192 alleviating routing loops hop limits on packets if

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: sly unreachable part of the network still remains that way. ￿ 19.2 Alleviating Routing Loops: Hop Limits on Packets If a packet is sent along a sequence of routers that are part of a routing loop, the packet will remain in the network until the routing loop is eliminated. The typical time scales over which routing protocols converge could be many seconds or even a few minutes, during which these packets may consume significant amounts of network bandwidth and reduce the capacity available to other packets that can be sent successfully to other destinations. To mitigate this (hopefully transient) problem, it is customary for the packet header to include a hop limit. The source sets the “hop limit” field in the packet’s header to some value larger than the number of hops it believes is needed to get to the destination. Each switch, before forwarding the packet, decrements the hop limit field by 1. If this field reaches 0, then it does not forward the packet, but drops it instead (optionally, the switch may send a diagnostic packet toward the source telling it that the switch dropped the packet because the hop limit was exceeded). The forwarding process needs to make sure that if a checksum covers the hop limit field, then the checksum needs to be adjusted to reflect the decrement done to the hoplimit field.1 Combining this information with the rest of the forwarding steps discussed in the previous lecture, we can summarize the basic steps done while forwarding a packet in a besteffort network as follows: 1. Check the hop-limit field. If it is 0, discard the packet. Optionally, send a diagnostic packet toward the packet’s source saying “hop limit exceeded”; in response, the source may decide to stop sending packets to that destination for some period of time. 2. If the hop-limit is larger than 0, then perform a routing table lookup using the destination address to determine the route for the packet. If no link is returned by the lookup or if the link is considered “not working” by the switch, then discard the packet. Otherwise, if the destination is the present node, then deliver the packet to the appropriate protocol or application running on the node. Otherwise, proceed to the next step. 3. Decrement the hop-limit by 1. Adjust the checksum (typically the header checksum) if necessary. Enqueue the packet in the queue corresponding to the outgoing link 1 IP version 4 has such a header checksum, but IP version 6 dispenses with it, because higher-layer protocols used to provide reliable delivery have a checksum that covers portions of the IP header. CHAPTER 19. NETWORK ROUTING - II 4 ROUTING AROUND FAILURES returned by the route lookup procedure. When this packet reaches the head of the queue, the switch will send the packet on the link. ￿ 19.3 Neighbor Liveness: HELLO Protocol As mentioned in the previous lecture, determining which of a node’s neighbors is currently alive and working is the first step in any routing protocol. We now address thi...
View Full Document

This document was uploaded on 02/26/2014 for the course CS 6.02 at MIT.

Ask a homework question - tutors are online