ENARSI Chapter 16
ENARSI Chapter 16
Instructor Materials
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 2
Redistribution Overview
• An organization might use multiple routing protocols, split up the routing
domain between multiple instances (processes) of the same routing protocol,
or need to merge networks with another organization that uses a different
routing protocol.
• The routes from one routing protocol process need to be exchanged with a
different routing protocol process to provide full connectivity.
• Redistribution is used to inject routes from one routing protocol into another
routing protocol.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 3
Redistribution Overview
Multiple Routing Protocol Topology
Figure 16-1 illustrates a network that has
multiple routing protocols that are not working
together.
• R1, R2, and R3 exchange routes using
Enhanced Interior Gateway Routing Protocol
(EIGRP), and R3, R4, and R5 exchange
routes with Open Shortest Path First (OSPF).
• R1 and R5 advertise their Loopback 0
interfaces (192.168.1.1/32 and
192.168.5.5/32) into their appropriate routing
protocol, but they cannot establish
connectivity to each other.
• Only R3 can connect to R1 and R5 because
it is the only router that participates with both
routing protocols and has a complete view of
the network.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 4
Redistribution Overview
Mutual Redistribution Topology
The routes are not automatically
redistributed between the routing protocols.
• Redistribution must be configured so
that EIGRP routes are injected into
OSPF and OSPF routes are injected
into EIGRP.
In Figure 16-2, the R3 router is performing
mutual redistribution.
The OSPF routes are present in EIGRP as
external routes, and the EIGRP routes are
present in the OSPF routing domain as
external routes (Type 5 link-state
advertisements [LSAs]).
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 5
Redistribution Overview
Redistribution Source Protocols
Redistribution always encompasses two
routing protocols:
• A source protocol and a destination
protocol.
• The source protocol provides the network
prefixes that are to be redistributed, and
the destination protocol receives the
injected network prefixes.
• The redistribution configuration exists
under the destination protocol and
identifies the source protocol.
• Using a route map allows for the filtering
or modification of route attributes during
the injection into the destination protocol.
Table 16-2 provides a list of source protocols
for redistribution.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 6
Redistribution Overview
Redistribution Is Not Transitive
When redistributing between two or more routing
protocols on a single router, redistribution is not
transitive.
• When a router redistributes protocol 1 into
protocol 2, and protocol 2 redistributes into
protocol 3, the routes from protocol 1 are not
redistributed into protocol 3.
• Example 16-1 provides sample logic for
EIGRP mutually redistributing into OSPF and
OSPF mutually redistributing into BGP.
Figure 16-3 illustrates redistribution on the router.
• The EIGRP route 172.16.1.0/24 redistributes
into OSPF but does not redistribute into BGP.
• The BGP route 172.16.3.0/24 redistributes
into OSPF but does not redistribute into
EIGRP.
• The prefix 172.16.2.0/24 redistributes to both
EIGRP and BGP.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 7
Redistribution Overview
Redistribution Is Not Transitive (Cont.)
For routes to be exchanged between all three
routing protocols, mutual redistribution must
be configured between all three protocols, as
shown in Example 16-2.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 8
Redistribution Overview
Sequential Protocol Redistribution
Sequential protocol redistribution is redistribution between multiple protocols over a series of
routers, as shown in Figure 16-5.
• R2 redistributes the EIGRP 192.168.1.1/32 prefix into BGP, and R4 redistributes the BGP
192.168.1.1/32 prefix into OSPF.
• All three routing protocols contain the 192.168.1.1/32 prefix.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 9
Redistribution Overview
Routes Must Exist in the RIB
A route must exist in the RIB in order for it to be
redistributed into the destination protocol. This
ensures that the route is deemed reachable by
the redistributing router.
• The source protocol that redistributes into the
destination protocol must be the source for
the route in the RIB. This ensures that the
router redistributes only the route it deems
the best for the destination protocol.
• The only exception for this logic is for directly
connected interfaces participating in the
source protocol because they have an
administrative distance (AD) of 0.
In Figure 16-6, R1 and R3 both advertise the
10.13.1.0/24 network, and R5 is redistributing
RIP and OSPF routes into EIGRP.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 10
Redistribution Overview
Routes Must Exist in the RIB (Cont.)
R5 receives route information for 10.13.1.0/24 from both
RIP and OSPF routing protocols.
• Example 16-3 provides verification that R5 contains an
entry for the 10.13.1.0/24 network in RIP’s database
and OSPF’s LSDB.
• R5 determines that the most desirable path to reach
10.13.1.0/24 is the OSPF route because it has a lower
AD than RIP. Example 16-4 displays R5’s routing table.
• The 10.13.1.0/24 OSPF route is inserted into the RIB.
