CCNP Route Lab 4-2, Redistribution Between EIGRP and OSPF

CCNP Route Lab 4-2, Redistribution Between EIGRP and OSPF

Topology

ccnp-route-lab-redistribution-eigrp-ospf

Objectives

  • Review EIGRP and OSPF configuration.
  • Redistribute into EIGRP.
  • Redistribute into OSPF.
  • Summarize routes in EIGRP.
  • Filter routes using route maps.
  • Modify EIGRP distances.
  • Modify OSPF distances.
  • Create passive interfaces in EIGRP.
  • Summarize in OSPF at an ABR and an ASBR.

Background
R1 is running EIGRP, and R3 is running multi-area OSPF. In this lab, you configure redistribution on R2 to enable these two routing protocols to interact, allowing full connectivity between all networks. In Appendix A of this lab, you explore black hole operation.

Note: This lab uses Cisco 1841 routers with Cisco IOS Release 12.4(24)T1 and the Advanced IP Services image c1 841 -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 additional addressing.
a. Start with the final configurations of Lab 4.1, “Redistribution Between RIP and OSPF.” On R1 and R2, remove the RIPv2 configuration and the static route with the following commands.

b. Configure the additional loopback interfaces on R2 and R3, as shown in the diagram.

Step 2: Configure EIGRP.
a. Configure R1 and R2 to run EIGRP in autonomous system 1. On R1, add in all connected interfaces either with classful network commands or with wildcard masks. Use a classful network statement on R2 and disable automatic summarization.

b. Verify the configuration with the show ip eigrp neighbors and show ip route eigrp commands on both routers.

Step 3: Create passive interfaces in EIGRP.
a. Issue the show ip eigrp interfaces command on R2.

Because you used the classful network command, both serial interfaces are involved with EIGRP.

b. To stop EIGRP from sending or processing received EIGRP packets on the serial interface going to R3, use the passive-interface interface_type interface_number command.

c. Verify the change with the show ip eigrp interfaces and show ip protocols commands.

How does preventing hello packets out of an interface affect the update capabilities of EIGRP out that interface?
The use of the passive-interface command in EIGRP prevents EIGRP packets from being sent out or processed on the interface. This disallows adjacencies that would normally be created through the exchange of hello packets.

Is this behavior more like RIP or like OSPF in regard to the passive-interface command?
This behavior is most similar to the behavior of the OSPF passive-interface command because it prevents adjacencies from forming through a passive interface.

Step 4: Manually summarize with EIGRP.
You can have EIGRP summarize routes sent out an interface to make routing updates more efficient by using the ip summary-address eigrp as network mask command.

a. Have R1 advertise one supernet for loopbacks 48 and 49 to R2. Do not summarize loopbacks 50 and 51 in this statement, because these will be summarized in Step 9.

b. Verify the configuration with the show ip route eigrp and show ip route 192.168.48.0 255.255.254.0 commands on R1. Notice the administrative distance for this route.

Why does EIGRP make the administrative distance different for summary routes?
EIGRP creates a route to the Null0 interface with an AD of 5 and advertises it to a list of assigned neighbors. As a summary, it will be preferred higher than other routing protocols such as BGP, but lower than overriding routes generated by connected interfaces or static routes. If a summary route is advertised, the router should have knowledge of all its possible subnets. If a subnet is down, it is unknown to the router, and the packets to these nonexistent subnets should be dropped rather than routed somewhere else. Making the administrative distance of the Null0 route low enough makes sure that this route gets into the routing table.

Step 5: Additional OSPF configuration.
OSPF is already partially configured on R2 and R3.

a. You need to add the area 10 configuration to R2 and the area 20 configuration to R3 to complete the configuration.

b. Verify that your adjacencies come up with the show ip ospf neighbor command, and make sure that you have routes from OSPF populating the R2 routing table using the show ip route ospf command.

Notice that for the newly added loopback interfaces, OSPF advertised /32 destination prefixes (for example, R2 has a route to 192.168.8.1/32 in its routing table).

c. Override this default behavior by using the ip ospf network point-to-point command on the OSPF loopback interfaces on R2 and R3. You can copy and paste the following configurations to save time.

