Original subtitles

>> Up till now, we have talked about subnetting based on the number of networks you need,

the number of hosts per network that you need, and now we've fallen

into what I call reverse engineering a subnet problem

which is actually probably the most common thing that you will do.

It is extremely common to walk into an environment where you see an IP address

and a subnet mask and you need to find an answer to the question

of what network is that IP address on?

Or what would be another host on the same network as that IP address?

It's best explained with an example so let's jump back in the pool again, right?

So we've got a computer here - let's say a happy computer - had the IP address 1921681.127.

Now in the real world, you may be saying, "Okay well, this guy maybe can't connect

to the internet" or you know "Can't ping some other device.

What's wrong?

What's...?"

And so you want to find out what other devices are on the same network as this one?

Or on a certification exam, you might be asked a question,

you know you've got a happy computer here, "What would be the broadcast for his network?"

Or "What would be his network ID?"

Or "What computer would be on the same host?

A-B -- or on the same network, A-B-C-D?"

And you've got to pick it from a lineup.

So regardless of what situation you're in, you have to work backwards.

In my opinion, you can do these much faster than the previous examples that we've seen

in the last [inaudible] - finding the number of networks and finding the number of hosts -

because they've actually given you the answer.

Well at least from the perspective of what we were trying to find previously.

Remember previously we were like, "Okay, we've got this network

and we need 30 hosts per network, so what subnet mask will do this and what other [inaudible]?"

Well in this kind of thing, they're kind of like,

"Well we haven't really given you any of that."

We just say, you know somebody figured out that they wanted to use this subnet mask,

so now they're asking you to work backwards or essentially maybe work forward one step

because remember Step 1 was you know converted to binary.

So if you add 30 hosts, you would convert that number to binary.

Step 2 would be to reserve the bytes in the mask and find your increment.

And then Step 3 would be to find your network ranges.

So essentially with this problem, you're starting right here.

You've found the subnet mask if you will and now you just have to find the increment

and then find your network ranges.

So let me ask you this.

If somebody through a subnet mask at you and said,

"Hey here's a subnet mask, what increment does it have?

What increment would I use for my Step 3?"

How would you approach it?

Well you would say, "Okay, well what is the increment?"

The increment is the lowest network bid converted back to a decimal number.

So looking at this, I need to find out what the lowest network bid is.

Now if I take you know this, it'd be a whole bunch of ones.

Well I'm not really that concerned about all of these ones right here

because they're not going to give me anything.

I'm really concerned at where the lowest network bid is: that 224.

So if I convert that to a decimal number, it would be 1, 2, 3, 0, 0, 0, 0, 0.

That's the binary version -- did I say the decimal number?

If I converted that decimal number to a binary version, it would end up being this.

So then I would look and say, "Okay well lowest network byte as a decimal number?

That would be what?

Thirty-two.

Right? Okay, okay so I've not got this magic number if you will.

Let's jump back to the original.

You know I would look at that and say, "Well at some point they had

to start 192 dot 168 dot 1 dot 0.

I mean I would -- they have to start from zero at some point.

So I would then say, "Okay well let's do 32, 64, 96."

I'm just adding my increment right here.

One twenty-eight, 1 -- well actually stop right there because I pass 127.

That's my IP address.

So I would then -- again if you want to you can fill in all the ranges

but really the only one I'm concerned with is this one.

So this happy computer comes from the network 192168 dot 1 dot 96 thru 127.

That's his network and whoa, hello, major problem.

What's wrong with this picture?

One twenty-seven is what?

The broadcast address of that happy computer.

He's not happy at all.

He's not working because we assigned him the broadcast IP address

for the network, he is not going to function.

Now matter of fact, if you try to assign that IP address in Windows or OSX

or whatever operating system, it will give you an error because Windows is smart enough to say,

"Whoa, that's not a valid IP address.

Try that again."

But that would be a great test question where they're

like you know, "What's wrong with that guy?"

And do you see?

You wouldn't be able to tell what's wrong with that guy

without really figuring out the network ranges.

I mean if you didn't see this, if that was hidden from you,

and they're like, "Well what's wrong with that?"

And you're like, "I don't know.

What's wrong with that?

You tell me because from my perspective, it's just an IP address,"

but the subnet mask puts it in such context.

Okay, let's do another example of this so you can see just how --

just how key this skill is and how tricky the questions can be.

Let's just say this is an exam situation or maybe even a real world situation

where they're saying, "Hey, this host cannot surf the internet.

They're not able to get out and you know do their day to day functions.

Please identify what's wrong with this picture?"

Now of course in the real world and on the exam, of course it wouldn't be just left to this.

