This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Path Determination
• Static Routes• Dynamic Routing Protocols
– Routing Information Protocol (RIP)– Interior Gateway Routing Protocol (IGRP)– Open Shortest Path First (OSPF)– Enhanced Interior Gateway Routing Protocol
(EIGRP)– Border Gateway Protocol (BGP)
Routing Overview• In order to travel from one network to
another, some device must know to transport that information
• Routing is the process by which information gets routed from one location to another:– mail– telephone calls– trains
Router InformationA router (or entity performing routing) needs to know:
• Destination Address – What is the destination address of the item to be routed?
• Information Sources – From which source (i.e., other routers) can the router learn paths to a given destination
• Possible Routes – What are the initial possible routes or paths to the intended destination
• Best Routes – What is the best path to the intended destination
• Routing Information Maintenance and Verification – A way to verify that known paths to destinations are current and valid
Connected Routes
10.120.2.0 E0
172.16.0.0
S0
Network Protocol
Destination Network
Exit Interface
Connected
Learned
10.120.2.0
172.16.0.0
E0
S0
Table Construction
• If destination is directly connected, router knows which port to use when forwarding packets
• If destination networks are not directly attached, router must learn best route– Manually by network administrator– Dynamically by collecting information about
processes running through the network
Forwarding Packets
• Static Routes – Routes learned by router when administrator manually establishes route. The administrator must update these routes when topology changes occur
• Dynamic Routes – Routes automatically learned by router after administrator configures a routing protocol that helps determine routes. Route knowledge is automatically updated whenever topology changes
Enabling Static Routes
• Static routes are administratively defined routes that specify the explicit path packets must take to destination
• They are administratively defined and allow very precise control over routing behavior
• Important if Cisco IOS software cannot build a route to destination
• Gateway of “last resort” – address a router would send a packet destined for a network not listed in the routing table
Stub Network
• Static routes are commonly used when routing from a network to a stub network
• Stub network (aka “leaf node”) is a network accessed by a single route
172.16.1.0172.16.2.
1
172.16.2.2S0Network
Stub Network
End-to-End Connectivity
• Static route is configured for connectivity to data link, not directly to router
• End-to-end connectivity is configured in both directions
Configuring Static Routes
• Enter ip route in global configuration mode.• Parameters for ip route further define the static
route• Static route allows manual configuration of
routing table• Entry will remain in routing table as long as path
is active• Only exception is permanent option – route will
remain in table even if path is not active
Static Route To StubStatic route from Router A to stub network is configured as follows:
• 0.0.0.0 – routes to non-existent subnet (with special mask, it denotes the default network)
• 0.0.0.0 – Specifies special mask indicating default route
• 172.16.2.2 – Specifies IP address of next-hop router to be used as default for packet forwarding
172.16.1.0172.16.2.
1
172.16.2.2S0Network
Stub Network
BA
Learning Routes
• Static routes are useful in some situations
• It is not satisfactory that the network administrator reconfigure routers to accommodate change
• Another method is to learn available routes automatically accommodating changes
Routing Protocols
10.120.2.0 E0
172.16.1.0
S1
S0
172.17.3.0
Network Protocol
Destination Network
Exit Interface
Connected
RIP
IGRP
10.120.2.0
172.16.1.0
172.17.3.0
E0
S0
S1
Routing Protocols
• Dynamic routing relies on a routing protocol to dissiminate and gather knowledge
• Routing protocol defines set of rules used to communicate with neighboring routers
• Routing protocol is a network layer protocol that intercepts packets from other routers to learn and maintain a routing table
