CCNP Route Lab 4-1 , Redistribution Between RIP and OSPF

CCNP Route Lab 4-1 , Redistribution Between RIP and OSPF

Topology

ccnp-route-lab-redistribution-rip-ospf

Objectives

  • Review configuration and verification of RIP and OSPF.
  • Configure passive interfaces in both RIP and OSPF.
  • Filter routing updates using distribute lists.
  • Redistribute static routes into RIP.
  • Redistribute RIP routes into OSPF.
  • Redistribute OSPF routes into RIP.
  • Originate a default route into OSPF.
  • Set a default seed metric.
  • Modify OSPF external network types.
  • Configure summary addresses.

Background
Two online booksellers, Example.com and Example.net, have merged and now need a short-term solution to inter-domain routing. Since these companies provide client services to Internet users, it is essential to have minimal downtime during the transition.

Example.com is a small firm running RIP, while Example.net has a somewhat larger network running OSPF.The diagram identifies R2 as the router that will bridge the two networks. Because it is imperative that the two booksellers continuously deliver Internet services, you should bridge these two routing domains without interfering with each router’s path through its own routing domain to the Internet.

The CIO determines that it is preferable to keep the two protocol domains shown in the diagram during the transition period, because the network engineers on each side need to understand the other’s network before deploying a long-term solution. Redistribution will be a short-term solution.

In this scenario, R1 and R2 are running RIPv2, but the 172.16.23.0/24 network between R2 and R3 is running OSPF. You need to configure R2 to enable these two routing protocols to interact to allow full connectivity between all networks.

Note: This lab uses Cisco 1841 routers with Cisco IOS Release 12.4(24)T1 and the Advanced IP Services image c1841 -advipservicesk9-mz.124-24.T1 .bin. You can use other routers (such as 2801 or 2811) and Cisco IOS Software versions if they have comparable capabilities and features. Depending on the router model and Cisco IOS Software version, the commands available and output produced might vary from what is shown in this lab.

Required Resources

  • 3 routers (Cisco 1841 with Cisco IOS Release 12.4(24)T1 Advanced IP Services or comparable)
  • Serial and console cables

Step 1: Configure loopbacks and assign addresses.
a. Configure all loopback interfaces on the three routers in the diagram. Configure the serial interfaces with the IP addresses, bring them up, and set a DCE clock rate where appropriate.

b. (Optional) On each router, create an enable secret password. Configure the console line for synchronous logging and no timeout. Configure the vty lines to allow Telnet to and remote configuration of network devices.

c. Verify that you can ping across the serial links when you are finished. Use the following Tcl script to check full and partial connectivity throughout this lab.

At this point, the only pings that you should receive back are those of the connected networks of the router from which you are pinging.

Step 2: Configure RIPv2.
Configuring RIPv2 on a router is fairly simple:

  • Type the global configuration command router rip to enter RIP configuration mode.
  • Enable RIPv2 with the version 2 command.
  • Enter the no auto-summary command to disable automatic summarization at classful network boundaries.
  • Add the networks you want using the network network command.

Unlike EIGRP and OSPF, the RIP network command only requires the classful network address to be entered and does not support a wildcard mask. This behavior is inherited from the classful RIPv1 protocol configuration and is kept for backward compatibility with older Cisco IOS versions that would not otherwise be able to process network commands with wildcard masks. Classful protocols do not support subnets; therefore, subnet or wildcard masks are unnecessary.

Based on the topology diagram, which major networks need to be advertised into RIP for R1?
172.16.0.0/16
192.168.48.0/24
192.168.49.0/24
192.168.50.0/24
192.168.51.0/24
192.168.70.0/24

Which major networks need to be advertised into RIP for R2?
Only 172.16.0.0/16

a. Apply the following commands to R1 and R2.

b. Verify that the RIP routes were learned from the other routers using the show ip route rip command on each router.

c. You can also verify which routes are coming in from RIP advertisements with the show ip rip database command.

Step 3: Configure passive interfaces in RIP.
a. On R1, use the show ip route rip command to view the RIP routes in the routing table. Notice that the network for the serial interface of R2 that connects to R3 is present, even though you do not have a RIP neighbor on that interface. This is because the entire class B network 172.16.0.0 /16 was added to RIP on R2.

b. Issue the show ip protocols command to verify that RIPv2 updates are being sent out both serial interfaces.

For security reasons and to reduce unnecessary traffic, RIP updates should not be propagated into the OSPF domain. You can disable sending updates with the passive-interface interface_type interface_number router configuration command

c. On R2, configure the serial interface connecting to R3 as passive. Notice that the interface is no longer listed in the output of the show ip protocols command.

d. On R1, issue the show ip route rip command. Notice that the 172.16.23.0 network is still in the routing table and being sourced from RIP.

