Not very Feasible...

volfkhat

Hola,

I recently started reading about EIGRP (more in-depth)... and have immediately gotten lost.

Can someone please explain to me the difference/relation between

~ Successor Route
~ Feasible Successor Route

~ Feasible Distance
~ Reported Distance

?

i understand that the "Successor" route is your main route.
And the the Feasible Successor is your backup route.

That makes sense.

But i immediately get confused when it comes to understanding the Feasible/Reported/Administrative Distance aspect.

Not too happy

verbhertz

Feasible distance is the metric EIGRP uses for the successor route to a prefix. It is a combination of reported distance (what EIGRP learns from the next-hop router on that path) and the cost of the link between routers (or the first hop so to speak). Administrative Distance is what the router uses when comparing routes to the same prefix from different protocols (or static, connected, etc routes). If you were only using one routing protocol and had no static routes, then AD wouldn't come into play.

The terminology for EIGRP is definitely confusing. The reported distance is important because for a route to be a backup (feasible successor route) the reported distance must be less than the current feasible distance. This is a loop prevention mechanism of EIGRP.

volfkhat

verbhertz wrote: »

The reported distance is important because for a route to be a backup (feasible successor route) the reported distance must be less than the current feasible distance. This is a loop prevention mechanism of EIGRP.

WOW... that is Deep.

I studied your post for about 10 minutes... but i think i Finally get it.

So... To put it in my own words:
1st-hop Distance = My 'cost' to get to You.
Reported Distance = Your 'cost' to get to the Destination.

Feasible Distance = the total of Both Costs together.
?

It really makes sense when i draw out a picture:


Based on this example... it's easy to see why 'Red' can Not be a backup route.

TWX

Yeah, honestly, the terminology sucks. "sucessor route" is the route chosen, based on "feasible distance" from your router all of the way to the destination network. "feasible successor route" is a route whose reported distance (ie, minus your first-hop) is better than the feasible distance (ie, with the first-hop included in the calculation) of the successor route (ie, the current route) but whose feasible distance is worse than that of the successor route.

CiscoWayne

TWX wrote: »

Yeah, honestly, the terminology sucks. "sucessor route" is the route chosen, based on "feasible distance" from your router all of the way to the destination network. "feasible successor route" is a route whose reported distance (ie, minus your first-hop) is better than the feasible distance (ie, with the first-hop included in the calculation) of the successor route (ie, the current route) but whose feasible distance is worse than that of the successor route.

Oh no I've gone cross-eyed......

That explanation is spot on, but my god did they make it difficult to explain when they named these things lol

volfkhat

TWX wrote: »

Yeah, honestly, the terminology sucks. "sucessor route" is the route chosen, based on "feasible distance" from your router all of the way to the destination network. "feasible successor route" is a route whose reported distance (ie, minus your first-hop) is better than the feasible distance (ie, with the first-hop included in the calculation) of the successor route (ie, the current route) but whose feasible distance is worse than that of the successor route.

and JUST like that... i'm LOST again.
lol

In one of the videos on Danscourses.com,
he used the terminology:

Successor route = BEST route.
Feasible Successor route = NEXT Best route(s).

Feasible Distance = TOTAL Cost.


So,
to rephrase your post:
the BEST route is the route chosen, based on the "TOTAL Cost" from your router all of the way to the destination network.
The "NEXT BEST" route is a route whose reported COST is LESS than the TOTAL COST of the CURRENT route.

~ This makes sense to me.... but i bet its only adding to the Confusion :]


Oh, and,
"(ie, the current route) but whose feasible distance is worse than that of the successor route."
i just had to completely ignore this last part ;]

pevangel

You got it. The last part is a given.

The terms can be confusing especially when, instead of reported distance, they sometimes use advertised distance or AD which people gets confused with administrative distance.

james43026

I'm going to break this down in what I think is a easy to read and understand way. Let say we are talking about getting to network 10.0.0.0/16 here

Reported distance: This is the distance that your neighbor reports that have to go in order to reach the network in question
Feasible distance: This is the reported distance that your neighbor sent to you, plus the distance between you and that same neighbor.