Routing Protocols (Cont)• Routing protocols describe the following
information– How updates are conveyed– What knowledge is conveyed– When to convey knowledge– How to locate recipients of updates
• Examples of routing protocols are:– RIP– IGRP
Routed Protocols
• Routed protocols such as TCP/IP and IPX define the format and use fields within a packet to provide a transport mechanism for user traffic
• As soon as routing protocol determines a valid path between routers, the router can route a routed protocol
Types of Routing Protocols
• Interior Gateway protocols (IGP) – Used to exchange routing information within an autonomous system. Examples:– RIP– IGRP
• Exterior Gateway Protocols (EGP) – Used to exchange routing information between autonomous systems. Example:– BGP
• EGPs are not discussed in this book
IGP Vs EGP
EGP: BGP
IGP: RIP, IGRP
Autonomous System 100
Autonomous System 200
Autonomous System
• Collection of networks under a common administrative domain
• Internet Assigned Numbers Authority (IANA) allocates autonomous system numbers
• Using IANA-assigned autonomous system number is necessary only if organization plans to use EGP public network such as the internet
Administrative Distance• Multiple routing protocols and static routes may be used at
the same time• If several routing sources provide common routing
information, an administrative distance value is used to rate trustworthiness of each routing source
• Allows Cisco IOS software to discriminate between sources of routing information
• For each network learned, IOS selects route from routing source with lowest administrative distance
• It is a number between 0 and 255.• Routing protocol with lowest administrative distance has
most likelihood of being used
Administrative Distance (Cont)
Router D
Router B
Router C
Router A
RIP
Administrative
Distance = 120
IGRP
Administrative
Distance = 100
Send packet from router A to network E by best route
E
Default ValuesRoute Source Default Distance
Connected interface 0
Static route address 1
EIGRP 90
IGRP 100
OSPF 110
RIP 120
External EIGRP 170
Unknown/Unbelievable 255 (Will not be used to pass traffic)
Non-Default Values
• Non-default values may be necessary when redistributing routes
• Network administrator can use Cisco IOS to configure administrative distance values on a per-router, per-route basis
• See ACRC coursebook available from CISCO press
Classes of Routing Protocols• Distance Vector – Determines direction (vector)
and distance of any link in the internetwork. Examples include RIP and IGRP
• Link-state – (also called shortest path first) recreates exact topology of entire internetwork for route computation (or at least component where router is located). Examples include OSPF and NLSP
• Balanced Hybrid – Combines aspects of link-state and distance vector algorithms. Example is EIGRP
Comparison
• There is no single best routing algorithm for all internetworks
• All routing protocols provide information differently
Distance Vector Protocols
• Also known as Bellman-Ford-Fullerton algorithms• Pass periodic copies of routing table from router
to router and accumulate distance vectors– Distance means how far– Vector means which direction
• Regular updates between routers communicate topology changes
• Each router receives routing table from its direct neighbor
Table Updates
B
C
A
D
Routing
TableRouting
Table
Routing
Table
Routing
Table
A
B
C
Algorithm Activities
• Identify sources of information
• Discover routes
• Select best route
• Maintain routing information
Information Exchange
S0 S0 S1E0 S0 E0
10.1.0.0
10.2.0.0
10.3.0.0
10.4.0.0
A B C
10.2.0.0
S0 0
10.3.0.0
S1 0
10.4.0.0
S1 1
10.1.0.0
S0 1
Routing Table
10.1.0.0
E0 0
10.2.0.0
S0 0
10.3.0.0
S0 1
10.4.0.0
S0 2
Routing Table
10.3.0.0
S0 0
10.4.0.0
E0 0
10.2.0.0
S0 1
10.1.0.0
S0 2
Routing Table
Multiple Paths
• There might be multiple paths to any given destination network
• When table is updated, primary objective is to determine best path
• Each distance vector routing protocol uses a different routing algorithm to determine best route
• Algorithms generate a number called metric value for each path through the network
• Smaller the metric, the better the path
