Wendell odoms Subnetting not as good as i'd hoped.

pandimus

So i'm trying to get a refresher on my subnetting. Heck 216 was like 2 months ago.. Anyway, I am going through the book and I realize both books have a replica of the same chapter for subnetting.. (no biggy)..
But when i start reading it, i realize he makes it harder than it is. (IMHO). I think the best place's ive found to learn to subnet are.

www.Learntosubnet.com
http://www.mcsefreak.com/subnetting.htm

I guess i will review it again when less tired. But by what i could tell, he just didnt make it easy for me.. Perhaps someone else would like his style better. Or even someone who hasnt learned this stuff yet..

BTW, does anyone know if there is more to subnetting for cisco than there was for 216?

mbeaven

www.learntosubnet.com is a great resource. I used that quite awhile ago and it made some of the bizarre methods I've run across since understandable. I suppose I can't go into much detail about what was on the exam but you should probably get subnetting down to where you don't have to think about it. Also VLSM and wildcard masks.

Now if someone can explain hex numbers to me in a way that makes sense...

2lazybutsmart

pandimus: I'm reading Todd Lammel's book; I was supposed to buy Wendell's, but I chanced to read the first chapters of both books at a book store and I liked the style and writing of Todd's--Well personal preference, but he's great!.

2lazybutsmart

mbeaven: I've used that site as well; but asides from the repetition (the guy repeates and re-repeates everything over and over), I find Todd Lammles' method (I'm using his 4th version book) of subnetting with your head a masterpiece.

Now everything has a drawback--and I'm not blaming Todd for this--, but I just don't understand the VLSM part of the game. I do understand the basic concept, but I don't get this whole Block stuff and all. Just like how a file cabinet placed incorrectly onto the cabinet won't slide in; this VLSM isn't just sliding into place. Perhaps some of you guys can help me out.

One of the things that confused me was the author took an example of a network and apparently gave it a wrong number of bock sizes.

This was the network .

Network ID:A--Mask:/27--Hosts:30
ID:B--Mask:/28--Hosts:10
ID:C--Mask:/28--Hosts:12
ID:D--Mask:/30--Hosts:2
ID:E--Mask:/30--Hosts:2
ID:F--Mask:/30--Hosts:2
ID:G--Mask/28--Hosts:12
ID:H--Mask:/26--Hosts:60
ID:I--Mask:/28--Hosts:14
ID:J--Mask:/30--Hosts:30
ID:K--Mask:/28--Hosts:8

Now does this network have the following block sizes (As specified on the book):
2 block sizes of 64, two of 32, four of 16, and three of 4,

OR

2 block szies of 64, one of 32, five of 16, and 3 of 4.

And what is the purpose of this block size in the first place ??

2lbs.

mbeaven

I don't think I've seen the Lammle book I was going to hit the book store tonight so I'll check it out. A few years ago when I was first doing subnetting I had 3 different books with 3 different methods and got confused so that site helped alot. He does repeat info a lot and he's kind of hard to listen to for more than a few minutes.

I haven't seen VLSM questions put that way either. So I'd also be interested in the result.

2lazybutsmart

yeah... that multiple source stuff is sometimes confusing. Authors don't always put the theory for a given subject the same way. Not to mention their different examples and all that.

While we wait for someone to answer that VLSM question (and elaborate)... here's one other problem you should consider in the exam.

The subnet mask /25 has only one bit available for subnetting. Now that bit is either 0 or 1--subnets 0 or 128 respectively. This subnet mask allows 2 subnets and 126 hosts per subnet. Now the strange part is that the author says "Cisco" says we shouldn't use this subnet mask (and consider it invalid in exams). At the same time, my router--and perhaps all routers-- support the ip-subnet zero command. To add to the confusion, the author says we can make use of this (and should) in real time production.

There's always a reason Cisco says something; so what's the moral behind this "do as we say, not as we do" story.

2lbs.

pandimus

The book itself is not bad regarding other topics.. But just to show you what i mean..

He refers you to a website to learn how to change binary numbers into ip address's and vice versa.

mbeaven

Everything I've seen says ip-subnet zero is valid for production networks as long as all the routers on that subnet support it. My exam prep books all say that Cisco will no use it in the exam. Some of the practice exams I've used do use it however.