Router R2:

Router R3:
(Only configure the point-to-point network type for the newly added loopbacks in area 20. The area 0 loopbacks were configured in Lab 4-1.)

Note: You can also use the interface range command to configure multiple interfaces simultaneously, as shown below.

d. Verify the configuration with the show ip route command on R2. Notice that the routes now each show on one line with the /24 major network mask.

Notice that R2 is the only router with knowledge of all routes in the topology at this point, because it is involved with both routing protocols.

Step 6: Summarize OSPF areas at the ABR.
Review the R2 routing table. Notice the inter-area routes for the R3 loopbacks in area 20.

a. Summarize the areas into a single inter-area route using the area area range network mask command on R3.

b. On R2, verify the summarization with the show ip route ospf command on R2.

Where can you summarize in OSPF?
In OSPF, you can summarize at the Area Border Router (ABR) and Autonomous System Border Router (ASBR). Inter-area summary LSAs created at the ABR are embedded in Type 3 LSAs. External summary LSAs created at the ASBR are embedded in Type 5 LSAs.

Compare and contrast OSPF and EIGRP in terms of where summarization takes place.
EIGRP allows summarization at any EIGRP router interface in the domain. OSPF can summarize only at the ABR and the ASBR.

Explain the synchronization requirement in OSPF that eliminates other routers as points of summarization.
OSPF requires that all routers in an OSPF area have synchronized link-state databases (LSDBs). Filtering cannot be applied within an OSPF area, as a result some of the area’s LSAs on one router are not propagated to another area router. Therefore, the two entry and exit points to an area, ABRs and ASBRs, are the only places where summarization in OSPF can occur.

Why or why not does EIGRP have this requirement?
The synchronization requirement in OSPF is a critical part of the loop-prevention mechanism for Dijkstra’s algorithm. The DUAL algorithm has a different set of rules to prevent loops, primarily based on the feasibility condition, which guarantees that EIGRP will not advertise looped paths through the local router. Because each router must function by the same principle, the entire autonomous system avoids loops. Although EIGRP has a similar type of multi-area functionality with its assortment of autonomous systems, the database requirements are not the same as OSPF.

Step 7: Configure mutual redistribution between OSPF and EIGRP.
a. Under the OSPF process on R2, issue the redistribute eigrp 1 subnets command. The subnets command is necessary because, by default, OSPF only redistributes classful networks and supernets. A default seed metric is not required for OSPF. Under the EIGRP process, issue the redistribute ospf 1 metric 10000 100 255 1 1500 command, which tells EIGRP to redistribute OSPF process 1 with these metrics: bandwidth of 10000, delay of 100, reliability of 255/255, load of 1/255, and a MTU of 1500. Like RIP, EIGRP requires a seed metric. You can also set a default seed metric with the default-metric command.

b. Issue the show ip protocols command on the redistributing router, R2. Compare your output with the following output.

c. Display the routing tables on R1 and R3 so that you can see the redistributed routes. Redistributed OSPF routes display on R1 as D EX, which means that they are external EIGRP routes. Redistributed EIGRP routes are tagged in the R3 routing table as O E2, which means that they are OSPF external type 2. Type 2 is the default OSPF external type.

d. Verify full connectivity with the following Tcl script:

Step 8: Filter redistribution with route maps.
One way to filter prefixes is with a route map. When used for filtering prefixes, a route map works like an access list. It has multiple statements that are read in a sequential order. Each statement can be a deny or permit and can have a match clause for a variety of attributes, such as the route or a route tag. You can also include route attributes in each statement that will be set if the match clause is met.

a. Before filtering the R3 loopback 25 and 30 networks from being redistributed into EIGRP on R2, display the R1 routing table and verify that those two routes currently appear there.

There are multiple ways to configure this filtering. For this exercise, configure an access list that matches these two network addresses and a route map that denies based on a match for that access list.

b. Configure the access list as follows:

c. Configure a route map with a statement that denies based on a match with this access list. Then add a permit statement without a match statement, which acts as an explicit permit all.

d. Apply this route map by redoing the redistribute command with the route map under the EIGRP process.

e. As an alternative, if you previously configured a default metric under EIGRP, you can simply use the following command.

f. Verify that these routes are filtered out in the R1 routing table.

