Routing Operation

Routing Operations in Cisco
IOS Routers
BRKRST-2350
Scott Custer
Team Lead – RTP TAC Routing Protocols
2011
BRKARC-2350 © 2010 Cisco and/or its affiliates. All rights reserved. Cisco Public 1
Routing Operation in Cisco IOS Routers
Topics Covered:
• The Routing Table (RIB)

• Overriding the Routing Table
• Load Sharing
• Routing Segmentation and Separation
• Routing and Router Resources
The Routing Table
© 2011 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 3
The Routing Table
• Basic Structure
• Route Selection
• Interface Down Events
• Backup Routes
• Static Routes
• Discard Routes
The Routing Table
Basic Structure
55
© 2011 Cisco and/or its affiliates. All rights reserved. Cisco Confidential
BRKARC-2350 © 2010 Cisco and/or its affiliates. All rights reserved. Cisco Public
• The Routing Information Routing Protocols
Base or RIB
Interfaces
• Routing Protocols*
Other Sources
Install routes into the RIB
• Interfaces
RIB
• Other Sources
-Performance Routing (PFR)
-Reverse-Route Injection (RRI)
Basic Structure
• Routing Protocols Routing Protocols

Pull routes from the RIB Interfaces
for redistribution
• Cisco Express Forwarding (CEF) Other Sources

CEF maintains the FIB, Forwarding Information
Base, and the Adjacency tables
A copy of the RIB is sent down to the FIB RIB
A copy of the RIB is sent down to the hardware
forwarding component
FIB Adj
Hardware
Forwarding
Tables
Basic Structure
router#show ip route
Codes: L - local, C - connected, S - static, R - RIP, M - mobile, B - BGP

D - EIGRP, EX - EIGRP external, O - OSPF, IA - OSPF inter area
N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2
E1 - OSPF external type 1, E2 - OSPF external type 2
i - IS-IS, su - IS-IS summary, L1 - IS-IS level-1, L2 - IS-IS level-2
ia - IS-IS inter area, * - candidate default, U - per-user static route
o - ODR, P - periodic downloaded static route, + - replicated route
10.0.0.0/8 is variably subnetted, 2 subnets, 2 masks

C 10.1.20.0/25 is directly connected, Ethernet1/0
L 10.1.20.1/32 is directly connected, Ethernet1/0
D 172.16.24.0/24 [90/307200] via 10.1.20.2, 00:23:36, Ethernet1/0
C 192.168.10.0/24 is directly connected, Serial2/0
L 192.168.10.1/32 is directly connected, Serial2/0
S 200.15.0.0/16 is directly connected, Null0
Basic Structure
Network

D 172.16.24.0/24 [90/307200] via 10.1.20.2, 00:23:36, Ethernet1/0
Basic Structure
Route

D 172.16.24.0/24 [90/307200] via 10.1.20.2, 00:23:36, Ethernet1/0
Basic Structure
Network + Route

D 172.16.24.0/24 [90/307200] via 10.1.20.2, 00:23:36, Ethernet1/0
Basic Structure
Major networks with subnets show up under a single network with

multiple routes
D 172.16.24.0/24 [90/307200] via 10.1.20.2, 00:23:36, Ethernet1/0
Native mask routes and supernets show up as separate networks

Basic Structure
Administrative distance Metric Last Modification Time
D EX 192.168.254.0/24 [170/3072256] via 208.0.246.10, 00:58:45, Serial3/0
• The administrative distance is locally significant

• The metric is a protocol specific measure
• The time shown is the amount of time since the route
was last touched
EIGRP recalculation of any type, including losing an alternate path, resets
this timer
OSPF SPF run resets this timer
IS-IS SPF run resets this timer
Admin Distance
Route Source Default Distance Values
Connected interface 0
Static route 1
Enhanced Interior Gateway Routing Protocol (EIGRP)

5
summary route
External Border Gateway Protocol (BGP) 20
Internal EIGRP 90
IGRP 100
OSPF 110
Intermediate System-to-Intermediate System (IS-IS) 115
Routing Information Protocol (RIP) 120
Exterior Gateway Protocol (EGP) 140
On Demand Routing (ODR) 160
External EIGRP 170
Internal BGP 200
Unknown (infinity) 255

Routing Table:
Route Selection
Route Selection
How does the RIB decide which route is best among
various sources?
EIGRP 1
(a) EIGRP 1 (c) The RIB (d) EIGRP
installs a sends the decides if
Actually, it doesn‘t. route into new route the new
the RIB to EIGRP 1 route should
replace the
existing one
D EX 192.168.254.0/24 ....
(b) OSPF 2 (e) The RIB

attempts to notifies OSPF
install the 2 if the route
same route in was installed
the RIB or not
OSPF 2
Route Selection
• Each route is marked with the installing routing process
• When another process attempts to install an overlapping
route in the RIB, the RIB allows the owner of the current route
to decide if it should be installed or not
• Generally, this decision is made using the administrative
distance of the two routing processes
Route Selection
The RIB receives OSPF’s new route, calls into EIGRP, and EIGRP
determines if the OSPF route should be installed
The RIB receives the EIGRP reply and flushes the

EIGRP route
RT: closer admin distance for 192.168.254.0, flushing 1 routes
EIGRP-IPv4(1):Callback: lostroute 192.168.254.0/24
RT: add 192.168.254.0/24 via 208.0.245.11, ospf metric [110/65]
EIGRP receives a callback stating the RIB has

removed one of its routes
The RIB installs OSPF’s route

The RIB notifies OSPF its route has been installed
Route Selection
How is administrative distance used to determine which route
should be installed?
RIPv2 EIGRP EIGRP
10.1.1.0/24 10.1.1.0/24 10.1.1.0/25

These two routes
are identical
EIGRP internal AD = 90
RIP AD = 120
EIGRP internal route is installed
The lowest administrative distance wins.