2lazybutsmart

pandimus wrote:

He refers you to a website to learn how to change binary numbers into ip address's and vice versa.

But I think you should be able to do that in your head . It should be a piece of cake.

You know... I think if someone's overqualified, he can't stay focused on one topic and explain it thoroughly. The guys a CCIE --keep that in mind. He probably knows 222 much (And then every reader sees it a different way ).

2lazybutsmart

mbeaven wrote:

all say that Cisco will not use it in the exam.

and... of course, the question "Why?" still remains.

mbeaven

2lazybutsmart wrote:

and... of course, the question "Why?" still remains.

Because we won't be smart enough to deal with that concept until CCNP? Another problem I have is that Cisco says for the CCNA standard ip access lists should be placed as close to the destination as possible and extended should be placed close to the source, yet I had 2 questions on the exam that had scenarios where it wasn't possible to place the standard access-list on the destination router.

2lazybutsmart

mbeaven wrote:

2lazybutsmart wrote:

and... of course, the question "Why?" still remains.

Because we won't be smart enough to deal with that concept until CCNP?

haha..was that a "randomly-generated" answer. Well then we should get some CCNP guy to answer it. One thing I know is that big companies might just say something now because of something and stick to it even when things have changed. subnet-zero was probably a problem when they laid down that rule..... But then... exam contents aren't there for fun.... There has to be a real reason behind everything. (break; let's not make a fuss outa it. we'll pass if we do it Cisco's way... so why not?)

Access-lists.... haven't reached them yet.
2lbs.

mbeaven

I thought older versions of the IOS didn't support ip-subnet zero so that's probably part of the equation. After failing the CCNA twice I'm not sure what Cisco is trying to do. Unfortunately I can't get them to just come watch me at work for a day.

2lazybutsmart

mbeaven wrote:

Unfortunately I can't get them to just come watch me at work for a day.

hahaha. Well a lot of people wish they could "wish" that very wish.

And something to laugh at: Cisco systems (probably) came from SanFransico Systems and the legal paper got crampled up--or teared-- en-route to the lawyers office only to reveal cisco Systems.

pandimus

But I think you should be able to do that in your head

I dont think it is a question of weather or not i can do that in my head, its the fact that his book is a learning tool, and if he only teach's you one part, and tells you to go to a website to learn the other part, he really isnt teaching you the whole thing. I honestly cant convert binary in my head yet, but i can write it on paper and have it down withing 30 seconds. Sofar i havent used binary except for my 216 test.

Needless to say, i think for a book to be a good book, they should teach you the whole concept.

2lazybutsmart

pandimus wrote:

Needless to say, i think for a book to be a good book, they should teach you the whole concept.

No one could put it better. I totally agree.

I'm not saying this ostentatiously, but after reading Todd's explanation of binary to decimal conversion and subnetting; I can almost do it perfectly in my head. I hope I don't wake up tomorow morning with a blank memory (Looks like I shouldn't forget copy running-config startup-config )

2lbs.

pandimus

but after reading Todd's explanation of binary to decimal conversion and subnetting; I can almost do it perfectly in my head.

i wish i could, my book refers me to the internet to learn how to convert binary to decimals.. I dont have access to the internet.. So now ive got to go to the library or buy another book. But if he would have included that...

2lazybutsmart

pandimus wrote:

i wish i could, my book refers me to the internet to learn how to convert binary to decimals..

Well is the internet link helpful at least....??

I dont have access to the internet..

I thought people had to be on the net in order to post on forums . Did you mean "no access to the internet website" on the book??

2lbs.

pandimus

Some of the websites refered to

http://www.goshen.edu/compsci/mis200/decbinary.htm

http://www.doit.ort.org/course/inforep/135.htm

http://www.ibilce.unesp.br/courseware/datas/numbers.htm#mark2.

pandimus

I thought people had to be on the net in order to post on forums . Did you mean "no access to the internet website" on the book??

just an example, not everyone in the world does have access.. I am sure people can get it regardless, and that is not really the point i am getting at.. The process is not hard and would have probably taken one more extra page.. At least...... in the bare minimum... put it in the intro book.....

pandimus

Just outta curiousity.. is Todd lamles way by using the chart?


128 64 32 16 8 4 2 1

and then putting the 1's and zero's in place and calculation the number?