• Example 16-5 shows the EIGRP topology table for the
10.13.1.0/24 network.
• R5 checks the Routing Information Base (RIB) for the
10.13.1.0/24 network and verifies its existence in the
RIB and confirms that the source protocol is the
protocol that installed the route.
• EIGRP identifies the source protocol as OSPF, with a
path metric of 3.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 11
Redistribution Overview
Seed Metrics
Every routing protocol uses a different methodology for calculating the best path for a
route.
• For example, EIGRP can use bandwidth, delay, load, and reliability, whereas OSPF primarily uses
the path metric for calculating the shortest path first (SPF) tree (SPT). OSPF cannot calculate the
SPT using EIGRP path attributes, and EIGRP cannot run diffusing update algorithm (DUAL) using
only the total path metric.
• The destination protocol must provide relevant metrics to the destination protocols so that the
destination protocol can calculate the best path for the redistributed routes.
• Every protocol provides a seed metric at the time of redistribution that allows the destination
protocol to calculate a best path. The seed metric is a baseline and may reflect a loss of
information during redistribution when redistribution occurs between two different protocol types.
• A route map modifies the seed metric for a route during redistribution. Table 16-3 provides a list of
seed metrics for the destination routing protocol.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 12
Protocol-Specific Configuration
Every routing protocol has a unique redistribution behavior.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 13
Protocol-Specific Configuration
Route Map Match Command
Every routing protocol has a unique redistribution behavior.
• IOS and IOS XE routers use the following command syntax in the destination protocol to
identify the source routing protocol:
redistribute {connected | static | eigrp as-number | ospf process-id [match
{internal | external [1|2]}] | bgp as-number} [destination-protocoloptions]
[route-map route-map-name].
• Redistribution commonly uses route maps to manipulate or filter routes on the redistributing
router. Table 16-4 lists additional conditional matching commands for route selection during
redistribution.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 14
Protocol-Specific Configuration
Route Map Match Set Command
Table 16-5 lists the route map set actions that modify the route as it is redistributed
into the destination protocol.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 15
Protocol-Specific Configuration
Source Specific Behaviors: Connected Networks
Connected networks are networks associated with primary and secondary IP
addresses for any up interfaces that are not participating with the destination protocol.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 16
Protocol-Specific Configuration
Source Specific Behaviors: BGP
By default, BGP redistributes only eBGP routes into IGP protocols.
• In Figure 16-7, R3 advertises the 192.168.3.3/32 network, and R4 advertises the
192.168.4.4/32 network into BGP.
• R2 is redistributing BGP into OSPF, but only the 192.168.4.4/32 address is redistributed
because it is an eBGP route. The iBGP route from R3 was not included because of BGP loop-
prevention rules. It is assumed that the IGP routing topology already has a path to reach the
network.
• BGP’s default behavior requires that a route have an AS_Path to redistribute into an IGP
routing protocol.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 17
Protocol-Specific Configuration
Source Specific Behaviors: BGP (Cont.)
BGP behavior can be changed so that all BGP routes are redistributed with the BGP
configuration command bgp redistribute-internal. To enable the iBGP route 192.168.3.3/32
to redistribute into OSPF, the bgp redistribute-internal command is required on R2.
BGP is designed to handle a large routing table, whereas IGPs are not. Redistributing
BGP into an IGP on a router with a larger BGP table (for example, the Internet table with
800,000+ routes) should use selective route redistribution. Otherwise, the IGP can become
unstable in the routing domain, which can lead to packet loss.
Redistributing iBGP routes into an IGP could result in routing loops. A more logical
solution is to advertise the network into the IGP.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 18
Protocol-Specific Configuration
Destination Specific Behaviors: EIGRP
External EIGRP routes are given an AD of 170 and use a default seed metric of infinity,
which prevents the installation of the routes into the EIGRP topology table.
• If an EIGRP autonomous system redistributes into another EIGRP autonomous system, all the
path metrics are included during redistribution.
• The default path metric can be changed from infinity to specific values for bandwidth, load,
delay, reliability, and maximum transmission unit (MTU), thereby allowing for the installation
into the EIGRP topology table.
• Routers can set the default metric with the address family configuration command default-
metric bandwidth delay reliability load mtu.
• Delay is entered in tens of microseconds (μs).
The metric can also be set within a route map or at the time of redistribution with this command:
redistribute source-protocol [metric bandwidth delay reliability
load mtu] [route-map route-map-name]
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 19
Protocol-Specific Configuration
Destination Specific Behaviors: EIGRP Topology
Figure 16-8 provides a topology example in
which R2 mutually redistributes OSPF into
EIGRP and R3 mutually redistributes BGP into
EIGRP.
