eigrp routes with the same metric

aueddonline

ok to there are two links from r6 to r10 out of s0/0 and s0/2 I was looking at the route to subnet 10.10.10.20/30 and the metric is the same for both routes even though the bandwidths on the links are different?


R6_2611xm#show int s0/0
Serial0/0 is up, line protocol is up
Hardware is PowerQUICC Serial
Internet address is 10.2.2.33/27
MTU 1500 bytes, BW 768 Kbit, DLY 20000 usec,
reliability 255/255, txload 1/255, rxload 1/255
Encapsulation HDLC, loopback not set
Keepalive set (10 sec)
CRC checking enabled
Last input 00:00:01, output 00:00:00, output hang never
Last clearing of "show interface" counters never
Input queue: 0/75/0/0 (size/max/drops/flushes); Total output drops: 0
Queueing strategy: weighted fair
Output queue: 0/1000/64/0 (size/max total/threshold/drops)
Conversations 0/1/256 (active/max active/max total)
Reserved Conversations 0/0 (allocated/max allocated)
Available Bandwidth 576 kilobits/sec
5 minute input rate 0 bits/sec, 0 packets/sec
5 minute output rate 0 bits/sec, 0 packets/sec
2418 packets input, 188658 bytes, 0 no buffer
Received 465 broadcasts, 0 runts, 0 giants, 0 throttles
1 input errors, 0 CRC, 1 frame, 0 overrun, 0 ignored, 0 abort
1815 packets output, 111101 bytes, 0 underruns
0 output errors, 0 collisions, 3 interface resets
0 output buffer failures, 0 output buffers swapped out
0 carrier transitions
DCD=up DSR=up DTR=up RTS=up CTS=up

R6_2611xm#show int s0/2
Serial0/2 is up, line protocol is up
Hardware is PowerQUICC Serial
Internet address is 10.2.2.65/27
MTU 1500 bytes, BW 8000 Kbit, DLY 20000 usec,
reliability 255/255, txload 1/255, rxload 1/255
Encapsulation HDLC, loopback not set
Keepalive set (10 sec)
CRC checking enabled
Last input 00:00:00, output 00:00:03, output hang never
Last clearing of "show interface" counters never
Input queue: 0/75/0/0 (size/max/drops/flushes); Total output drops: 0
Queueing strategy: weighted fair
Output queue: 0/1000/64/0 (size/max total/threshold/drops)
Conversations 0/1/256 (active/max active/max total)
Reserved Conversations 0/0 (allocated/max allocated)
Available Bandwidth 6000 kilobits/sec
5 minute input rate 0 bits/sec, 0 packets/sec
5 minute output rate 0 bits/sec, 0 packets/sec
1058 packets input, 71542 bytes, 0 no buffer
Received 320 broadcasts, 0 runts, 0 giants, 0 throttles
0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored, 0 abort
1557 packets output, 127768 bytes, 0 underruns
0 output errors, 0 collisions, 3 interface resets
0 output buffer failures, 0 output buffers swapped out
3 carrier transitions
DCD=up DSR=up DTR=up RTS=up CTS=up

R6_2611xm#show ip eigrp topology
IP-EIGRP Topology Table for AS(1)/ID(6.6.6.6)

Codes: P - Passive, A - Active, U - Update, Q - Query, R - Reply,
r - reply Status, s - sia Status

