1) Control Plane

● Activities related to maintaining routing table, performing NAT, QOS, ACL, etc., belongs to the
control plane.
● CPU is involved in these activities of a device.

● Basically it is designated to provide information and assistance to data plane.

2) Data Plane
● Activities related to make forwarding decision on the basis of information provided by the
control plane.
● All interface of a device residence in the plane.

3) Management Plane
● Activities related device management like configuring device from CLI via console , telnet , ssh
and SNMP

Switching Methods
1. Process Switching
2. Fast switching
3. CEF (cisco express forwarding)

Process switching

● When frame is received that frame is always is sent to the processor (CPU)

● This is done repeatedly same every time a frame is received even if it is from same source
and for same destination
● It will increase CPU utilization high
 ● Device will stop functioning.

● Per packet load balancing.

Fast switching

● This method is knows for process once and switch many times

● RP :- Route Processor – CPU

o To process the frame so table can be populated
● SE :- Switch Engine – Shortcut
o To forward the traffic on the basis of entry in table.
● For every new connection switch always look into route processor.

● Load balancing per destination basis.

CEF: - Cisco Express Forwarding

● A mechanism of packet switching developed by cisco

● It is a fast packet switching method in comparison of process & fast switching

● By default , CEF is enabled on cisco devices like router , Switch (if routing is configured)

● It can disable but not recommended.

● It enabled by default on cisco devices.

● RIB :- Routing Information base

o The routing Table
● FIB :- Forwarding information base
o Its duplicate copy of RIB
o Is always synchronised with RIB
o Every info in RIB will automatically copied in FIB
o FIB is always updated through RIB
▪ If any routes goes down then only from FIB will automatically get deleted + vice
versa
o Layer 3
 ▪ Generation 1 🡪 switch process / switching engine

▪ Generation 2 🡪 RP / CEF

When use CEF PUNT – RIB

● An entry cannot be located in the FIB.

● The FIB table is full.

● The IP time-to-live (TTL) has expired.

● When TTL is one (1).

● The maximum transmission unit (mtu) is exceeded, and the packet must be fragmented.

● An Internet control Message Protocol (ICMP) redirect is involved.

● The encapsulation type is not supported.

● Vlan ACL, Port ACL, filtering.

● QOS

● If destination route is not exist.

● If adjacency table is not completed.

● SH ip CEF
Adjacency

● Sh adjacency

● Sh adjacency detail

● Sh ip cef 10.0.0.2 255.255.255.255 detail

Type of Adjacency table

1) Null adjacency: - Null adjacency table will be responsible to handle those packets which are
forwarded toward null interface.
2) Drop adjacency: - This table is basically responsible to handle those packets which encountered
with mismatch of encapsulation or crc error.
3) Discard adjacency: - This table is responsible to handle those packets which are discarded
(forced) by an acl.
4) Glean adjacency: - The table is responsible to have information about all directly connected
networks and whenever a packet will move to any directly connected network then those
packet will be handled by glean adjacency.
5) Punt adjacency: - The table is responsible to handle those packets which is not processed by cef
and forwarded to control plane to process these packet.

How to disable CEF

● #no ip cef

C-CEF

● Central CEF

D-CEF

Distribute CEF

Routing
● It is a process by which router will forwarded packets.
 ● Router have two table to take forwarding Decision
o Routing table
▪ It find out exit interface

▪ It have L3 information
o ARP table
▪ It have MAC Address

▪ It will change L2 information

● Routing is a process in which if router receive traffic on one interface after receiving traffic
router will check routing table just to find out the exit interface , after checking the exit interface
it will switch the traffic on that exit interface , now it will change L2 information by the help of
ARP table and after that traffic will forward.

● Type of Communication

o Unicast
o Broadcast
o Multicast

● Different type of network

o Point to point network (Serial)

o Point to multipoint(Ethernet)

Type of Routing

1) Static Routing
2) Dynamic Routing

Static Routing
● In this Routing administrator manually specify the route information.

● There are three way to configure static route

1. With using next hop address
 ▪ When Route is assigned through next hop

● Solution 🡪no proxy ARP

● Problem 🡪 recursive Lookup

● R1(config)#IP Route <destination network > <destination subnet

mask > <next hop ip >
2. With using exit interface
▪ When route is assigned through exit interface

● Problem 🡪Proxy ARP

● Solution 🡪 no recursive Lookup

o Ip route <Destination Prefix ><destination subnet
mask><exit interface>
o Ip route < 30.0.0.0>< 255.0.0.0>< fastethernet 0/1>
3. With using combination of both Exit Interface and Next hop.
▪ Ip route <Destination Prefix ><destination subnet mask><exit
interface><next hop ip >

● Recursive Lookup
o Repetition of routing table lookup for a single destination.
● Proxy Arp
o An ARP reply is given on behalf of other devices.

● Serial Link 🡪
o Does not Support ARP
o Serial Link Does not have mac address
o Static route can be assigned through exit interface if that interface is serial 🡪
so that will no problem of proxy ARP.

● Advantages
o CPU utilization less
o Security

● Disadvantage
o Burden on an administrator
o Works in small network
 o Time consuming

Route Preference criteria

● Purpose 🡪 Redundancy (Backup )
o Primary path
o Secondary Path
● When there are multiple paths for a single destination, so router should prefer any
one path to reach that destination.

Highest Prefix length (CIDR)

o Higher the prefix , less the number of host IPs

o Less the number of host IPs, lesser the cpu utilization.
o When multiple routes are assigned on a single router, both the routers will be
installed in routing table.

Lowest administrative Distance (AD)🡪 8 bit (0-255)

o When prefix (CIDR) is length is same.

o Measure of trustworthiness.
o Lower the AD value, the higher the trust/Preference.
o Connected = 0 (always)
o Static = 1 (always )
o RIP= 120
o EIGRP=90 , 170
o OSPF=110
o BGP=20,200
o 255= route will be considered as invalid.
 o Router having lower AD value will only get into routing table.

Lowest metric

o If AD value is also same then

o Connected and static route does not have metric = always (0)
o It is useful in dynamic routing protocol
o RIP 🡪Hop count
o EIGRP🡪Composite metric weights
o OSPF 🡪Cost
● When all criteria are same then routers will perform “Load Balancing”

● Both the routes will be installed in routing table.

● Packets will be distributed among both the routes equally and alternating.

Dynae automatically.
● Routers automatically exchange the routes b/w each other.

