Routing Protocols


Routing protocols are a necessity in the networking world. Learning how routers use routing protocols to make a routing decision and how each routing protocol works is necessary to overcome this part of the exam. Interior routing is implemented at the Internet layer of the TCP/IP suite of protocols. Interior routing protocols use IP as a routing protocol and use a specific algorithm for different protocols. Each type of interior routing protocol uses different algorithms and mechanisms to accomplish routing. The following are a few examples of interior routing protocols:

  • Routing Information Protocol (RIP) ” RIP is one of the most commonly used routing protocols for the Internet. RIP uses a maximum hop count of 15 to calculate a routing path . RIP is a distance-vector routing protocol, and has a default distance of 120.

  • Interior Gateway Routing Protocol (IGRP) ” IGRP is a protocol proprietary to Cisco equipment. It was designed to overcome some of RIP's limitations. IGRP uses a hop count of a maximum of 255 to calculate a routing path. IGRP is a distance-vector routing protocol, and has a default distance of 100 (you learn more about distance-vector and link-state routing protocols in upcoming sections of this chapter).

  • Open Shortest Path First (OSPF) ” OSPF is a link-state routing protocol that uses an autonomous system to accomplish routing. It is used for large networks, because its maximum metric limit is 65,535. OSPF has a default distance of 110.

    graphics/note_icon.gif

    An autonomous system is defined as a set of routers under a single technical administration, using an interior gateway protocol and common metric to route packets within the autonomous system (AS), and using an exterior gateway protocol to route packets to other autonomous systems (ASs).


  • Enhanced Interior Gateway Routing Protocol (EIGRP) ” EIGRP is a proprietary routing protocol for Cisco. EIGRP combines link-state routing and distance-vector routing to achieve a balanced hybrid routing protocol. EIGRP has a default distance of 90.

Different routing protocols use different techniques, algorithms, and type of protocols to dynamically learn routes, share routes with other routers, and make sure that the information the router keeps is as current as possible. The following sections discuss these routing protocol features:

  • Link-state protocols

  • Convergence time

  • Distance-vector routing protocols

  • Split horizon

  • Split horizon- poison reverse

  • Hold-down timers

  • Triggered updates

Link-State Routing Protocols

Protocols using the link-state routing algorithm possess a complex table of network topology information for routing. The link-state routing process uses link-state packets (LSPs) to inform other routers of distant links. The routers all use these "hello packets" to inform the other routers on the network where they are and their proximity to each other. After the routers have all updated their fellow routers, each router will possess a routing table to refer to when it needs to make routing decisions that incorporate the best path.

One of the benefits of using a link-state routing algorithm is how it reports the best path. The best path may not be the shortest distance, but instead is the fastest way. Link-state uses many factors when declaring the best path, such as hop count, bandwidth, congestion, and link speed.

Convergence Time

Convergence is the time required for all the routers to update their routing tables after a change to the network. When one router is informed of a change on the network, it reports the change to its neighbors, and its neighbors report to their neighbors, and so on. Convergence time is not a set period of time, but depends on the number of routers and the size of the network.

For a router to converge with its neighbors, it must remember its name and the cost (distance) of the path to the neighboring router. Then it must send an LSP with the information to a neighboring router. Routers must also receive LSPs so that they can update their own routing tables. After they have exchanged the information, the routers will have a topology of the existing network and the best paths.

Distance-Vector Routing Protocols

Distance-vector routing protocols are designed to send a copy of a router's entire routing table to all of its neighbors. This enables all the routers to know the routing paths and determine the lowest cost (shortest distance) for the traffic before forwarding the packets. This information comprises the local routing table and is re-advertised to the router's neighbors for an optimal route on the network. When operating a large campus environment, finding the best path for traffic will make better use of the network.

Using distance-vector routing does have a drawback: It does not update the routers at the same time. Distance-vector routing protocols update the routers on the network every 30 to 90 seconds, so when a router fails, the other routers on the network may not receive the information that it has failed. Because of the time difference in route updates, there may be a router or routers that don't realize a path isn't working.

It is very important for your network to learn and update its neighboring routers on the best path for traffic in order to prevent problems of a routing loop. A routing loop is data that makes a virtual circle through the network and passes through an interface repeatedly until something occurs to stop the looping data.

Routing loops can destroy an internetwork and can multiply for a long time trying to find the packet destination. Distance-vector protocols use a few different methods to prevent routing loops . These include split horizon, poison reverse, triggered updates, and hold-down timers.

Split Horizon

The split horizon method can reduce incorrect routing information and excessive routing with a simple rule. The split-horizon rule uses the premise of not sending information back the same way that it came. With the split-horizon rule, if a router receives an update from network A through interface B, the router will not update that route to network A through interface B.

Split Horizon-Poison Reverse

The split horizon method is designed to stop routing loops between routers that have formed nearby relationships with other routers. The poison-reverse technique allows networks to be advertised with a hop count set to infinity, thus causing all routes to be flushed from the routing tables. In this type of environment, the split-horizon-poison-reverse method will stop a two-node routing loop and reduce the possibility of large routing loops.

Hold-Down Timers

Hold-down timers are used to prevent routers from sending inaccurate routing updates. It is common in networking for a router or interface to go down for various reasons, including loss of power, a loose cable, or an accidental hit of the power switch. When a router or interface goes down, the down router will trigger the other routers to update their routing tables.

If the down router or interface suddenly comes back online, the rest of the routers on the network will still be trying to update their tables. Hold-down timers help eliminate this problem. You can set the routers' hold-down timers for a period of time to prevent routers from trying to update as a result of brief network glitches. With the hold-down timer, if a router or interface goes down, the router's hold-down timer will wait a specified amount of time before it tries to update the routing table. In most cases, if the down router is caused by a minor glitch, the router will come back online and the network will not be affected.

Triggered Updates

Triggered updates are used to allow routers to inform their neighbors immediately of routing changes, so they don't have to wait for the regular timed updates. Some protocols, such as EIGRP, use only triggered updates to let the network know of a change.



CCNP CIT Exam Cram 2 (642-831)
CCNP CIT Exam Cram 2 (Exam Cram 642-831)
ISBN: 0789730219
EAN: 2147483647
Year: 2003
Pages: 213
Authors: Sean Odom

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net