Webmaster

I hope the following will clear some things up about VLSM and the 'minus 2':

Traditionally, it was strongly recommended that subnet zero and the all-ones subnet not be used for addressing. According to RFC 950 , "It is useful to preserve and extend the interpretation of these special (network and broadcast) addresses in subnetted networks. This means the values of all zeros and all ones in the subnet field should not be assigned to actual (physical) subnets." This is the reason why network engineers required to calculate the number of subnets obtained by borrowing three bits would calculate 23-2 (6) and not 23 (8 ). The "-2" takes into account that subnet zero and the all-ones subnet are not used traditionally.

"

microsoft wrote:

RFC 950 forbade the use of the subnetted network IDs where the bits being used for subnetting are set to all 0's (the all-zeros subnet) and all 1's (the all-ones subnet). The all-zeros subnet caused problems for early routing protocols and the all-ones subnet conflicts with a special broadcast address called the all-subnets directed broadcast address.

However, RFC 1812 now permits the use of the all-zeros and all-ones subnets in a CIDR*-compliant environment. CIDR-compliant environments use modern routing protocols that do not have a problem with the all-zeros subnet and the all-subnets directed broadcast is no longer relevant.

The all-zeros and all-ones subnets may cause problems for hosts or routers operating in a classful mode. Before you use the all-zeros and all-ones subnets, verify that they are supported by your hosts and routers. Windows 2000 and Windows NT support the use of the all-zeros and all-ones subnets.

As mentioned in RFC1878:
"For the sake of completeness within this memo, tables 2-1 and 2-2
illistrate some options for subnet/host partions within selected
block sizes using calculations which exclude all-zeros and all-ones
subnets [2]. Many vendors only support subnetting based upon this
premise. This practice is obsolete! Modern software will be able to
utilize all definable networks."

I wrote:

In traditional subnetting you have to substract the 2. (this is the way it is described in most books, tutorials and online references, especially those related to certification) If your network equipment supports CIDR you do not have to substract the 2, allowing you to assign the Subnet Zero and the All-Ones Subnet.

You 'may' run into questions on the exam where you don't have to subtract the two. But then it will be explicitely mentioned, i.e. "the entire network supports CIDR" or "all OSPF routers are configured with the ip-subnet zero command." In all other exam question scenarios you have to apply 'traditional subnetting'. But:

At the same time, my router--and perhaps all routers-- support the ip-subnet zero command. To add to the confusion, the author says we can make use of this (and should) in real time production.

this is correct.

Pandimus wrote:

BTW, does anyone know if there is more to subnetting for cisco than there was for 216?

A bit more, not all subnetting questions in the CCNA (and CCNP exams) focus just on subnetting. As routing is the primary topic, subnetting is an elementary part of several questions including sims.

mbeaven wrote:

I don't think I've seen the Lammle book I was going to hit the book store tonight so I'll check it out.

Before Odom's Cisco Press book arrived on the scene, which seem to have become very popular, Todd Lammle always used to be 'the' author for CCNA books. He's been elected best tech author for several years in a row. If you want to use two books, I think combining Wendell's and Todd's books is the best way to go.

2lbs wrote:

And what is the purpose of this block size in the first place??

So we don't have to waste entire IP networks on small (sub-)networks. Without VLSM every subnet in an IP range would have to use the same subnetmask. Hence, if one subnet needs 64 hosts, you can't use the remaining hosts in a single subnet, you would have to use the block size 64. So VLSM allows an IP network, ie. class C network 192.168.10.0 to be divides into several subnets with a different block size. A common example, is the lab example we usually discuss in our forums, and TechLabs. Say you got one class C IP network for all three networks:


Without VLSM you would have to use the same subnetmask on network 2 as network 3, you could assign 192.168.10.0/26 to network 1, and 192.168.10.64/26 to network 2 and 192.168.10.128/26 to network 3(assuming it supports the ip subnet-zero). Leaving 192.168.10.192/26 unused. But more importantly, this would waste 62 addresses on network 2, because it only needs 2 (both ends of the serial link). With VLSM however, you could assign the mask 255.255.255.252 (/30) to network 2 and use other mask to use the remaining addresses for network 1 and 2.
VLSM is developed because of the shortage of IPv4 addresses.