Route Selection
• Only identical routes are compared
• Identical prefixes with different prefix lengths are not the same route
• The route from the protocol with the lower administrative distance is
installed
Route Selection
• What happens if the administrative distance of the two routes are
equal?
• It depends on the
routing protocol RIPv2 EIGRP EIGRP
10.1.1.0/24 10.1.1.0/24 10.1.1.0/25
These two routes

are identical
EIGRP internal = 90
RIPv2 = Configured to 90
It depends. Usually the route with the default

AD is installed into the RIB
Route Selection
OSPF and IS-IS
OSPF 1 OSPF 2 IS-IS

AD=110 AD=110 AD=115
10.1.1.0/24 10.1.1.0/24 10.1.1.0/24

IS-IS default AD 115
OSPF default AD 110
OSPF wins
Default AD is the same between the

two OSPF processes
The older route remains in the RIB

Route Selection
OSPF and IS-IS
• The default administrative distance of each route is
compared
• If these are the same, the older route remains in the
routing table
Route Selection
EIGRP
EIGRP 1 EIGRP 2
AD=90 AD=90 OSPF
Metric=1000 Metric=500 AD=110
10.1.1.0/24 10.1.1.0/24 10.1.1.0/24

EIGRP def. AD is 90
OSPF def. AD is 110
EIGRP wins
EIGRP 2 route metric is

lower than EIGRP 1
EIGRP 2 route is installed in the RIB
Tie breaker is the lowest AS number

Route Selection
• What happens if the same routing process has multiple identical
routes with the same metric?
EIGRP 1 EIGRP 1
AD=90 AD=90 OSPF
Metric=500 Metric=500 AD=110
10.1.1.0/24 10.1.1.0/24 10.1.1.0/24

EIGRP def. AD is 90
OSPF def. AD is 110
EIGRP wins
Both routes have the same When multiple paths exist

metric and the same AS
number within the same routing
process with equal cost, both
BOTH routes are installed in the RIB are presented to the RIB for
equal cost load-sharing
Route Selection
EIGRP
• Default administrative distance of each route‘s
protocol is compared
• If these are the same, both routes must be EIGRP
• Compare the metric type and metric, the lower cost
route is installed
• If the metric and metric type are the same, compare the EIGRP
AS number
• The lower AS number wins
The Routing Table:
Interface Down Events
How does the RIB interact with the routing protocol when an interface
fails?
interface Ethernet1/0
ip address 10.1.20.1 255.255.255.128
router eigrp 1
network 10.0.0.0
A
E1/0
E1/0
B
interface Ethernet1/0
ip address 10.1.20.2 255.255.255.128
router eigrp 1
network 10.0.0.0
A#show ip route

D 172.16.24.0/24 [90/307200] via 10.1.20.2, 00:00:05, Ethernet1/0....
The RIB tells EIGRP the interface is down 20:46:58.151: EIGRP-IPv4(1): Callback: route_adjust
Ethernet1/0
20:46:58.151: RT: interface Ethernet1/0 removed from routing
10.1.20.0, connected, is removed from the table
RIB 20:46:58.151: RT: del 10.1.20.0 via 0.0.0.0, connected metric
[0/0]
20:46:58.151: RT: delete subnet route to 10.1.20.0/25
172.16.24.0, learned through EIGRP, is 20:46:58.151: RT: delete route to 10.1.20.1 via 0.0.0.0,
removed from the RIB (before EIGRP Ethernet1/0
takes the neighbor down) 20:46:58.151: RT: no routes to 10.1.20.1, flushing
20:46:58.151: RT: delete route to 172.16.24.0 via 10.1.20.2,
Ethernet1/0
20:46:58.151: RT: no routes to 172.16.24.0, flushing
The EIGRP neighbor goes down
20:46:58.151: %DUAL-5-NBRCHANGE: EIGRP-IPv4 1:
Neighbor 10.1.20.2 (Ethernet1/0) is down: interface down
20:47:00.139: %LINK-5-CHANGED: Interface Ethernet1/0,
The interface changes to down state changed state to administratively down
20:47:01.139: %LINEPROTO-5-UPDOWN: Line protocol on
Interface Ethernet1/0, changed state to down
The Routing Table:
Backup, Static, and Discard Routes
Backup Routes
Backup Routes
RIPv2 EIGRP EIGRP
10.1.1.0/24 10.1.1.0/24 10.1.1.0/25
These two routes

are identical
EIGRP internal AD = 90
RIP AD = 120
EIGRP
RIPinternal installed
route installed
RIB calls backups

Backup Routes
• If a route with a low administrative distance fails…
• The routing table calls each routing process asking for backup routes
• Each routing process attempts to install its
matching routes
• The route with the lowest administrative distance wins
Backup Routes
router-b#show ip route
Codes: D - EIGRP, EX - EIGRP external, O – OSPF....
.... The route is installed
O 10.0.16.0/24 [110/1064] via 10.0.12.10, Serial0/3
by OSPF
router-b#show ip eigrp topo EIGRP has the same route

IP-EIGRP Topology Table for AS(100)/ID(208.0.17.11) in its topology table, but
.... it’s not installed because
P 10.0.16.0/24, 0 successors, FD is Inaccessible it has a higher AD (170)
router-b#debug ip routing
router-b#debug ip eigrp notifications The OSPF route fails…
....
RT: delete route to 10.0.16.0/24 EIGRP gets a callback for
10.0.16.0/24, which is the
IP-EIGRP: Callback: callbackup_routes 10.0.16.0/24
OSPF route that failed
IP-EIGRP: Callback: reload_iptable
RT: add 10.0.16.0/24 via 10.0.12.10, eigrp metric [170/3072256] EIGRP installs the
existing 10.0.16.0/24 route
from its topology table
router-b#show ip route
Codes: D - EIGRP, EX - EIGRP external, O – OSPF....
.... The route is now installed
D EX 10.0.16.0/24 [170/3072256] via 10.0.12.10, Serial0/3
.... by EIGRP
Static Routes
Static Routes
Static Routes Can Have a Next Hop of an IP Address
• ip route 10.1.1.0 255.255.255.0 10.1.2.1
• This causes the RIB and CEF to do a recursive lookup to find the
correct Layer 2 header to rewrite onto the packet
• Recursive lookup: For each packet destined to 10.1.1.0/24:
1) Look up the destination (10.1.1.0/24)
2) Find the next hop is 10.1.2.1
3) Look up how to get to 10.1.2.1
4) Find 10.1.2.1 is via connected interface
5) Look up the layer 2 header out connected interface to next-hop 10.1.2.1
• As long as the next hop is reachable, the router assumes the destination
through that next hop is reachable
Static Routes
Static Routes Can Have a Next Hop of a Point-to-Point
interface
• ip route 10.1.1.0 255.255.255.0 serial0
• The RIB and forwarding tables point the route directly out the
point-to-point interface
- No need to do a recursive lookup
• For each packet destined to 10.1.1.0/24, the Layer 2 rewrite

header is set up to reach the other end of the point-to-point
link
• Faster, less complicated lookup
• As long as the interface is up, the router assumes the
destination is reachable through that interface
Static Routes
Static Routes Can Have a Next Hop of a broadcast
interface
• ip route 10.1.0.0 255.255.0.0 fa0/0
• The RIB and CEF will point this route directly to the broadcast
interface
....
10.0.0.0/16 is subnetted, 1 subnets
S 10.1.0.0 is directly connected, FastEthernet0/0
Static Routes
ip route 0.0.0.0 0.0.0.0 FastEthernet 0/0
10.2.1.0/24 via internetwork
A B
10.2.1.25
Packet to ARP for
10.2.1.25 10.2.1.25
Best route is Route to Proxy ARP must be enabled
through broadcast 10.2.1.25 through
interface FE0/0 internetwork
ARP reply
10.2.1.25 via B This happens for each
destination in which host
Send packets to
10.2.1.25 via B’s
A does not have a route
MAC address
Note: Proxy ARP disabled by default
Static Routes
• For a default route (0.0.0.0/0), this could result
in 232 ARP entries in A‘s local tables
This would overflow the ARP cache, and crash A
• Control static routes to broadcast interfaces