• R1 is advertising the Loopback 0 address
192.168.1.1/32, and R4 is advertising the
Loopback 0 address 192.168.4.4/32.
• Example 16-7 shows the relevant EIGRP
configuration. R2 uses the default-metric
configuration command and is displayed with
classic and named mode configurations.
• Notice where the default metrics are placed
with EIGRP named mode configuration. R3’s
configuration specifies the seed metrics with
the redistribution command.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 20
Protocol-Specific Configuration
Destination Specific Behaviors: EIGRP Metrics
You can overwrite EIGRP seed metrics by setting K
values with the route map command set metric
bandwidth delay reliability load mtu.
• Example 16-8 shows the configuration without
the use of the default-metric command but by
setting the EIGRP metric using a route map.
• Example 16-9 shows the EIGRP topology table
with the locally redistributed routes highlighted.
• The redistributed routes are shown in the routing
table with D EX and an AD of 170, as shown in
Example 16-10.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 21
Protocol-Specific Configuration
EIGRP-to-EIGRP Redistribution
Redistributing routes between EIGRP
autonomous systems preserves the path
metrics during redistribution.
• Figure 16-9 shows a topology that
includes multiple EIGRP autonomous
systems.
• R2 mutually redistributes routes
between AS 10 and AS 20, and R3
mutually redistributes routes between
AS 20 and AS 30.
• R1 advertises the Loopback 0 interface
(192.168.1.1/32) into EIGRP AS 10.
• R4 advertises the Loopback 0 interface
(192.168.4.4/32) into EIGRP AS 30.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 22
Protocol-Specific Configuration
EIGRP-to-EIGRP Redistribution Configuration
Example 16-11 shows the configurations
for R2 and R3.
• The default seed metrics do not need
to be set because they are maintained
between EIGRP ASs.
• R2 is using classic configuration
mode, and R3 is using EIGRP named
configuration mode.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 23
Protocol-Specific Configuration
EIGRP-to-EIGRP Redistribution Verification
Example 16-12 provides verification that
R1 has routes learned from AS 20 and
AS 30, and R4 has learned routes from
AS 10 and AS 20.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 24
Protocol-Specific Configuration
EIGRP-to-EIGRP Redistribution Verification (Cont.)
Example 16-13 shows the EIGRP
topology table for the route
192.168.4.4/32 in AS 10 and AS 20.
• The EIGRP path metrics for
bandwidth, reliability, load, and delay
are the same between the
autonomous systems.
• Notice that the feasible distance
(131,072) is the same for both
autonomous systems, but the
reported distance (RD) is 0 for AS 10
and 130,816 for AS 20.
• The RD was reset when it was
redistributed into AS 10.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 25
Protocol-Specific Configuration
Destination Specific Behaviors: OSPF
The AD is set to 110 for intra-area, interarea, and external OSPF routes. External OSPF
routes are classified as Type 1 or Type 2, with Type 2 as the default setting.
• The seed metric is 1 for BGP-sourced routes and 20 for all other protocols.
• The exception is that if OSPF redistributes from another OSPF process, the path metric is
transferred.
The main differences between Type 1 and Type 2 External OSPF routes follow:
• Type 1 routes are preferred over Type 2.
• The Type 1 metric equals the redistribution metric plus the total path metric to the
autonomous system boundary router (ASBR). In other words, as the LSA propagates
away from the originating ASBR, the metric increases.
• The Type 2 metric equals only the redistribution metric. The metric is the same for the
router next to the ASBR as the router 30 hops away from the originating ASBR.
• If two Type 2 paths have exactly the same metric, the lower forwarding cost is preferred.
This is the default external metric type used by OSPF.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 26
Protocol-Specific Configuration
Redistribution into OSPF
For redistribution into OSPF, you use this command:
• If the optional subnets keyword is not included, only the classful networks are redistributed.
• The optional tag keyword allows for a 32-bit route tag to be included on all redistributed routes.
• The metric and metric-type keywords can be set during redistribution.
Figure 16-10 provides a topology example in which R2 mutually redistributes EIGRP into
OSPF and R3 mutually redistributes RIP into OSPF. R1 is advertising the Loopback 0 interface
192.168.1.1/32, and R4 is advertising the Loopback 0 interface 192.168.4.4/32.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 27
Protocol-Specific Configuration
Redistribution into OSPF Configuration
Example 16-14 shows the relevant OSPF configuration. Notice that R2 and R3 use
different OSPF process numbers but are still able to form an adjacency. The OSPF
process numbers are locally significant in linking the OSPF-enabled interfaces to a
process, as shown later in this section.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 28
Protocol-Specific Configuration
Redistribution into OSPF LSDB
Example 16-15 shows the Type 5 LSAs
for the external networks in the OSPF
domain.
