Flexibility– Rapid convergence


26

Routing Protocol Optimality

• Optimality refers to the capability of the routing protocol to select the best route, which depends on the metrics and metric weightings used to make the calculation


27

Routing Protocol Overhead

• Low overhead refers to simple and efficient overhead


28

Routing Protocol Robustness

• Routing protocols must be robust, which means that they should perform correctly in the face of unusual or unforeseen circumstances, such as hardware failures, high load conditions, and incorrect implementations


29

Routing Protocol Flexibility

• Routing protocols should also be flexible, which means that they should quickly and accurately adapt to a variety of network circumstances– Assume, for example, that a network segment

has gone down– As they become aware of the problem, many

routing protocols will quickly select the next-best path for all routes normally using that segment


30

Routing Protocol Convergence

• Routing protocols must converge rapidly, which is a process of agreement, by all routers, on optimal routes

• When a network event causes routes either to go down or become available, routers distribute routing update messages that permeate networks, stimulating recalculation of optimal routes and eventually causing all routers to agree


31


• Routing protocols that converge slowly can cause routing loops or network outages

• An example of the need for rapid convergence is seen in the use of distance vector protocols

• Routers using routing protocols based on the distance vector method receive their neighbor’s routing table


32


• Using this they build a network map• This approach to learning can cause

problems such as routing loops and counts to infinity

• Routing loops can occur if the internetwork is slow to converge on a new configuration after a route fails

• This situation will produce inconsistent entries in the router tables


33


• Let’s organize the different types of routing protocols and then discuss each one


34



Intradomainor

Interior

Interdomainor

Exterior

Distance Vector Link State Path Vector

Standard Proprietary Standard Proprietary Standard

HELLO 1 IGRP1 OSPF NLSP1 EGP1

RIP V11 EIGRP IS-IS BGP

RIP V2

1 No Longer Used


35

No Longer Used

• HELLO– The original NSFnet backbone consisted of

six Digital Equipment Corporation LSI-11 computers located across the United States

– These computers ran special software colloquially called fuzzball that enabled them to function as routers

– These fuzzball routers connected various networks to the NSFnet and the ARPAnet


36

No Longer Used

– The six NSFnet routers worked as an autonomous system and like any AS, used an interior routing protocol to exchange routing information

– The routing protocol used in these early routers was called the HELLO protocol

– It was developed in the early 1980s and documented in RFC 891 published December 1983


37

No Longer Used

– The name HELLO is capitalized, but is not an acronym; it simply refers to the word hello, since the protocol uses messages that are sort of analogous to the routers talking to each other

– The HELLO protocol uses a distance-vector algorithm, like the RIP

– Unlike RIP, HELLO does not use hop count as a metric


38

No Longer Used

– Instead, it attempts to select the best route by assessing network delays and choosing the path with the shortest delay

– One of the key jobs of routers using HELLO is to compute the time delay to send and receive datagrams to and from its neighbors

– On a regular basis, routers exchange HELLO messages that contain clock and timestamp information


39

No Longer Used

– By comparing the clock value and timestamp in the message to its own clock using a special algorithm, a receiving device can compute an estimate for the amount of time it takes to send a datagram over the link

– HELLO messages also contain routing information in the form of a set of destinations that the sending router is able to reach and a metric for each


40

No Longer Used

– However in this case, the metric is an estimate of the round-trip delay cost for each destination

– This information is added to the computed round-trip delay time for the link over which the message was received, and used to update the receiving router's own routing table


41

No Longer Used

• RIP Version1– This version of RIP only supports FLSM

based on address classes– As address classes no longer exist version 1

is useless


42

No Longer Used

• IGRP– With the deployment of EIGRP and OSPF

there is no longer any need for IGRP– Therefore, no one uses it any longer


43

No Longer Used

• NLSP– NLSP - NetWare Link Services Protocol is a

link-state routing protocol in the Novell NetWare architecture

– NLSP is based on the OSI IS-IS or Intermediate System-to-Intermediate System protocol and was designed to replace IPX RIP and SAP, Novell's original routing protocols that were designed for small scale internetworks


44

No Longer Used

– Compared to RIP and SAP, NLSP provides improved routing, better efficiency, and scalability