Small range of reachable addresses
Don‘t use with proxy ARP, just for reaching hosts
actually connected to that segment
• Static routes to point-to-point interfaces don‘t

have this problem
Static Routes
• For a static route to an interface, the destination network is shown in
the routing table as connected:
router(config)#ip route 10.1.0.0 255.255.0.0 fa 0/1
....
S 10.1.0.0 is directly connected, FastEthernet0/1
• Static routes to interfaces will be included if you configure redistribute

connected
• How do routing protocols handle this in relation to the network
statement?
Static Routes
• OSPF:
Static routes to interfaces are not advertised as a result
of a network statement
• IS-IS:
IS-IS doesn‘t use network statements, so static routes to
interfaces are not advertised without redistribution
• EIGRP:
Static routes to interfaces are considered connected routes
They will be picked up and advertised if they are contained
within a network statement
• BGP:
Static routes to interfaces are installed in the routing table
They will be picked up and advertised if they match a
network statement
Static Routes
Floating Static Routes
ip route 10.1.1.0 255.255.255.0 <B> 250
show ip route
....
D EX 10.1.1.0/24 via <C>
EIGRP
A learned route
No
dynamic
routing C
10.1.1.0/24
Static Routes
• The concepts of administrative distance and backup routes are
used to create floating static routes
• Configuring a static route with a very high administrative
distance ensures it won‘t be installed as long as there is a
dynamically learned route installed in the RIB using the default
AD
• 255 = unreachable
Static Routes
Floating Static Routes
ip route 10.1.1.0 255.255.255.0 <B> 250
show ip ip
route
route
....
....
S10.1.1.0
10.1.1.0
via
via
<C>
<B>
EIGRP
A learned route
No
dynamic
routing C
10.1.1.0/24
Static Routes
• When the dynamically learned route fails, the RIB calls the processes
looking for a backup route
• Since no other processes have routes to install, the static route with
an administrative distance of 250 wins
• This assumes that the primary route will be removed from the table in
a failure event. If a failure event will not remove the primary route from
the RIB then the floating static backup will not be installed.
Static Routing with Object Tracking
How can we get dynamic failover with no dynamic routing?
ip route 10.1.1.0 255.255.255.0 <C>
ip route 10.1.1.0 255.255.255.0 <B> 10
show ip route
....
10.1.1.0 via <C>
A 172.16.12.0/24
172.16.13.0/24
C
.2
B 172.16.24.0/24
.3
10.1.1.0/24
Static Routing with Object Tracking
ip route 10.1.1.0 255.255.255.0 172.16.12.2 track 1
ip route 10.1.1.0 255.255.255.0 172.16.13.3 10
track 1 ip sla 1 reachability
ip sla 1
icmp-echo 172.16.24.2 source-interface Ethernet1/0
frequency 5
ip sla schedule 1 life forever start-time now
Probes are being sent to 172.16.24.2 02:34:12.106: ICMP: echo reply rcvd, src 172.16.24.2, dst
172.16.12.1, topology BASE, dscp 0 topoid 0
02:34:17.114: ICMP: dst (172.16.12.1) host unreachable rcv
from 172.16.12.2
The track object goes down 02:34:17.306: Track: 1 Change #9 ip sla 1, reachability Up-
when reachability fails >Down
02:34:17.306: %TRACKING-5-STATE: 1 ip sla 1 reachability
The routing table is updated to remove Up->Down
02:34:17.306: RT: del 10.1.1.0 via 172.16.12.2, static metric
the route to the destination through the
[1/0]
tracked path
02:34:17.306: RT: delete subnet route to 10.1.1.0/24
02:34:17.306: RT: updating static 10.1.1.0/24 (0x0) via
The floating static route is installed into 172.16.13.3
the routing table 02:34:17.306: RT: add 10.1.1.0/24 via 172.16.13.3, static
metric [10/0]
02:34:17.310: RT: updating static 10.1.1.0/24 (0x0) via
Probes are still sent to determine when 172.16.13.3
02:34:22.114: ICMP: dst (172.16.12.1) host unreachable rcv
this path is available again
from 172.16.12.2
Object Tracking Tip - Use EEM
event manager applet static_tracking
event syslog pattern "%TRACKING-5-STATE: 1 ip sla 1 reachability Up->Down"
action 1 wait 3
action 2 cli command "enable"
action 3 cli command "term len 0"
action 4 cli command "term exec prompt timestamp"
action 5 cli command "show log | append flash:log_output"
action 6 mail server "<mail_server_ip>" to "<email_address>" from "<sender>" subject “Link C-D is down."
end
A
Connectivity is unstable.
Any helpful tools?
C
We can use Embedded B

Event Manager (EEM)
to notify us of the issue D
www.cisco.com/web/go/eem 10.1.1.0/24
Static Routing with BFD
• BFD – Bidirectional Forwarding Detection

BFD builds its own neighbor relationship with adjacent routers to provide fast
peer failure detection independent of media type, encapsulation, or routing
protocols
• Static routing has no method of peer discovery

