In-depth analysis of RIP routing protocols

 

The RIP Routing protocol (Routing Information Protocols) is the most widely used distance vector protocol. It was developed by Xerox in 1970s. At that time, RIP was part of the Xerox Network Service (Xerox Network Service) protocol cluster. The TCP/IP version of RIP is the license version of Xerox protocol. The biggest feature of RIP is that the implementation principle and configuration method are very simple.

The measurement method of RIP is based on the number of hops (hops count). Each time a router passes through, the number of hops in the path is incremented by one. As a result, the larger the number of hops, the longer the path, and the RIP algorithm will first select a path with fewer hops. The maximum number of hops supported by RIP is 15, and the number of hops is 16.

Route update in the RIP route protocol is implemented through scheduled broadcast. By default, the router broadcasts its route table to the network connected to it every 30 seconds. The router connected to the broadcast adds the received information to its route table. Every vro is broadcast so that all vrouters on the network will know all route information. Under normal circumstances, the router can receive a route message confirmation every 30 seconds. If the route entry is not confirmed after 180 seconds, that is, six update cycles, the router considers it invalid. If the route entry remains unconfirmed after 240 seconds, that is, eight update cycles, it will be deleted from the routing table. The above 30 seconds, 180 seconds, and 240 seconds are all controlled by the Timer, which are the update Timer (_ updateTimer) and the Invalid Timer (Invalid Timer) and Flush Timer ).

The routing loop distance vector algorithm is prone to route loops. The RIP routing protocol is a distance vector algorithm, so it is no exception. If there is a route loop on the network, information will be transmitted cyclically and never reach the destination. To avoid this problem, the RIP vector algorithm implements the following four mechanisms.

Split horizon ). Horizontal segmentation ensures that the router remembers the source of each route information and does not send it again on the port that receives this information. This is the most basic measure to prevent route loops.

Toxicity reversal (poison reverse ). When a path becomes invalid, the router does not immediately delete it from the routing table, but broadcasts it with a 16, that is, an inaccessible measurement value. In this way, although the route table size is increased, it is very helpful to eliminate the routing cycle. It can immediately clear any loops between neighboring routers.

Trigger update ). When the route table changes, the update packet is immediately broadcast to all adjacent routers, instead of waiting for a 30-second update cycle. Similarly, when a router just starts the RIP routing protocol, it broadcasts request packets. The adjacent router that receives the broadcast immediately responds to an update message, instead of waiting for the next update cycle. In this way, the changes in the network topology will spread on the network as quickly as possible, reducing the possibility of loop routing.

Hold down timer ). After a route entry is invalid, the route entry is in the suppression State for a period of time, that is, it no longer receives route updates for the same destination address within a certain period of time. If the router learns that a path is invalid from one CIDR block, it immediately learns that the route is valid on the other CIDR block. This effective information is often incorrect. Blocking timing avoids this problem. In addition, when a link starts and stops frequently, blocking timing reduces route fluctuations and increases Network stability.

Even if the above four methods are used, the routing loop problem cannot be completely solved, but the problem is minimized. Once a route loop occurs, the metric value of the route entry is counted to Infinity. This is because route information is cyclically transmitted. Each time a vro is transmitted, the measurement value is added to 1 and the value is always added to 16, and the path becomes inaccessible. It is clever to select 16 as an inaccessible metric value for the RIP routing protocol. It is large enough to ensure that most networks can run normally and small enough, it takes the shortest time to count to infinity.

Some neighboring networks are NBMA (Non-Broad_cast MultiAccess, non-broadcast multiplexing), that is, broadcast is not allowed to transmit data on the network. For such a network, RIP cannot rely on the broadcast route table. There are many solutions. The simplest solution is to specify a neighbor (neighbor), that is, to specify to send a route table to a specific router.

Although the defects of the RIP routing protocol are simple and time-tested, there are also some important defects, mainly including the following:

It is too simple to calculate the metric value based on the number of hops and often obtain the non-optimal route. The Metric value is limited to 16, which is not suitable for large networks. The security is poor and routes from any devices are updated; classless IP addresses and VLSM (Variable Length Subnet Mask, Variable-Length Subnet Mask) are not supported. The convergence speed is slow, usually longer than 5 minutes, and the bandwidth consumption is large.

The RIP routing protocol is a classic network protocol and is widely used. Therefore, it should be the most familiar to network engineers. The above content is a summary and statement of RIP, hoping to help readers.