– As no one uses NetWare anymore, no one uses NLSP any longer


45

No Longer Used

• EGP– EGP – Exterior Gateway Protocol was the first

routing protocol used to allow autonomous systems to talk to each other

– It was developed in 1982 by Eric C. Rosen and David L. Mills

– It was first formally described in RFC 827 and formally specified in RFC 904 in 1984

– EGP is no longer used

No Longer Used

– BGP - Border Gateway Protocol is now the accepted standard for Internet routing and has essentially replaced the more limited EGP


46

Interior and Exterior Protocols

• What is the difference between the various classes of routing protocols

• Where are exterior and interior protocols used


47


48

Interior and Exterior ProtocolsAUTONOMOUS SYSTEM

INTRADOMAININTERIOR

RIPIGRP

EIGRPOSPF

USED INSIDE AN AUTONOMOUS SYSTEM

ALL ROUTERS THAT ARE DIRECTLYCOMMUNICATING WITH EACH OTHER MUST ALLUSE THE SAME ROUTING PROTOCOL.THIS MUST BE BE INTRADOMAIN, IN OTHERWORDS INTERIOR, ROUTING PROTOCOL.IT CAN BE EITHER A DISTANCE VECTOR ORLINK STATE TYPE OF ROUTING PROTOCOL.IF THESE ROUTERS ARE ALL FROM DIFFERENTMANUFACTURERS, THEN THE ROUTINGPROTOCOL MUST BE STANDARDS BASED, NOTPROPRIETARY.

INTERDOMAINEXTERIOR

BGPUSED BETWEEN

AUTONOMOUS SYSTEMS

ALL ROUTERS THAT ARE DIRECTLYCOMMUNICATING WITH EACH OTHER MUST ALLUSE THE SAME ROUTING PROTOCOL.THIS MUST BE BE INTRADOMAIN, IN OTHERWORDS INTERIOR, ROUTING PROTOCOL.IT CAN BE EITHER A DISTANCE VECTOR ORLINK STATE TYPE OF ROUTING PROTOCOL.IF THESE ROUTERS ARE ALL FROM DIFFERENTMANUFACTURERS, THEN THE ROUTINGPROTOCOL MUST BE STANDARDS BASED, NOTPROPRIETARY.

AUTONOMOUS SYSTEMINTRADOMAIN

INTERIORRIP

IGRPEIGRPOSPF

USED INSIDE AN AUTONOMOUS SYSTEM


49



50



51

Intradomain v Interdomain

• Some routing protocols work only within domains

• Others work between domains• A domain in these terms is an autonomous

system, which is a group of routers under a single administrative control


52


• The nature of these two types of routing protocols is different

• In that the intradomain routing protocols are concerned with talking to only their close relatives

• Whereas interdomain routing protocols are concerned with talking to strangers


53


• Using these two types enables the organization to control the type and amount of outside traffic that comes in and goes out of its network

• The terms are also expressed as interior – intradomain and exterior – interdomain


54



55

Distance Vector Protocols

• A distance vector protocol is so named because its routes are advertised as vectors - distance and direction - where distance is defined in terms of a metric and direction is defined in terms of the next hop router

• These, known as Bellman-Ford protocols, call for each router to send all or some its routing table, but only to its neighbors


56


• In this arrangement each router depends on its neighbors for information, which its neighbors may have learned from their neighbors, and so on

• An individual router has no way of knowing if the information in the routing table it receives is accurate

• These routers just believe everything they hear


57


• As such distance vector routing protocols are sometimes referred to as routing by rumor

• A typical distance vector routing protocol uses a routing algorithm in which routers periodically send routing updates to all neighbors by broadcasting their entire routing table


58


• In this case periodically means to transmit on a regular schedule

• Neighbors are those routers at the other end of a data line

• The originating router sends its update to this neighbor

• It expects the neighbor to send the information on to that router's neighbors, and so on


59


• This update includes everything the router knows

• In other words its entire routing table with a few exceptions is sent out


60

Link State Protocols

• The information that a distance vector protocol has available has been likened to a road sign

• That is it is just one more step on the journey

• Whereas the information available to a link state protocol is more like a road map


61


• A link state routing protocol cannot be easily fooled into making a bad routing decision because - with the map - it has a complete picture of the network

