< Free Open Study > |
Tuning, Redistribution, and Control of RIP UpdatesRIP offers several parameters for tuning timers, controlling broadcasts, and controlling routes. The following is a list of some of the common parameters adjustable within RIP:
The following commands are not exclusive to RIP and can be used with other routing protocols:
To see how some of these concepts work, apply some to the existing lab. Figure 9-3 changes the subnet between rogue and cyclops to 128.200.3.16/29. Instead of RIP on this segment, run EIGRP as the routing protocol. Redistribute between EIGRP and RIP on the rogue router. As Figure 9-3 shows, another router, storm, has been integrated into the network. Instead of running broadcast updates on the Token Ring segment between the storm and wolverine routers, run unicast updates. Figure 9-3. Route Redistribution and Unicast Example
To configure RIP unicast between storm and wolverine, you must first use the passive-interface command to prevent RIP broadcasts from entering the Token Ring network. Next , add a neighbor statement to point to the router in which RIP updates are to be sent. Example 9-8 lists the RIP configuration of wolverine and storm. Example 9-8 RIP Unicast Configurationhostname wolverine ! router rip passive-interface TokenRing1 network 128.200.0.0 neighbor 128.200.10.2 _______________________________________________________________________ hostname storm ! router rip passive-interface TokenRing0 network 128.200.0.0 neighbor 128.200.10.1 The next step is to perform mutual redistribution between EIGRP and RIP on the rogue router. Because there is only one redistribution point in the network, there is no need to perform any route filtering when using mutual redistribution. To redistribute between EIGRP and RIP, use the redistribution command, along with adding a default metric. The default metric chosen for RIP is 2, whereas 10000 1000 254 1 1500 is the default metric for EIGRP. Because the routing domains overlap (which means that RIP broadcasts will be heard on cyclops), use the passive-interface command to prevent excessive broadcasts from occurring. Example 9-9 lists the configuration of EIGRP and RIP on the rogue router. Example 9-9 EIGRP and RIP Configuration on roguerouter eigrp 2001 redistribute rip passive-interface Ethernet0/0 network 128.200.0.0 default-metric 10000 1000 254 1 1500 no auto-summary ! router rip redistribute eigrp 2001 passive-interface Ethernet0/1 network 128.200.0.0 default-metric 2 At this point, the network is close to being complete. However, if you observe the routing table on storm, you see that it does not have a route to 128.200.3.16/29, as illustrated in Figure 9-4. Figure 9-4. Route Table Discrepancies: RIP/EIGRP Network Before Summarization
Because the Ethernet segment between rogue and cyclops is on a 29-bit boundary, and because the segment between rogue and wolverine is on a 24-bit boundary, the rogue router will not forward any 29-bit network updates out its E0/0 port. Performing a debug ip rip command on the rogue router as done in Example 9-10 substantiates this. Example 9-10 debug ip rip Command Output from roguerogue# debug ip rip RIP protocol debugging is on RIP: sending v1 update to 255.255.255.255 via Ethernet0/0 (128.200.1.2) suppressing null update no 128.200.3.x route RIP: received v1 update from 128.200.1.1 on Ethernet0/0 128.200.10.0 in 1 hops 128.200.2.0 in 1 hops RIP: sending v1 update to 255.255.255.255 via Ethernet0/0 (128.200.1.2) suppressing null update RIP: received v1 update from 128.200.1.1 on Ethernet0/0 128.200.10.0 in 1 hops 128.200.2.0 in 1 hops rogue# To allow full IP reachability between cyclops and storm, the route between them must be summarized on a 24-bit boundary. To accomplish this, use the following command on the cyclops router: cyclops(config-if)# ip summary-address eigrp 2001 128.200.3.0 255.255.255.0 Observing the routing table on rogue, as in Figure 9-5, an EIGRP route 128.200.3.0/24 now exists. Because the route has a 24-bit mask, rogue can forward it out its E0/0 port, where storm will eventually receive the route. Figure 9-5. Route Table Discrepancies Resolved: RIP/EIGRP Network After Summarization
CAUTION Caution should be used whenever redistributing one routing protocol into another. Potential routing loops can occur if there are two or more redistribution points within the network. When multiple redistribution points are in the network, distance vector protocols are vulnerable to loops. Use the route maps with a well-thought-out IP addressing scheme to control loops . TIP Whenever you are working with routing protocols that don't carry a subnet mask within the routing update, such as RIP-1 and IGRP, be extremely careful to preserve a consistent bit mask throughout the entire internetwork. For example, a RIP domain might be operating on 24-bit boundary, whereas an OSPF domain might have some LAN networks at the 24-bit boundary, but its entire WAN is using a 30-bit boundary. The RIP domain will have no problem reaching the LANs because of the bit boundary match; however, it will not be capable of reaching the WAN interfaces. To allow full reachability between the two domains, the OSPF WAN networks that reside on a 30-bit boundary must be summarized to a 24-bit boundary before redistribution into RIP. |
< Free Open Study > |