Successor and Feasible Successor-EIGRP

closetgeek

I'm have a bit of trouble understanding the EIGRP successor concept...can anyone give some insight on the matter.

luke_bibby

The successor route is the best route to reach a subnet, based on the advertised distance (AD) from the neighbor plus the distance to reach that neighbor. This is the route which is installed in the routing table.

The feasible successor route is a route which has a higher metric than the successor route to reach a subnet but meets the feasibility condition and can be used in the event that the successor route goes down. This route does NOT get installed in the routing table but is kept in the topology table. The feasibility condition states that the AD from a neighbor must be less than the metric of the successor route (the feasible distance [FD]) because routing through a feasible successor when the AD > FD may cause a routing loop.

All other routes (non-successor and non-feasible successor) are called possibilities are kept in the topology table but are only visible when u supply the all-links argument to the show ip eigrp topology command.

Hope that helped

EDIT: just noticed you didn't ask about feasible successor and possibility routes :P oh well

closetgeek

Thanks that helped...

Forsaken_GA

First thing to make sure of when you're dealing with EIGRP - Do *NOT* confuse Administrative Distance, which is what IOS regards as the 'believeability' of a routing protocol, with Advertised Distance or Feasible Distance. I see that mistake all the time.

This is the way it works in EIGRP, in a nutshell -

The successor is the best route to a given network. It's what you will see in the routing table output of show ip route. It has the best metric to that network.

A feasible successor is a backup route. This is an important concept in EIGRP - a backup route can be used IMMEDIATELY if the primary route (the successor) goes down. There's no need to disrupt traffic flow, it just starts shoving traffic down that other route instead. If a route does not have a feasible successor, the route transitions to the active state and starts asking other routers if they have a route to the network.

To analogize -

Let's say you're a smoker. And you go outside to have a smoke. Most of the time, you have 2 packs on you. You smoke out of the pack you already have opened (your opened pack is your successor route). But every once in awhile you go downstairs to smoke and you forget your opened pack (your successor route goes down). Since you have a 2nd pack of smokes, you open that up and you start smoking (you make use of your feasible successor). If you didn't have your second pack of smokes, you'd have to ask the people around you for a smoke (you've gone into active for the network and are querying your neighbors for a route).

Now, what determines a feasible successor is a bit trickier.

Let's say your successor route has an advertised distance of 90, and your distance from that router is 10. 90 + 10 = 100, so the feasible distance is 100. In order for a route to qualify as a feasible successor, the advertised distance of the route has to be less than the feasible distance of the successor. This is necessary to ensure a loop free path. So if another router is advertising the same route with an Advertised Distance of 95, that's less than the Feasible Distance of the current successor, so it would qualify as a feasible successor. If a third router was advertising the route with an Advertised Distance of 100 or greater, that would not qualify as a feasible successor, since the Advertised Distance is not less than the Feasible Distance.

Hopefully, that's not too confusing. Once you understand how this works, it'll be a little easier to glom onto the concepts of multiple successors, either via equal cost load balancing, or the variance settings, but trying to gather it all in at once can result in the utterance of an explicative or nine.

luke_bibby

Love the smokers analogy

Forsaken_GA

Honestly, that's one of the most effective ways I learn. When I'm having trouble grasping a concept, I sit down, and force myself to think how I would explain the concept to someone who has absolutely no background in networking, in a way they would understand. This is easy for me to do, fortunately, as the majority if my coworkers are very smart people with lots of experience in IT, but weak networking backgrounds. They tolerate me educating them so I can educate myself because it betters their own understanding. Putting it in terms a layman can understand and using real world situations as analogies forces me to learn the material well enough to be able to articulate it.

It's also a damned good customer service skill. When you start spouting tech terms at customers who aren't on the same page, the result is usually along the lines of 'Look, I don't care what the problem is, just FIX IT!!!'. If you can explain something to them in terms that doesn't make them feel like an idiot (even if they are) the reaction is more along the lines 'oh, that makes sense. Let me know when it's fixed, ok?'

closetgeek

That is a great concept, I will try that. Also, that Smokers analogy was awesome can you give one for equal cost and load balancing...

kryolla

Equal cost means the metric is the same for the same prefix. Load balancing is done per the switching method, process or cef. Load balancing is done for the data traffic and not control traffic such as routing updates.

jovan88

Forsaken_GA wrote: »

First thing to make sure of when you're dealing with EIGRP - Do *NOT* confuse Administrative Distance, which is what IOS regards as the 'believeability' of a routing protocol, with Advertised Distance or Feasible Distance.