● Some packets are there for each protocol which is advertised by routing protocols to
get the routes from their neighbouring routers🡪 Directly Connected Routers.
● Dynamic Routing
o IGP(Interior Gateway Protocols)
o EGP(Exterior Gateway Protocols)
● IGP🡪Routing Protocols that share routes among same autonomous system.

o Ex🡪 RIP , EIGRP , OSPF mic Routing

o Routing in which routes are maintained and calculat
● EGP🡪Routing Protocols that share routes between different autonomous systems.
o BGP

Autonomous System Numbers (IANA)

 o Group of router or network in single administration is called AS number
o This is share by same routing policies & controlled by a single admin or team.
o It is identify by decimal number
o 16 bit (0-65535) 0 & 65535
o 1-64511(public AS)
o 64512-65534(Private AS)

● IGP(Interior Gateway Protocols)

o RIP(Routing Information Protocols)
o EIGRP(Enhanced Interior Gateway Protocols)
o OSPF(Open Shortest Path First)
● EGP(Exterior Gateway Protocols)

1. BGP ( Border Gateway Protocol)

EIGRP
1) It stand for Enhanced Interior Gateway Protocols
2) It is an Interior Gateway Protocols
3) It is Hybrid Routing Protocols
a. It will not use distance (hop count) but still hop count will be forward
in EIGRP updates.
b. Next-hop = source of update
c. Table
4) It is Advance Distance Routing Protocols
5) It is Open standard.

Metric
a. Bandwidth = 1*K1 (by default)
i. Capacity of link
ii. Higher value is preferable
iii. Sh int fa0/0 | s BW
b. Load= 0*K2
i. Link utilization
 ii. Lower value is preferable
c. Delay= 1*K3 (by default)
i. Time taken during transmission from one device to another
ii. Lower value is preferable.
d. Reliability= 0*K4
i. Link having least downtime.
ii. Lower value is preferable
e. MTU= 0*K5
i. Interface through which we can send more data.
ii. Lower value is preferable.
iii. R1(Config-router)#metric weights 0 1 0 1 1 0
● Eigrp uses composite metric

● In hello packets both routers will tell each other what k values they are using for metric
calculation.
● By default only (k1) bandwidth and (k3) delay are used for metric
calculation.

Metric calculation = 256(10^7/least b/w) kpbs + total delay /10)

● Standard IEEE
o Serial
▪ Bw =1.544 mbps

▪ Delay = 20000
 o Ethernet

Bandwidth Delay

Ethernet 10,000 kbps 1000

Fast Ethernet 100,000 kbps 100

Gig Ethernet 10,00,000 kbps 10

10 gig 10,000,000 kbps -

Loop back 8,000,000 kbps 5000

1) It supports equal load balancing, by default 4, max 16/32

a. R1 (Config-ROUTER) #MAXIMUM-PATHS 10.
2) Hello timer 5 sec and hold timer 15 sec.
a. After 15 sec, neighbour ship will break and routes will be removed
from routing table.
b. In slow link (frame-relay) hello is 60 sec and hold is 180 sec
3) Hop count by default 100 and maximum 255
a. Hop count not use for metric calculation.
b. R1(config-router)#metric maximum-hops
4) AD VALUE
a. 90(Internal)
b. 170(External)
c. 5(Summary Route)
d. Routes exchanged within the AS will be installed with the AD
value of 90 and between the AS with the help of redistribution
will be installed with the AD value of 170.
5) Network can be advertised with wildcard mask also
a. Opposite of subnet mask
6) RID should be unique within AS.
 a. Redistributing router RID should be unique for all the AS it is
connected to.
b. Statically
c. Highest Loopback IP
d. Highest Physical Interface IP.
e. 32 bit identity always in IPV4
f. R1(config-router)# Eigrp Router-id 1.1.1.1
7) It is a classless routing Protocol.
a. It send network Id with subnet mask
8) It support VLSM
9) By default auto-summary is enabled / disable in IOS version 15.0 and
above.
10) It uses multicast address 224.0.0.10
11) It encapsulate in IP Packet.
12) It uses Protocols Number 88
13) Passive interface
a. Hello will not be sent or received
14) To create multiple domain of Eigrp we use AS ( Autonomous System
Number)
15) It uses dual algorithm for best path selection (Diffusing Update
Algorithm)

DUAL (Diffusing Update Algorithm)

o It is used to store metric Information.
o Calculated of a route from source to destination.
● CD (Cumulative Distance )
o Total metric from source to destination of all paths available.
● FD(Feasible Distance)
o Best/least metric from source to destination/ least CD.
● RD(Reported Distance)
o Total metric from neighbour of source to destination.
o Metric adv. By neighbouring Router
● Terminologies
o Successor
▪ Best route / path having least CD. Will be

▪ Copied to the routing table.

o Feasible Successor
▪ Second Possible Best path
o Feasibility condition🡪 RD of another path < FD of Successor
o Topology table keeps the record of route that matches or fulfils the
feasibility condition.
o Whenever a router will receive the update, it will compare the B.W
carried in the update to the B.W of received interface To calculate FD
it will use the least B.W
o Delay is cumulative in nature. It will add delay carried in the update
with delay of interface on which update is received to calculate F.D
● EIGRP Table
o Routing
o Neighbour
o Topology
▪ Sub-table 🡪 topology all links
o Topology all-links = Successor, F. Successor, Alternate. CD/RD/FD
o Topology = Successor , F. Successor CD/RD/FD
o Routing= Successor FD
 o Neighbour = directly connected Neighbour Information

Types of Msg in EIGRP

● Hello

● Update

● Query

● Reply

● Acknowledgement

Notes:-

1) Router will start dynamic Neighbour discovery on that interface

2) Router will add network present in that interface to local topology table.
3) Router will discover neighbour using Hello message.
4) Router will send multicast hello on address 244.0.0.10 .

● Hello
o Hello packet is used to form and maintain Neighbour.
o Hello packet in Eigrp is used to check whether neighbour is alive or
not
▪ bandwidth utilization is reduced
o This packet carries neighbour parameters which need to be matched
b/w routers with in a broadcast domain.
o Hello packet does not carry subnet mask information
o Multicast (ether , serial ) 224.0.0.10

● Content of Hello Msg

 1. Version = 2
2. OP code = 5
3. Checksum Value
4. Seq=0
5. Ack=0
6. AS number
7. K value
8. Hold = 15 sec
9. IOS version
10.Authentication (optional)

● EIGRP Neighbour ship parameter

o AS(Autonomous System ) Num should match

o Subnet Should match (on the basis of AND Operation)
o Metric weight should match
o Authentication Should match(optional)
▪ Eigrp doesn’t have its any authentication parameter it will take
help router key chain feature.
▪ Key ID and Key chain is same but key name can be different.

▪ R1(Config)#Key chain R1

▪ R1(config-keychain-key)#Key 1

▪ R1(config-keychain-key)#Key-string ccna

▪ R1(config)#int fa0/0

▪ R1(Config-if)ip authentication mode Eigrp 100 md5

 ▪ (Config-if)ip authentication key-chain Eigrp 100 R1

▪ Show key chain

● Changing key according to time.

