EIGRP K values
Hello Experts,
I am a little bit confused about the the EIGRP metric weight command. The syntax is :
metric weights tos k1 k2 k3 k4 k5
I undertand that EIGRP can use Bandwidth, Delay, Reliability, and Load to compute the metric of a route and by default uses only Bandwidth and Delay. My question is we have 5 K-values and only 4 "attributes" (Bandwidth, Delay, Reliability, and Load).
Which K-value represent which "attribute"?
Best regards
Comments
k1 = bw
k2 = load
k3 = dly
k4 = reliability
where "=" sign means "corresponds to"
if routes have equal metric, than route with largest MTU win.
k5 is a strange attribute, it computes in sync with k4.
This may be helpful...
http://www.cisco.com/en/US/docs/ios/12_3t/ip_route/command/reference/ip2_c1gt.html#wp1094645
It's a different command, so the accepted values may differ slightly, but it does show the definition of each of the k-values.
things.
decisions. The default metrics are used as the seed values for
redistributed routes that inherently have no "history" of interfaces to
pass through!
delay] * [K5 / (reliability + K4)]
that K1 = bandwidth ('cause the default value of 1 means 1 * bandwidth =
bandwidth) and K2 is often referred to as load, but actually is a more modified
version of bandwidth/(256-load). K2 defaults to 0 which results in an
unused number. K3, much like K1 gives a straight out value. K3
defaults to 1, so delay is used as a primary metric as
well.
default-metric command, the third value is actually reliability. The
second value there is delay.
talked about as modifiers. That has a lot to do with the idea that the
"reliability" of an interface is kind of a floating thing anyway. It all
has to do with errors seen. More errors = Less reliable. But the
math involved here of K5/(reliability + K4) has absolutely nothing to do with
MTU size.
what the lowest MTU along path to a route happens to
be.
more basic information on the algorithm.
Scott Morris, CCIE4 #4713, JNCIE-M #153, JNCIS-ER, CISSP, et al.
CCSI/JNCI-M/JNCI-ER
Senior CCIE Instructor
<?xml:namespace
prefix = st1 ns = "urn:schemas-microsoft-com:office:smarttags" />[email protected]
<?xml:namespace prefix = o ns ="urn:schemas-microsoft-com:office:office" />
Internetwork Expert, Inc.
http://www.InternetworkExpert.com
Toll
Free: 877-224-8987
Outside US: 775-826-4344
Online Community:
http://tinyurl.com/6dmnsu
CCIE Blog:
http://tinyurl.com/2nxxaq
Knowledge is power.
Power corrupts.
Study hard and be
Eeeeviiiil......
From: [email protected] [mailto:[email protected]]
On Behalf Of roastdog
Sent: Thursday, September 18, 2008 8:01
AM
To: [email protected]
Subject: Re: [CCIE
R&S] EIGRP K values
This may be helpful...
http://www.cisco.com/en/US/docs/ios/12_3t/ip_route/command/reference/ip2_c1gt.html#wp1094645
It's a different command, so the accepted values may differ slightly, but it
does show the definition of each of the k-values.
Internetwork
Expert - The Industry Leader in CCIE
Preparation
http://www.internetworkexpert.com
Subscription information
may be found
at:
http://www.ieoc.com/forums/ForumSubscriptions.aspx
Hmm.. do we actually need to know that whole formula? Every book I've ever read has basically said "bw and delay are the only ones that get used anyway, so ignore the rest"
Config Guide for EIGRP anyway, so in case you DO need it, you can always look it
up in about 15 seconds.
the rest, but I wouldn't spend a lot of time stressing about
it!
From: [email protected] [mailto:[email protected]]
On Behalf Of IPv6Freely
Sent: Thursday, September 18, 2008
10:01 AM
To: [email protected]
Subject: Re:
[CCIE R&S] RE: EIGRP K values
Hmm.. do we actually need to know that whole formula? Every book I've ever
read has basically said "bw and delay are the only ones that get used anyway, so
ignore the rest"
Internetwork
Expert - The Industry Leader in CCIE
Preparation
http://www.internetworkexpert.com
Subscription information
may be found
at:
http://www.ieoc.com/forums/ForumSubscriptions.aspx
Thanks for the advice