Can use BFD to track the reachability of the peer
• Static route only installed in RIB if BFD session is up allowing

us to consider the Gateway reachable
Interface GigabitEthernet0
ip address 2.2.2.1 255.255.255.252
bfd interval 500 min_rx 500 multiplier 5
ip route static bfd GigabitEthernet0 2.2.2.2

ip route 192.168.1.1 255.255.255.255 GigabitEthernet0 2.2.2.2
Discard Routes
Discard Routes
• Discard routes are created when a router

aggregates routing information
(EIGRP) ip summary-address eigrp 100 10.1.0.0 255.255.0.0 5
(OSPF) area 1 range 10.1.0.0 255.255.0.0
(IS-IS) summary-address 10.1.0.0 255.255.0.0 level-2
....
Router_A#show ip route
....
D 10.1.0.0/16 is a summary, 00:04:03, Null0
• A discard route has an administrative distance

of five by default
Discard Routes
Why is this discard route created?
A
• Suppose
10.1.0.0/16
A is advertising a default
route to B
B is advertising the summary
10.1.0.0/16 to A
B is not building a discard route 0.0.0.0/0
10.1.3.0/24
for this summary
interface eth 0/0 10.1.1.0/24

ip summary-address eigrp 100 10.1.0.0 255.255.0.0
…
ip route 0.0.0.0 0.0.0.0 <A>
Discard Routes
A receives a packet for 10.1.3.1
A
10.1.3.1 via 10.1.0.0/16
10.1.0.0/16
0.0.0.0/0
B 10.1.3.0/24
10.1.3.1 via 0.0.0.0/0
10.1.1.0/24
We have a permanent routing loop.

Routing Loop avoided if B had a discard route
for 10.1.0.0/16
Discard Routes
• A receives a packet for 10.1.3.1
A examines its local routing table, and finds the best path is through B,
using the route to 10.1.0.0/16
A forwards the packet to B
• B receives the packet for 10.1.3.1

B examines its local routing table, and finds the best path is through A,
using the default route
B forwards the packet to A
• We have a permanent routing loop!

• If B builds a discard route for 10.1.0.0/16, it will discard the
packet, rather than forwarding it through the default route back
to A
Discard Routes
• Can you prevent the routing protocol from creating
a discard route?
OSPF
router ospf 100
no discard route
EIGRP
interface serial0
ip summary-address 10.1.0.0 255.255.0.0 255
• Why would you want to get rid of the discard route?

Be very careful with removing discard routes—this can create routing loops
Discard Routes
Be careful when using non default AD values for summary

routes.
D#show ip route 10.1.1.0

A Routing entry for 10.1.1.0/24
Known via "eigrp 1", distance 250, type internal
Redistributing via eigrp 1
Routing Descriptor Blocks:
C * directly connected, via Null0
10.1.1.0/24
AD = 90
B
D receives the route from C via
D EIGRP with an AD of 90. This is
ip summary-address eigrp 1 better than the installed route to
10.1.1.0 255.255.255.0 250
10.1.1.0/30 Null with an AD of 250.
10.1.1.4/30
10.1.1.8/30…
Overriding the Routing Table
Overriding the Routing Table
What if I do not want to route my traffic based only on destination?
• Policy-Based Routing (PBR)
Route-map
• Performance Routing (PfR)

Source IP
ToS
Application
Link Utilization
Overriding the Routing Table:
Policy Based Routing
Route Maps
• Route maps allow you to:
Combine more than one type of filter into a single phrase
Use some rudimentary forms of logical ―AND‖ and ―O
R‖
to filter routes
Set some route attributes, rather than just permitting or
denying routes
• Route maps can be used to:

Set IP next-hop
Filter BGP updates
Filter EIGRP updates
Filter routes being redistributed between two protocols
Etc.
Route Maps
One of the two results of a route map is
whether the route is permitted or denied
through the filter
Type of phrase (permit or deny) Phrase sequence
route-map networkers permit 10 If this match
Phrase 1 match ip address 10 succeeds, the
route-map networkers permit 20 route map exits
Phrase 2 match ip address 20 If this match with permit
set ip next-hop 10.1.1.1 succeeds, set
route-map networkers permit 30 next-hop is If not, the route
Phrase 3 set ip next-hop 10.2.2.2 executed, and map continues
the route map with the next
exits with phrase
If the route map makes it to this permit
phrase, set next-hop is
executed, and the route map If not, the route
exits with permit map continues
with the next
phrase
Route Maps
Phrase Match Route map result

type result
Permit Permit Set statements within the phrase are executed and
the route map exits with permit
Deny Permit Set statements within the phrase are not executed
and the route map exits with deny
Permit/ Deny Set statements within the phrase are not executed
Deny and route map continues with the next phrase
If there is no next phrase, route map exits with deny

Route Maps
• PBR proceeds through the route map until a match is
found. If no match is found in the route map, the packet
will be forwarded according to normal destination-based
routing
• If the route-map statement is marked as a deny, the
packets meeting the match criteria are forwarded
according to normal destination-based routing
• If the statement is marked as permit and the
packets do not meet the match criteria, the packets
are forwarded according to normal destination-
based routing
• If the route-map statement is marked as permit and the
packets meet the match criteria, the set clauses are
applied and policy routing is performed
Route Maps
Match Description
metric Metric of the route
In BGP’s case, this is the MED
Must match exactly!
route-type OSPF or EIGRP route type
Internal, External OSPF external type 1 or 2
tag Route tag
ip address Standard or extended access list
Applied against the prefix
Numbered or named
ip address prefix-list Prefix list
Applied against the prefix and prefix length
ip next-hop Standard or extended access list
Applied against the next hop (via in the routing table)
Numbered or named
length Packet length
ip route-source Standard or extended access list
Applied against the neighbor this route was learned from
(from in the routing table)
Numbered or named
Route Maps
• Not all set statements work with all protocols or

in all situations
• It is recommended to test what you want to do before you try to
use it
Set Description
ip next-hop Set the next hop in the routing table or transmitted route
ip next-hop recursive Set the next hop to a subnet which is not directly connected
ip next-hop verify Set the next hop and verify availability using tracking
interface Set the output interface
metric Set the metric of the redistributed or transmitted route
metric-type Set the type of external route
External type 1 or type 2 for OSPF
tag Sets the route tag
Route Map Logic (AND)
route-map networkers permit 10

match ip address 10 Must match BOTH
match tag 1000 ‘match’ conditions to
set ip next-hop 10.1.1.1 successfully match
route-map phrase
Logical AND
Some types of matches cannot co-exist in the same route

map phrase, such as an access list and a prefix list
Route Map Logic (OR)

match ip address 10 20 Can match any ACL in
set ip next-hop 10.1.1.1 the ACL list under the
single ‘match’ statement
Logical OR
Route Map Logic (NOT)

match ip address 10 If we match either ACL in
route-map networkers permit 20 phrase 10 or phrase 20
match ip address 20 then we exit the route-map
route-map networkers permit 30 and don’t fall to catch-all
set ip next-hop 10.1.1.1 phrase 30.
Therefore we must NOT