Metrics• Hop count – Number of routers through which packet will
pass• Ticks – Delay on a data link using IBM PC clock ticks (
55 milliseconds)• Cost – Arbitrary value, usually based on bandwidth, dollar
expense, or another measurement assigned by network administrator
• Bandwidth – Data capacity of link. 10 Mbps Ethernet is better than 64Kbps leased line
• Delay – Length of time to move from source to destination• Load – Amount of activity on a network resource such as
router or link• Reliability – Usually refers to bit-error rate of each
network link• MTU – Maximum transmission unit. Maximum frame
length in octets that is acceptable to all links on path
Transmission from A to B
C
A
B
56T1
56
T1IPX
Ticks, Hop Count
IGRP
Bandwidth
Delay
Load
Reliability
MTU
RIP
Hop Count
Methods
• IGRP – Bases decision on combined characteristics, such as bandwidth, delay, reliability, and MTU. Emphasis is on bandwidth and delay, so it would choose T1 lines
• RIP – Hop counts are equal, so it would load balance between paths
Distance Vector EventsThe following occurs step-by-step from processor to processor
• Topology change
• Network discovery process
• Topology change updates– The entire routing table is sent to each adjacent or directly connected
neighbor
– Routing table contains information about total path cost (defined by metric) and logical address of the first router on the path to each network it knows about
• Updates are compared to own routing table
• Router adds cost of reaching neighbor to path cost reported by neighbor
• Finding a better route results in update of routing table
Maintaining Routes
Topology
change
causes
routing
table
update
Router A sends
out updated routing
table at the end of
next period
Process to
update this
routing table
Process to
update this
routing table
A B
“Converged” Network
S0 S0 S1E0 S0 E0
10.1.0.0
10.2.0.0
10.3.0.0
10.4.0.0
A B C
10.2.0.0
S0 0
10.3.0.0
S1 0
10.4.0.0
S1 1
10.1.0.0
S0 1
Routing Table
10.1.0.0
E0 0
10.2.0.0
S0 0
10.3.0.0
S0 1
10.4.0.0
S0 2
Routing Table
10.3.0.0
S0 0
10.4.0.0
E0 0
10.2.0.0
S0 1
10.1.0.0
S0 2
Routing Table
“Slow” Convergence
If network 10.4.0.0 fails, the routing tables should change so that the network slowly re-converges
Start of Counting to Infinity
S0 S0 S1E0 S0 E0
10.1.0.0
10.2.0.0
10.3.0.0
10.4.0.0
A B C
10.2.0.0
S0 0
10.3.0.0
S1 0
10.4.0.0
S1 1
10.1.0.0
S0 1
Routing Table
10.1.0.0
E0 0
10.2.0.0
S0 0
10.3.0.0
S0 1
10.4.0.0
S0 2
Routing Table
10.3.0.0
S0 0
10.4.0.0
E0 down
10.2.0.0
S0 1
10.1.0.0
S0 2
Routing Table
X
Router C Update
S0 S0 S1E0 S0 E0
10.1.0.0
10.2.0.0
10.3.0.0
10.4.0.0
A B C
10.2.0.0
S0 0
10.3.0.0
S1 0
10.4.0.0
S1 1
10.1.0.0
S0 1
Routing Table
10.1.0.0
E0 0
10.2.0.0
S0 0
10.3.0.0
S0 1
10.4.0.0
S0 2
Routing Table
10.3.0.0
S0 0
10.4.0.0
S0 2
10.2.0.0
S0 1
10.1.0.0
S0 2
Routing Table
X
Router A and B Updates
S0 S0 S1E0 S0 E0
10.1.0.0
10.2.0.0
10.3.0.0
10.4.0.0
A B C
10.2.0.0
S0 0
10.3.0.0
S1 0
10.4.0.0
S1 3
10.1.0.0
S0 1
Routing Table
10.1.0.0
E0 0
10.2.0.0
S0 0
10.3.0.0
S0 1
10.4.0.0
S0 4
Routing Table
10.3.0.0
S0 0
10.4.0.0
S0 2
10.2.0.0
S0 1
10.1.0.0
S0 2
Routing Table
X
Next Iteration
S0 S0 S1E0 S0 E0
10.1.0.0
10.2.0.0
10.3.0.0
10.4.0.0
A B C
10.2.0.0
S0 0
10.3.0.0
S1 0
10.4.0.0
S1 5
10.1.0.0
S0 1
Routing Table
10.1.0.0
E0 0
10.2.0.0
S0 0
10.3.0.0
S0 1
10.4.0.0
S0 6
Routing Table
10.3.0.0
S0 0
10.4.0.0
S0 4
10.2.0.0
S0 1
10.1.0.0
S0 2
Routing Table
X
Troubleshooting
• IP distance vector routing algorithms have inherent limits via Time To Live (TTL) value in IP header
• Router reduces TTL by 1 each time it gets a packet
• When 0, router discards packet
• However, routing loop might count to infinity first
Maximum Metric
S0 S0 S1E0 S0 E0
10.1.0.0
10.2.0.0
10.3.0.0
10.4.0.0
A B C
10.2.0.0
S0 0
10.3.0.0
S1 0
10.4.0.0
S1 16
10.1.0.0
S0 1
Routing Table
10.1.0.0
E0 0
10.2.0.0
S0 0
10.3.0.0
S0 1
10.4.0.0
S0 16
Routing Table
10.3.0.0
S0 0
10.4.0.0
S0 16
10.2.0.0
S0 1
10.1.0.0
S0 2
Routing Table
X
Maximum Metric Setting
When value reaches maximum, network is considered unreachable
Split Horizon
It is never useful to send information about a route back in the direction from which the original update came
Split Horizon Example