● First time should me same

▪ R1(Config)#Key chain R1

▪ R1(config-keychain-key)#Key 1

▪ R1(config-keychain-key)#Key-string ccna

▪ R1(config-keychain-key)#accept-lifetime 01:23:00 2 sep 2016

2:00:00 2 sep 2016
▪ R1(config-keychain-key)#send-lifetime 01:23:00 2 sep 2016
2:00:00 2 sep 2016
o Clock set
o Interfaces should not be passive
▪ Hello will not be sent nor received in EIGRP

▪ R1(Config)#router Eigrp 100

▪ R1(Config-if)#passive-interface fa0/0
o It advertise periodically in every 5 sec

Commands
● Router (config)#Router Eigrp 100

● Router (Config-router)#Network 12.0.0.0

● Router(config-if)#IP hello-Interval Eigrp 100 10 (1-65535)

 ● Router(config-if)#IP hold-Interval Eigrp 100 15 (1-65535)

● Router Eigrp 100

● Router (config-router)#Metric weights <TOS> 0 <k value >1 1 1 1 1

● R1(Config)#do sh ip Eigrp interfaces

● Show ip Eigrp interfaces detail fastethernet 0/0

RTP (reliable Transport Protocols)

● Cisco Proprietary

● Is used for reliable delivery of update

● Since Eigrp is a n/w layer protocols

1. Directly encapsulate in IP header
2. IP is a connection less protocol 🡪 No ack

Adjacency
● Null update is used to verify bidirectional connectivity b/w routers which hello cannot provide

● Hello does not provide ack but update packet does.

 ● Update Message
3. It is used to share prefix information
4. Unicast (serial)
5. Multicast (ethernet)
6. Update packet is separate from hello packet.
▪ Update are not periodic in Eigrp but only trigged
7. OP code = 1
8. Ack = 5
9. When two router will form Neighbourship and share update it start-up
10. In that case router will share full Update on multicast address
11. When any new network is added in topology in that case router will generate
partial Update.

● If there is no ack of update , query and reply is received then

1. A router will send the same update to the neighbour from which ack
is not received sixteen times as unicast. (re transmission throttling)
1st 🡪 update 🡪 multicast 🡪 Seq 1

(Multicast flow timer) 🡪 318 ms 🡪 neighbour table

2nd 🡪 update 🡪 unicast🡪retry (1) 🡪same seq number 1

 RTO 🡪 468 ms

3rd 🡪update 🡪 Unicast 🡪 retry (2) 🡪same seq number 1

� If there is no ack of update , query and reply is received then:

� A router will send the same update to the neighbour from which ACK is not received
sixteen times as unicast
� If ack is received after 8 unicast then it will stop sending further unicast messages.

� If ack is not received even after 16 unicast message then the router will flap the
neighbour ship.
o Hello 🡪 No ACK 🡪 unreliable 🡪 def multicast 🡪 can be unicast
o Update 🡪ACK 🡪 reliable
o ACK 🡪 No ACK 🡪 unreliable 🡪 Always unicast
o Query 🡪 ACK 🡪 reliable 🡪 def Multicast
o Reply 🡪 ACK reliable 🡪 Always 🡪 always Unicast

● SRTT (Smooth round Trip Timer)

o Avg. Time of sending the update and getting the ack in return
● RTO
o Time difference between two unicast during RTP
o Max = 5000ms

● Q Count
o Number of pending acknowledgement from a particular neighbour.
● Seq num
o Seq number of update last received from neighbour.
● Query
o OP Code = 3
o Ethernet = Multicast
o Serial = Unicast
o It is sent when successor goes down and feasible successor is not available for a
particular route.
o Delay infinity
o If neighbour ship is static then will be unicast
o Positive reply
▪ When router have redundant path
o Negative Reply
▪ When router does not have redundant Path
 o 3 min wait for Query msg
o Stuck in active state

Split horizon
● If you receive any update on interface you cannot send back on that interface

● The split Horizon feature prevents a route learned on one interface from being advertised back
out of that same interface. It is loop in EIGRP

Route Poison
● Hop count 16 (RIP)

● Delay Infinite value ( EIGRP)

Split Horizon with poison reverse

 ● It is a loop avoidance Technique by default it is enable on interface.

● Bounded Update.

● The same prefix Update which router get form interface router will not send back that prefix to
another router on same interface.
● It is used to confirm the sending router that spilt horizon is enabled on receiving router’s
interface.
● R1(config)#interface ethernet 1/0

● R1(config-if)#no ip split-horizon Eigrp 100

Type of NBR Discovery

● Dynamic NBR
o Is useful when there are more than two routers in a single broadcast
domain
o Purpose
▪ Neighbourship can be formed without knowing Neighbourship

▪ b/w utilization reduction , CPU utilization reduction

● Static NBR
o Is useful when there are only two routers with in a broadcast
domain.
▪ Security purpose
 o R1(config)#Router Eigrp 100
o R1(Config-Router )#network 192.168.1.0
o R1(Config-Router )#neighbor 192.168.1.2 ethernet 1/0
o Router will send unicast msg in Eigrp
● If you want to configure static Neighbour you have to configure both side

● Static Neighbour is not compatible with Dynamic Neighbour ship.

● In static neighbour hello msg sends as Unicast.

● Hub and Spoke technique

● NBMA ( Non Broadcast Multi access)

Next-Hop Rule in Eigrp

● When Router advertised any Prefix to other router it will send it ip as Next-hop

● R1(config)#interface ethernet 1/0

● R1(config-if)#no ip split-horizon Eigrp 1

● R1(config-if)# No ip Next-hop-self Eigrp 100

How to reduce query

1) Feasible successor
2) Summarization
3) Eigrp Stub

EIGRP Stub Router

● Non stub router will never advertise query toward Stub router.

● Stub Router will never receive Query message but can send Query msg

● Stub router will never forward one neighbour routes to another neighbour

● Connected and summary routes advertise only by stub router

 ● Stub router can generate query toward non Stub Router
o Information go in Hello msg
● R2(config-router)#router Eigrp 100

● R2(config-router)# Eigrp stub

Equal cost Load Balancing and Unequal

Cost Load Balancing
● Equal cost Load Balancing
o When router have multiple path but I have same metric that is known as Equal cost load
Balancing
a. By changing metric (delay and bandwidth)
b. By using offset list

Offset-list
● It used to perform Load-balancing and for path manipulation.

● With help of offset-list we can perform load-balancing for particular Prefix.