There would be a lot more distractors in there.

There would be, "Well is the internet up?

Is the -- you know how many other routers?

How many other hosts are having this issue?"

I mean there should be a lot of other distracting things

but let's just focus on the core.

We've got this host who has this IP address: 172 16 68 65.

His default gateway - and they line up - you know is 172 16 68 62 which looks good you know

because that's the IP address assigned to the router.

You know and down here is the internet or whatever network we're connecting to.

Okay, well it seems okay.

Well let's reverse engineer and se f we can figure out.

We look at that subnet mask, we've got 3 octets of ones.

That's the Class C. And then let's see, this last octet would be 1, 2, 3, 4, 0, 0, 0, 0.

That'd be if we're converting 240 to binary.

And you will get used to a lot of like well 240 is 4 bytes,

you know 224, 3, bytes, those kind of things.

So we look and we go, "Okay well the increment - that means the increment - is 16.

Okay. Well let's do this.

So we go 172 dot 16 dot 68 dot -- well they would have had to have started at zero

at some point so I can go, "Okay 16, 32."

I'm just counting up until I can pass those networks.

So it would be 48, 48, plus 16.

That's where my muscle memory wears out.

So it would 64 plus again, that would be 80.

And okay, okay, I've stopped.

I've passed those so I can start filling in the [inaudible].

It just goes through 47.

This goes through 63.

This goes through 79.

Whoa, wait a sec.

Whoa, wait a sec.

Okay, [inaudible] stop the train.

First off, are either of these IP addresses invalid?

No, they're not.

The router happens to have the last valid IP address from this range.

The router is in 172 dot 16 dot 68 dot 48 through 60 [inaudible].

He's got a valid IP address.

It's not a broadcast.

It's not a network.

But it's the last one from that range and this computer happens

to have the first valid IP address from this range.

What's the problem?

You know 65 is not the network.

It's the first real IP address.

The problem is they're in different networks.

Even though they're plugged into the same network, they're in different networks.

Now that doesn't work.

That's what a router is supposed -- a router is supposed

to the one that moves you between networks.

As in the router should be in your network like this host should be

in the same network as the router.

And then the router would move you to say that network

or some other network that you're trying to reach.

You can't have the computer be on a different network

than the router that's supposed to get him off his network.

Does that make sense?

So that is -- therein lies the problem.

Now do you see how tricky that could be because you're looking and you're like --

you know first off, if we hadn't broken that out, this totally looks okay.

We would look at this and go, "Well 65, 62, they seem close together.

They're only 3 IP addresses apart, right?

It seems like that should work okay."

So you wouldn't even know until you reverse engineer it and find this.

Now expect if you're planning to get certified that this will be embedded in a much bigger,

more complex situation that involves [inaudible]

and all the other technologies that are out there.

And they're like, "Okay, well try and figure out what's wrong."

And you'll be thinking of really advanced stuff but then you're like, "Oh wait.

These guys are in different networks."

That's why I say this form of subnetting is probably the most common kind of subnetting

that you're going to encounter is because it's so easy to write a question for it,

but also so common of a situation where you want to figure out what network a device comes from.

Okay, at this point you have seen subnetting based on the number of networks,

hosts and reversed engineering or subnet mask.

If you've got all that, if you're like, "Okay, I've got those skills,"

I would call you a subnetting quasi Jedi.

The only reason that quasi is in there is

because you probably just need more practice to solidify those skills.

There is one more method and that's VLSM but you'll find

in the next nugget that it's nothing new.

It's just doing what we've seen already again and again and again and again.

So before I launch you into VLSM, I want to make you aware of a rule.

And the only reason this rule exists is because we as humans don't think about zero.

As in -- let's say you had a cell phone in your hand and somebody looked at it and they're like,

"Oh how many of those cell phones do you have?"

You would look at him and you'd say, "I just have 1 cell phone."

Now the person you're talking to wouldn't interpret that as 2 because you know you said 1.

And they wouldn't say, "Oh well there's the zero value as well

so you must really have 2 cell phones."

You see what I mean?

But a computer does.

A computer counts zero as a value even though it's really a zero value.

Or for example, if there are paper clips on your desk and you said, "I have 10 paper clips."

The person wouldn't think, "Well you must really mean 11

because zero through 10 gives you 11 values."

But computers do.

And it's the same thing when you're working in binary, computers count that binary zero value.

So these values might throw off your calculations but if you're aware

of the rule, it really becomes a nonissue.

So for example, let me give you a scenario here.

If somebody said, "Take 192 dot 168 dot 1 and break that into 4 networks."