• This example shows that the routes
10.12.1.0/24, 10.34.1.0/24,
192.168.1.1/32, and 192.168.4.4/32
successfully redistributed into OSPF.
• The redistributed networks are Type 2
with a metric of 20.
• The redistributed routes appear in the
routing table with O E2 for Type 2 and
O E1 for Type 1 external routes.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 29
Protocol-Specific Configuration
Redistribution into OSPF Verification
In Example 16-16, the routers do
not explicitly set a metric type,
so all the redistributed routes
from the topology are type E2.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 30
Protocol-Specific Configuration
OSPF-to-OSPF Redistribution
Redistributing routes between OSPF processes preserves the path metric during
redistribution, independent of the metric type.
Figure 16-11 shows a topology with multiple OSPF processes and areas.
Although the topology in Figure 16-11 looks discontiguous, OSPF is redistributing the routes
between processes. This technique can be used to advertise routes over discontiguous OSPF
networks, but it results in the loss of path information as the Type 1, Type 2, and Type 3 LSAs
are not propagated through route redistribution.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 31
Protocol-Specific Configuration
OSPF-to-OSPF Redistribution (Cont.)
Example 16-17 shows the relevant
configurations for R2 and R3. Notice
that the metric type is set at the time
of redistribution into the destination
protocol so that you can see the
metric increase as the route travels
between the OSPF processes.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 32
Protocol-Specific Configuration
OSPF Forwarding Address
OSPF Type 5 LSAs include a field known as the forwarding
address that optimizes forwarding traffic when the source
uses a shared network segment.
• The scenario defined in RFC 2328 is not common but is
shown in Figure 16-12.
• Example 16-19 shows the Type 5 LSA for the AS 100
route for 192.168.1.1/32.
• Network traffic from R3 (and R5) takes the suboptimal
route (R3→R5→R4→R2→R1), as shown in Example
16-20. The optimal route would use the directly
connected 10.123.1.0/24 network.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 33
Protocol-Specific Configuration
OSPF Forwarding Nondefault
The OSPF forwarding address changes from 0.0.0.0 to
the next-hop IP address in the source routing protocol
when:
• OSPF is enabled on the ASBR’s interface that points
to the next-hop IP address.
• That interface is not set to passive.
• That interface is a broadcast or nonbroadcast OSPF
network type.
When the forwarding address is set to a value besides
0.0.0.0, the OSPF routers forward traffic only to the
forwarding address.
Example 16-21 provides the Type 5 LSA for the
192.168.1.1/32 network. Now that OSPF has been
enabled on R2’s 10.123.1.2 interface and the interface
is a broadcast network type, the forwarding address has
changed from 0.0.0.0 to 10.123.1.1.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 34
Protocol-Specific Configuration
OSPF Forwarding Address Verification
Example 16-22 verifies that connectivity from
R3 and R5 now takes the optimal path to R1
because the forwarding address has changed
to 10.123.1.1.
• If the Type 5 LSA forwarding address is not
a default value, the address must be an
intra-area or interarea OSPF route.
• If the route does not exist, the LSA is
ignored and is not installed into the RIB.
Figure 16-14 shows a topology in which R2 mutually redistributes between OSPF and
BGP, and R3 mutually redistributes between EIGRP AS 65100 and BGP.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 36
Protocol-Specific Configuration
Redistribution into BGP Configuration
Example 16-23 shows R2’s BGP
configuration for redistributing OSPF into
BGP on R2 and R3’s configuration for
redistributing EIGRP into BGP.
• R3 has disabled the default IPv4 address
family configuration.
• Notice that R2 and R3 have used the
command bgp redistribute-internal,
which allows for any iBGP learned
prefixes to be redistributed into OSPF or
EIGRP.
Example 16-24 shows the BGP table for AS
65100. Notice that the 192.168.1.1/32 and
192.168.4.4/32 networks have been installed
into the BGP table. The metric is carried over
from the IGP metric during redistribution.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 37
Protocol-Specific Configuration
Redistribution into BGP Verification
Example 16-25 shows detailed BGP path
information for the redistributed routes. The
origin is incomplete, and the BGP metric
matches the IGP metric at the time of
redistribution.
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 38
Prepare for the Exam
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 39
Prepare for the Exam
Key Topics for Chapter 16
Description Description
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 40
Prepare for the Exam
Key Terms for Chapter 16
Terms
destination protocol
mutual redistribution
source protocol
seed metric
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 41
Prepare for the Exam
Command Reference for Chapter 16
Terms
© 2016 Cisco and/or its affiliates. All rights reserved. Cisco Confidential 42