ccna/032/w/8 norgren

ccna/032/w/8 norgren.One of the mechanisms used by notification update routing protocols is to share routing information and reachability information with neighboring routers. To disseminate this information, some protocols use local broadcasts, some use multicasts, and some use unicasts. Distance vectoring protocols typically use periodic local broadcasts with a destination IP address of 255.255.255.255 to share routing information, and these protocols always insist on doing so regardless of changes: they broadcast their routing information to all devices connected to their interfaces as soon as the timer expires. Note that distance vectoring protocols do not care at all about who is listening for these updates, nor do they verify that neighboring routers have received broadcast updates. A router running the distance vectoring protocol listens for routing broadcasts on its interface to learn what neighbors are available. In this process, there is no formal handshake or Hello process. The distance vectoring protocol assumes that the neighbors are informed via broadcasts, and if a neighbor fails, it must be discovered that the neighbor did not send a broadcast. Even if the situation changes and the router does not receive an update from the neighbor, it is assumed that the router will be informed of the change by the next broadcast update.Handling Updates After receiving a routing update from a neighboring router, the distance vectoring protocol performs the following steps. 1. increase the metric value of the route in the announcement (RIP adds 1 to the number of route hops to be announced). 2. compare the network number in the neighbor's routing update with the network number in the routing table. 3. If the neighbor provides better information, add it to the routing table and remove the old entry. 4. If the neighbor provides worse information, ignore it.5. if the neighbor provides exactly the same information as an existing entry in the routing table, reset the timer for that entry (in other words, the router(in other words, the router has been informed of the route from that neighbor)6. if the neighbor provides another path to a known destination network but with the same metric as an existing route in the routing table, the routingIf the neighbor provides another path to a known destination network with the same metric as the route already in the routing table, the router will add it to the routing table and keep the original route; provided, of course, that it does not exceed the maximum iso-overhead route for that destination network.path number for that destination network. This means that the router has learned the same network number from two different neighbors, and that both neighbors advertise the network numberindicate the same metric value.Step 6 is often referred to as the Bellman-Ford algorithm. As you can see from step 6, Cisco supports load balancing between equal-overhead paths of the same type, such as RIP paths. Equal overhead load balancing allows the routing protocol to list the same destination network multiple times in the routing table - each time for one path to that destination network - and to use those paths simultaneously. The advantage of this approach is that if one path fails, convergence will be completed quickly because there are other paths in the routing table. Most vendors use connection-based load balancing: all traffic from the same connection is routed using the same path to reduce the likelihood of causing problems to the connection. Equal-overhead load balancing is discussed in more depth in Chapters 19 and 20.Distance vectoring protocols are the simplest of the three types and are easy to configure and troubleshoot. They impose low overhead on the router and require few CPU cycles and memory to process updates: they receive incoming updates, increase the metric, compare the results with existing routes in the routing table, and update the routing table as necessary.