● Make Acl
o Ip access-list standard 10
▪ PERMIT 192.168.3.0 0.0.0.255

● Then apply offset list in EIGRP

o ROUTER EIGRP 1
▪ Offset list 10 in 504320 fa0/0
Unequal Cost Load Balancing
● When router have multiple path but I have different metric that is known as Unequal cost load
Balancing
o Variance
o Traceroute 3.3.3.3 probe 10
● A best route (successor) and backup route (F.s) in topology table.

● Prefix length will check first

o It prefix length is different it will add both in routing table
o It Prefix length is same , then it will check AD value
o Prefix that have least AD value that will install in routing table.
o If AD value is same then will come with metric.
o Prefixes that have least metric that prefix are installing in routing table.
o If metric is also same then they both prefix is install in routing table and both are
Successor.
o In Topology table we have both successor and F.Successor
o Routing table we have only successor.
o Variance 🡪 (1 to 128) by default is 1
o R1(config)#Router Eigrp 100
o R1(config-router)variance 2

Route – filtering
1) Access-list
a. Standard
i. Ip access-list standard 11
1. Permit 30.10.0 0 0.0.255.255
2. exit
● R2(config)#router Eigrp 1

● R2(config-router)#distribute-list 11 in fastethernet 1/0

b. Extended ACL
i. <Action> permit /deny
ii. <protocol> IP
iii. <source >any
iv. <Destination > n/w you want to match.
1. Ip access-list ex 101
 a. Deny ip any 10.1.0.0 0.0.255.255
b. Permit
2. distribute-list 101 in fastethernet 1/0
2) Prefix list
a. In acl we match network part but we doesn’t match subnet mask in ACL
b. In Prefix list we match subnet mask also.
c. R2(config)#ip prefix-list forty deny 40.1.0.0/16 ge 27 le 3
d. R2(config)#ip prefix-list forty permit 0.0.0.0/0 le 32
e. R2(config-router)#distribute-list prefix forty in
f. R2(config)#ip prefix-list abs deny 0.0.0.0/0 ge 24 le 27
g. R2(config)#ip prefix-list abs permit 0.0.0.0/0 le 32
h. R2(config-router)#distribute-list prefix abs in

Route-Map
● Which parameter we can match in route-map
1) Metric
2) Metric Range
3) Next hope
4) Route-type
5) Source Protocol
6) Interface
a. Metric
i. R2(config)#Route-map <name> ccna deny <seq > 10
ii. R2(config-route-map)#match metric 33280
iii. R2(config)#Route-map <name> ccna permit <seq > 20
iv. R2(config-route-map)#exit (match all 🡪 permit any)
v. R2(Config)#router Eigrp 100
vi. R2(config-router)#distribute-list route-map ccnp in
b. Metric range
i. (Max metric + min metric)/2
ii. (32768 + 32000 )/2 = 32384
iii. 32384 – (min metric ) 32000 = 384
iv. R2(config)#route-map DAKU
v. R2(config-route-map)#match metric 32384 + - 384
vi. R2(Config)#router Eigrp 100
vii. R2(config-router)#distribute-list route-map DAKU in
c. Next hope
 i. Ip access-list standard 10
1. Permit 192.168.123.1 0.0.0.0
ii. Route-map nb deny 10
1. Match ip next-hop 10
iii. Route-map nb permit 20
1. Exit
iv. Router Eigrp 100
1. Distribute-list route-map nb in
d. Route type
i. Internal (90) same domain
ii. External ( 170) different domain
1. Route-map nb deny 10
a. Match route-type < internal > <external>
2. Route-map nb permit 20
a. Exit
3. Router Eigrp 100
a. Distribute-list route-map nb in

e. Source protocol
i. Route-map source-protocol
1. Match source-protocol ospf 4
f. Interface
i. Route-map nb
1. Match metric 32384 + - 384
2. Match interface fa0/1
3. Exit
ii. Router Eigrp 100
1. Distribute-list route-map nb in
g. Tag
1. R2(config)#route-map ccna permit
2. R2(config-route-map)#set tag 500
3. R2(config)#router eigrp 1
4. R2(config-router)#redistribute ospf 1 route-map ccna metric 1 1 1 1
1

Summarization
a. Auto-summary
b. Manual -summary
Automatic-Summary
1) At least one subnet should be same.
2) Subnet mask default
3) Summary
a. Local topology
b. directly connected
4) Routing topology
5) Whenever any router perform summarization then router will create entry in the Eigrp table
with NULL 0 (AD=5)
a. AD🡪 5 (to prevent control plane)
b. Null = 0 (to prevent forwarding / data plane loop)
6) With summary we will always forward least metric

Manual Summary

1) At least in local topology

2) Subnet mask (defined by us)
3) Routing topology
a. AD🡪 5 (to prevent control plane)
b. Null = 0 (to prevent forwarding / data plane loop)
4) With summary we will always forward least metric
5) Interface basis
6) Switch(config#)ip summarize-address Eigrp 100 3.3.0.0 255.0.0.0

Redistribute default
a. Redistribute static
i. R1(config)#router eigrp 1
ii. R1(config-router)#redistribute static
iii. R1(config-router)#exit
b. Summarization
i. Ip summary-address Eigrp 100 0.0.0.0 0.0.0.0
c. Network 0.0.0.0
i. Default route should be given by exit interface
IPV6
1) It is 128 bits address.
2) It is represent in Hexadecimal from.
3) Separated with the help of colon “:” only.
4) IPV4
a. 32 bits divided in 4 octets
b. Each octet there is 8 bits
c. 2^8 = 0-255 ( in decimal form)
5) IPV6
a. 128 bits divided on 8 hextets
b. Each hextets there is 16 bits
c. 2^16 = 0-65535 ( in hexadecimal form)
d. 1 hex digit = 4 bits
e. 1 hextets = 16 bit ( 4 digits)
 ● No need of NAT

● No need of Subnetting

● DHCP Stateless (No need DHCP Server)

o No separate configuration is needed.
o No binding are maintained
● DHCP State full (Need DHCP Pool)
o Dhcp server (config)
o Dhcp server will maintain all binding table

16 bit: 16 bit: 16 bit: 16 bit: 16 bit: 16 bit: 16 bit: 16 bit

Starting 64 bits = Network Bits

Last 64 bits = Host bits

Identify the network Id portion / Address type

● In IPV6 there is no broadcast address.