match either ACL 10 nor
20 in order for the ‘set’ to
apply
Logical NOT
Route Maps
Route map Logic Notes

route-map networkers permit 10 Both matches must succeed for the set
match ip address 10 to be executed
match tag 1000 AND
set ip next-hop 10.1.1.1 Some types of matches cannot co-exist
in the same route map phrase, such as
an access list and a prefix list
route-map networkers permit 10 If the route matches either access list 10
match ip address 10 20 OR or 20, the set will be executed
set ip next-hop 10.1.1.1
route-map networkers permit 10 The route must not match access list 10
match ip address 10 or 20 for the set to execute
route-map networkers permit 20 NOT If the access lists deny routes, then the
match ip address 20 routes must not exist for a specific action
route-map networkers permit 30 to be taken (useful in conditional
set ip next-hop 10.1.1.1 advertisement)
Route Maps
• Route map AND and OR rules
If two different types of matches are configured in the same phrase, they must both
succeed for the set to be executed and the route map to exit (logical AND)
If two of the same type of match are configured (where permitted) in the same
phrase, either one can succeed for the set to be executed and the route map to exit
(logical OR)
All matches in a single phrase must fail for the route map to fall through to the next
phrase (logical NOT)
• Route maps can become very complicated based on these parsing

rules
Route Maps
Gathering Policy with Continue
If this match
succeeds
Execute this set

match ip address 10
set ip next-hop 10.1.1.1 Continue to 30
continue 30
route-map networkers permit 20 Execute this set If this match
match ip address 20
set ip next-hop 10.1.2.1 succeeds
continue 30
route-map networkers permit 30 Execute this set
set ip precedence critical
set ip df 0
Continue to 30
Execute this set

Route Maps
• In normal processing, if all matches fail, the route map falls through to
the next phrase
• Route map continue allows you to continue to another phrase if the
matches succeed
• Sets are executed before the continue is followed
• Use for:
Gathering policy (matches and sets) into a single phrase
More complex logical constructions
• Allows packets to be filtered through route maps containing policies
that selectively determine the next hop to which packets are to be
forwarded
• Policy routes can be determined based on such things as the source
of the packet, protocol types, port numbers, and the size of the packet
• Must be applied on the interface on which the packet is received. ―ip
policy route-map <name>‖ in interface configuration mode
1.1.1.0/24
Router_B#show ip route
....
D 10.1.3.0/24 via <E>
1.1.2.0/24
Router_B#show run A
interface Ethernet1/0 E1/0

ip policy route-map networkers
B .1 10.2.2.0/24
access-list 10 permit 1.1.1.0 0.0.0.255 .1 .2
access-list 20 permit 1.1.2.0 0.0.0.255
10.1.1.0/24
D
route-map networkers permit 10 .2
match ip address 10
route-map networkers permit 20 C
match ip address 20
set ip next-hop 10.2.2.2 E
10.1.3.0/24
• PBR applied to an interface only affects traffic that comes in that
interface
• Can configure a PBR policy local to the router
Router#show run
ip local policy route-map networkers
access-list 101 permit ip any 1.1.1.0 0.0.0.255

match ip address 101
• Local PBR policy only affects traffic generated from the router itself
• Can PBR be Dynamic?
You can use Object Tracking with PBR to track the availability of the next-hop
track 1 ip sla 1 reachability •Tracking object tied to IP SLA

track 2 ip sla 2 reachability
object
ip sla 1
icmp-echo 10.10.10.2 source-interface Ethernet1/0
ip sla 2 •Route-map ties next-hop to
icmp-echo 10.10.10.3 source-interface Ethernet1/0 tracking object so next-hop is
ip sla schedule 1 life forever start-time now
ip sla schedule 2 life forever start-time now only valid if the tracking object
is UP
match ip address 101
set ip next-hop verify-availability 10.10.10.2 1 track 1
set ip next-hop verify-availability 10.10.10.3 2 track 2
•If both tracking objects are
DOWN, normal routing is used
Sequence number determines priority of next-hops

• Load sharing—Supplemental to dynamic load-sharing capabilities
offered by Cisco IOS, PBR allows traffic to be administratively
distributed among multiple paths based on the traffic characteristics
• Quality of Service (QoS)—Using IP Precedence or type of service
(ToS) values to prioritize differentiated traffic
• Source-sensitive routing—Route traffic originating from different users
through different paths
• Cost—Route traffic across low-bandwidth, low-cost permanent paths
or high-bandwidth, high-cost,
switched paths
• Security—Route certain types of traffic (like http) to
firewall/IPS/content filtering device and allow other traffic to follow
normal routing
Overriding the Routing Table:
Performance Routing
Performance Routing
• Traditional routing based on destination of packet
• Policy-based routing allows routing based on more information about
the packet
Source IP, Protocol, Ports Used, QoS markings, etc
• Performance Routing (PFR) allows for routed path decisions to be

made on path characteristics (like reachability, delay, packet loss,
jitter, Mean opinion Score) so application traffic can be given the
optimum path given it‘s path requirements
www.cisco.com/go/pfr
Performance Routing
 Learn traffic and applications
 Discovers traffic going through network via Netflow
 Measure traffic and application performance

 Tracks characteristics like loss/delay/jitter about paths either
passively (via netflow) or actively (via IP SLA probes)
 Apply policies to the traffic based on measurements

 Allows definitions of policies so certain applications or traffic classes
given required network service
 Reroute traffic
 Dynamically alters path of application traffic if current service not in
line with specified policy to sustain performance
Performance Routing
Enterprise
Information
ISP1
BR Internet
Command
MC Voice
Apps
BR
Information ISP2
 Border Routers collect traffic information and pass the

information to a central router (called a Master Controller)
 Master Controller receives information about flows and
determine if they are within configured policy for traffic class
 If measurements of traffic class is out of policy or less-
optimum, Master controller can send commands to the
borders to re-route traffic www.cisco.com/go/pfr
Load Sharing
Load Sharing
• Assume the same routing process attempts to install two routes for
the same destination in the RIB
• The routing process may allow the second route to be installed based
on its own rules
OSPF IS-IS EIGRP

