From: Kenneth Wygand (KWygand@customonline.com)
Date: Mon Apr 19 2004 - 13:29:06 GMT-3
Hey everyone,
I know this is an old post, but I just found some things out that seem
very important to clarify. In the below listed requirements, it says to
change the K values of the EIGRP weights such that RELIABILITY is
reflected in the EIGRP metric (in other words). The answer agreed upon
seemed to imply that you just need to change the K4 value to one and
leave the K5 value at 0 if you do not care about MTU and that should
satisfy the requirements. However, straight from the documentation CD,
K5 has to be set to a non-zero value if you want to include K4 in the
EIGRP metric calculation...
<snip>
If k5 equals 0, the composite EIGRP metric is computed according to the
following formula:
metric = [k1 * bandwidth + (k2 * bandwidth)/(256 - load) + k3 * delay]
If k5 does not equal zero, an additional operation is performed:
metric = metric * [k5/(reliability + k4)]
</snip>
So in summary, if you want to use reliability as a factor in determining
an EIGRP metric, you must not only set K4 to a non-zero value, but you
must also set K5 to a non-zero value.
Kenneth E. Wygand
Systems Engineer, Project Services
CISSP #37102, CCNP, CCDP, ACSP, Cisco IPT Design Specialist, MCP, CNA,
Network+, A+
Custom Computer Specialists, Inc.
"The only unattainable goal is the one not attempted."
-Anonymous
-----Original Message-----
From: Snow, Tim [mailto:timothy.snow@eds.com]
Sent: Monday, September 15, 2003 10:47 PM
To: 'Hunt Lee'; Snow, Tim
Cc: 'ccielab@groupstudy.com'
Subject: RE: EIGRP metrics
If you wanted to take reliability into consideration, you could set r4
=1
(aka, the variable is taken into consideration). The part to keep in
mind
is that Reliability changes if the line starts taking errors the
reliability
flucuates and would cause routing changes and introduce stability issues
in
the routing protocol.
As I read your question, I think I answered the part within quotes was
answered in my first paragraph. I believe by using "metric weights 0 0
0 0
1 1" will only take reliability and MTU (offset to reliability)
Yes, your statement is correct if you were to use the numbers I listed
above.
Yes, your answer about requirement appears to be correct, I would set k4
= 1
and k5 = 0
Where's Brian from internetworkexpert.com, he explained this awhile
back, I
wonder if he has his old post.
Tim
#12042
-----Original Message-----
From: Hunt Lee [mailto:huntl@webcentral.com.au]
Sent: Monday, September 15, 2003 8:39 PM
To: 'Snow, Tim'
Cc: 'ccielab@groupstudy.com'
Subject: RE: EIGRP metrics
Hi Tim,
Is there any scenario where you may wanted to set k4 & k5 to 1?
I have been working on a scenario, and it has both k4 & k5 set to 1.
"It has been seen lots of times, an ATM interface generates CRC errors
on
either R6 or R9 routers of this lab. Configure EIGRP such upon getting
such
errors on any particular router will automatically reflect to EIGRP
topology
of all routers in EIGRP AS."
Yet they only set k4 and k5 to 1, but not k2...
So does that mean the metric will only care about reliability and mtu,
but
not load?
Also, to fulfill the requirement, wouldn't it be better if I just set k4
=
1, but leave k5 as 0?
Thanks,
Hunt
-----Original Message-----
From: Snow, Tim [mailto:timothy.snow@eds.com]
Sent: Tuesday, 16 September 2003 11:29 AM
To: 'Nir Wittenberg'
Cc: 'ccielab@groupstudy.com'
Subject: RE: EIGRP metrics
Hopefully this well help, some of my eigrp notes when studying.
EIGRP uses a composite metric that is derived from the five submetrics.
When
EIGRP computes the composite metric, it uses a formula that involves
five
constants or "k" values. The constant values have default value such as
the
following:
k1 = k3 = 1 and k2 = k4 = k5 = 0
By setting k2, k4, and k5 to 0, it essentially nullifies the submetrics
of
load (k2) , reliability (k4) , and MTU (k5). This is precisely why you
should first use delay and then bandwidth when trying to influence which
routes EIGRP prefers. The formula EIGRP uses to calculate the composite
metric is as follows:
Metric = 256 x ([k1 x BWmim + (k2 x BWmim ) / (256-LOAD) + k3 x
DELAYsum ]
x X)
where the following is true:
BWmim = 107 / bandwidth_of_slowest_link DELAYsum =
(delays_along_the_path) X = k5 / (reliability + k4) if and only if
k1<>1,
if k1 = 1 then X = 1 With the k values set at the default value you have
k1 = k3 = 1 k2 = k4 = k5 = 0 CM = 256 x (BWmim + DELAYsum )
Using the default values of constants, k1 = k3 = 1 and k2 = k4 = k5 =0,
the
formula quickly breaks down to this:
(256 x [BWmim and DELAYsum])
Substituting the constants, you have the following:
CM = 256 x ([1 x BWmim + (0 x BWmim) / (256-LOAD) + 1 x
DELAYsum] x 1) CM = 256 x ([BWmim + (0) / (256-LOAD) +
DELAYsum] x 1) CM = 256 x (BWmim + DELAYsum)
NOTE
For reference, the metric is computed the same way for IGRP, except the
result of bandwidth and delay is not multiplied by 256, and the DELAY
sum
variable is divided by 10.
CM = (k1 x BWmin + [k2 x BWmin] / [256-LOAD] +
[k3 x DELAYsum] x X)
where the following is true:
BWmin = 107/ bandwidth_of_slowest_link DELAYsum
= S(delays_along_the_path) /10 X = k5 / (reliability + k4) if and only
if
k1<>1, if k1=1 then
X=1 k1=k3=1 k2=k4=k5=0
With k values set at the default value, you have:
CM = BWmin + DELAYsum
In short, TOS is always 0, "metric weights TOS BANDWIDTH LOAD DELAY
RELIABILITY MTU"
Tim
#12042
-----Original Message-----
From: Nir Wittenberg [mailto:nwittenberg@msncomm.com]
Sent: Monday, September 15, 2003 7:40 PM
To: ccielab@groupstudy.com
Subject: EIGRP metrics
Do K1, K2, K3, K4, K5 correspond to Bandwidth, Delay, Reliability, Load,
and
MTU respectively?
I ask because on a practice lab I was asked to have EIGRP take load into
account when determining a route metric.
My answer was:
metric weight 0 1 0 1 1 0
Their answer is:
metric weight 0 1 1 1 0 0
I have searched the archives and cisco, but still no clear answer.
TIA
Nir
***Get your CCIE and a FREE vacation: Shop.GroupStudy.com***
This archive was generated by hypermail 2.1.4 : Mon May 03 2004 - 19:48:49 GMT-3