• This is because link state routers have first hand information from all of their peer routers, those that speak the same routing protocol


62


• Each of these routers originates information about itself, its directly connected links, and the state of those links

• This information is passed around from router to router, each router making a copy, but no router changing the information

• How does this all work


63

Link State Concepts


64


• Like this– Each router establishes a relationship - an

adjacency - with each of its neighbors– Each router sends link state advertisements

to each neighbor– One link state advertisement is created for

each of the router's links, identifying the link, the state of the link, the metric cost of the link, and the neighbors that are connected to the link


65


– Each router receiving this information in turns forwards it to its neighbors

– Each router stores the link state advertisements it has received in a database

– Since all routers receive all link state advertisements, all routers have the same information

– The algorithm for the routing protocol is then applied to the information in the link state database to create a routing table

Dijkstra Algorithm

• Link-state protocols use the Dijkstra SPF - Shortest Path First algorithm to calculate and add routes to the IP routing table

• The SPF algorithm calculates all the possible routes to each destination network, and the cumulative metric for the entire path

66Copyright 2005-2009 Kenneth M. Chipps Ph.D. www.chipps.com

Dijkstra Algorithm

• Each router views itself as the starting point, and each subnet as the destination, and use the SPF algorithm to look at the LSDB - Link State Database to create a roadmap and pick the best route to each subnet


67

Characteristics of Link State

• The main features of link-state routing protocols– All routers learn the same detailed information

about the states of all the router links in the internetwork

– The individual pieces of topology information are called LSAs, with all LSAs stored in RAM in the LSDB


68


– Routers flood LSAs when they are created, on a regular but long time interval if the LSAs do not change over time, and immediately when an LSA changes

– The LSDB does not contain routes, but it does contain information that can be processed by the Dijkstra SPF algorithm to find a router’s best routes


69


– Each router runs the SPF algorithm, with the LSDB as input, resulting in the best - lowest cost - routes being added to the IP routing table

– Link-state protocols converge quickly by immediately reflooding LSAs and rerunning the SPF algorithm


70


– Link-state protocols consume much more RAM and CPU than do distance vector routing protocols

– If the internetwork changes a lot, link-state protocols can also consume much more bandwidth due to the relative to distance vector protocols large number of bytes of information in each LSA


71


72

Routing Metrics

• Routing protocols use metrics to determine the best or optimal route

• The following metrics are often used– Path Length– Reliability– Delay– Bandwidth– Load– Cost


73

Routing Metrics


74

Routing Metric Components

Example Routing Metrics


75

Administrative Distance

• A single router may learn routes from many different sources

• For example, from static routes and from running multiple routing protocols

• When a router learns more than one route to the same subnet, from different sources, the router needs to decide which route is best and then add that route to the IP routing table


76


• Because each routing protocol uses a different metric, a router cannot use the metric to determine which route is the best route

• When choosing between multiple routes to the same destination but learned from different sources, the router picks the route with the lowest administrative distance


77


• The administrative distance is a number assigned to all the possible sources of routing information, routing protocols and static routes included


78

Default Distance Values


79


• The administrative distance is shown by issuing the show ip route command

• The show ip route command output lists the administrative distance for most routes, with the notable exception of connected routes, which default to an administrative distance of 0

• The example shown next shows the output of the show ip route rip command


80


• The output highlights the administrative distance for the one RIP route known on router R1, which defaults to RIP’s setting of 120


81



82

Multiple Equal Cost Routes

• A single router may learn several routes to the same subnet, but the metrics may tie

• These routes are typically called equal-cost routes

• When this occurs, the router uses the following logic to choose which route to add to its routing table


83


– It can add up to four of the routes to the routing table, which is the default

– The number of equal-cost routes added to the routing table can be changed to between one and six by using the maximum-path number command as a subcommand of the routing protocol

• After routes are added to the routing table, the router then load-balances the traffic over various routes


84



85


86

Populating Routing Tables

• Let’s now look a little closer at the details of the two ways of populating routing tables– Static Routes– Dynamic Routes


87

Static Routes

• To use static routes an entry is made directly into the router's routing table from the command line of the router's operating system

• For example, to make such an entry into the routing table of a Cisco router the following is done