Step 9: Summarize external routes into OSPF at the ASBR.
You cannot summarize routes redistributed into OSPF using the area range command. This command is effective only on routes internal to the specified area. Instead, use the OSPF summary-address network mask command.

a. Before you make any changes, display the R3 routing table.

Notice the three external routes for the R1 loopback interfaces 48 through 51. Two of the loopbacks are already summarized to one /23.

Which mask should you use to summarize all four of the loopbacks to one prefix?
Use the 22-bit mask with the network address of 192.168.48.0.

b. You can summarize this all into one supernet on R2 using the following commands.

c. Verify this action in the R3 routing table.

What would happen if loopback 50 on R1 were to become unreachable by R2?
R2 would still advertise the 22-bit summary address to R3 until all the subnets included in the summary become inaccessible.

Would data destined for 192.168.50.0/24 from R3 still be sent to R2?
Yes. Data packets destined for 192.168.50.0/24 from R3 will still be sent to R2.

Would data destined for 192.168.50.0/24 from R2 continue to be sent to R1?
No. Because R2 has no prefixes longer than 22 bits that match the 192.168.50.0/24 subnet, packets will be routed to the Null0 virtual interface on R2.

d. If you are unsure of the outcome, shut down the interface on R1. Issue the ICMP traceroute command to 192.168.50.1 from R3 and then from R2. Check your output against the output and analysis in Appendix A. Remember to issue the no shutdown command when you are finished checking.

Is this a desirable outcome? Explain.
It is a desirable outcome because summarization allows routing tables to be reduced in size. However, a result which sometimes might be considered mildly undesirable is that data traffic is forwarded beyond where it would be forwarded without summarization.

Step 1 0: Modify EIGRP distances.
a. By default, EIGRP uses an administrative distance of 90 for internal routes and 170 for external routes. You can see this in the R1 routing table and in the output of the show ip protocols command.

b. You can change the administrative distance with the distance eigrp internal external command. This
command is only applicable locally. Change the distance to 95 for internal routes and 165 for external
routes.

Note: The EIGRP neighbor adjacency will be re-negotiated:

c. Verify the change in the routing table with the show ip route eigrp and show ip protocols commands.

Step 11: Modify OSPF distances.
You can also modify individual OSPF distances. By default, all OSPF distances are 110, but you can change the intra-area, inter-area, and external route distances using the distance ospf intra-area distance interarea distance external distance command. All the command arguments are optional, so you can change only what you need to.

a. Before changing anything, display the R3 routing table.

b. Change the intra-area distance to 105, inter-area distance to 115, and external routes to 175 on R3.

c. Verify the change in the routing table. Unfortunately, the only information that you can get from the output
of the show ip protocols command is the default distance, which is the intra-area distance.

Challenge: Change the Administrative Distance on R2

The previous two steps demonstrated using the distance command in a fairly inconsequential environment.
In which types of scenarios would the distance command be more valuable?
The distance command is more valuable when paths to destination networks exist through more than one routing protocol. The distance command allows a router to decide which routing protocol to prefer to destination networks.

On R2, you are running both EIGRP and OSPF. Imagine a fourth router, R4, connected to both R1 and R3. R4 is redistributing between the two routing protocols.

Using the default administrative distances for EIGRP and OSPF, which protocol would be preferred in the
routing table for destination prefixes in native OSPF networks and why?
OSPF networks will be preferred because the AD for OSPF routes is 110, as compared to the external AD through EIGRP, 170.

Which protocol would be preferred in the routing table for destination prefixes for native EIGRP networks?
EIGRP networks will be preferred through EIGRP because the EIGRP internal AD is 90. The OSPF AD is 110.

Instead of adding the 172.16.1.0/24 networks natively to EIGRP using a network statement, add the networks using the redistribute connected command in EIGRP configuration mode on R1.

With the default administrative distances set, what would the administrative distance be for that prefix on R2 in EIGRP and in OSPF? Explain why.
The route through EIGRP would have an AD of 170, because it is external to EIGRP. The route through OSPF would have an AD of 110 and would be preferred over EIGRP. By default, OSPF does not assign different AD values when installing AS internal and external routes in the routing table.

How could you make the EIGRP path prefer this route? Is there more than one way?
You could either lower the external EIGRP administrative distance to below 110, or raise the external OSPF administrative distance to above 170, or manipulate distance by gateway.