You would look at it and go, "Okay, great.

Let me lay out my binary chart, 128, 64, 32, 16, 8, 4, 2, 1."

So you'd go, "Okay, Step 1 is to say how many bytes to get the number 4?"

And I'd look and I'd go, "Okay, well -- 4 zero, zero.

Then it would be 4."

And you would say, "Okay, well 3 bytes to get to the number 4."

Well in actuality because of the zero value, you can get the number 4 with 3 bytes.

And I'd say, "Well how's that possible?"

Well if you look at this, you've got 2 in 1 which gives you 3.

You know that's the biggest number you can get with 2 bytes.

But then we start working it down.

We go, "Okay, well there's 2.

There's 1 and then there is zero you know or binary values would be 0- 0, 0-1, 1-0 and 1-1."

Those are the binary values.

So you can actually -- the truth is you can get 4 networks

with 2 bytes because of counting from zero.

So essentially if you look at the binary chart and there's any kind of question asking you

to figure out how many networks and it's an exact binary number, you just have to remember,

"Well okay binary always counts from zero.

So we can just simply subtract 1 when we're doing that."

Now it's similar on the host side.

Slightly different but similar on the host side because of the zero but it's somewhat offset.

And let me explain.

So let's say you've got the same situation: 192168 dot 1 dot zero.

And someone says, "Okay I want you to subnet that into networks of 7 hosts per network."

And I would say if they're asking that they're probably trying to trick you or something.

Seven hosts per network.

And you would look at that and you would go, "Okay, well Step 1, let's convert 7 to binary."

And I would go, "Okay, let's see, okay.

One, that's where my first one would go.

Two, three -- so all three of those bytes.

One, 0, 0, 0, 0, 0, 1, 1, 1 is how I get the number 7."

Well therein lies a problem that you know we're after number 7.

So we would say, "Okay, it takes 3 bytes to get to number 7 so in the next one,

you know Step 2, we would save 3 bytes."

We would go, "Okay, well we started with the slash 24

which is [inaudible] you know a whole bunch of ones.

And we go 0, 0, 0, 0, 0, 0, 0, 0" -- is that 8?

That's amazing.

How'd I do that?

So we would save 3 bytes so I'd say, "One, 2, 3 are saved.

So these would all flip over to the network side."

So that'd be right there.

So our increment would be an 8, right?

Well there's our problem because when we started doing our increments,

we have to start realizing we're going to go 192 dot 168 dot 1 dot 0 dot 1 dot 8 dot 1 dot 16

and you know down and down we would go.

And we'd fill this in and we'd go, "Okay well it looks like we've got you know 8 [inaudible]"

but therein lies the problem that we actually only have 6 valid addresses.

We end up with one short.

Now how did that happen?

Well we ended up 1 short because we always subtract 2 from that value, right?

We always -- you know whenever we're saying 7 hosts per network, we always have to think well,

but we also have to account for the network and the broadcast.

But it somewhat gets accounted for because of the zero value.

Because of zero, but you know it's --

so zero covers one of the IP addresses that we would be short but we are still one more short.

Does that make sense?

Because there's a network and there's a broadcast.

So zero covers one but we still end up one short so here's the exception.

To be safe, always subtract one when finding networks.

It will never hurt you to subtract one when you're finding the number of networks

or always add one when finding the number of hosts per network.

[Inaudible] you know if they ask you to calculate 7 hosts per network, calculate 8.

If they ask you to calculate 101 hosts per network, calculate 102.

It will never harm you to add 1 to the value nor will it harm you to subtract 1

when finding the number of networks.

And that will be the rule.

And when you get to the point where you just get so familiar with how binary works,

you just look at it, what -- it will sound a red flag.

When you see something that you're like, "Wow, they're asking for exactly the binary number?"

Okay, that's going to give me -- like it will just send --

like you know that spidey-sense [phonetic] that you get inside of yourself

when evil is about to attack you?

It's that same kind of thing.

When you're seeing enough of these subnetting problems and you're like, "Ooh that's --

they're asking me for something weird like why would they ask me for 15 hosts per network?

And when I figure that out, it comes out to a binary value of a solid 1 block.

Okay let me just -- you know spidey-sense is going off here.

Something's wrong with that situation."

So you'll start to navigate and figure it out on your own but if you want a rule to remember,

that will be a solid rule that will never fail.

Alright, well you are well on your way to subnetting mastery.

One more topic coming up and that variable length subnet masking,

but for now we have seen reverse engineering which is the third and final method

of subnetting that is very unique and then we saw of course the great exception.

I hope this has been informative for you and I'd like to thank you for viewing.