Another way to not get confused, the term Advertised Distance is also known as Reported distance.

Jaz_10

So that's what it is!

Dilbert65

I have to use analogy's all the time when talking to customers that can barely use a mouse.

I have to give you a 9.75 on the smoker analogy to explain eigrp very nice

mrengineer

thnx man, i have some questions

which command view the FD
which command view cost for feasible route
what do u mean by distance in (10 is the distance) metric or cost?
if the distance for the feasible succ route is less than the distance for the succ route, why it is not the succ route?

i may make no sence but any answer can help

Project2501

Successor is the king. Feasible successor is the Prince waiting for the thrown.

CBT nuggets esq.

outrunred

You're obviously not form the UK - no one here can afford to have to packs of smokes.....not at over £6 per pack!

great analogy though....the beauty of networking is being able to fit it into real world, every day examples, something I employ a lot too... I imagine it's not so easy for coders, that'll teach them!

alexinthis

Hi,

I was wondering if someone could clear this up. I under stand why the route must be less than the feasible distance to avoid routing loops. However wouldn't this make it a better route to the network you are trying to get to? Why wouldn't it want to use this one, then add the old route as a backup?

Newbie z7

Forsaken_GA wrote: »

Now, what determines a feasible successor is a bit trickier.

Let's say your successor route has an advertised distance of 90, and your distance from that router is 10. 90 + 10 = 100, so the feasible distance is 100. In order for a route to qualify as a feasible successor, the advertised distance of the route has to be less than the feasible distance of the successor. This is necessary to ensure a loop free path. So if another router is advertising the same route with an Advertised Distance of 95, that's less than the Feasible Distance of the current successor, so it would qualify as a feasible successor. If a third router was advertising the route with an Advertised Distance of 100 or greater, that would not qualify as a feasible successor, since the Advertised Distance is not less than the Feasible Distance.

Forsaken,

I am a little overwhelmed with the whole concept of FD, but your description helped a lot. Could you expalin a little further for me on how figure out the distance from the router? for example if I have an IP of 10.10.8.0/23 (90/3847680) and need the successor network 1) how would I start to figure it out and then 2) how would I figure out the FD?

Thanks

Forsaken_GA

Newbie z7 wrote: »

Forsaken,

I am a little overwhelmed with the whole concept of FD, but your description helped a lot. Could you expalin a little further for me on how figure out the distance from the router? for example if I have an IP of 10.10.8.0/23 (90/3847680) and need the successor network 1) how would I start to figure it out and then 2) how would I figure out the FD?

Thanks

It's not in the routing table, you need to look at the eigrp topology with show ip eigrp topology. Stretch over at packetlife has an example readily available, so I'll steal his!


R3# show ip eigrp topology
IP-EIGRP Topology Table for AS(1)/ID(3.3.3.3)

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

P 192.168.5.0/24, 1 successors, FD is 3072
via 10.0.35.5 (3072/512), Serial1/3
via 10.0.34.4 (4352/1792), Serial1/2


Ok, the top part of the route tells you the Feasible Distance for the route is 3072.

In the two routes listed below that, there are two numbers. These are the feasible distance and the advertised distance. The advertised distance is what the remote router says is the distance to that route, and the feasible distance is the remote distance + the cost of the link to reach the router that's advertising that advertised distance.

So for the path through 10.0.35.5, the AD is 512, and when the router figures its own cost for that link in, the Feasible Distance becomes 3072.