• Router B has access to network 10.4.0.0 through Router C, so it makes no sense for Router B to announce to Router C that it has access to 10.4.0.0 through Router C
• Router B announced 10.4.0.0 network to Router A, so it makes no sense for Router A to announce its distance to Router B
• Having no alternate path to 10.4.0.0, Router B concludes 10.4.0.0 is inaccessible
Split Horizon Routing Tables
S0 S0 S1E0 S0 E0
10.1.0.0
10.2.0.0
10.3.0.0
10.4.0.0
A B C
10.2.0.0
S0 0
10.3.0.0
S1 0
10.4.0.0
S1 down
10.1.0.0
S0 1
Routing Table
10.1.0.0
E0 0
10.2.0.0
S0 0
10.3.0.0
S0 1
10.4.0.0
S0 down
Routing Table
10.3.0.0
S0 0
10.4.0.0
S0 down
10.2.0.0
S0 1
10.1.0.0
S0 2
Routing Table
XXX
Route Poisoning
• Route Poisoning is another form of Split Horizon
• Attempts to eliminate routing loops caused by inconsistent updates
• Router sets a table entry that keeps network state consistent while other routers gradually converge
• Frequently used with “holddown timers (described in next section)
Poisoning Example
• When 10.4.0.0 goes down,– Router C poisons its link to network 10.4.0.0 by
entering infinite cost (indicating network is unreachable)
– By poisoning route to network, Router C is not susceptible to other incorrect updates about 10.4.0.0
Poisoned Route
S0 S0 S1E0 S0 E0
10.1.0.0
10.2.0.0
10.3.0.0
10.4.0.0
A B C
10.2.0.0
S0 0
10.3.0.0
S1 0
10.4.0.0
S1 1
10.1.0.0
S0 1
Routing Table
10.1.0.0
E0 0
10.2.0.0
S0 0
10.3.0.0
S0 1
10.4.0.0
S0 2
Routing Table
10.3.0.0
S0 0
10.4.0.0
E0 infinity
10.2.0.0
S0 1
10.1.0.0
S0 2
Routing Table
X
Router B Actions
• Router B notices that 10.4.0.0 jumps to infinity
• Router B sends update called poison reverse back to Router C
• Poison reverse overrides Split Horizon direction
• Poison reverse serves as acknowledgement that poison message was received
Poison Reverse
S0 S0 S1E0 S0 E0
10.1.0.0
10.2.0.0
10.3.0.0
10.4.0.0
A B C
10.2.0.0
S0 0
10.3.0.0
S1 0
10.4.0.0
S1 possibly
down
10.1.0.0
S0 1
Routing Table
10.1.0.0
E0 0
10.2.0.0
S0 0
10.3.0.0
S0 1
10.4.0.0
S0 2
Routing Table
10.3.0.0
S0 0
10.4.0.0
E0 infinity
10.2.0.0
S0 1
10.1.0.0
S0 2
Routing Table
XPoison Reverse
Holddown Timers
• Holddown times are used to prevent regular update messages from inappropriately reinstating a bad route
• Tell routers to hold any changes that might affect routes for some period of time
• Holddown period is usually just grater than time necessary to update entire network with a routing change
Holddown Timer Operation• Step 1 – When router receives update indicating a network
is inaccessible, the router marks the route as inaccessible and starts holddown timer
• Step 2 – If update arrives from neighboring router with better metric than originally recorded, the router marks the network as accessible and removes holddown timer
• Step 3 – If a poorer metric update is received from a neighboring router at any time before the holddown timer expires, the update is ignored. Ignoring poorer updates allows more time for knowledge of the change to propagate
• Step 4 – During holddown period, routes appear in routing table as “possibly down”
Holddown Example
S0 S0 S1E0 S0 E0
10.1.0.0
10.2.0.0
10.3.0.0
10.4.0.0
A B C
X
Network 10.4.0.0
is unreachable
Update after holddown time
Update after holddown time
Triggered Updates
• If routers wait for regularly scheduled updates, before notifying neighbors of catastrophes, the following serious problems can occur– Loops– Dropped traffic
• A triggered update is sent immediately• Detecting router immediately sends messages to
adjacent routers which in turn notify their neighbors
Triggered Updates
S0 S0 S1E0 S0 E0
10.1.0.0
10.2.0.0
10.3.0.0
10.4.0.0
A B C
X
Network 10.4.0.0
is unreachable
Network 10.4.0.0
is unreachable
Network 10.4.0.0
is unreachable
Triggered Update Problems
• Packets containing update message can be dropped or corrupted by some link in the network
• Triggered updates do not occur instantaneously• It is possible that a router issue a regular update
just before it is about to receive a triggered update, causing a bad route to be reinserted into a neighbor who already received the triggered update
• Therefore, we couple triggered updates with holddowns
Holddown Timers with Triggered Updates
