Eigrp.... confused :(

kaynaan

hi guyz.

i'm new to this forum, i registered because i could not get any help beyond a basic level from my instructor. and i was hoping some of the more experienced fellows here may give me a hand.

some Questions are bugging me ..... the first 1 is:
-i'm using Todd lammle's book "CCNA study guide" 4th ED. (couldnt get my hands on 5th. !) for my exam preparation.

on chapter #6"Eigrp" it is says that Eigrp routing protocol has a maximum hop count of 255.

maximum hop count is easy to understand in RIP, when hop count is used as the metric to select the optimal path.

but eigrp uses Bandwidth,Delay ,Reliability,Load,MTU. altogether or some of them, in a very complex formula.

this formula generates 'costs of path', which are much greater than 255.
if u have seen in the routing table [90/*****].

my Q is, what is the relationship between cost of a path, and max hop count, and how do u count hops in eigrp ???.



thanks.

mikej412

kaynaan wrote:

on chapter #6"Eigrp" it is says that Eigrp routing protocol has a maximum hop count of 255.

maximum hop count is easy to understand in RIP, when hop count is used as the metric to select the optimal path.

but eigrp uses Bandwidth,Delay ,Reliability,Load,MTU. altogether or some of them, in a very complex formula.

Hop count is also used to determine if there is a routing loop -- exceed the allowed hop count, and the packet is dropped.

RIP has a small hop count, so that limits the size of the network.

EIGRP has a larger maximum hop count, so the networks can be bigger. That's where the routing metrics become important for routing the network traffic as to its destination as fast as possible.

Hop count as a metric for routing doesn't work that great.
Hop count to drop packets to prevent routing loops works okay.

So remember, hop count is used by RIP as a routing metric, but hop count is also used as a loop prevention method.

Yankee

Real world answer is that EIGRP uses BW and Delay only to determine a loop free network using the DUAL algorithm. No one should alter those basics without a damn good reason, which I could not fathom.

To answer your question, because EIGRP does calculate a loop free network the hop count should never become an issue (feel free to memorize such numbers for the exam). That is the offical answer but reality is that a poorly designed network may cause EIGRP problems where loops may occur as the network attempts to stabilize especially due to delayed response from fully participating neighbors but that is well beyond the CCNA scope.

Yankee

kaynaan

thnx for u'r replies but my question was ....

1- how are hops counted in eigrp ?
2- what's the relation between path cost and hop count e.g. in rip the path cost is hop count, and the max hop count is 15.
but in eigrp the path cost are usually many times the max hop count ... (my main source of confusion !)

if any of u can give me more detailed technical explanation i would be very gratefull, or just point me to a source.

thnx guyz.

darkuser

pointing ....

http://www.cisco.com/univercd/cc/td/doc/product/software/ios122/122cgcr/fipr_c/ipcprt2/1cfeigrp.htm

rossonieri#1

kaynaan wrote:

thnx for u'r replies but my question was ....

1- how are hops counted in eigrp ?
2- what's the relation between path cost and hop count e.g. in rip the path cost is hop count, and the max hop count is 15.
but in eigrp the path cost are usually many times the max hop count ... (my main source of confusion !)

if any of u can give me more detailed technical explanation i would be very gratefull, or just point me to a source.

thnx guyz.

hello,

answer :
1. hops counted - how many router you have passed to reach final destination. yes - eigrp metric could be a hop count or BW/delay(fast link vs slow link, or simply link quality) - which one is better to destination will be used.
2. path cost function is to determine which path in multipath environment will be the best path to destination.

cheers...

myk_roque

I think I know where the confusion came from. Just disassociate the Metrics used by the routing protocol to the Methods of preventing routing loops.

The Metrics used by EIGRP is not related at all to the hop count and is based entirely (by default) to the BW and Delay. Remember that the metrics used for RIP is just the hop count. for IGRP and EIGRP, it would be delay and BW.

The Max Hop count as being referred to by Todd Lamlle is the number of routers the routing protocol can traverse. I believe it is to make you realize just how scalable the IGRP and EIGRP is as compared to RIP. As you can see with RIP having only 15 hops max, it cannot traverse any more routers greater than this number, as opposed to IGRP and EIGRP which has (by default) 100.

kaynaan wrote:

thnx for u'r replies but my question was ....

1- how are hops counted in eigrp ?
2- what's the relation between path cost and hop count e.g. in rip the path cost is hop count, and the max hop count is 15.
but in eigrp the path cost are usually many times the max hop count ... (my main source of confusion !)

if any of u can give me more detailed technical explanation i would be very gratefull, or just point me to a source.

thnx guyz.

EdTheLad

kaynaan wrote:

on chapter #6"Eigrp" it is says that Eigrp routing protocol has a maximum hop count of 255

Just to be a sticklier, the hop count for Eigrp is 224 not 255!

kaynaan

thnx 4 bringing it up.

i also read that somewhere , max hop count of eigrp is 244, but i thought it was a type, since it says 255 in many other places.

if anybody can clarify..............

wildfire

I always thought I was 255, and according to The Cisco router field config manual Ed is right.

However read this document about Eigrp on Cisco.com and it suggests the max by default is 100 but can be configured to 240.

http://www.cisco.com/warp/public/103/eigrp-toc.html

Would love to put it to test but I dont have 250 odd routers to test it on Nor the time for that matter.

However I looked through the BSCI 3rd Edition Exam certification guide and there is no mention of it in there.

Sorry I couldnt clarify maybe someone like Kenny will have a definitive answer for us!