Route cost Must be equal to Must be equal to Must be less than or
installed route installed route equal to the lowest
cost route times the
variance
Maximum Paths Must be less than or equal to the maximum-paths configured
under the routing process
Load Sharing
Load sharing performed in the CEF (Cisco Express
Forwarding) path
CEF has 2 forms of load sharing
• Per-session (Per-destination)
Eth1 –
Eth2 –
Flow 1 -
Flow 2 -
• Per-packet
Eth1 –
Eth2 –
Load Sharing
Per-Session Load Sharing
• Often referred to as per-destination load sharing, even within Cisco
IOS
• This method is the default behavior and does not require any
additional configuration
• A session is a flow that shares the same source and destination.
Traffic with different source to destination pairs tend to take different
paths
• This method ensures that traffic for a given session arrive in order
• Has the potential for traffic polarization and is more effective as the
number of source to destination pairs increase
Load Sharing
Per-Packet Load Sharing
• To utilize this method, configure ―ipcef load-sharing per-packet‖
in interface configuration mode. Each outgoing interface must
have this command configured
• Uses a round-robin method to determine which path each
packet takes to the destination without consideration of source
to destination sessions
• Ensures traffic is more evenly distributed over
multiple paths
• Packets for a given source to destination session may take
different paths, introducing a greater potential for packets to
arrive out of sequence. Not advisable for all types of traffic
• Method used when process-switching
Load Sharing
router#show ip route 192.168.239.0

Routing entry for 192.168.239.0/24
Known via "eigrp 100", distance 170, metric 3072256, type external
Last update from 192.168.245.11 on Serial3/1, 00:18:17 ago
* 192.168.246.10, from 192.168.246.10, 00:18:17 ago, via Serial3/0
Route metric is 3072256, traffic share count is 1
....
192.168.245.11, from 192.168.245.11, 00:18:17 ago, via Serial3/1
....
The traffic share count is critical to

understanding the actual load sharing of
packets using these two routes
How is this calculated?
Load Sharing

....
....
The metric of each route is divided into

the highest metric among the available
metrics
3072256/3072256 == 1
The resulting number is the traffic share count

Load Sharing

....
....
If one metric is less than another metric, the
traffic share count will be something other
than 1 (only for EIGRP and requires variance
to be configured)
3072256/3072256 == 1
3072256/1536128 == 2
The resulting number is the traffic share count
Load Sharing

....
....
When process switching, traffic share count packets is

sent down one path, and then the process moves to the
next available path
The route with the * beside it is the current in use path for
process-switching
Load Sharing Router#sh ip cef 1.1.1.3 internal
[snip]
1.1.1.3/32, epoch 0, RIB[I], refcount 5, per-destination sharing
Ethernet1/0(7): 10.3.3.1, 10.3.3.2
 CEF uses 16 hash buckets nexthop 10.3.3.1 Ethernet1/0, adjacency IP adj out of Ethernet1/0, addr 10.3.3.1
and assigns hash buckets to nexthop 10.3.3.2 Ethernet1/0, adjacency IP adj out of Ethernet1/0, addr 10.3.3.2
each next-hop flags: Per-session, for-rx-IPv4
Router#sh ip route 1.1.1.3 16 hash buckets
Routing entry for 1.1.1.3/32 < 0 > IP adj out of Ethernet1/0, addr 10.3.3.1 044C4608
< 1 > IP adj out of Ethernet1/0, addr 10.3.3.2 044C44E8
Each
Known via "ospf 10", distance 110, metric 20, type extern 2,
forward metric 30 < 2 > IP adj out of Ethernet1/0, addr 10.3.3.1 044C4608 next-hop
Last update from 10.3.3.2 on Ethernet1/0, 00:01:04 ago < 3 > IP adj out of Ethernet1/0, addr 10.3.3.2 044C44E8 has 8 hash
Routing Descriptor Blocks: < 4 > IP adj out of Ethernet1/0, addr 10.3.3.1 044C4608 buckets
10.3.3.2, from 70.70.70.70, 00:01:04 ago, via Ethernet1/0 < 5 > IP adj out of Ethernet1/0, addr 10.3.3.2 044C44E8
Route metric is 20, traffic share count is 1 < 6 > IP adj out of Ethernet1/0, addr 10.3.3.1 044C4608
* 10.3.3.1, from 70.70.70.70, 00:01:24 ago, via Ethernet1/0 < 7 > IP adj out of Ethernet1/0, addr 10.3.3.2 044C44E8
The result is
Route metric is 20, traffic share count is 1 < 8 > IP adj out of Ethernet1/0, addr 10.3.3.1 044C4608 a 50/50
< 9 > IP adj out of Ethernet1/0, addr 10.3.3.2 044C44E8 chance of
Each packet that comes in gets
<10 > IP adj out of Ethernet1/0, addr 10.3.3.1 044C4608 getting each
<11 > IP adj out of Ethernet1/0, addr 10.3.3.2 044C44E8
measured against the HASH, and the next-hop
<12 > IP adj out of Ethernet1/0, addr 10.3.3.1 044C4608
HASH result determines which hash <13 > IP adj out of Ethernet1/0, addr 10.3.3.2 044C44E8
bucket the packet uses <14 > IP adj out of Ethernet1/0, addr 10.3.3.1 044C4608 1:1
<15 > IP adj out of Ethernet1/0, addr 10.3.3.2 044C44E8 load-sharing
Load Sharing
Hash
Hash Inputs
1
2
3 via Serial3/0
4
5 via Serial3/1
Hash Result 6
.
 Per-destination Load-sharing takes a set of inputs, runs those

inputs into the hash algorithm, and the result of the algorithm
determines which load-sharing Hash bucket that packet will use
 Per-destination load-sharing algorithm used will determine which
inputs are put into the hash
Load Sharing
Source 10.1.1.1
Destination 192.168.239.1 Hash
1
2
3 via Serial3/0
4
5 via Serial3/1
6
CEF hashes the source =5
.
and destination Hash Result
addresses, and chooses
a bucket from the load The load share table
share table points to an adjacency
corresponding to one of
the next hops in the
routing table
Load Sharing
• How do I tell which next-hop a particular packet will take?
router#show ip cef exact-route 10.1.1.1 192.168.239.1

10.1.1.1 -> 192.168.239.1 : Serial3/0 (next hop 192.168.246.10)
•‗exact-route‘ command in CEF takes hash inputs

(source/destination IP) and puts them through the hash to
result the egress interface
•Useful in tracing path of packet during troubleshooting
Load Sharing
Polarization
 If the same input into the hash algorithm produces the same result,
then what if there are many routers using the same algorithm?
Packet 1 = src 1.1.1.1 dst 2.2.2.2 If the hash for packet 1

Packet 2 = src 1.1.1.1 dst 3.3.3.3 always results in path 1
And if the hash for packet 2

