EIGRP K Values

n3twrk0p

[FONT=Verdana, Arial, Helvetica, sans-serif]Hello everyone,[/FONT]

[FONT=Verdana, Arial, Helvetica, sans-serif]I would like to raise a question regarding the K values of EIGRP. When manipulating the delay parameter in EIGRP most of the forums I see everyone taking off the bandwidth [ K1 ] and working only with the delay value [ K3 ] ...why is that? should the bandwidth always be OFF when working with delay or you can leave the K1 on?[/FONT]

f0rgiv3n

I'm not sure why they turn it off but you can turn on ALL the K values if you want. No restrictions on "either or". Any or all more applies in this situation.

networker050184

Most people turn off the bandwidth when labbing because it makes it easier to manipulate the metric as its a simple cumulative number that you can wrap your head around.

MrXpert

You can just leave K3 delay value on by itself as was mentioned earlier it is easier manipulate
I found using bandwidth and delay was better because if you start using these from day one it encourages you to work out the distances and i think it really helps your understanding better. It is also more likely this is what you will encounter by default in real life. Know harm trying both but focus more on the defaults. Just my opinion of course.

n3twrk0p

The reason I'm asking is because when I work with the Delay [ K3 ] parameter I'm able to performed load balancing with whatever ratio. Now when I include the bandwidth there is no way I can come up with a desire ratio either I'm using an incorrect formula or adding the bandwidth just wont work.

For example:

I'm setting a ratio of 2:1 for every 2 packets sent over an Ethernet interface, 1 will be sent over a Serial interface to destination network x.x.x.x with a FD of 153600.

Formula being used when only the K3 parameter is used:

153600 x (2/1)=Less preferred metric
307200=256x(delay/10)
307200/256=delay/10
1200x10=delay
12000=delay

12000 is the delay from end to end

12000 - 5000 = 7000 in tens of microseconds

Under the serial interface I added delay of 700 giving me a 2:1 ratio from R2 to R1 P2P links. This only works when using the DELAY parameter (Metric weight 0 0 0 1 0 0)


Now when I add the bandwidth (Metric weight 0 1 0 1 0 0) with the following formula:
(Metric/256 - 10^7/Bw)x10=Delay

the numbers just don't add up. Could some one please show me how could I do this using the bandwidth as well to manipulate the delay. Again I see too many forums just showing how to perform the ratio calculation showing only with the delay parameter "for clarity" but in a live environment i don't think you will be allow to turn off the bandwidth just because it will be easier to get the desire ratio.

Any help will be greatly appreciated. Thanks

n3twrk0p

Anyone?

I just need for someone with the bellow information to demonstrate how for example get a 2:1 Ratio using both parameters bandwidth and delay in their formula to come up with a DELAY Value:

I currently have two routers facing each other, using a serial interface with a delay of 25000 and an Ethernet interface using a delay of 6000 with a feasible distance of 409600 to reach destination network 10.10.10.0/24

Could someone please show how to get the right value in order to achieve a 2:1 ratio

R1 Config where destination network 10.10.10.0/24 resides:

R1#show run int e0/0Building configuration...


Current configuration : 82 bytes
!
interface Ethernet0/0
ip address 192.168.108.1 255.255.255.0
full-duplex
end


R1#




R1#show run int s1/3
Building configuration...


Current configuration : 85 bytes
!
interface Serial1/3
ip address 192.168.200.1 255.255.255.0
clockrate 128000
end


R1#


R1#show run int lo100
Building configuration...


Current configuration : 66 bytes
!
interface Loopback100
ip address 10.10.10.1 255.255.255.0
end


R1#

R2 Config where we will be changing the delay value to obtain a 2:1 ratio:

R2#show run int e0/0Building configuration...


Current configuration : 82 bytes
!
interface Ethernet0/0
ip address 192.168.108.2 255.255.255.0
full-duplex
end


R2#




R2#show run int s1/1
Building configuration...


Current configuration : 67 bytes
!
interface Serial1/1
ip address 192.168.200.2 255.255.255.0
end


R2#


R2#show ip protocol | i metric weight
EIGRP metric weight K1=1, K2=0, K3=1, K4=0, K5=0
R2#




R2#show ip eigrp top 10.10.10.0/24
IP-EIGRP (AS 100): Topology entry for 10.10.10.0/24
State is Passive, Query origin flag is 1, 1 Successor(s), FD is 409600
Routing Descriptor Blocks:
192.168.108.1 (Ethernet0/0), from 192.168.108.1, Send flag is 0x0
Composite metric is (409600/128256), Route is Internal
Vector metric:
Minimum bandwidth is 10000 Kbit
Total delay is 6000 microseconds
Reliability is 255/255
Load is 1/255
Minimum MTU is 1500
Hop count is 1

192.168.200.1 (Serial1/1), from 192.168.200.1, Send flag is 0x0
Composite metric is (20640000/128256), Route is Internal
Vector metric:
Minimum bandwidth is 128 Kbit
Total delay is 25000 microseconds
Reliability is 255/255
Load is 1/255
Minimum MTU is 1500
Hop count is 1
R2#




R2#show ip eigrp nei
IP-EIGRP neighbors for process 100
H Address Interface Hold Uptime SRTT RTO Q Seq
(sec) (ms) Cnt Num
0 192.168.108.1 Et0/0 13 00:03:36 403 2418 0 1
1 192.168.200.1 Se1/1 10 00:03:36 270 1620 0 2
R2#

Danielh22185

I just did a lab on this from the CCNP lab book found on this form. It was setting up a 2:1 ratio. I don't remember the exact commands but I know it was accomplished via manipulating the bandwidth K values. I'll go back and check when I can tonight to try and get the answer. Someone else might chime in by then. Also you can look at the same lab book for an answer.