For the second route through 10.0.34.4, the remote router is advertising a distance of 1792, and when the current router adds its own cost of the link to that, the feasible distance becomes 4352. 3072 is "closer" (ie, lower) than 4352, so it becomes the successor route. But since the second link's Advertised distance is 1792, and 1792 is less than 3072, the route becomes a feasible successor (remember, the comparison is advertised distance vs. feasible distance). If the advertised distance for the second route was, say, 4092 instead of 1792, then an AD of 4092 is greater than the FD of 3072, so it wouldn't qualify as a feasible successor

chX

alexinthis wrote: »

Hi,

I was wondering if someone could clear this up. I under stand why the route must be less than the feasible distance to avoid routing loops. However wouldn't this make it a better route to the network you are trying to get to? Why wouldn't it want to use this one, then add the old route as a backup?

Let's say we have two routes. Two different routers, both 100 away from the router you're on.

First route has AD of 200. 200 + 100 = 300 FD.

Second route has AD of 250. 250 + 100 = 350 FD.

The first route becomes the successor (lower FD - 300).
The second route can become a feasible successor, as its AD (250) is less than the FD (300) of the successor.

The second route wouldn't become the successor, because it still has a higher FD.

HTH, and someone let me know if any of the above is wrong.

technomania

can some one explain to me how a routing loop would be formed if the AD of the feasable successor is greater than the FD? Hoping to get an answer soon.

bermovick

Wow; major thread necro.

Just to clarify - it's not that a routing loop WOULD be formed, but that one COULD be formed. EIGRP doesn't keep a link-state database like OSPF, it just knows the cost (FD) to the destination. If the AD received from a neighbor were higher than the FD, then it could be because that route has gone through the router already, been advertised out, then looped around and come back in.

mautoncman

The below example confuses me. It clearly states that there should be an example of a reported distance that is not less then the feasible distance in the "show ip eigrp topology table below. According to what I thought i knew before this lab, I cannot see one in the below example. Could someone be as kind as to point out what it is that i am missing ? I have bold faced the way i am comparing. Thank you in advance.

These unequal-cost routes also show up in the EIGRP topology table, even
though they are not considered successor routes (their reported distance is not
less than the feasible distance). Check this with the output of the show ip eigrp
topology command.
R3# show ip eigrp topology
IP-EIGRP Topology Table for AS(100)/ID(10.1.3.9)

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

P 10.1.3.8/30, 1 successors, FD is 128256
via Connected, Loopback39
P 10.1.2.8/30, 1 successors, FD is 20642560
via 10.1.103.1 (20642560/156160), Serial0/0/0
via 10.1.203.2 (40640000/128256), Serial0/0/1
P 10.1.1.8/30, 1 successors, FD is 20640000
via 10.1.103.1 (20640000/128256), Serial0/0/0
via 10.1.203.2 (40642560/156160), Serial0/0/1
P 10.1.3.0/30, 1 successors, FD is 128256
via Connected, Loopback31
P 10.1.2.0/30, 1 successors, FD is 20642560
via 10.1.103.1 (20642560/156160), Serial0/0/0
via 10.1.203.2 (40640000/128256), Serial0/0/1
P 10.1.1.0/30, 1 successors, FD is 20640000
via 10.1.103.1 (20640000/128256), Serial0/0/0
via 10.1.203.2 (40642560/156160), Serial0/0/1
P 10.1.3.4/30, 1 successors, FD is 128256
via Connected, Loopback35
P 10.1.2.4/30, 1 successors, FD is 20642560
via 10.1.103.1 (20642560/156160), Serial0/0/0
via 10.1.203.2 (40640000/128256), Serial0/0/1
P 10.1.1.4/30, 1 successors, FD is 20640000
via 10.1.103.1 (20640000/128256), Serial0/0/0
via 10.1.203.2 (40642560/156160), Serial0/0/1
P 10.1.103.0/29, 1 successors, FD is 20512000


P 10.1.102.0/29, 1 successors, FD is 21024000
via 10.1.103.1 (21024000/20512000), Serial0/0/0
via 10.1.203.2 (41024000/20512000), Serial0/0/1
P 10.1.203.0/29, 1 successors, FD is 40512000
via Connected, Serial0/0/1
P 10.1.200.0/29, 1 successors, FD is 20514560
via 10.1.103.1 (20514560/28160), Serial0/0/0
via 10.1.203.2 (40514560/28160), Serial0/0/1

EdTheLad

mautoncman wrote: »

The below example confuses me. It clearly states that there should be an example of a reported distance that is not less then the feasible distance in the "show ip eigrp topology table below. According to what I thought i knew before this lab, I cannot see one in the below example. Could someone be as kind as to point out what it is that i am missing ? I have bold faced the way i am comparing. Thank you in advance.