1 D always results in path 2
1 B
2 E
Then all routers will make the same
A
path decision and as a result the
2 1 links between B=>E and C=>F will
F
C never be used!
2
G
Load Sharing
Polarization
 We can fix this if we change the inputs on each router by looking at
something else besides just the src/dst IP
-But this extra input would need to be unique per router, otherwise
every router will pick the same path again
1 D Universal Algorithm
1 B Each router adds a unique

2 E
random number to the hash
A algorithm resulting in the
2 1 possibility that the hash result on
F
C each hop may be different
2
G
Load Sharing
Algorithms
• Original algorithm—Src/Dst IP
-could potentially produce distortions in load sharing across multiple routers because
the same algorithm was used on every router (polarization)
• Universal algorithm—Src/Dst IP and Universal ID

-allows each router on the network to make a different load sharing decision for each
source-destination address pair (resolves polarization)
-default algorithm in IOS.
• Tunnel algorithm—
-designed to balance the per-packet load when only a few source and destination
pairs are involved.
• Include-ports algorithm— Src/Dst IP Src/Dst Port and Universal ID

-allows you to use the Layer 4 source and destination ports as part of the load-
balancing decision.
Load Sharing
Performance Routing
Enterprise 20%
Routing Updates
ISP1
GW1
Internet
CORE Email
Voice
Routing Updates GW2
ISP2 Apps
60%
 In addition to load sharing traffic based on application

policies, PfR can also load share or load balance traffic based
on link utilization.
Load Sharing
Performance Routing
Enterprise 39%
Information
ISP1
BR
Command
MC
Internet
Email
Voice
BR
Information ISP2 Apps
41%
 CE2 link was 60% utilized and CE1 only 20%. PfR can identify
this and move traffic to better balance out the egress link
utilizations.
 Done by configuring a policy to keep link utilization within a % of
each other, so one link isn’t utilized more than the other
Routing Segmentation and Separation
• When would you want to separate routing operations?
Prevent any potential exchange of data or routing information.
• Why not use ACLs or other security features?

Helpful but limited
• A VRF can help prevent the exchange of routes as well as data and
does not have to be constantly updated.
What is a VRF?
• A VPN Routing and Forwarding (VRF) is an IOS routing instance.
All tables (routing/cef) maintained in routing instance (vrf)
All protocols/features run independently in each VRF instance
Allows for logical separation at Layer-3
• Originally designed for MPLS VPN so multiple MPLS customers can

use overlapping IP space and be logically separated from each other
This presentation will be referring to VRF outside of an MPLS VPN
context. Also known as VRF-Lite.
Router_A#show run vrf blue Router_A#show run vrf red
ip vrf blue ip vrf red

! !
! !
interface Ethernet0/1 interface Ethernet0/2
ip vrf forwarding blue ip vrf forwarding red
ip address 172.16.12.1 255.255.255.0 ip address 172.16.12.1 255.255.255.0
! !
interface Loopback1 interface Loopback2
ip vrf forwarding blue ip vrf forwarding red
ip address 1.1.1.1 255.255.255.255 ip address 2.2.2.2 255.255.255.255
! !
router eigrp 1 router eigrp 1
! Router_A
!
address-family ipv4 vrf blue address-family ipv4 vrf red
network 172.16.12.0 0.0.0.255 network 172.16.12.0 0.0.0.255
no auto-summary no auto-summary
autonomous-system 1 autonomous-system 1
exit-address-family exit-address-family
! !
How to configure and identify a VRF
Router#show ip vrf
Name Default RD Interfaces
blue <not set> Et0/1
Lo1
red <not set> Et0/2
Lo2
Router#show ip vrf int

Interface IP-Address VRF Protocol
Et0/1 172.16.12.1 blue up
Lo1 1.1.1.1 blue up
Et0/2 172.16.12.1 red up
Lo2 2.2.2.2 red up
Router#show ip route vrf blue
Routing Table: blue

C 1.1.1.1 is directly connected, Loopback1
• Routing principles are the same within a VRF as in the global routing
table.
• All routes and prefixes are unique to a given VRF unless route leaking
is configured.
• Features that affect forwarding (like Routing Protocols/static
routes/NAT/PBR) need to be configured on a per-VRF basis
• Interface-level features that affect traffic (like ACLs/QoS/uRPF) do not
need to be configured to be VRF aware because they inherit the VRF
of the interface on which they are configured
Routing and Router Resources:
CPU
CPU
• Central Processing Unit responsible for carrying out
instructions.
• IOS uses a priority run-to-completion model for executing
processes.
• The task scheduler is responsible for scheduling and executing
kernel processes on the CPU
• Process Priorities:
Critical
High
Medium
Low
• There is no preemption but higher priority processes have
more opportunity to access the CPU.
CPU
• What uses CPU resources?

Router processes
Packet switching
IP Input – Process switched packets
Interrupts – CEF switched packets
CPU utilization for five seconds: 5%/2%; one minute: 3%; five minutes: 2%
PID Runtime (ms) Invoked uSecs 5Sec 1Min 5Min TTY Process
[snip]
2 68 585 116 1.00% 1.00% 0% 0 IP Input
17 88 4232 20 0.20% 1.00% 0% 0 BGP Router
18 152 14650 10 0% 0% 0% 0 BGP Scanner
CPU
How does routing information affect CPU resources?

• BGP scanner walks the BGP table and confirms reachability of
the next hops as well as checks conditional-advertisement to
determine whether or not BGP should advertise conditional
prefixes.
• The larger the BGP table, the longer BGP scanner takes to run
which in turn may force lower priority processes to wait.
CPU
How does routing information affect CPU resources?
• BGP router calculates the best BGP path and processes any route
"churn". It also sends and receives routes, establishes peers, and
interacts with the routing information base (RIB).
• Does a large amount of work during initial convergence where large
amount of prefixes are exchanged.
• The larger the tables that are being exchanged, the more time BGP
router will have to use the CPU.
CPU
Monitoring CPU usage
• Show commands:
show process cpu sorted
show process cpu history
• CPU threshold monitoring/logging

process cpu threshold type {total | process | interrupt} rising percentage
interval seconds [falling percentage interval seconds]
18:41:20.934: %SYS-1-CPURISINGTHRESHOLD: Threshold: Total CPU