88

Static Routes

– At the enable level• routername#config terminal• routername(config)#ip route 192.168.1.0 255.255.255.224

S0• routername#CTRL Z

– This is read as follows– The command is ip route– The IP address is the address to be entered

into the table


89

Static Routes

– Next is a subnet mask to identify the network portion of the IP address

– Last is the address of the directly connected interface of the next hop router

– In this case out serial port 0• The above is done for all routes at each

router• This method is used for two main reasons


90

Static Routes

• The first is it is all that is needed for a private network in a hub and spoke arrangement

• The second reason is security• If no information is exchanged with any

outside entity, it is less likely that anyone will be able to determine the extent and layout of your network

Static Default Route

• There is a special kind of static route that is used when an entry cannot be found in the routing table for the network of interest

• This special type of static route is also used on stub networks when there is no other way out of the network

• In this case every packet that does not belong on the LAN is sent out to the default route


91


• When a router receives a packet whose network address is not found in the router’s IP routing table, the router discards the packet, unless a default route has been configured

• A default route tells a router where to send packets that do not match any of that router’s other IP routes


92


• Default routes can be most useful in two major cases– In routers that have only one possible

physical path to forward packets to the rest of the internetwork

– To route packets to the Internet, when there is a single connection to the Internet

– For example in the diagram that follows


93


• Each branch office has one router, with the only link back to the headquarters

• The enterprise network also has one link to an ISP for its Internet connection

• Configuring of static default route is similar for both cases

• On branch router R1, the command would be as follows– ip route 0.0.0.0 0.0.0.0 S0/0


94



95

Gateway of Last Resort

• This type of route is also called a gateway of last resort, since without a default route, a router discards packets whose destination address does not match the router’s routing table


96

Floating Static Routes

• A floating static route is a static route that the administrator wants to be used some of the time

• The term floating comes from the idea that the static route leaves the routing table under some conditions and comes back into the routing table under other conditions


97

Floating Static Routes

• Floating static routes can be very useful for dial backup, using the following logic– When a WAN connection is up, the router

should ignore the static route and instead use the routes learned by the routing protocol

– These routes will forward packets out the permanent WAN connection

– When the permanent WAN connection is down, use the statically defined route that sends traffic over the dial backup link


98

Advertising Default Routes

• In some cases, it makes sense to distribute a default route throughout an internetwork


99


• For example in the diagram that follows– All routers in the enterprise internetwork learn

about all subnets of Class B network 130.1.0.0 via RIP

– Router R-core defines a static default route pointing to the Internet

– Router R-core advertises a default route to the rest of the routers in the enterprise


100



101


102

Dynamic Routes

• Unlike static routes, which point one way and only one way, a dynamic routing protocol can compensate for changes in the network without someone having to go to the command line of each router and make the change

• There are only a few major routing protocols that can do this work for you


103

Dynamic Routes

• Of course all the routers must speak the same language for this to work

• Recall as well that dynamic routing protocols fall into two general classes– Distance Vector– Link State


104


• Distance vector routing protocols include– RIP – Routing Information Protocol– EIGRP – Enhanced Interior Gateway Routing

Protocol


105


• Link state routing protocols include– OSPF – Open Shortest Path First– IS-IS – Intermediate System to Intermediate

System


106

Path Vector Protocols

• The final type is the path vector sort, of which there is only one– Border Gateway Protocol


107

Which One to Use

• Cisco will spend more time on the routing protocols they invented or prefer

• In the real world the two main interior routing protocols are– OSPF

• OSPF is used by both Cisco only and mixed vendor shops

– EIGRP• EIGRP is used by Cisco only operations


108

Which One to Use

• There is a reasonable amount of RIP in use still

• IGRP is rarely used, but is seen in small operations

• IS-IS is used by some ISPs and the like

Review

• What is the difference between static and dynamic routing

• What is the difference between distance vector and link state routing protocols

• What dynamic routing protocols are commonly used


109

Copyright 2005-2009 Kenneth M. Chipps Ph.D. Routing and Routing Protocols Last Update 2009.07.17...

Documents

Transcript of Copyright 2005-2009 Kenneth M. Chipps Ph.D. Routing and Routing Protocols Last Update 2009.07.17...