• Holddown rule says that when a route is invalid, no new route with the same or worse metric will be accepted at the destination for holddown time
• Therefore, triggered updates have time to propagate throughout the network
Multiple Solutions
• Routers have multiple routes to each other• Routers A and D receive triggered update• Router B removes its route to network 10.4.0.0
X
10.4.0.0
EB
A
D
C
Route Fails
X
10.4.0.0
EB
A
D
C
Holddown
Holddown
Holddown• Routers A and D receive triggered update
• Set their own holddown timers• Routers A and D in turn send triggered updates to Router E
Route Holddown
X
10.4.0.0
EB
A
D
C
Holddown
Holddown
Holddown
• Routers A and D send poison reverse to Router B• Router E sends poison reverse to Routers A and D
Poison Reverse
Poison Reverse
Poison Reverse
Poison Reverse
Holddown Duration
Routers A, D and E remain in holddown until
one of following occurs:• Holddown timer expires• New route with better metric is received• A flush timer (the time a route is held before being
removed) removes the route from the routing table
Packets During Holddown
X
10.4.0.0
EB
A
D
C
Holddown
Holddown
Holddown
Packet for 10.4.0.0
Packet for 10.4.0.0
• Router E sends message to 10.4.0.0• Router B will drop packet and send ICMP “network
unreachable” message
10.4.0.0 Returns to Operation
10.4.0.0E
B
A
D
C
• Network 10.4.0.0 returns to operation• Router B sends a trigger update to Routers A and D
notifying them that the link is active• After holddown timer expires, Routers A and D add
10.4.0.0 back to their routing table
Link Up!
Network Converges
10.4.0.0E
B
A
D
C
• Routers A and D send Router E a routing update stating that network 10.4.0.0 is up
• Router E updates its routing table after hoddown timer expires
Link Up!
Link-State and Hybrid Routing Protocols
• Focus of this chapter has been “distance vector” routing
• Link-State and Hybrid are alternative routing protocols
Link-State Protocol Diagram
Link State Packets
TopologicalDatabase
SPF
Algorithm
Shortest-Path-First Tree
Routing Table
Link-State Protocol
• Link-state protocols build routing tables based on a topology database
• Topology database is built from link-state packets that are passed between all routers to describe the state of a network
• Database is used by the shortest-path-first algorithm to build the routing table
Shortest-Path Algorithms• Link-state algorithms (also known as shortest-path
algorithms) maintain a complex database of topology information
• Distance vector algorithm has non-specific information about distant networks and no knowledge of distant routers
• Link-state maintains full knowledge of distant routers and how they interconnect
• Link-state routing uses Link-State Packets (LSPs), a topological database, the SPF algorithm, the resulting SPF tree, and the routing table of paths and ports to each network
Link-State Advantages
As networks become larger in scale, link-state becomes more attractive because:• Link-state protocols only send updates of a topology
change• Periodic updates are more infrequent for distance vector
protocols• Networks running link-state can be segmented into area
hierarchies, limiting the scope of route changes• Networks running link-state support classless addressing• Networks running link-state support summarization
Balanced Hybrid• A third protocol, called “Balanced Hybrid,” combines
distance vector and link-state protocols
• Balanced hybrid uses distance vectores with more accurate metrics to determine the best paths to destination networks
• However, it uses topology changes to trigger routing database updates as opposed to periodic updates
• Balanced hybrid converges more rapidly, like link-state but emphasizes economy in the use of resources such as bandwidth, memory and processor overhead
• CISCO’s Enhanced Interior Gateway Routing Protocol is a balanced hybrid protocol
Configuring Dynamic Routing Protocols
• To enable dynamic routing protocol, perform the following
• Select a routing protocol, such a RIP or IGRP• Select IP networks to be routed• Dynamic routing uses broadcasts and multicasts to
communicate with other routers• When information from other routers is received,
it uses routing metric to find the best path to each network or subnet.
