CCNA Questions...

Poontos

Platinum Member
Mar 9, 2000
2,799
0
0
Can someone please elaborate on the answers of these questions and why they are that way?

Question: Which class of network addresses allows the borrowing of 15 bits to create subnets?
A. Class A
B. Class B
C. Class C
D. No class of network allows the borrowing of 15 bits.

Question: How large is the network/subnet portion of a Class B license having a subnet mask of 255.255.240.0?
A. 18 bits
B. 19 bits
C. 20 bits
D. That is an invalid mask for a Class B network.

How many total subnets can be created if four bits are borrowed from the host field?
A. 8
B. 16
C. 32
D. 64


Do all routers use ARP?

What is described when a router acts as a default gateway, allowing a device on one subnet to find the MAC address of a host on another subnet?
A. proxy ARP
B. proxy RARP
C. RARP
D. subnetting



Thanks!
 

ScottMac

Moderator<br>Networking<br>Elite member
Mar 19, 2001
5,471
2
0
Class A - if you borrowed 15 bits from a class B, there wouldn't be any room for host addresses

20 Bits - 16 bits of network address, four bits of subnet address.

16 Subnets - You have four bits to work with. 2^4=16 (two to the fourth power). 0000,0001,0010,0011,0100,0101,0110...1111. For a host-portion, by convention, you throw away two of the addresses, the "zero" address, and the "all ones" address, which are the "Network Address" and the "Broadcast Address" respectively.

All ((( -- IP --))) routers can ARP, but they don't need to if the address has already been resolved and is still resident in the ARP cache/table. I believe the default aging time is 3 minutes. IPX doesn't need to do ARP, because the MAC address is part of the host address. AppleTalk has it's own thing.

Proxy ARP.

Good Luck

Scott
 

me19562

Senior member
Jun 27, 2001
374
0
0
Actually the right equation for calculate the numbers of subnets is (2^(# of borrowed bits))-2
Something like this (2^4)-2=14
 

ScottMac

Moderator<br>Networking<br>Elite member
Mar 19, 2001
5,471
2
0
Yup, it is.

The question asked for -total- subnets (including broadcast & network number: assumed because there was no "14" choice)

The questions on the test can be tricky. Some will want total, some will want useful (or usable) subnets...gotta read carefully and check the answers.

2^BITS - 2 is the usual way to get a usable subnet count.

I was just taking a guess anyway, all this technical stuff is way over my head.

FWIW

Scott
 

bigshooter

Platinum Member
Oct 12, 1999
2,157
0
71
Actually woiuldn't a subnet mask of anything other than 4,8, or 12 bits be classless and therefore not a class a,b, or c address?
 

ScottMac

Moderator<br>Networking<br>Elite member
Mar 19, 2001
5,471
2
0
I dunno. I s'pose that's true, though, any address/block using other than a natural mask has subnets (or supernets), but the class identifier (first octet, first four bits) still exists...so there is a class definition that doesn't change (unless your doing some kind of Super-Duper Supernet). I guess it depends on your definition of classless.

If you have a 172.16.X.0 with a 255.255.255.0 mask, applied consistantly across you network, you can still use RIPv1 which is a classfull protocol. I think it's when you kick in the variable-length subnet masking (VLSM) where the problems begin, and then only in some scenarios.

Technically, I believe you could be right.

Practically, I don't think it'll stick. People still call tissue "Kleenex," xerographic duplicates "Xeroxes," and anything vaguely resembling UTP as "Cat 5."

In the same vein, folks continue to call a "255.255.255.0" mask applied to a Class A or B network numer as a "Class C subnet mask," (and similar for a double octet as a "Class B" subnet mask) which would also be inaccurate.

I even read a Microsoft "White Paper" where the author(s?) said the Class of the IP address was determined by the applied mask (instead of the "First Octet Rule").

I s'pose it boils down to appliing an "accuracy mask" to the statement: Put it in context, and decide if it's "accurate enough" for the context of the discussion.

I an aggregation of serious NetGeeks, two minute discussions turn into half-hour discussions of minutia while the pecking order is define, then the loose terms fly fast & furious....because they know what the other person meant, even though it's not 100% accurate.

That's my take on it anyhow.....


FWIW

Scott
 

xyyz

Diamond Member
Sep 3, 2000
4,331
0
0
muhahahhahaha....

these are cisco adademy test questions... what sem1 chapter10 or something? I remeber these last thursday... some guy totally bombed on this chapter and was going over these with the instructor.