Successor: The best route to a network, IE lowest feasible distance.
Feasible Successor? The next best route to the same network. There can be multiple Feasible Successors, you aren't limited to just one. But for a route to become a feasible successor it must pass the feasibility condition= the reported distance of a feasible successor must be less than the feasible distance of the successor route. As long as a route meets this feasibility condition is is placed into the EIGRP topology table and used as either a backup route in case the successor route fails, or load balancing will occur based on several parameters.

TWX

CiscoWayne wrote: »

Oh no I've gone cross-eyed......

That explanation is spot on, but my god did they make it difficult to explain when they named these things lol

volfkhat wrote: »

and JUST like that... i'm LOST again.
lol

I guess that I was able to type that out without having to refer back to the reference material means I've got the concept including the naming conventions down... *grin*


I also dislike some of the terminology for spanning-tree. WHO thought it was a good idea to use "root" in two related but dissimilar ways?!

volfkhat

pevangel wrote: »

You got it. The last part is a given.

The terms can be confusing especially when, instead of reported distance, they sometimes use advertised distance or AD which people gets confused with administrative distance.

No kidding. This terminology Sucks!!

But i think i GOT it! i really do :]
In fact, i have tweaked my previous post (based on the feedback):

Feasible Successor route = NEXT Best route(s)
~added the 's' on the end.


Feasible Distance = TOTAL Cost.
~ It's not good to define a Term by using the SAME term in the definition.
Thus, I switched 'Distance' into 'Cost'.

Which also means:
the BEST route is the route chosen, based on the "TOTAL Cost" from your router all of the way to the destination network.
The "NEXT BEST" route is a route whose reported COST is LESS than the TOTAL COST of the CURRENT route.

~ this 'translation' suddenly become more comprehendable :]
(or not)

verbhertz

It sounds like you've got the hang of it!

theodoxa

Reported Distance (RD): Also known as "Advertised Distance". The metric reported by the neighbor.

Computed Distance (CD): The metric for the route through a given neighbor. The RD is the metric the neighbor is using, while the CD is the metric you are using.

Feasible Distance (FD): A locally significant (not advertised) value used to determine if a route is loop free. It is the lowest Computed Distance for a destination since the last time the route went from the Active state into the Passive state. This will typically be the same as the Computed distance of the Successor, but could be lower if the lowest Computed Distance changed without the route going Active. The specifics of this (CD changing without a route going Active) are covered in the CCIE OCG and are beyond the scope of the CCNA. The Feasible distance will be listed under the "show ip eigrp topology" command.

Feasible Successor: All routes to a destination which meet the Feasibility Condition. When a neighbor goes down, EIGRP will first determine the next best route and then check whether or not it is a Feasible Successor. If it is, the Feasible Successor will become the Successor route. If it isn't (even if there are Feasible Successors available), then the route will go Active (the Active and Passive states and their details are a CCNP topic if I remember correctly and are beyond the scope of the CCNA).

Successor: The current best route for a destination in EIGRP.

Feasibility Condition: Is the Reported Distance for a destination through a given neighbor lower than the current Feasible Distance for that destination? If it is, then a routing loop cannot exist. If the reported distance is higher than the feasible distance, then it is possible that using the neighbor would result in a routing loop.

james43026

theodoxa wrote: »

Reported Distance (RD): Also known as "Advertised Distance". The metric reported by the neighbor.

Computed Distance (CD): The metric for the route through a given neighbor. The RD is the metric the neighbor is using, while the CD is the metric you are using.

Feasible Distance (FD): A locally significant (not advertised) value used to determine if a route is loop free. It is the lowest Computed Distance for a destination since the last time the route went from the Active state into the Passive state. This will typically be the same as the Computed distance of the Successor, but could be lower if the lowest Computed Distance changed without the route going Active. The specifics of this (CD changing without a route going Active) are covered in the CCIE OCG and are beyond the scope of the CCNA. The Feasible distance will be listed under the "show ip eigrp topology" command.