● FF00 ::/8= Multicast address ( null 0 )

● FF02::/8 = Multicast address

● FC00 ::/7= Private Address(Unique local Unicast) these are used for Intra
network communication
● 2000::/3 = Public Address(Global unicast address) these are used for
communication in global (Internet)
● ::1 /128 = loopback
 ● FE80::/10 = Link Local Address
o Link local address is made with the help of EUI ( Extended Unique
Identifier) – 64
o Subnet is always /64
o Fe80 ::/64 = network id ( by default)
o Rest 64 bits will be considered for host IP
o This host Ip is derived from physical int mac address
o Step 1 – FE80:: 🡪 64 bits
● FE80:0000:0000:0000:
o Step 2—🡪Auto calculate by MAC Address of interface
▪ 🡪 1234.ABCD.0128 + FFFE

▪ 1234.AB FF:FE CD.0128

o Step 3—Change 7th bits of 1st Hextet
▪ 1234.ABFF:FECD.0128

▪ 1034.ABFF:FECD.0128

▪ FE80:: 1034.ABFF:FECD.0128

● Manually IPV6 Address

Rule 3:- Identify the host ID portion.

🡪It is in 64 bits.

🡪10AA.01AO.190A

🡪10AA:01 FF: FEAO: 190A

🡪IF 7th bit is 0 then it converts into 1.

🡪IF 7th bit is 1 then it will convert into 0.

How to Write in Short form of IPV6:-
1) If there is leading Zero in a field we can avoid them to write
a. 2001:0014:0001:0002:3003:0008:7000:0080
b. 2001:14:1:2:3003:8:7000: 80
2) If multiple fields are zero then write ten with double colon [::] only once in
a IPV6 address.
a. Ex- FC00:0000:0000:0001:0000:0000:0000:0001
● FC00::1:0:0:0:1

● FC00:0:0:1::1

NDP (Neighbor Discovery Protocol (ICMPv6)

1) Neighbour solicitation ( 135 )

1. It is like ARP REQUEST
2. It is Multicast Msg

2) Neighbour advertisement ( 136)

1. It is like ARP REPLY.
2. It is Unicast

3) Router solicitation (133)

4) Router Advertisement (134)

NDP feature

1) Automatic address configuration (auto-config)

 2) Prefix discovery
3) Duplicate address (NS, NA)
a. Neighbour solicitation
i. SIP (any) ::
ii. DIP ( multicast) FF02::1:FFxx:xxxx
iii. First 104 bit are fixed
iv. Last 24 bit are derived from IPV6 address
b. Neighbour advertisement
i. SIP = IPV6 add
ii. DIP = FF02 ::1
c. If IPV6 is same then it will give error(DAD) duplicate add
detection
4) Mtu discovery
5) Mac address Resolution (RA , NA)

How to assign IPV6

● R1(config-if)#ipv6 enable ( to enable IPV6)

● R1(config-if)#ipv6 address 192:168:101:1::1/64

o FE80::C801:9FF:FE20:8
o ca01.0920.0008
● R1(config)# IPV6 unicast-routing ( to enable IPV6 Routing)

● R2(config-if)#ipv6 address autoconfig

Create loopback IPV6

● R1(config)#int l0

● R1(config-if)#ipv6 address 1::1/128

Configure Static Routing

● R1(config)#ipv6 route 2::2/128 fastethernet 0/0

FE80::C802:4FF:FE70:8
● R2(config)#ipv6 route 1::1/128 fastethernet 0/0
FE80::C801:9FF:FE20:8

Autoconfig

● DHCP for IPV6

o Stateless Auto config
▪ Stateless dhcp which does not maintain any database.

▪ Cannot provide dns-server info.

▪ RS (client) , RA (server)

▪ Valid lifetime 30 days , preferred lifetime 7 days (always)

▪ If you want the client to install any ip received from router then the
prefix length should be /64
▪ Otherwise EUI-64 will not work
o Stateful dhcp (DHCPV6)
▪ DHCP will maintain database

▪ DNS info can be provided

▪ Valid lifetime 48 hrs , Preferred lifetime 24 hrs

▪ Solicit , advertise , request , reply

▪ Client 546 , server 547

o R1(config)#ipv6 dhcp pool R2
o R1(config-dhcpv6)#address prefix 10:1:1::/80
o R1(config-dhcpv6)#dns-server 100:1:1:1::1
o R1(config-dhcpv6)#exit
o R1(config)# int fa0/0
o R1(config-if)#ipv6 dhcp server R2
 o R1(config-if)#ipv6 address 100:1:1::1/80
o R2(config-if)#ipv6 address dhcp R2
o R2(config-if)#ipv6 address autoconfig default
o R1 (config-if)#ipv6 nd prefix 100:1:1::1/80 600 300

🡪Eigrp (Named Mode)

● R1(config)#ipv6 unicast-routing ( to enable IPV6 routing)

● R1(config)#ipv6 router eigrp 100

● R1(config-rtr)#eigrp router-id 1.1.1.1

● R1(config)#int fa0/0

● R1(config-if)#ipv6 eigrp 100

● R1(config-if)#exit

● R1(config)#int fa1/0

● R1(config-if)#ipv6 eigrp 100

● R1(config-if)#exit

● R2(config)#ipv6 router eigrp 100

● R2(config-rtr)#eigrp router-id 2.2.2.2

● R2(config-rtr)#exit

● R2(config)#int fa0/0

● R2(config-if)#ipv6 eigrp 100

● R2(config-if)#exit

● R2(config)#int fa1/0
● R2(config-if)#ipv6 eigrp 100

● R2(config-if)#exit

● R3(config)#ipv6 router eigrp 100

● R3(config-rtr)#eigrp router-id 3.3.3.3

● R3(config-rtr)#exit

● R3(config)#int fa1/0

● R3(config-if)#ipv6 eigrp 100

● R3(config-if)#exit

● R3(config)#int fa0/0

● R3(config-if)#ipv6 eigrp 100

● R3(config-if)#exit

● R4(config)#ipv6 router eigrp 100

● R4(config-rtr)#eigrp router-id 4.4.4.4

● R4(config-rtr)#exit

● R4(config)#int fa1/0

● R4(config-if)#ipv6 eigrp 100

● R4(config-if)#exit

● R4(config)#int f0/0

● R4(config-if)#ipv6 eigrp 100

● R4(config-if)#exit

● R4(config)#int range loopback 1-4

● R4(config-if-range)#ipv6 eigrp 100

● R4(config-if-range)#exit

● R1#sh ipv6 eigrp neighbours

● R1# sh ipv6 eigrp interfaces

● R1# sh ipv6 eigrp interfaces detail

● R1(config)#int fa0/0

● R1(config-if)#ipv6 hello-interval eigrp 100 10

● R1(config-if)#ipv6 hold-time eigrp 100 30

● R1#sh ipv6 eigrp topology

● R1#sh ipv6 eigrp topology all-links

● Offset list is not used in IPv6

● Prefix list
o R1(config)#ipv6 prefix-list NB deny 172:168:101:1::/64
o R1(config)#ipv6 prefix-list NB permit 0::0/0 le 128
o R1(config)#ipv6 router eigrp 100
o R1(config-rtr)#distribute-list prefix-list NB in
o R1(config-rtr)#exit