A reported distance that is not less than the feasible distance, i.e. a reported distance greater than the feasible distance, this breaks the feasible condition, which means the route is not a feasible successor.
Only successors and feasible successors appear when using the command "show ip eigrp topology"

If you want to see all routes including non feasible successors, you must issue the command "show ip eigrp topology all-links".

mautoncman wrote: »

These unequal-cost routes also show up in the EIGRP topology table, even
though they are not considered successor routes (their reported distance is not
less than the feasible distance). Check this with the output of the show ip eigrp
topology command.
R3# show ip eigrp topology
IP-EIGRP Topology Table for AS(100)/ID(10.1.3.9)

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

P 10.1.3.8/30, 1 successors, FD is 128256
via Connected, Loopback39
P 10.1.2.8/30, 1 successors, FD is 20642560
via 10.1.103.1 (20642560/156160), Serial0/0/0
via 10.1.203.2 (40640000/128256), Serial0/0/1
P 10.1.1.8/30, 1 successors, FD is 20640000
via 10.1.103.1 (20640000/128256), Serial0/0/0
via 10.1.203.2 (40642560/156160), Serial0/0/1
P 10.1.3.0/30, 1 successors, FD is 128256
via Connected, Loopback31
P 10.1.2.0/30, 1 successors, FD is 20642560
via 10.1.103.1 (20642560/156160), Serial0/0/0
via 10.1.203.2 (40640000/128256), Serial0/0/1
P 10.1.1.0/30, 1 successors, FD is 20640000
via 10.1.103.1 (20640000/128256), Serial0/0/0
via 10.1.203.2 (40642560/156160), Serial0/0/1
P 10.1.3.4/30, 1 successors, FD is 128256
via Connected, Loopback35
P 10.1.2.4/30, 1 successors, FD is 20642560
via 10.1.103.1 (20642560/156160), Serial0/0/0
via 10.1.203.2 (40640000/128256), Serial0/0/1
P 10.1.1.4/30, 1 successors, FD is 20640000
via 10.1.103.1 (20640000/128256), Serial0/0/0
via 10.1.203.2 (40642560/156160), Serial0/0/1
P 10.1.103.0/29, 1 successors, FD is 20512000


P 10.1.102.0/29, 1 successors, FD is 21024000
via 10.1.103.1 (21024000/20512000), Serial0/0/0
via 10.1.203.2 (41024000/20512000), Serial0/0/1
P 10.1.203.0/29, 1 successors, FD is 40512000
via Connected, Serial0/0/1
P 10.1.200.0/29, 1 successors, FD is 20514560
via 10.1.103.1 (20514560/28160), Serial0/0/0
via 10.1.203.2 (40514560/28160), Serial0/0/1

Which routes? they all obey the feasible condition.

mautoncman

Thank you, I was just trying to confirm. The lab states that there are routes in the topology table were reported distance that is not less then the feasible distance, but I do not see any. They all look good to me. I needed a second opinion. Thank you.

Magic Johnson

mautoncman wrote: »

Thank you, I was just trying to confirm. The lab states that there are routes in the topology table were reported distance that is not less then the feasible distance, but I do not see any. They all look good to me. I needed a second opinion. Thank you.

Even if you issue the 'all-links' command?

Magic Johnson

p.s you're not the only one struggling with this. I think the really large metric numbers don't help things!

Magic Johnson

Quick Q: If EIGRP has two exact same metric routes to a subnet, do both of them go in to the routing table?

2nd Quick Q: If there are two exact same routes that meet the FS condition do they both become feasible successors? The chances of this are slim though aren't they? But what if...?

I've seen some things and some questions asked that have confused the hell out of me when trying to look in to this further on Google.