The general-purpose distance command could be used to manipulate the external EIGRP distance:

The general-purpose distance command could be used to manipulate the external OSPF distance:

The gateway-specific distance command could be used to manipulate the distance in OSPF:

The distance command will be used in more detail in Lab 4-3.

Could using the distance command in this situation cause asymmetric routing? Explain.
Before applying the distance command, you will have asymmetric routing. R2 will forward packets destined for 172.16.1.1 along the path from R3 to R4 to R1. Because all the networks connected to R2 are advertised to R1 via EIGRP, R1 will send data back to R2 directly, without taking the path involving R3 and R4. The distance command can be used in many situations to solve asymmetric routing problems, but can be the source of such problems as well.

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.

a. Configure R1 and shut down the loopback 50 interface:

b. On R2, you should see the following output.

Notice the absence of 192.168.50.0/24 in a specific route in the R2 routing table.

c. Begin debugging all incoming IP packets on R2, and then issue the ping 192.168.50.1 command.

The summary route, pointing to the Null0 interface as the next hop, acts as a “catch all” for any traffic generated by R2 or forwarded to R2 with the destination network 192.168.48.0/24. R2 sends traffic to the Null0 virtual interface, as shown by the IP packet debugging output highlighted above.

R2 is not able to ping the R1 shutdown loopback interface because the 192.168.50.0/24 route no longer exists in the routing table.

Check to see if network 192.168.50.0/24, or a supernet of it, is in the routing table of R3.

d. Begin debugging all IP and ICMP packets on R3. Ping the address 192.168.50.1 from R3. Try to trace the route from R3 to 192.168.50.1.

Analyze the process indicated by the ICMP responses. You might also want to refer to debugging messages for ICMP and IP packets on R2.

  1. R3 generates an ICMP echo request (ping) to 192.168.50.1.
  2. R3 looks up the (next-hop address, outgoing interface) pair for the longest matching prefix containing 192.168.50.1 in the IP routing table. It finds (172.16.23.2, Serial0/0/1).
  3. R3 routes the IP packet to (172.16.23.2, Serial0/0/1).
  4. R2 receives the IP packet from R3 on interface Serial0/0/1.
  5. R2 looks up the (next-hop address, outgoing interface) pair for the longest prefix matching containing 192.168.50.1 in the IP routing table. The longest matching prefix that the routing table returns is 192.168.48.0/22, for which the routing table responds with (null, Null0) because it has no next-hop address or physical outgoing interface.
  6. R2 realizes that this packet was routed remotely to it but that it has no route, so it sends an ICMP Type 3, Code 1 (host unreachable) packet to the source address of the packet, 172.16.23.3.
  7. R2 looks up the (next-hop address, outgoing interface) pair for 172.16.23.3 and resolves it to (172.16.23.3, Serial0/0/1).
  8. R2 then routes the ICMP packet for destination 172.16.23.3, normally 172.16.23.3 through
    Serial0/0/1.
  9. R3 receives a packet destined for its local address 172.16.23.3 and reads the packet, sending the ICMP “Host Unreachable” message to the ping output.

Note: For more information about how routers respond to unreachable hosts, see RFC 792 (ICMP) at
http://tools.ietf.org/html/rfc792 and RFC 4443 (ICMPv6) at http://tools.ietf.org/html/rfc4443.

Notice that R2 sends R3 an ICMP Type 3, Code 1 reply indicating that it does not have a route to the host
192.168.50.1. This ICMP “Host Unreachable” message is not only sent in response to pings or traceroutes (also a form of ICMP) but for all IP traffic.

e. If you were to use Telnet to 192.168.50.1, you would receive the following message based on the ICMP response from R2:

This is not an example of Telnet timing out, but of intelligent network protocols responding to routing issues in the network.

This summarization problem is a classic example of a “black hole” in a domain, which simply means traffic passing through the network destined for that subnet is discarded at some point along the way. Thankfully, ICMP informs sources of when their traffic is being discarded.

f. Do not forget to issue the no shutdown command on the R1 loopback 50 interface to re-enable routing to this network.

Device Configurations (Instructor version)

Router R1

Router R2

Router R3

More Resources

About the author

Scott

Leave a Comment