● R1(config)#int fa0/0

● R1(config-if)#ipv6 summary-address eigrp 100 172:168:101::/61

 ● Classic mode = IPV4 , ipv6
o Micro second
● Named mode = ipv4 , Ipv6
o Pico second
o R1(config)# router eigrp ccnp
o R1(config-router)#address-family ipv4 autonomous-system
200
o R1(config-router-af)#exit
o R1(config-router)#address-family ipv6 autonomous-system
200
o R1(config-router-af)#eigrp router-id 1.1.1.1
o R1(config-router-af)#exit

🡪Ospf V3

● R1 (config)#ipv6 router ospf 10

● R1(config-rtr)#router-id 1.1.1.1

● R1(config-if)#int r fa0/0,l0

● R1(config-if-range)#ipv6 ospf 10 area 1

● R3(config)#router ospfv3 10

● R3(config-router)#router-id 3.3.3.3

● R3(config)#int r fa0/0,l0

● R3(config-if-range)#ospfv3 10 ipv6 area 1

Eigrp (Enhanced Interior Gateway routing Protocol)

● Mode
o Classic mode (32 bit)
o Name Mode (64 bit)
● Metric
o Fixed
▪ Bandwidth

▪ Delay
o Variable
▪ Load

▪ Reliability

● Minimum delay of interface 10 micro second

o 1 gig
o 10 gig
o 20 gig
o 40 gig
o 100 gig
● Bandwidth
o 10^7 / least bandwidth
o 1 gig = 10000000 / 100000 = 10
o 10 gig = 10^7/10^7 = 1
o 20 gig = 10^7/20*10^6= 1
o 40 gig = 10^7/ 40*10^6=1

● 10 gig =
o (10^7 /least bandwidth + total delay /10 )*256
o (10^7/1 0000000+10/10)*256 = (1+1)*256=512
● Metric use in named mode
o Bandwidth
o Load
o Latency (delay)
 o Reliability
o Mtu
o Energy jitter (Reserved)
● Well known

● Lagging

Eigrp state
●

Classic metric problem

● If link is more than 1 gig metric calculation is same

● You have to use multiple process for IPV4 and IPV6

Named mode use wide metric

● Eigrp named mode use one process for IPV4 and IPV6

● Delay use in picoseconds second in named mode

Configuration

● ROUTER EIGRP CCNA

o Address-family ipv4 unicast autonomous-system 100
▪ Network 12.1.1.1 0.0.0.0

● Address-family ipv6 unicast autonomous-system 100

● Show Eigrp address-family ipv4 neighbors.

● Show Eigrp address-family ipv4 interfaces.

● Show Eigrp address-family ipv4 topology

● (Eigrp formula)*256 *256 = metric 64 bit

● Metric 64 bit / RIB scale (by default 128 ) (1 -255)

● Metric rib-scale (1-255)

● Metric version 32 bit

● Formula of Named Mode Eigrp

● (10^7/ B.w + delay pico second /10 ^6 )*65536

● How device calculate delay in pico second

o Method 1
▪ Bandwidth <= 1 Gbps

▪ Delay *10^6 = pico sec

o Method 2
▪ Bandwidth >1 Gbps

▪ 10 ^13 / b.w = pico sec

▪ 10 gbps = 10^13/10^7
o Method 3
 ▪ Manual Delay configured

▪ Delay * 10^6 = pico sec

● How Eigrp calculate metric

OSPF (Open Shortest Path First)

● It is an IGP Protocol.

● Link State Routing Protocols (LSRP)

● Link 🡪

● How many links are their between Routers.

● State 🡪

● How many routers are connected with each link

● In ospf, routes are not advertised routes Infact complete database in the form of link
state advertisements(LSA) are advertised among the complete topology
● Routers calculate their best path themselves.

● OSPF metric = cost = Reference bandwidth /link bandwidth

● Reference Bandwidth (100 mbps)🡪default

● Maximum path by default = 4 maximum = 16/32

● Hello = 10 sec , hold = 40 sec

● OSPF create a map of complete topology on the basis of “AREAS”

 ● Area ID = 32 bit = 0 -- 4.3 billion

● Backbone area (area id =0)

● Non backbone area (1-4.3 billion)/Normal Area

Routers identify with in area

o Backbone Routers
o Non backbone Routers/Normal routers
o ABR(area Border Routers)
o ASBR(Autonomous System Boundary Routers)
● Backbone Routers

● Router have All interface in AREA 0

● Non backbone

● Router have All interface in non backbone area

● Complete database is exchanged between routers within area only.

● Area Border routers(ABR)

o This connected to different area together.
o A router will be ABR only when at least one interface of that is connected to
A0
o ABR will automatically exchanged best routes of one area into another area
without redistribution
o ABR’s will have the complete database of each area they are connected to.
o OSPF work in hierarchical structure(tree)
▪ Root

▪ Stems/Branches
o A0 should be centralized located between all non backbone areas
o There should not be more than one area 0 in ospf domain.
● Algorithm = SPF(Shortest path first ) for best path selection.
● AD=110

● By default auto summary disable

● Classless routing Protocol

● Hop count = unlimited

● Metric = Cost

● ASBR(Autonomous System Boundary Router)

o At least one interface should be in ospf
● Neighbour ship
o Area Id should be same
o Subnet and subnet mask should same
o Hello & dead timers should be same
o Router id should be unique within area
o Authentication should match (optional)
▪ Type 0 = null

▪ Type 1= plain text

▪ Type 2 = md5
o Ospf network type should match
o OSPF stub flag must match

● Process ID 🡪 Locally significant (1-65535)

o You can have same or different process ids on different routers.
o Never advertises into updates.
● Router-id 🡪to differentiate between routers within an area
o Should be unique within area of all routers
o Must not be unique among routers of different areas.

● Election of RID
o Manual defined
o Highest loopback Int IP
 o Highest physical int ip no matter that int is enabled with ospf or not
● Hello = 10 sec

● Hold = 40 sec

● OSPF tables
o Neighbour table
o Routing table
o Database table

● OSPF message / packets

o Hello
▪ It is used to discover the neighbour

▪ Keep Alive

▪ Content in Hello MSG

● OSPF version = 2

● Msg type = 1

● Packet length = header info.

● ROUTER ID

● Area id

● Packet checksum

● Auth type

● Auth data
o Type 0 = null
o Type 1= plain text
o Type 2 = md5
● Network mask

● Hello
 ● Hold

● Priority

● Neighbour Router id

● DR router id

● BDR router-id
o DBD(data base description)
o LSR (Link state Request
o LSU (Link state Update)
o LSACK(Link state ACK)

OSPF state
o Down
▪ No hello will be sent or received
o Init
▪ Hello is sent but not received in return

▪ Active neighbour ship

o 2-way
▪ hello is sent and received
o Exstart
● Null DBD is exchanged between to verify bidirectional connectivity
for updates.
▪ Master/Slaves (Election)

● Highest RID

▪ To decide who will start sending the database first

▪ Once the election will be completed they will move to “ exchange”

state
▪ Seq number

▪ MTU size negotiate

o Exchange
 ▪ Asking about each other’s database.

