EIGRP – IMPORTANT INFO FOR EXAM DAY on K weights and Redistribution, and using an offset-list for Metric tuning!

Only the NBMA and Ethernet segments will be used for quick demonstrations and clarity, unless R4 or R5 is needed for demonstration.

There are two different ways to change the Metric for EIGRP, one of those is by changing the “bandwidth” or “delay” on the interface, or by using an offset list.

Changing the Metric values of bandwidth or delay on an interface:

R4(config-if)#bandwidth ?
<1-10000000>   Bandwidth in kilobits
inherit        Specify how bandwidth is inherited
qos-reference  Reference bandwidth for QOS test
receive        Specify receive-side bandwidth

R4(config-if)#delay ?
<1-16777215>  Throughput delay (tens of microseconds)

R4(config-if)#do sh int fa0/1
FastEthernet0/1 is up, line protocol is up
Hardware is Gt96k FE, address is 001b.5336.f2cd (bia 001b.5336.f2cd)
Internet address is 172.12.34.4/24
MTU 1500 bytes, BW 100000 Kbit/sec, DLY 100 usec,
reliability 255/255, txload 1/255, rxload 1/255
Encapsulation ARPA, loopback not set

So “Delay” should be your first change if required to change either, as many things rely on “Bandwidth” such as QoS and other protocols, as well as changing the Feasible Distance to upstream routers.

However I’ve highlighted the odd Delay difference in configuration on the interface, and how it is displayed. It is displayed in Microseconds, but when configured is shown to be changed in units of “tens of microseconds” so to put this interfaces default delay of 100 on it the command would be “delay 10” on the interface.

Watch this Cisco gotcha on exam day, because that little oddity is begging for a question.

Speaking of oddities, I want to quickly demonstrate the curiosity of this is when you do “sh ip proto” by default you will see K1 and K3 on:

R4(config-router)#do sh ip proto
*** IP Routing is NSF aware ***

Routing Protocol is “eigrp 100”
  Outgoing update filter list for all interfaces is not set
  Incoming update filter list for all interfaces is not set
  Default networks flagged in outgoing updates
  Default networks accepted from incoming updates
  EIGRP-IPv4 Protocol for AS(100)
    Metric weight K1=1, K2=0, K3=1, K4=0, K5=0

Being that EIGRP Metric by default is determined by Bandwidth and Delay, when setting the metric for redistributing routes into EIGRP that it should be the first and third value according to “sh ip proto”, but that is not the case as shown redistributing RIP into EIGRP:

R4(config-router)#redistribute rip metric ?
  <1-4294967295>  Bandwidth metric in Kbits per second

R4(config-router)#redistribute rip metric 1 ?
  <0-4294967295>  EIGRP delay metric, in 10 microsecond units

R4(config-router)#redistribute rip metric 1 2 ?
  <0-255>  EIGRP reliability metric where 255 is 100% reliable

R4(config-router)#redistribute rip metric 1 2 3 ?
  <1-255>  EIGRP Effective bandwidth metric (Loading) where 255 is 100% loaded

R4(config-router)#redistribute rip metric 1 2 3 4 ?
  <1-65535>  EIGRP MTU of the path

So if you are being asked to redistribute into EIGRP this is a huge red flag as to whether you know this difference, the attributes go in this order:

• Bandwidth
• Delay
• Reliability
• Load
• MTU

So on exam day when it asks you to change the Bandwidth or Delay in Redistribution, you will know that it is first Bandwidth, then Delay, then all the other stuff that you really shouldn’t touch because we have offset-lists that work much better than changing parts of the EIGRP Metric Formula.

You can also use “metric weight 0 k1 k2 k3 k4 k5” in router configuration mode, however the formula is so complex it might make my brain collapse in on itself right now, so we will move on to the most feasible of all our options!

 

OFFSET-LISTS!

 

An offset-list is a way for EIGRP to come up with its own Metric, but having the local router add value to that Metric to adjust it, without changing any values in EIGRP itself calculates (seen below). This uses an access-list to define the traffic you want the offset-list to effect.

This allows you to modify the Distance for specific routes without effecting all of them like you would if you changed Bandwidth or Delay on an interface, in fact you don’t need to even specify an interface that the network is learned on.