Pandimus, maybe this older discussion we had can be of some use again:
www.techexams.net/forums/viewtopic.php?t=2935

mbeaven wrote:

Another problem I have is that Cisco says for the CCNA standard ip access lists should be placed as close to the destination as possible and extended should be placed close to the source, yet I had 2 questions on the exam that had scenarios where it wasn't possible to place the standard access-list on the destination router.

I'm not entirely sure, but I think this has been discussed in earlier posts in the CCNA forum, please try a search. If you can't find anything please start a new topic.

pandimus

What are we gonna do when IPV6 comes intoo play??

2lazybutsmart

Some of the websites refered to

http://www.goshen.edu/compsci/mis200/decbinary.htm

http://www.doit.ort.org/course/inforep/135.htm

http://www.ibilce.unesp.br/courseware/datas/numbers.htm#mark2.

They're not bad. But that info in a consice form should've been in the book. But still... you know this stuff already, so if it covers other chapters neatly, I don't think you should buy a new book.

How much did you pay for it??

2lbs.

Webmaster

pandimus wrote:

What are we gonna do when IPV6 comes intoo play??

We stop subnetting and won't use NAT anymore

pandimus

They're not bad. But that info in a consice form should've been in the book. But still... you know this stuff already, so if it covers other chapters neatly, I don't think you should buy a new book.

How much did you pay for it??

The reason i said that was so that people who dont know subnetting, might swerver towards a more thourough book.

I think i paid 40 bucks for the 2 book series, brand new, with a 10 dollar voucher and some decent practice questions. with a small version of boson router simulator.

2lazybutsmart

pandimus wrote:

Just outta curiousity.. is Todd lamles way by using the chart?


128 64 32 16 8 4 2 1

and then putting the 1's and zero's in place and calculation the number?

Basicly, what he does is he makes you look at the byte as 2 nibbles. He then tells you that since the most important taks for binary to decimal conversion is subnetting, turning 1's and 0's off can only happen in a single file in the right most nibble (first 4 bits) starting from the right most bit.

i.e. 1111 would be 240. and 1011 (176) would not be used in subnetting masks. so essentially, the /25 to the /30 masks in class C would only include 1's and 0's switched on or off in a single file. The same will apply for the A and B classes. only the slash(/) notaion number would decrease.

the hard part is when you want to convert a binary-decimal for the host portion. since hosts will contain any numbers possible in the range of their bits, they won't have 0's and 1's switched on and off in a single file. But you won't be doing that headache anyways. host ID's are determined by multiplying 2 with the exponent of the number of bits you have left for the hosts after taking away the subnet bits and subtracting 2. And then valid subnet's would be determined by subtracting the subnet mask (in decimal) from 256. The result multiplied by itself untill you reach the subnet mask (excluding the mask) would be the available subnets. And the available hosts would be the numbers available between one subnet ID and it's broadcast address (which is always the next subnet ID minus 1).

2lbs.

pandimus

Do you memorize you roots? 2^16=

If so how far up do you go?

2lazybutsmart

He doesn't encourage the memorization of all the roots possible (which is 32 for the whole 32 bits and that=2 much headace).

Instead, you should memorize only the roots up tol 8. So if we had a Class B Network with a /26 subnet mask, you won't find the root of 2^10, but rather the root of 2^8 (which is 256--and you should know all the roots from 1 to 8 by rote). You would then multiply the result with the root of the remaining bits (2 in our example which means 4). So that would be 256X4=1024.

This method is basicly reducing the number of roots you have to do in your head. But if you a mathematician, then you won't have to worry . But I find it extremely helpful to multiply 256 with 2 rather than finding it's 10 or 11th root.

2lbs.

pandimus

One thing i do like in this book, is the fact that it recognizes people might have labs with a bit more outdated equipment. It recommends using 2950 switch's but has an appendix on 1900 switch's..

2lazybutsmart

I just hit the chapter that talks about Switching today. And I was a bit surprised to find that he explains the configurations of both (Catalyst 1900 and 2950) switches simultaneously.

I was thinking about stopping there are reading the whole book over again untill I nail the previous stuff and have it dialed in. Do you think I should do that; or just move on??

How are you studying?? I know Wendell's book is 2 in one. Are you going to take the exams like that too?

2lbs.