▪ LSA header send from master

▪ Then slave will send LSA header in response

▪ But will not have network information in that DBD.

o Loading
▪ Slave will send LSR to ask detail information about LSA Header.

▪ Master will send LSU

▪ Slave will send LSACK

▪ When the database is completely synchronised they will move to full

state and will said to form “adjacency”
▪ Neighbour ship / adjacency -> 40 sec (wait timer )

2 –Way State
Network type = BMA (Broadcast Multi Access)

= P2P (Point to point)

Number of Neighbour ship = n (n-1)/2 n= Number of router

● To reduce the number of adjacencies, router will elect DR & BDR between the router

● DR(Designated Router)

● BDR(Backup Designated Router)

● Election Criteria
o HIGHEST Priority (def=1) = 0 -255
o Highest RID
One broadcast domain can have only one DR and one BDR

o Every DRother will form adjacency with DR and BDR both.

o DR and BDR will form adjacency with each other.
o DRother will never form adjacency with DRother.

* 224.0.0.5 🡪 When ever DR send any update all will received from this multicast add. Every
router of ospf can accept that.

* 224.0.0.6🡪Only DR and BDR can accept the update.

Clear IP OSPF PROCESS (Y) 🡪 Simultaneously with in msec

🡪DR (only two router in one BMA) currently

🡪DR/BDR (when more than two router in BMA)

🡪No election takes place of DR and BDR on P2p Segment.

LSA (LINK STATE ADVERTISEMENT)

1) LSA 1(Router LSA) 🡪 with in Area
● Generated by each router with in an area.

● Link State ID
o RID of router generated the LSA
● Adv. Router
o RID of Router advertise the LSA
● Number of router LSA = Number of router in an area

● Link count
o Broadcast (BMA)= 1 segment = 1 link count
o P2P = 1 Segment = 2 link count
o Per loopback = 1 link count

2) LSA2 (NETWORK LSA) 🡪 with in area

 ● Generated by DR

● LSID =DR physical interface IP

● Adv. Router= DR RID

● Number of DR within AREA = number of broadcast multi Access segment

● Number of network LSA = Number of DR

● Since in Router LSA, Network subnet mask does not get advertised so in order to advertise the
subnet mask for the transmit links n/w LSA is used.

3) LSA3 (Summary LSA) 🡪 Inter Area LSA

o Generated by ABR
o To advertise best routes of one area into another area.
o This is calculated on the basis of cost.
o LSID= network ID of route advertised by ABR
o ADV. Router= ABR RID
o Number of summary LSA= number of routes of another area.
● Area 0 should be only one and centralised in a single ospf domain.

● Two ABR’s will not accept and forward summary LSA of each other until they both have RLSA of
each other learned trough Area 0

Virtual Link

● Virtual link is always considered as P2P

● No DR/BDR

● No dead Timer

● Update sent over virtual link for Area will never get aged out
o Will be sent with a DNA bit to notify neighbour that it will get updates with DNA bit set
periodically
o Authentication is applicable on virtual link.
o
4) LSA 5 (AS-External LSA)
● Generated by ASBR
 o when external AS other than ospf are redistributed into OSPF.
o Number of LSA 5 = Number of external routers adv into ospf
o LSID = network id of external routes
o Adv router= ASBR RID
o RID of ASBR does not change throughout ospf domain in LSA 5

5) LSA 4 (Summary – ASBR LSA)

1. Generated by ABR
2. To let other router to know how to reach ASBR
a. LSID = ASBR RID
b. ADV. RID = ABR RID
c. LSA 4 exists in those areas where router LSA of ASBR does not exist.

● LSA5 (Metric)
o E1 = External Type 1
o E2 = External Type 2 (Default)
▪ For redistribute routes = Seed metric Value = 20 (Default)

▪ For default route = Metric value = 1 (default)

o E1 = Total path cost
▪ # Redistribute Eigrp 100 subnets metric-type 1

▪ Show ip ospf border-routers

▪ O>OIA>E1>E2 🡪Route preference ospf

▪ O = Intra Area Route

▪ OIA = Inter Area Route

▪ E1 =

● for redistributed routes = metric value = 20 (def) + forward metric

● For default route = metric value = 1 (def) + forward metric

● Forward metric = metric to reach ABR will be calculated on the basis of

LSA 4.
6) LSA 6 (Group membership)
7) NSSA External LSA
Types of Area in OSPF
1) Standard Area
2) Stub Area
3) Totally stub Area
4) NSSA Area (Not so stub Area)
5) Totally NSSA Area

Standard Area

● By default all area are standard Area

Stub Area

● To filter LSA5 into stub area.

● Since ABR filtered LSA 5 into stub area, it will automatically filter LSA 4 into stub area.

● ABR will continue to advertise LSA 3 from non-stub area to stub area and vice versa.

● ABR will generate default route towards stub area routers in the form of LSA3.

● Stub is run on both sides’ means ABR and all other router of this area.

● Stub area not allow virtual link

● Stub area is not allowed ASBR router so we can’t configure stub area near ASBR router.

● We can’t make stub to that area which is attached with ASBR because it can’t do
redistribution.
● Area 0 cannot be configured as Stub

● Neighbourship parameter
 o Stub flag should be same
o By default = 1 ( when stub is not configured)

● Area 1 stub

Totally stub Area

● Implement only on ABR router.

● ABR not only filters LSA 5 & LSA 4 but also filter LSA 3 but continue to advertise only
default route in the form of LSA 3.
● ABR will continue to advertise LSA3 of stub area into non-stub areas.