P 10.10.10.10/32, 0 successors, FD is Inaccessible
via 10.2.2.66 (960000/128256), Serial0/2
via 10.2.2.34 (3973120/128256), Serial0/0
P 4.4.4.4/32, 1 successors, FD is 156160
via 10.1.3.129 (156160/128256), FastEthernet0/1
P 3.3.3.3/32, 1 successors, FD is 2300416
via 10.1.4.1 (2300416/156160), Serial0/1
P 1.1.1.1/32, 1 successors, FD is 2300416
via 10.1.4.1 (2300416/156160), Serial0/1
via 10.1.3.193 (2809856/2297856), Serial0/3
P 5.5.5.5/32, 0 successors, FD is Inaccessible
via 10.1.3.193 (2297856/128256), Serial0/3
via 10.1.4.1 (2809856/2297856), Serial0/1
P 2.2.2.2/32, 1 successors, FD is 2297856
via 10.1.4.1 (2297856/128256), Serial0/1
P 9.9.9.9/32, 2 successors, FD is 4485120
via 10.2.2.34 (4485120/3973120), Serial0/0
via 10.2.2.66 (4485120/3973120), Serial0/2
P 8.8.8.8/32, 0 successors, FD is Inaccessible
via 10.2.2.66 (46866176/46354176), Serial0/2
via 10.1.3.129 (46868736/46866176), FastEthernet0/1
via 10.2.2.34 (46866176/46354176), Serial0/0
via 10.1.3.193 (47890176/47378176), Serial0/3
via 10.1.4.1 (47892736/47380736), Serial0/1
P 6.6.6.6/32, 1 successors, FD is 128256
via Connected, Loopback6
P 10.1.3.0/26, 1 successors, FD is 2684416
via 10.1.4.1 (2684416/2172416), Serial0/1
via 10.1.3.193 (3193856/2681856), Serial0/3
P 10.1.2.0/25, 1 successors, FD is 2681856
via 10.1.4.1 (2681856/2169856), Serial0/1
P 10.2.2.0/27, 1 successors, FD is 2681856
via 10.2.2.66 (2681856/2169856), Serial0/2
via 10.2.2.34 (4357120/2169856), Serial0/0
P 10.1.4.0/26, 1 successors, FD is 2169856
via Connected, Serial0/1
P 10.10.10.24/30, 2 successors, FD is 47250176
via 10.2.2.34 (47250176/46738176), Serial0/0
via 10.2.2.66 (47250176/46738176), Serial0/2
P 10.10.10.16/30, 2 successors, FD is 46738176
via 10.2.2.34 (46738176/46226176), Serial0/0
via 10.2.2.66 (46738176/46226176), Serial0/2
P 10.10.10.20/30, 2 successors, FD is 4357120
via 10.2.2.34 (4357120/3845120), Serial0/0
via 10.2.2.66 (4357120/3845120), Serial0/2
P 10.2.2.32/27, 1 successors, FD is 3845120
via Connected, Serial0/0
P 10.1.3.64/26, 1 successors, FD is 2172416
via 10.1.3.193 (2681856/2169856), Serial0/3
P 10.2.2.64/27, 1 successors, FD is 832000
via Connected, Serial0/2
P 192.168.2.0/24, 1 successors, FD is 2172416
via 10.1.4.1 (2172416/28160), Serial0/1
P 10.2.2.96/27, 1 successors, FD is 28160
via Connected, FastEthernet0/0
P 10.1.3.128/26, 1 successors, FD is 28160
via Connected, FastEthernet0/1
P 10.1.2.128/25, 1 successors, FD is 2172416
via 10.1.4.1 (2172416/28160), Serial0/1
P 10.1.1.128/25, 1 successors, FD is 2681856
via 10.1.3.193 (2681856/2169856), Serial0/3
P 10.1.3.192/26, 1 successors, FD is 2169856
via Connected, Serial0/3

networker050184

EIGRP metric is calculated from the lowest bandwidth on the path and the cumulative delay. The best way to tweak EIGRP metrics is to use the delay.

aueddonline

That’s why, damp I knew that, these details slip your mind in the heat of the moment, I was thinking cumulative bandwidth, thanks man for the fast response

aueddonline

so what is best pratice for changing the delay on the interface

for instance I have a 768k link that I have changes the delay to 25,000 then I can use the vaience commad to allow eigrp to unequal cost load using the cofigure bandwith to determine the balance over the two.

but should I then change all the 768k links in my topology to delay 25,000

kryolla

if it is load balancing already why do you want to tweak it and then use the variance command the outcome is the same

networker050184

kryolla wrote:

if it is load balancing already why do you want to tweak it and then use the variance command the outcome is the same

For learning purposes. Thats the fun part of having a lab!

kryolla

networker050184 wrote:

kryolla wrote:

if it is load balancing already why do you want to tweak it and then use the variance command the outcome is the same

For learning purposes. Thats the fun part of having a lab!

it would make more sense to use the variance command on a prefix that only has 1 successor

P 10.2.2.0/27, 1 successors, FD is 2681856
via 10.2.2.66 (2681856/2169856), Serial0/2
via 10.2.2.34 (4357120/2169856), Serial0/0

networker050184

It would make more sense to leave all the defaults and have a perfect world, but you don't learn much that way

kryolla

networker050184 wrote:

It would make more sense to leave all the defaults and have a perfect world, but you don't learn much that way

It'll make things a lot more simpler LOL

dtlokee

aueddonline wrote:

so what is best pratice for changing the delay on the interface

for instance I have a 768k link that I have changes the delay to 25,000 then I can use the vaience commad to allow eigrp to unequal cost load using the cofigure bandwith to determine the balance over the two.

but should I then change all the 768k links in my topology to delay 25,000

Remember how EIGRP calculates it's metric and use that as the basis for how you will change the delay value. If you want to have some fun take 2 routers and connect them together with 2 serial links. Then set the bandwidth on one to 1544 (the default) and the other one to 768. Now tweak the delay values so that EIGRP will load share in a proportion of 2 to 5, 5 packets on the link with a bandwidth of 1544 for every 2 packets on the link that is 768.

You an see the load sharing ratio in the output of a "show ip route x.x.x.x" command

aueddonline

that's what I have set up two serials between two routers. one at 8mb one at 768k, the delay does have to be adjusted a lot for it to effect the metric proportionatly to the difference in the link speeds.

It's definatly worth getting in deep on this topic I think because it helps make sence of a lot of elements of eigrp

I'm going to play around with the 'traffic-share min across-interfaces' command

that 'show ip route (subnet)' command is just what i have been looking for, is there a way to see it in real time I tried a 'debug ip icmp' but it was only showing the reply's and not the interface the pings where being sent on.

networker050184

You can do a debug ip packet, but you will have to disable cef to see the load balancing happen. By default cef load balances on a per destination basis so all your pings will be to the same destination.