Feasible Successor: All routes to a destination which meet the Feasibility Condition. When a neighbor goes down, EIGRP will first determine the next best route and then check whether or not it is a Feasible Successor. If it is, the Feasible Successor will become the Successor route. If it isn't (even if there are Feasible Successors available), then the route will go Active (the Active and Passive states and their details are a CCNP topic if I remember correctly and are beyond the scope of the CCNA).

Successor: The current best route for a destination in EIGRP.

Feasibility Condition: Is the Reported Distance for a destination through a given neighbor lower than the current Feasible Distance for that destination? If it is, then a routing loop cannot exist. If the reported distance is higher than the feasible distance, then it is possible that using the neighbor would result in a routing loop.

You can see discrepancies between computed distance and feasible distance if you were to change the delay for a route, and then check the topology table of a neighbor, you would see that the feasible distance hasn't changed, yet the metric directly next to the route in the topology table has changed, you can even see this with the use of a offset list as well, and then you only need one router to check it out. It's important to also know that the metric (calculated distance) included with the route in the topology table is the metric that the router will use in the route table, and is the distance that will be advertised to neighbors, not the feasible distance.

volfkhat

theodoxa wrote: »

Computed Distance (CD): The metric for the route through a given neighbor. The RD is the metric the neighbor is using, while the CD is the metric you are using.

Okay...... so what does that mean exactly?

Is it saying that the Computed Distance is:
from "ME" to "YOU".

or is it saying:
from "ME" to the "DESTINATION".
?

EdTheLad

ME to Destination

volfkhat

Thanks Ed!

Thenyo

I can configure EIGRP and have been in thorough lab environments for about a year yet I've NEVER understood all this successor/feasible successor mumbo jumbo. After reading this thread, I finally have a grasp on whats taking place here. Thank you guys SO much. I am happy to have found this community!

TWX

Thenyo wrote: »

I can configure EIGRP and have been in thorough lab environments for about a year yet I've NEVER understood all this successor/feasible successor mumbo jumbo. After reading this thread, I finally have a grasp on whats taking place here. Thank you guys SO much. I am happy to have found this community!

Great!

So now, can you explain it to us? *grin*

Thenyo

TWX wrote: »

Great!

So now, can you explain it to us? *grin*

Let me try lol. Correct me if I am completely wrong here.


The feasible distance is the total distance between you and and the destination plus the total distance between your neighbor and the destination. A feasible successor is selected based on if its meets the feasible condition, which is if the RD of that successor is less than the current FD... Did I just totally butcher this explanation?

EdTheLad

Thenyo wrote: »

Did I just totally butcher this explanation?

Yes

The feasible distance is your lowest/best cost path to the destination.
The reported distance is your neighbors feasible distance i.e. their lowest/best path to the destination.

Why do i care about my neighbors feasible distance?
1) i need to go via a neighbor unless i'm directly connected to the destination, so their best cost plus my exit interface cost to them equals my lostest cost via them.
2) If the neighbor has a lower best cost to the destination then i know i'm 100% sure i'm not looping back to myself.


My advise is, get the logic of why and how the feasible condition works before learning the terms, it's very easy to understand and remember after that.
Ask yourself, what is a loop? how does the feasible condition avoid a loop.

Thenyo

EdTheLad wrote: »

Yes

The feasible distance is your lowest/best cost path to the destination.
The reported distance is your neighbors feasible distance i.e. their lowest/best path to the destination.

Why do i care about my neighbors feasible distance?
1) i need to go via a neighbor unless i'm directly connected to the destination, so their best cost plus my exit interface cost to them equals my lostest cost via them.
2) If the neighbor has a lower best cost to the destination then i know i'm 100% sure i'm not looping back to myself.


My advise is, get the logic of why and how the feasible condition works before learning the terms, it's very easy to understand and remember after that.
Ask yourself, what is a loop? how does the feasible condition avoid a loop.