● ABR Router will generate default Route Automatically in stub Area

● Area 1 stub no-summary

NSSA Area (Not so stub Area)

● In NSSA Area ASBR can exist in form of TYPE 7 LSA

● NSSA ASBR type 7 NSSA-External LSA (Area Dependent)

● Network ID , Subnet Mask , Metric default “20” , Route – type N2 (default) and N1

● ABR will translate type 7 into type 5 N2 into N1 (vice versa)

● Then ABR will be ASBR

● There is no Need of type 4 LSA

● Whenever NSSA ABSR advertise route. It will add interface ip as forwarding address.
 ● Default route will be manually configured

● Then ASBR will copy and forward

i. Lowest loopback ip
ii. Highest physical interface ip
iii. Area 1 NSSA
iv. Area 1 nssa default – information originate.
v. R1(config-router)#redistribute eigrp 100 subnets metric-type 1

Totally NSSA Area

b. Area 1 nssa no-summary
1) Stub
a. Lsa 4 & Lsa 5 filter
b. Default route generate automatically in LSA3 by ABR
2) Totally stub
a. Lsa 5 & Lsa4 and LSa 3 filter
b. Default route generate automatically in LSA3 by ABR

3) NSSA
a. LSA 5 convert into LSA 7
b. Manually generate default route on ABR
4) Toally NSSA
a. Lsa 5 & Lsa4 and LSa 3 filter
b. Default route generate automatically in LSA3 by ABR
How to Originate default route in OSPF

● Type 3
o Stub
o Totally Stub
● Type 5
o ASBR
o Default –information originate
o Default –information originate metric-type 1
o Default –information originate metric-type 1 metric 100
o Default –information originate metric-type 1 100 always
o Conditional default route advertisement
▪ IP access-list standard 10

● Permit 172.168.1.0 0.0.0.255

▪ Route-map 10

● Match ip add 10

▪ Router ospf 1

● Default-information originate route-map 10

Filtering
 1) Route filtering
a. Distribute-list
i. ACL
ii. Prefix-List
iii. Route-map
b. AD value
i. Distance ospf intra-area
c. Summarization
i. Summary-address 50.1.0.0 255.255.0.0 not-advertise
2) Area filter-list
a. LSA 3 filter / Inter Area Route
b. Apply on ABR only
c. Prefix-list
d. #ip prefix-list ccnp deny 60.1.0.0/16 ge 32 le 32
e. #route ospf 100
f. #area 1 filter-list prefix ccnp out
3) LSA filtering
a. Types of Area
b. Area Filter-list (LSA 3)
c. Distribute-list on ASBR Router for outgoing direction (LSA 5)
d. Summarization
i. ABR (LSA 3)
ii. ASBR (LSA 5)

Authentication in Ospf
1) Type 0 = Null
2) Type 1 = Plain text
3) Type 2 = MD5 authentication

OSPF Network Type

1) Broadcast
a. Default network type on ethernet port
b. Hello 10 , hold 40
c. Dynamic Neighbourship will form.
d. DR and BDR elect
e. # IP OSPF Network Broadcast
 2) Non-Broadcast
a. Full Mesh Topology
b. Default network on serial Int (Frame-relay)
c. Hello 30 , Dead 120
d. Not support multicast so dynamic Neighbourship will not form
e. Statically Neighbourship will support
f. DR & BDR will elect
3) Point to Point
4) Point-to-multipoint
a. Hello 30 , dead 120
b. Dynamic NBR ship
c. No DR/BDR election
5) Point-to-multipoint Non-Broadcast
a. Hello 30 , dead 120
b. No Dynamic NBR ship
c. No DR/BDR election

6) Loopback
a. It will always advertise in /32
7) Virtual Link

OSPFv3
● OPSF v2 (IPV4)

● OSPF V3(IPV6)

● Router OSPFV3 process id ( Address family config)

● IPV6 router ospf process id (IPV6)

● Number of SPF calculation reduce in OSPFV3

● Hello
1 Source IPV6
▪ Int (link local address)
2 Destination IPV6
▪ FF02::5 , FF02::6
3 Cost calculation same
 4 Layer protocols and it encapsulate in IPV6
● LSA
1 Router LSA
▪ Which Neighbour connected on which link
2 Network LSA
3 Inter Area Prefix LSA
4 Intra Area router LSA
5 External
6 NSSA External
7 TYPE 8 Link LSA
▪ It will share Link local address

▪ It will be local in link

8 TYPE 9 Intra area Prefix LSA
▪ If any network add type 9 will generate

● IPv6 router ospf 1

1 Router id 1.1.1.1
● IPV6 ospf 1 area 1

● Show pv6 ospf database link adv-router 1.1.1.1

● Show ipv6 ospf database router adv-router 1.1.1.1

● Show ipv6 ospf database prefix adv-router 1.1.1.1

Policy Based Routing

 1) Traffic match
a. Match in ext. acl
2) Policy
a. Route-map
3) Policy
a. Apply in Interface
● IP access-list extended 101

● Permit ip 192.168.1.128 0.0.0.127 192.168.2.0 0.0.0.255

● Route-map policy

● Match ip address 101

● Set ip next-hop 23.0.0.3

● Exit

● Int fa0/0

● Ip policy route-map policy

● Show ip int fa0/0 | section Policy

● Ip access-list extended 101

● Permit tcp any any eq 23

● Route-map Policy

● Match ip address 101

● Set ip next-hop 23.0.0.3

● Int fa0/0

● Ip policy route-map Policy

URPF
1) Unicast reverse path forwarding
 2) R1(config)#ip access-list extended 101
3) R1(config-ext-nacl)#permit ip host 10.0.0.130 host 36.0.0.6
4) R1(config-ext-nacl)#exit
5) R1(config)#route-map 101
6) R1(config-route-map)#match ip address 101
7) R1(config-route-map)#set ip next-hop 12.0.0.2
8) R1(config-route-map)#exit
9) R1(config)#ip local policy route-map 101
10) R2(config)#int fa1/1
11) R2(config-if)#ip verify unicast source reachable-via rx
12) R2(config-if)#exit
13) R2(config)#ip access-list standard 10
14) R2(config-std-nacl)#deny 1.1.1.1 0.0.0.0
15) R2(config-std-nacl)#permit any
16) R2(config-std-nacl)#exit
17) R2(config)#router ospf 1
18) R2(config-router)#distribute-list 10 in
19) R2(config-router)#exit
20) R2(config)#ip route 0.0.0.0 0.0.0.0 23.0.0.3
21) R2(config-if)#ip verify unicast source reachable-via any allow-default

PBR + IP SLA (service Level Agreement)

● R1(config)#ip Sla 10

● R1(config-ip-sla)#icmp-echo 12.0.0.2 source-ip 12 .0.0.1

● R1(config-ip-sla-echo)#Frequency <1-604800>

● R1(config)# ip sla schedule 10 start-time now life forever

● #show ip sla statistics

● R1(config)#track 11 ip sla 10 reachability

● Show track

● R1(config)#route-map Policy

● R1(config-route-map)#set ip next-hop verify-availability 12.0.0.2 100 track 11

● R1(config)#int fa0/0

● R1(config-if)#ip policy route-map policy

● R1(config-if)#exit

● Show ip policy

● R1#show ip int fa0/0 | section policy

● Show route-map Policy

● R3(config)#ip access-list extended 101

● R3(config-ext-nacl)#deny icmp host 12.0.0.2 host 12.0.0.1

● R3(config-ext-nacl)#permit ip any any

● R3(config-ext-nacl)#exit

● Set interface

● Set ip default next-hop