Some very important notes regarding offset-lists before we get into configuration:

• Must use a ‘permit’ on the ACL for the offset-list to work, deny is invalid for offset-lists, so if you see this it is incorrect
• Offset value is added to both Feasible Distance and Reported Distance of Networks defined in the Access-List
• ONLY STANDARD ACL’S CAN BE USED – NO EXTENDED ACL’S FOR OFFSET-LISTS!
• Direction is also specified, whether you want incoming or outgoing routes to have their metric offset
• Offset-Lists can only be used to increase the Metric, they CANNOT be used to decrease it

So lets look at the configuration of this in EIGRP:

We will use 172.12.23.0/24 for this offset-list config, note its FD

R1#sh ip eigrp top
EIGRP-IPv4 Topology Table for AS(100)/ID(1.1.1.1)
Codes: P – Passive, A – Active, U – Update, Q – Query, R – Reply,
       r – reply Status, s – sia Status

P 172.12.123.0/24, 1 successors, FD is 2169856
        via Connected, Serial0/0/0
P 172.12.15.0/24, 1 successors, FD is 28160
        via Connected, FastEthernet0/1
P 172.12.23.0/24, 2 successors, FD is 2172416
        via 172.12.123.2 (2172416/28160), Serial0/0/0
        via 172.12.123.3 (2172416/28160), Serial0/0/0

Now we create the access-list, and apply it to our offset-list

R1#conf t
Enter configuration commands, one per line.  End with CNTL/Z.
R1(config)#access-list 23 permit 172.12.23.0 0.0.0.255
R1(config)#router eigrp 100
R1(config-router)#offset-list ?
  <1-99>       Access list of networks to apply offset (standard range)
  <1300-1999>  Access list of networks to apply offset (extended range)
  WORD         Access-list name

R1(config-router)#offset-list 23 ?
  in   Perform offset on incoming updates
  out  Perform offset on outgoing updates

R1(config-router)#offset-list 23 in ?
  <0-2147483647>  Offset

R1(config-router)#offset-list 23 in 1000

Now lets take another look at the FD for our paths to 172.12.23.0/24

R1(config-router)#do sh ip eigrp top
EIGRP-IPv4 Topology Table for AS(100)/ID(1.1.1.1)
Codes: P – Passive, A – Active, U – Update, Q – Query, R – Reply,
       r – reply Status, s – sia Status

P 172.12.123.0/24, 1 successors, FD is 2169856
        via Connected, Serial0/0/0
P 172.12.15.0/24, 1 successors, FD is 28160
        via Connected, FastEthernet0/1
P 172.12.23.0/24, 2 successors, FD is 2172416
        via 172.12.123.2 (2173416/29160), Serial0/0/0
        via 172.12.123.3 (2173416/29160), Serial0/0/0

Notice that the original / actual EIGRP Metric shows in the Topology Tables Successor entry for the route, however the offset FD is what will be shown in the IP Route Table:

R1(config-router)#do sh ip route eigrp

Gateway of last resort is not set

      2.0.0.0/32 is subnetted, 1 subnets
D        2.2.2.2 [90/2297856] via 172.12.123.2, 00:50:34, Serial0/0/0
      3.0.0.0/32 is subnetted, 1 subnets
D        3.3.3.3 [90/2297856] via 172.12.123.3, 00:50:34, Serial0/0/0
      172.12.0.0/16 is variably subnetted, 5 subnets, 2 masks
D        172.12.23.0/24 [90/2173416] via 172.12.123.3, 00:09:57, Serial0/0/0
                        [90/2173416] via 172.12.123.2, 00:09:57, Serial0/0/0
R1(config-router)#

Also, the offset Feasible Distance is what will be advertised to downstream routers, not the actual EIGRP Metric calculated by the router (for the networks defined in the ACL).

Another good thing to note, offset-lists cause Adjacencies to bounce, so applying them will cause that quick reset in neighbor relationships.

The off-set list is configured the same going “out” rather than in, except it will not take effect on the local router, but will change the Metric when advertising the network(s) outbound to other EIGRP routers.

One final note about offset-lists and I will be done with this post