EIGRP Help

zol192000

Hi i have a query please explain

How r the routing loops being made if the condition of feasible successor on EIGRP will not be fulfilled i.e
AD of feasible successor <FD if successor.....

Please explain it with example....???

Thanks

EdTheLad

A feasible successor is the next hop router i.e. the downstream router, the router that is closer to the destination.The AD of the FS is the cost to get from the FS to the destination.
The local FD is the best cost to the destination, this cost is made up of the "cost of the link between the local router and the next-hop router plus the AD of the next-hop router".

A

---

B

---

C

So if my path is from A to C, B will advertise its AD to A, A will add the AD and the link cost to get the FD.Its not possible that the AD can be greater than the FD in this case as the FD is comprised of the AD.If there was a loop on B and B was actually sending the traffic back to A the AD could be larger than the FD on A.So as long as you abide by the rule FD > AD you will be sure there are no loops.

zol192000

Dear u have mixed my question......


My question is how loops will be created if

Feasible distance(FD) of successor <Advertise distance(AD) of feasible successor


suppose this is the topology

s1 interface A

---

B

---

C
s2 interface A

---

D

---

C


our goal is to reach from A to C
Cost from A---B=40 B----c=30

Cost from A---D=30 D

---

C=80

Hence s1 interface will be successor
whose FD=70 and its AD=30

And s2 interface will not be feasible successor
whose FD=110 and its AD=80

now according to condition
AD of feasible successor <FD of successor
80!<70 (hence condition voilated)
and it ll not be feasible successor

now question arises
how the loops are created if AD of feasible successor >FD of successor and its still a feasible successor.
My mind says that router will confuse in which one should be in the routing table but how....???
Please ny1 xplain it to me....Its very urgent dudes

EdTheLad

In the topology you describe there is no loop, the rule is present to avoid a possible loop.
Since this rule applies to eigrp as a protocol it doesnt matter what your config or topology is like, the rule will still apply.
By having this rule in place you can be 100% sure the FS will not be in a feedback loop.If you cant understand this concept go back and study rip with poison reverse and holddown timers.

kryolla

if your neighbor has a higher metric to reach the final destination than you do then it is assumed the router is routing through you to get there so it will not be a FS. But if that is not the case then when the router sends a query for an alternate route, if it responds with I have an alternate way so use me, it will use it as a backup via a query and not via the topology table. So it is a loop prevention but IF there is no loop it can be used as a backup.

zol192000

Question arises why this rule is there and how routing loops will be made if the rule will be voilated.Its very important and urgent guys please help me

networker050184

I think they have both explained it pretty well. If you are still not understanding then maybe you should get a book with explanations and diagrams to help.

zol192000

Dear the question is still unanswered that how loops will be created if

Feasible distance of successor<Advertise distance of feasible successor

What happens if this rule is voilated and still feasible successor will be there in toplogy table....???


i knw it cannot be voilated but how loops will be made if in case the rule is voilated and still the other route is a feasible successor...??


My teacher who is CCIE has asked me this question n i m thinking over it frm last 2 days ..I hv time till tomorrow....

Plzz help me

Xenz

zol192000 wrote: »

i knw it cannot be voilated but how loops will be made if in case the rule is voilated and still the other route is a feasible successor...??

If it cannot be achieved, how do you expect us to break it? Your English is a major problem here as I have no clue what you're really asking. There is no way that a route could be a feasible successor to create a loop as DUAL would run when something changes and you could not have this condition. Why not ask your teacher for the answer?

mikej412

zol192000 wrote: »

What happens if this rule is voilated and still feasible successor will be there in toplogy table....???

You call TAC and report a bug in the IOS image and install a version where EIGRP works properly.

zol192000

My teacher will most probably give me reply of this question today or may be on monday

he asked two questions from me

first one is if null0 interface will not be in the routing table when we summarize eigrp how loops will be made then...??

and 2nd one is

why there is condition of
feasible successor AD<successor FD
in order to be a feasible successor....???

I have a clear idea of first question but i m not sure of 2nd and still imbiguity is there

---

nyhow thanks for everyone

suffah

I think people in this thread answered your second question already, but maybe this will help clarify.

You must realize that if a route has a higher AD than a successor's FD, it does not automatically mean there will be a loop. It only means that the POSSIBILITY of a loop exists.

zol192000

Yup u r ryt....but in case of Unequal path load sharing ,i guess there will always be a chance of loop if this law is ruled out and we have a feasible successor in toplogy table who is voilating this law......
correct me please i m wrong