Making an interface in RIP passive only disables updates from being sent through RIP. It does not affect routes being received through it.

What are some reasons to prevent RIP from sending updates out a particular interface?
RIPv2 does not use an adjacency system. Therefore, RIPv2 floods all of its routing updates out of its interfaces rather than developing adjacencies with short hello packets like EIGRP and OSPF. Because RIPv2 could flood packets out interfaces where there are no RIPv2 receivers, it is a best practice to prevent RIP packets from being sent out these interfaces.

Putting a RIPv2 interface in passive mode saves the router from sending multicast RIP packets out an
interface that has no neighbors.

Does RIPv2 send advertisements out loopback interfaces?
By default, loopback interfaces act like normal interfaces in RIPv2 when sending and receiving packets. However, a router will never have RIPv2 neighbors out its loopback interface, so loopback interfaces can always be configured as passive interfaces, thereby saving CPU resources.

e. If you are unsure, monitor the output of the debug ip rip command to verify your answer. On R1 and R2, configure all loopbacks from which RIPv2 is sending advertisements in passive state with the passiveinterface command.

When running RIPv2, implement passive interfaces as a common practice to save CPU processor cycles and bandwidth on interfaces that do not have multicast RIPv2 neighbors.

Note: An alternative to making each loopback interface on R1 passive is to make all interfaces passive with the passive-interface default command in router configuration mode. Then make any interfaces that need to send updates, such as S0/0/0, nonpassive.

Step 4: Summarize a supernet with RIP.
a. On R2, issue the show ip route rip command. Notice that you can see all prefixes from R1 in the R2 routing table.

In preparing for redistribution, you want to redistribute the minimum number of destination prefixes into each of the routing protocols. Which RIP routes should you summarize because they are contiguous and

which mask should you use?
Summarize 192.168.48.0/24 through 192.168.51.0/24. Use a 22-bit mask to create the supernet.

Under normal circumstances, you could simply summarize the four consecutive class-C networks with the ip summary address rip command on the R1 serial 0/0/0 interface. However, the RIP implementation in the Cisco IOS Software does not allow summarizing to a mask length that is less than the classful network prefix (in this case, 24 bits). This limitation does not affect other routing protocols. If you do try, you receive the following error message:

Recall from the EIGRP labs that summary routes display in the summarizing device’s routing table as having the next hop being the Null0 interface. You can create an entry manually using the ip route command and redistribute it into RIP, thereby emulating the approach of EIGRP to a certain extent.

b. To get around the ip summary-address rip message error, create a static route on R1 to summarize the networks of loopbacks 48 through 51. Then redistribute the route on R1.

This solution might seem unusual, but for RIPv2, it resembles many effects of summarization as performed in other routing protocols like EIGRP or OSPF. Again, this is not a limitation of RIPv2, but rather a Cisco IOS implementation issue.

c. On R1 and R2, verify that the RIP supernet has been added to the routing table with the show ip route command.

Will this route to Null0 affect routing to prefixes with longer addresses on R1? Explain.
The routing table first matches based on longest IP prefix. If any of the summarized networks are routable on R1, as in this situation, R1 uses the connected route with the longer mask to reach those networks. If one of those interfaces were to be shut down, R1 would send traffic for that network to its Null0 virtual interface.

Step 5: Suppress routes using prefix lists.
Sometimes you might not want to advertise certain networks out a particular interface, or you might want to filter updates as they come in. This is possible with distance-vector routing protocols, such as RIP or EIGRP. However, link-state protocols are less flexible, because every router in an area is required to have a synchronized database as a condition for full adjacency.

Distribute lists can be used with either access lists or prefix lists to filter routes by network address. With prefix lists, they can also be configured to filter routes by subnet masks.

In this scenario, you want to filter updates from R1 to R2, allowing only the networks of Loopback 0 and Loopback 70 and the summary route to be advertised. You want to suppress the more specific prefixes so that routing tables are kept small, and CPU processor cycles on the routers are not wasted.

The 22-bit summary and the 24-bit major network address both have the same address, so access lists will not accomplish the filtering correctly. Therefore, it is necessary to use prefix lists.

To create a prefix list or add a prefix list entry, use the ip prefix-list command in global configuration mode.

The ge keyword represents the “greater than or equal to” operator. The le keyword represents the “less than
or equal to” operator. If both the ge and le keywords are omitted, the prefix list is processed using an exact
match.

a. On R1, use a prefix list as a distribution filter to prevent the more specific routes to loopbacks 48 through 51 from being advertised. Allow all other destination networks, including the summary route.

