What is the point of poison reverse?

Really? What is the point of suspending split horizon to hear about a connected route that is down from a neighboring router? If that router knows that one of it's connected routes is down why would it need to have another router tell it something it already knows?

Find more posts tagged with

Free for TechExams community: Cybersecurity salary guide

Compare cert salaries and plan your next career move

Button

Comments

itdaddy

not sure if this is the technical version but it makes sense to report a dead route down the chain
to other routers so they do not send the packets in that directions...if it is not reported then it is wasted bandwidth to send packets to a router that is dead or route is dead 16 hops shows dead or posioned!

that is what I think by instinct but do a google on posion reverse RIPv1 hop count posioned 16 hops..

- In a computer network that uses the Routing Information Protocol (RIP) or other distance vector routing protocols, a poison reverse is a way in which a gateway node tells its neighbor gateways that one of the gateways is no longer connected. To do this, the notifying gateway sets the number of hops to the unconnected gateway to a number that indicates "infinite" (meaning "You can't get there"). Since RIP allows up to 15 hops to another gateway, setting the hop count to 16 would mean "infinite."

found this definition on the internet..kinda says same thing I said. Just a way to notify other gateways that a gateway down the route is down to stop from sending packets that way.

wat08

Great minds think alike: http://www.techexams.net/forums/viewtopic.php?t=38476

I think dtlokee sums it up best: "I think your original question may be based on the idea that the failed route is directly attached to the router, but what if it is not?"

We need poison reverse to know what routes have failed. Try to convince yourself of this by walking through the algorithm with a bunch of different topologies.

LBC90805

Well ok then. But in the ICND2 book from ODOM, and with the Topology shown, the attached network goes down; the router it is attached to poisons the route and sends a triggered update to its neighboring router. Then that neighboring router sends an update, POISON REVERSE, back to the Router that the "downed" network is attached to.

Its kind of like telling a Rumor to a friend, and that friend turns around and tells you the same exact Rumor that you just told them.

I can see the point, kind of, of poison reverse but I still want more of an explanation of it.

wat08

Referencing your example from Odom, R2 needs to get confirmation of the poisoned route in order to start the route-timeout timer.

Q—What is the purpose of the timeout timer?

A—The timeout timer is used to help purge invalid routes from a RIP node. Routes that aren't refreshed for a given period of time are likely invalid because of some change in the network. Thus, RIP maintains a timeout timer for each known route. When a route's timeout timer expires, the route is marked invalid but is retained in the table until the route-flush timer expires.

http://www.cisco.com/en/US/docs/internetworking/technology/handbook/RIP.html#wp1020565

My question was the same as yours: why not just start the timer immediately upon learning of a failed connected link? Why wait for confirmation of something you already know? The answer is because the routing protocol treats failed links as a general instance. It is repetitive to swap a poisoned route, but what's the alternative? Re-design the entire protocol based on one little annoyance? It's easier to receive a notification of a failed link you already know about then to create a special workaround solely for failed connected links. I think that's the jist of it.