frame relay question

[gleeb1234]

What is the advantage of using a multipoint interface instead of point-to-point subinterfaces when configuring a Frame Relay hub in a hub - and - Spoke topology?

A: It avoids split-horizon issues with distance vector routing protocols.

B: IP addresses can be conserved if VLSM is not being used for subnetting.

C: A multipoint interface offers greater security compared to point-to-point subinterface configurations.

D: The multiple IP network address required for a multipoint interface provide grater addressing flexibility

over point-to-point configurations.

some say B some say D.

dont wanna say what i think

are there any supper expert never wrong solutionfinders kind of person that can answer it so if i run into this at CCNA exam i can 100% say what the right answer is (and if you can explane why your answer is the right one)

thanks, love.

[pkhatri]

I would have to say that the answer is B, because you do not have to assign a separate subnet per sub-interface.

Not sure how you classify someone as a "supper expert never wrong solutionfinders kind of person" but I did get 983/1000 when I last sat the CCNA exam :-) You probably need to wait for a response from someone who got 1000 !

Paresh

[scottmac]

Well, personally, I don't think any of them are really great answers ...

A is wrong, because if you have a route advertisement come in fom one location, you can't propagate it back out to the other locations ... because split horizon says you can't forward an advertisement out the interface it came in on. Perhaps if you set up nothing but statics you could get around it (no dynamic routing protocols) ... and if you were using statics, then you wouldn't care about split horizon ... there's no route advertisement.

B sucks, because if you're using a Class A network (10.x.x.x/8), you blow the whole network on a few connections (if you used VLSM, you could use a better mask, limit the addresses used)

C has nothing to do with multi-point interfaces (security, if anything, would be worse because an intruder on one segment sees all the traffic from the other nodes)

D (compared to point-to-point) is less flexible (to me anyhow) because you must address all the nodes in the same network/subnetwork. With a point-to-point, each span can have a vastly different subnet from the others, usually (in practice) a /30 mask.

To me, all those answers are wrong ... vaguely resembling a Test King question ... many of which are flat wrong and / or stupid and / or misleading.

If this came from a Cisco publication, well, then, "sucks to be you," pick C, (or maybe A, if no routing protocol) because it's the least wrong of the bunch (but still wrong, IMHO).

That's my .02

Good Luck

Scott

[pkhatri]

Hi Scott,

Maybe a bit of discussion on option B is in order. Just because VLSM is not used does not mean that FLSM (Fixed Length Subnet Masks) cannot be used. In fact, that was the practice when using classful routing protocols like RIPv1 and IGRP. So taking your example, if you used 10.0.0.0/8, you would not be assigning the entire /8 to a single network. You would select a subnet mask for the network and then, you would have to use that mask with all subnets of the network. So if you chose a /24 mask, that would mean that you would have to use a /24 mask for even point-to-point links. And that is what I believe option B is getting at.....

However, I do agree that the question is badly worded and probably came from a TestKing-like source so there's always going to be a degree of doubt about the answers...

Paresh

[nedian123]

Hi,

among these wrong answers B is the least wrong...

Akhtar