Question on P.142 & 143 (640-802)

[tedbaby]

Dear All:
I am reading the 640-802 and I found some questionable points. On page 142 Figure 3.6 shows VLSM example 1 and hosts on Network A~D are 14, 30, 20, 6; on page 143 Figure 3.7 at left down has a chart where Lammle filled out the VLSM network example 1. Here is my question why Hosts in the chart NetworK A~D are 12, 20, 25, 4 but not 14, 30, 20, 6?? Did I miss some important concepts or misunderstand something? Plz anyone help me thanks.

[aguilera]

These numbers do seem more like lotto numbers than amount of hosts that are available to Network's A through D don't they?:D

Well, I picked up the New 640-802 book, flipped over to page 142 and 143 and see exactly what you are talking about.

Well to totally understand what is happening here you have to go back a couple of pages... first if we look at the network diagram on page 138 it gives us an example of a network and the "True" amount of hosts that the network actually has to support.

For instance, the network on the bottom right of the network diagram on page 138 labeled FIGURE 3.4 you will see that the network is supporting 12 Hosts.

Our documentation would be similar to the chart on the bottom of page 143 with the "true" amount of hosts that the network is actually supporting.

In order to support a network that has 12 host, we would need a subnet that would give us a block size of 16. Although we get a block size of 16 we would have to subtract two because one of those 16 IP addresses represents a Network number and one of those 16 addresses represents the broadcast on that network, which now leaves us with 14 valid IP addresses.

We only needed 12, as we documented in our chart... so the /28 would work.

Figure 3.8 and Figure 3.9 do a much better job of explaining the process.

Do me a favor and move on to page 144 and compare what I have written to Figure 3.8 and 3.9 and everything should click.

If not let me know!

Good luck!

- aguilera

[tedbaby]

In order to support a network that has 12 host, we would need a subnet that would give us a block size of 16. Although we get a block size of 16 we would have to subtract two because one of those 16 IP addresses represents a Network number and one of those 16 addresses represents the broadcast on that network, which now leaves us with 14 valid IP addresses.

We only needed 12, as we documented in our chart... so the /28 would work.

Dear aguilera, thanks for the replication. As you said "Figure 3.8 and Figure 3.9 do a much better job of explaining the process." And I totally agree with that.

But on page 142 Figure 3.6 told us that Network A need 14 hosts, network B need 30 hosts, network C need 20 hosts, and network D need 6 hosts.
And now we must jump to Page 143 and see the chart as I mentioned, the Block size and subnet and mask are fine and I have no question.
The chart shows us that Hosts on Network A are 12 (why not 14), hosts on network B are 20 (why not 30) and so on....
Need more clarifications, thanks.

[aguilera]

The numbers on the bottom of Fig 3.7 don't seem to represent any of the diagrams. I would say that you did not miss some important concepts or misunderstand anything. I think it is just a typo. You have the concept down.

d. aguilera

[tedbaby]

Many thanks for your help, aguilera. Maybe it is just a typo, because I found the 5th edition has the same problem, and at first I assume that Mr. Lammle will update in the 6th edition. Because my first intuition is that theres muse be a erratum and I hope Mr. Lammle will give more interpretation at the next update.

[aguilera]

Todd, can you shed some light on this for us.

Thanks.

[lammle]

Typo? Are you kidding?? :D

no, this is not a typo. If you read carefully, you can see that if you have 20 hosts, you need a block size of 32. If you need 12 hosts, you need a block size of 16.

What this is explaining is that you can't make up block sizes...and you must think about growth...so even if you have 20 hosts, you must use a block size of 32. What other block size can you use? I suppose you can use a block size of 64, which provides 62 hosts, but for my examples, a block of 32 works fine.

I think where your problem lies is you are trying to line them up - amount of hosts to a specific block size...it doesnt' work that way. Your block size and amount of hosts reserved for that LAN should always be higher then the amount of hosts you are using right now.

So, no, there is no typo here...that I promise!
Cheers!
Todd

[pinguino001]

Hi Todd,Aguilera,all,
what tedbaby told in the first post is that only first 4 hosts in table of fig. 3.7
on the bottom left are wrong, i.e. instead of 12,20,25,4 (Network A,B,C,D) we
should have 14,30,20,6 , is this right? :confused:
Everything in this example is correct. :D

Best Regards
/marco

[Stevep]

I can't understand the reasoning here by Ted and Marco, the network numbers all match with figure 3.6, why wouldn't you think the chart is for that example? The trouble is that network A,B & D are out of available host addresses so adding another employee or printer is going to become a mess. This example needs to be thought out better, real or not.

[lammle]

Steve, you are correct, the LAN's are out of IP host addreses. I failed to mention in the question that the VLSM design was for a company that was downsizing and are laying people off in all departments...:D

Other then that, considering the conversation string, I think this example has proved it's value.

Cheers!
Todd Lammle

[belly]

hi, i so, the network diagram in figure 3.6 is wrong right ? it supposed to be " 20 hosts network B" , "4 hosts network D" , " 12 hosts network A " ..

is it ?

[d404zeiq]

hi,

I understand the point but infect it is little bit confusing; as the bottom-left chart on page 145 is related to figure 3.8.....in the same way the chart on page 142 must be related to example(figure3.6) on page on 142. OR ???

may i suggest to please reconsider the smaller-chart in its 'host' and 'subnet' columns.

- first 4 rows in this column must be 14, 30, 20, 6 (to eliminate the confusion)
- subnet column must look like as it is on page 145

Am i right???

thankx a lot and waiting for comments

[lammle]

Please try and understand how hard it is to explain VLSM in a book, without using math like all other books do.
I try and make it as simple to understand as possible.

Here's what you need to do:

Get a network, put a physical network design together.
Using my chart, create a logical design and implement it using VLSM.

THEN this will start making sense for you.

Cheers!
Todd Lammle

[stik]

hi! im sorry but im having the same confusion with the thread starter.
i would just like to know if figures 3.5, 3.6, and 3.7 are inter related. :confused:
is figure 3.6 the basis to fill in the chart in the lower left part of figure 3.5?
then figure 3.7 shows the correct answers for figure 3.5?
is that it?
thanks.