Line 1 of the prefix list permits the summary route and nothing else, because no other route can match that network address with a mask of exactly 22 bits.

Line 2 denies all prefixes with a network address in the 192.168.48.0/22 block of addresses that have subnet masks from 22 bits to 24 bits. This removes exactly four network addresses matching the 22, 23, and 24 bits in length of the subnet mask. Line 2 would deny the 192.168.48.0/22 summary route you created if Line 1 did not explicitly permit the summary route.

Line 3 allows all IPv4 prefixes that are not explicitly denied in previous statements of the prefix list.

b. From the RIP configuration prompt on R1, apply this access list with the distribute-list command.

c. On R2, verify that the filtering has taken place using the show ip route rip and show ip rip database commands.

Note: You might need to issue the clear ip route * command on R2 to see the removal of the more specific R1 prefixes. Also, if the network 192.168.48.0/22 does not appear on R2, this is incorrect behavior and might be corrected in recent versions of Cisco IOS Software. A workaround is to remove the network 192.168.48.0 command from RIP and issue the clear ip route * command on R1.

Why would you want to filter updates being sent out or coming in?
The intention is to summarize addresses at R1, achieving benefits such as smaller routing tables and the prevention of route flapping disturbing network stability. By redistributing the static route into RIPv2, only the summary route was added, but the more specific routes were not filtered. To achieve these benefits, filter the outgoing routes to R2 by allowing all routes except the 24-bit routes.

Step 6: Configure OSPF.
a. Configure single-area OSPF between R2 and R3. On R2, include just the serial link connecting to R3. On R3, include the serial link and all loopback interfaces.

b. On R3, change the network type for the loopback interfaces to point-to-point so that they are advertised with the correct subnet mask (/24 instead of /32).

c. Verify the OSPF adjacencies on R2 and R3 with the show ip ospf neighbors command. Also make sure that you have routes from OSPF populating the routing tables with the show ip route ospf command.

Note that output of the show ip route ospf command on R3 is blank.

The network 192.168.0.0 0.0.255.255 area 0 command allows OSPF to involve interfaces that have IP addresses in that range.

A common misconception is that OSPF advertises the entire range of the network given in the router’s network statement; it does not. However, it does advertise any connected subnets in that range of addresses to adjacent routers. You can verify this by viewing the output of the show ip route command on R2. Do you see a 192.168.0.0/16 supernet?
No, because the network command selects interfaces by IP address. It does not set the exact prefix to be advertised.

R2 is the only router with all routes in the topology (except for those that were filtered out), because it is involved with both routing protocols.

Step 7: Configure passive interfaces in OSPF.
Passive interfaces save CPU cycles, router memory, and link bandwidth by preventing broadcast and multicast routing updates on interfaces that have no neighbors. In link-state protocols, adjacencies must be formed before routers exchange routing information. The passive-interface command in OSPF configuration mode prevents an interface from sending or processing OSPF packets on that interface.

OSPF included the R3 loopback interfaces in its network statements shown in Step 6.

a. On R3, configure Loopback0 as a passive interface in OSPF. At the OSPF router configuration prompt, use the passive-interface interface_type interface_number command.

How is this different from the RIP version of this command?
Passive interfaces in RIPv2 prevent outgoing routing information via multicast.

Because OSPF must create an adjacency before sending routing updates, the OSPF version of the passive-interface command prevents sending or processing OSPF packets and, therefore, prevents adjacencies.

b. Cisco IOS Software provides a quick way of selecting interfaces for passive mode. Use the passiveinterface default command to make all interfaces passive. Then use the no passive-interface interface interface_number command to bring the Serial0/0/1 interface out of passive mode.

c. You can verify the application of this command by issuing the show ip protocols command.

Step 8: Allow one-way redistribution.
a. On R2, configure OSPF to redistribute into RIP under the RIP configuration prompt with the redistribute ospf process metric metric command, where process is the OSPF process number, and metric is the default metric with which you want to originate the routes into RIP. If you do not specify a default metric in RIP, it gives routes an infinite metric and they are not advertised.

b. Verify the redistribution with the show ip protocols command.

c. On R1, look at the routing table with the show ip route rip command. It has all the routes in the topology.

d. On R1, ping a loopback on R3. Notice that it shows that R1 has a route to R3, but R3 does not have a route back to R1.

e. On R1, verify that R3 does not have a route back with the traceroute command.

To address this problem, you can originate a default route into OSPF that points toward R2 so that the pings are routed back toward R2. R2 uses its information from RIPv2 to send pings back to R1.

f. From the OSPF configuration prompt, issue the default-information originate always command to force R2 to advertise a default route in OSPF.

g. Verify that this route is present in the R3 routing table.