Utilization(Total/Intr): 72%/0%, Top 3 processes(Pid/Util): 79/64%, 140/6%, 75/1%
CPU
• Reporting CPU events via EEM

event manager applet highcpu
event snmp oid 1.3.6.1.4.1.9.9.109.1.1.1.1.3.1 get-type exact entry-op ge entry-val 90 poll-interval 10
action 1.0 cli command "enable"
action 2.0 cli command "show proc cpu sorted | redirect flash:highcpu.txt―
action 3.0 syslog msg ―H igh CPU DETECTED "show process cpu sort" written to > flash:highcpu.txt ―
action 4.0 mail server "<mail_server_ip>" to "<email_address>" from "<sender>" subject ―C PU
exceeded 90%."
 Event manager applet monitoring SNMP OID for CPU

utilization percentage every 10 seconds
 Actual OID will depend on platform/code
 If CPU OID value exceeds value 90 (90%), script will trigger

the specified actions
www.cisco.com/web/go/eem
Routing and Router Resources:
Memory
Memory
• Managed in 2 pools: Processor and I/O
• The processor memory pool is the general memory pool common to
all IOS systems including storage for routing information.
• The I/O pool or packet memory manages memory for interface packet
buffers.
Memory
How does routing information affect memory resources?
• Most common example of where we see this is the storing of BGP
prefixes.
• BGP generally carries the largest number of prefixes as well as the
potential to store multiple tables.
Memory
• How much memory do I need to store my routing information?
Full BGP table
Multiple feeds
Route filtering
Soft reconfiguration inbound
Default route
Memory Usage Example
• BGP Profile 1
Baseline memory usage with no BGP peers
• BGP Profile 2
1 BGP peer sending 300,000 routes
• BGP Profile 3
2 BGP peers both sending the exact same 300,000 prefixes
Unique AS path and next-hop IP information from each peer
• BGP Profile 4
2 BGP peers both sending 300,000 unique prefixes with zero overlap
Unique AS path and next-hop IP information from each peer
Memory Usage Example
BGP Routes vs. Memory Consumption
1000000000
900000000
800000000
Total Memory (Bytes)
700000000
600000000
Free
500000000
Used
400000000
300000000
600000
200000000
300000 600000 Routes
No Routes
100000000 Routes Routes
0
1 2 3 4
BGP Profile #Peers #Routes Memory
1 0 0 27.5MB
The amount of memory used BGP Profile
to store prefixes also depends 2 1 300,000 221.2MB
on the amount of overlap 3 2 600,000 245.1MB
between peers.
4 2 600,000 416.1MB
Memory
Monitoring memory availability
• Show commands
show process memory sorted
show memory statistics history
• Memory Threshold Notifications

memory free low-watermark {processor threshold | io threshold}
memory reserve critical kilobytes
22:31:19.559: %SYS-4-FREEMEMLOW: Free Memory has dropped below 2000k

Pool: Processor Free: 66814056 freemem_lwm: 204800000
Q&A
Recommended Reading
• Continue your Cisco Live learning

experience with further reading from
Cisco Press
• Check the Recommended Reading
flyer for suggested books
Available Onsite at the Cisco Company Store

Complete Your Online
Session Evaluation
• Receive 25 Cisco Preferred
Access points for each session
evaluation you complete.
• Give us your feedback and you
could win fabulous prizes. Points
are calculated on a daily
basis. Winners will be notified by
email after July 22nd.
• Complete your session evaluation
online now (open a browser
through our wireless network to
access our portal) or visit one of
the Internet stations throughout Don’t forget to activate your Cisco
Live and Networkers Virtual account
the Convention Center.
for access to all session materials,
communities, and on-demand and live
activities throughout the year. Activate
your account at any internet station or
visit www.ciscolivevirtual.com.
Thank you.

Routing Operation

Uploaded by

Copyright:

Available Formats

Routing Operation

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Routing Operation

Uploaded by

Copyright:

Available Formats

Routing Operations in Cisco

• The Routing Table (RIB)

• Routing Protocols Routing Protocols

• Cisco Express Forwarding (CEF) Other Sources

Codes: L - local, C - connected, S - static, R - RIP, M - mobile, B - BGP

10.0.0.0/8 is variably subnetted, 2 subnets, 2 masks

10.0.0.0/8 is variably subnetted, 2 subnets, 2 masks

10.0.0.0/8 is variably subnetted, 2 subnets, 2 masks

10.0.0.0/8 is variably subnetted, 2 subnets, 2 masks

Major networks with subnets show up under a single network with

Native mask routes and supernets show up as separate networks

Administrative distance Metric Last Modification Time

D EX 192.168.254.0/24 [170/3072256] via 208.0.246.10, 00:58:45, Serial3/0

• The administrative distance is locally significant

Enhanced Interior Gateway Routing Protocol (EIGRP)

External Border Gateway Protocol (BGP) 20

Intermediate System-to-Intermediate System (IS-IS) 115

Routing Information Protocol (RIP) 120

Exterior Gateway Protocol (EGP) 140

On Demand Routing (ODR) 160

External EIGRP 170

Internal BGP 200

Unknown (infinity) 255

(b) OSPF 2 (e) The RIB

The RIB receives the EIGRP reply and flushes the

RT: closer admin distance for 192.168.254.0, flushing 1 routes

EIGRP-IPv4(1):Callback: lostroute 192.168.254.0/24

RT: add 192.168.254.0/24 via 208.0.245.11, ospf metric [110/65]

EIGRP receives a callback stating the RIB has

The RIB installs OSPF’s route

RIPv2 EIGRP EIGRP

10.1.1.0/24 10.1.1.0/24 10.1.1.0/25

EIGRP internal route is installed

The lowest administrative distance wins.

10.1.1.0/24 10.1.1.0/24 10.1.1.0/25

These two routes

It depends. Usually the route with the default

OSPF 1 OSPF 2 IS-IS

10.1.1.0/24 10.1.1.0/24 10.1.1.0/24

Default AD is the same between the

The older route remains in the RIB

10.1.1.0/24 10.1.1.0/24 10.1.1.0/24

EIGRP 2 route metric is

EIGRP 2 route is installed in the RIB

Tie breaker is the lowest AS number

10.1.1.0/24 10.1.1.0/24 10.1.1.0/24

Both routes have the same When multiple paths exist

10.0.0.0/8 is variably subnetted, 2 subnets, 2 masks

RIPv2 EIGRP EIGRP

10.1.1.0/24 10.1.1.0/24 10.1.1.0/25

These two routes

RIB calls backups

router-b#show ip eigrp topo EIGRP has the same route

• For each packet destined to 10.1.1.0/24, the Layer 2 rewrite

• Control static routes to broadcast interfaces

• Static routes to point-to-point interfaces don‘t

router(config)#ip route 10.1.0.0 255.255.0.0 fa 0/1

• Static routes to interfaces will be included if you configure redistribute

ip route 10.1.1.0 255.255.255.0 <B> 250

track 1 ip sla 1 reachability