Use of IGRP and RIP at Same Router
172.30.0.0
160.89.0.0
172.16.0.0
IGRP
RIP
RIP
Router Command
• The router command starts the routing process
router(config)#router protocol [keyword]
• Protocol is RIP, IGRP, OSPF or EIGRP• Keyword refers to an autonomous system in
protocols that require an autonomous system such as IGRP
Network Command• The network command allows the routing process to
determine which interfaces it will participate in the sending and receiving of routing updates
• The network command starts the routing protocol on all of a router’s interfaces that have IP addresses within the specified network scope
• The network command also allows router to advertise that network to other routers
router(config-router) #network network-number
• The network-number parameter specifies a directly connected network number
• For RIP and IGRP, network-number must be based on major-class network numbers, not subnet numbers
Enabled Protocols
After the protocol is enabled and a networks path is chosen, the router begins to dynamically learn the networks and paths available in the internetwork
RIP Route Choice
• The above shows how RIP would choose routes based on hop count
C
T1T1
19.2 kbps
T1
Versions 1 and 2
• The book describes RIP version 1– Version 1 is described in RFC 1058– www.isi.edu/in-notes/rfc1058.txt
• We are using RIP version 2 in our project– Version 2 is described in RFCs 1721 and 1722– www.isi.edu/in-notes/rfc1721.txt– www.isi.edu/in-notes/rfc1722.txt
General Characteristics of RIP
• It is a distance vector protocol
• Hop count is the metric for path selection
• Maximum allowable hop count is 15
• Entire routing table is broadcast every 30 seconds by default
• Can load balance over as many as six equal-cost paths (four paths is the default)
RIP-1 vs RIP-2
• RIP-1 requires that only one network mask can be used per network number for each major classful network being advertised
• RIP-2 permits variable-length subnet masks (VLSM) on the internetwork
• Standard RIP-2 supports triggered updates• Standard RIP-1 does NOT support triggered
• router IGRP 100 – enables IGRP routing process for autonomous system 100
• network 172.16.0.0 – associates network 172.16.0.0 and its interfaces with IGRP routing process
• network 10.0.0.0 – associates network 10.0.0.0 and its interfaces with IGRP routing process
IGRP Updates
• IGRP sends updates out interfaces in networks 10.0.0.0 and 172.16.0.0
• It also advertises directly connected networks 10.0.0.0 and 172.16.0.0, as well as other networks it learns about through IGRP (198.168.1.0)
IGRP Load Balancing
• The variance router configuration command controls IGRP load balancing
router(config-router)#variance multiplier
• Multiplier parameter specifies the range of metric values acceptable for load balancing– Range is from lowest (best) metric value to the lowest multiplied
times the variance value– Acceptable values are nonzero, positive integers– Default value is 1, which implies equal-cost load balancing
IGRP Traffic-Share
The traffic-share {balanced | min} command is used to control how traffic is distributed among IGRP load sharing routes
I 172.16.0.0/24 is possibly down, routing via 10.1.1.1, Serial2
10.0.0.0/24 is subnetted, 2 subnets
C 10.1.1.0 is directly connected, Serial3
C 10.2.2.0 is directly connected, Serial3
I 192.168.1.0/24 [100/89056] via 10.2.2.3, 00:00:14, Serial3
Ping – From Router B
RouterB#ping 172.16.1.1
Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 172.16.1.1, timeout is 2 seconds:
…
Success rate is 0 percent (0/5)
RouterB#
If 172.16.0.0 Comes Back Up
• RouterA sends another triggered update to Router B stating 172.16.0.0 is accessible with metric 89056
• Even Though RouterB receives the update, the route continues in a holddown state
• RouterB will not remove route from holddown and update routing table until holddown timer expires
• However, RouterB COULD successfully ping network 172.16.0.0 and send traffic there
Unknown Subnet of a Directly Attached Network
• Router assumes that all subnets of a directly attached network are present in the IP routing table
• If a packet is received with a destination address within an unknown subnet of a directly attached network, the router assumes the subnet does not exist and drops the packet
• This holds true even if the routing table contains a default route
• This behavior can be changed with the ip classless global configuration command
IP Classless
E0 S010.0.0.0
10.2.2.210.1.0.0
172.16.0.0
Default Route
Router(config)#ip classless
Network
Protocol
Destination
Network
Exit
Interface
C
C
RIP
10.1.0.0
10.2.0.0
172.16.0.0 via
0.0.0.0
E0
S0
S0
E0
IP Classless
The middle router will forward a packet with 10.7.1.1 as the destination address out of the default interface, E0, because ip classless is enabled
Routing Command SummaryCommand Description
ip route network mask {address | interface} [distance] [permanent]