You should now have full connectivity between all networks in the diagram.

h. Use the Tcl script from Step 1 to verify full connectivity.

Step 9: Redistribute between two routing protocols.
You can substitute this default route with actual, more specific routes.
a. On R2, under the OSPF router configuration prompt, remove the default route advertisement with the no default-information originate always command. Next, use the redistribute rip command. You do not need to specify a default metric in OSPF. Notice the warning.

b. If you display the routing table on R3, the external OSPF routes that were added are the 192.168.70.0/24 and 192.168.48.0/22 networks.

This is because, by default, OSPF only accepts classful networks and supernets when redistributing into it. The only classful network coming into R2 from RIP is the class C network 192.168.70.0, and the only supernet is the 192.168.48.0/22.

c. You can modify this behavior by adding the subnets keyword to the redistribute command.

d. On R3, verify the configuration with the show ip route ospf command.

You should again have full connectivity between all networks in the diagram.

e. Run the Tcl script on each router to verify full connectivity.

Step 10: Set a default seed metric.
Under any routing protocol, you can specify a default seed metric to be used for redistribution instead of, or in addition to, setting metrics on a per-protocol basis. A seed metric is a protocol-independent feature of the Cisco IOS Software that is usually configured when redistributing into distance-vector protocols.

Notice that the metric listed in the R3 routing table is 20.

You can override the global creation of a default seed metric on a per-protocol basis by using the metric argument in a redistribution command. You can also use the metric command under other routing protocols.

a. On R2, in OSPF configuration mode, issue the default-metric metric command to configure a default metric for redistributed routes. The default metric for all OSPF redistributed routes is 20, except for BGP, which is 1. Setting the metric for RIP to a higher number makes it less preferable to routes redistributed from other routing protocols.

b. Verify the new metric in the R3 routing table. It might take some time for the new metric to propagate.

Step 11: Change the OSPF external network type.
Look at the R3 routing table. Notice that the external (redistributed) routes have O E2 as their type. In the output, O means OSPF, and E2 means external, type 2. OSPF has two external metric types, and E2 is the default. External type 1 metrics increase like a usual route, whereas external type 2 metrics do not increase as they get advertised through the OSPF domain. Also notice that the metric is exactly the same as the seed metric in the previous step.

Where would an external type 1 metric be useful?
If there are multiple paths through an OSPF domain to an external destination network, use E1 routes so that OSPF will evaluate the shortest cost path. Especially use this option if multiple ASBRs are advertising the same destination network to prevent suboptimal routing.

Where would an external type 2 metric be useful?
If there is only one ASBR advertising an external route, use E2 routes. In this scenario, use the E1 type even though only one ASBR is advertising the external routes.

a. You can change the external type using the metric-type argument with the redistribute command. Change the type to E1 for RIP redistributed routes.

b. Display the R3 routing table again.

Which attributes of the routes changed?
The external route type has been changed from E2 to E1. The metric, previously 10000, has been incremented to represent the path through the OSPF network to the ASBR.

Note: Be sure to save your final configurations through Step 11 for use in Lab 4-2, “Redistribution Between EIGRP and OSPF.”

Router Interface Summary Table

Router Interface Summary
Router Model Ethernet Interface
#1
Ethernet Interface
#2
Serial Interface
#1
Serial Interface
#2
1700 Fast Ethernet 0
(Fa0)
Fast Ethernet 1
(Fa1)
Serial 0 (S0) Serial 0/0/1
(S0/0/1)
1800 Fast Ethernet 0/0
(Fa0/0)
Fast Ethernet 0/1
(Fa0/1)
Serial 0/0/0
(S0/0/0)
Serial 0/0/1
(S0/0/1)
2600 Fast Ethernet 0/0
(Fa0/0)
Fast Ethernet 0/1
(Fa0/1)
Serial 0/0 (S0/0) Serial 0/1 (S0/1)
2800 Fast Ethernet 0/0
(Fa0/0)
Fast Ethernet 0/1
(Fa0/1)
Serial 0/0/0
(S0/0/0)
Serial 0/0/1
(S0/0/1)
Note: To find out how the router is configured, look at the interfaces to identify the type of router and how many interfaces the router has. Rather than list all combinations of configurations for each router class, this table includes identifiers for the possible combinations of Ethernet and serial interfaces in the device. The table does not include any other type of interface, even though a specific router might contain one. For example, for an ISDN BRI interface, the string in parenthesis is the legal abbreviation that can be used in Cisco IOS commands to represent the interface.

Device Configurations (Instructor version)

Router R1

Router R2

Router R3

More Resources

About the author

Scott

Leave a Comment