Thank you. Once my semester's projects and finals are over I'm gonna start sharpening my EIGRP knowledge first. Followed by a few other things I need to brush up on.



Feasible Distance: My lowest/best cost to the destination.
Reported Distance: Neighbor's Feasible distance.

So if the neighbor's FD (aka the RD) is lower than my FD, it is a viable successor. If the neighbors FD is higher than my current FD, then there is a possibility of a loop occurring, therefor making making it a bad candidate for being a successor. ???

TWX

I think this diagram is right. Please comment.

EdTheLad

Thenyo wrote: »

Feasible Distance: My lowest/best cost to the destination.
Reported Distance: Neighbor's Feasible distance.

So if the neighbor's FD (aka the RD) is lower than my FD, it is a viable successor. If the neighbors FD is higher than my current FD, then there is a possibility of a loop occurring, therefor making making it a bad candidate for being a successor. ???

Exactly, easy when understand the logic .

EdTheLad

[TWX] Worst route? there can be many different paths to a destination, so using the term worst route is bad. Your second worst statement is wrong, the AD of the successor is irrelevant to all other candidates in regards to the feasible condition.

If a neighbor advertises a route to the destination with a better cost than my best path to that same destination, i know that path cannot be looping back to me, hence it is a feasible successor. Stop read that statement again, don't move on until you understand that statement.
If the neighbor did advertise it's best cost i.e. its RD to me with a cost higher than my best cost FD, what would that tell you?

TWX

EdTheLad wrote: »

[TWX] Worst route? there can be many different paths to a destination, so using the term worst route is bad. Your second worst statement is wrong, the AD of the successor is irrelevant to all other candidates in regards to the feasible condition.

If a neighbor advertises a route to the destination with a better cost than my best path to that same destination, i know that path cannot be looping back to me, hence it is a feasible successor. Stop read that statement again, don't move on until you understand that statement.
If the neighbor did advertise it's best cost i.e. its RD to me with a cost higher than my best cost FD, what would that tell you?

Not worst, worse. If it were the best route it would be the successor route. We've already established the successor route, so now we're establishing if the worse route (or any given other worse route) is a feasible successor or not.

I understand the statement fine. Wouldn't, by nature of the math, it follow that if the AD of a lesser-route is less than the FD, that the route could not pass through the router doing the comparison, as being less than the FD would exclude the router doing the comparison from being able to be part of the advertised route?

EdTheLad

Correct, so then you shouldn't have any issue with understanding the FC.

volfkhat

TWX wrote: »

Not worst, worse. If it were the best route it would be the successor route. We've already established the successor route, so now we're establishing if the worse route (or any given other worse route) is a feasible successor or not.

I found that using better "verbage" helped me understand:

Successor route = BEST route.
Feasible Successor route = NEXT Best route(s).

Feasible Distance = your TOTALCost.
Reported Distance = your Neighbor's Total Cost.

(i think "worse/worst" mucks things up)

Plus with my mspaint illustration... it kind of makes sense
(for me, at least)

TWX

volfkhat wrote: »

I found that using better "verbage" helped me understand:

Successor route = BEST route.
Feasible Successor route = NEXT Best route(s).

Feasible Distance = your TOTALCost.
Reported Distance = your Neighbor's Total Cost.

(i think "worse/worst" mucks things up)

Plus with my mspaint illustration... it kind of makes sense
(for me, at least)

It might. Maybe "other route" would be a less confusing way of looking at it.

Thing of it is, questions could ask if a route actually is a feasible successor or not, so until comparing the advertised distances and feasible distances one can't really label a route other than as not being as good as the successor.

theodoxa

Thenyo wrote: »

I can configure EIGRP and have been in thorough lab environments for about a year yet I've NEVER understood all this successor/feasible successor mumbo jumbo. After reading this thread, I finally have a grasp on whats taking place here. Thank you guys SO much. I am happy to have found this community!

You're not alone. Cisco seems to find a way to go even deeper and in many cases cover exceptions to what you had previously learned with every higher certification.