Broadcast Subnet??

Andrew Newman

1) I'm looking at a Boson practice test for ICND1 where it states that a subnet with the subnet number 192.168.7.128 and mask 255.255.255.128 is a broadcast subnet. Is this because there are only 1's in the 4th octet?

2) Along the same lines is the subnet number 172.30.0.0 /21 considered to be a ZERO subnet because there are only 0's in the 3rd octet?

thanks for any help

Bl8ckr0uter

Andrew Newman wrote: »

1) I'm looking at a Boson practice test for ICND1 where it states that a subnet with the subnet number 192.168.7.128 and mask 255.255.255.128 is a broadcast subnet. Is this because there are only 1's in the 4th octet?

2) Along the same lines is the subnet number 172.30.0.0 /21 considered to be a ZERO subnet because there are only 0's in the 3rd octet?

thanks for any help

127 is the broadcast.

1-126 are valid hosts.

Andrew Newman

Hi, thanks for that.
Can you explain why 192.168.7.255 is not the broadcast and and why 127 is?

Forsaken_GA

There's a little bit of confusion.

255.255.255.128 is a /25, so you get two subnets out of that, 128 ip's each

the first subnet is 192.168.7.0 to 192.168.7.127

0 is the network identifier, 127 is the broadcast IP

192.168.7.128 to 192.168.7.255 is the second subnet

128 is the network identifier, 255 is the broadcast

so Boson is correct, 192.168.7.128 is a broadcast subnet, since it contains the broadcast IP for the /24 that was split up

Bl8ckr0uter

Andrew Newman wrote: »

Hi, thanks for that.
Can you explain why 192.168.7.255 is not the broadcast and and why 127 is?

Yes. The broadcast for the subnet is the last number before the next number. A /25 means 2 networks in the 4th octet.

0-126


127 is broadcast.

What is the broadcast for the network?

Review these: Learn To Subnet:* A Free, Lecture-Based Presentation on IP Addressing and Subnetting
subnettingquestions.com - Free Subnetting Questions and Answers Randomly Generated Online



@For

Ok I understand what you are saying. Forsaken is right and he explanation makes sense.

Andrew Newman

Hi, I'm getting closer to understanding it now.I can see why /25 for a class C address means 2 possible subnets but I still don't see why 192.168.7.128 is a broadcast subnet and not just a regular subnet.

I don't get what you mean when you say 'so Boson is correct, 192.168.7.128 is a broadcast subnet, since it contains the broadcast IP for the /24 that was split up'
Can you expand on that a bit please, some zeros and noughts would be good .I'm new to this and it's hurting my head something chronic but i'll keep at it until I've got it.

thanks

Forsaken_GA

Andrew Newman wrote: »

Hi, I'm getting closer to understanding it now.I can see why /25 for a class C address means 2 possible subnets but I still don't see why 192.168.7.128 is a broadcast subnet and not just a regular subnet.

I don't get what you mean when you say 'so Boson is correct, 192.168.7.128 is a broadcast subnet, since it contains the broadcast IP for the /24 that was split up'
Can you expand on that a bit please, some zeros and noughts would be good .I'm new to this and it's hurting my head something chronic but i'll keep at it until I've got it.

thanks

Because if it wasn't subnetted, that would normally be a /24, as 192.168 is, in classful terms, a class C

For that /24, the range would be 192.168.7.0 to 192.168.7.255, with .0 being the network identifier, and .255 being the broadcast.

If it wasn't subnetted, 192.168.7.255 would be the broadcast IP for the network.

Since it is subnetted and broken into two networks, 192.168.7.128/25 contains the broadcast IP of the classful allocation, hence it's a broadcast subnet

Forsaken_GA

You also need to understand that the concept is pretty obsolete, but the CCNA wants you to learn it. When I went through the CCNA net academy, we were taught that when you're busting up a classful network allocation into subnets, you can't use the first subnet (the all zeros) or the last subnet (the all ones) because the first subnet contains the network identifier for the entire network and the last subnet contains the broadcast for the entire network.

This was true when subnetting was first introduced, the equipment and the IP stacks didn't really know how to handle that (were you trying to send a broadcast to the entire network, or just the last subnet!) Today, it doesn't matter, the all-zeros subnet and the all-ones subnet are perfectly useable, since the classful method of operation has fallen by the wayside. It's possible that it could become an issue if you have legacy equipment and implementations in your network, however, which is why I guess they want you to learn about it.

tha_dub

I think foresaken has hit the nail on the head.

The Broadcast subnet is the highest subnet. so for 192.168.7.128 /25 you only have 2 subnets 192.168.7.0 and 192.168.7.128. the first is the zero subnet and the last (highest subnet) is the broadcast subnet. The only time you would really pay much attention to this is if the router had the "no ip subnet zero" command in which case both subnets would be unusable.

Assuming you have No ip subnet zero with say 192.168.1.1/26.

you'd have
192.168.1.0 (Zero subnet)
192.168.1.64 usable
192.168.1.128 usable
192.168.1.192 (broadcast subnet)

Basically you need to know this info for the test in case they tell you they want to know the broadcast or zero subnet or if they router has a no ip subnet zero command. Otherwise assume you can use the zero and broadcast subnets.

Andrew Newman

I get it now.
thanks

altjx

Bringing back an old thread. I am on the same question as this guy...

I understand his question, pretty simple for me. However, there's another subnet 192.168.6.0. Why isn't this a broadcast subnet? The broadcast address is 192.168.6.255, which is the broadcast address for a /24 (255.255.255.0) subnet mask.

I don't understand why 192.168.7.128 is a broadcast subnet (with a broadcast address of 192.168.7.255) but 192.168.6.0 isn't a broadcast subnet.

This doesn't make sense. This basically says 192.168.7.128/25 is a broadcast subnet but 192.168.7.0/24 isn't a broadcast subnet. They both have the same broadcast address, so what makes one a broadcast subnet and the other not? The only assumption I can make is that in order for subnets to be classified as a subnet zero (when they're using .0 subnet) is that the subnet has to be using VLSM. Subnets that are sitting on the .0 subnet and not using VLSM cannot be classified as subnet zeros. Correct?

Please help.

Forsaken_GA

altjx wrote: »

This doesn't make sense. This basically says 192.168.7.128/25 is a broadcast subnet but 192.168.7.0/24 isn't a broadcast subnet. They both have the same broadcast address, so what makes one a broadcast subnet and the other not? The only assumption I can make is that in order for subnets to be classified as a subnet zero (when they're using .0 subnet) is that the subnet has to be using VLSM. Subnets that are sitting on the .0 subnet and not using VLSM cannot be classified as subnet zeros. Correct?

Please help.

You're almost there, you just haven't figured out the why.

Subnet zero and the broadcast subnet are only considerations when you actually subnet.

192.168.7.0/24 is not a subnetted range, it is a native class C allocation, so if that's what you want, no further subnetting is needed.

192.168.7.128/25 *is* a subnet that falls outside of the classful definitions. The reason it's a broadcast subnet is because it contains the broadcast address (192.168.7.255) for the entire Class C allocation (192.168.7.0/24). This is where the problem used to be - the router didn't know how to differentiate between a broadcast destined for the entire Class C and one destined just for that subnet.

This is why for years the idea was to not use the broadcast subnet for assignment to hosts. That restriction went away awhile ago

altjx

Forsaken_GA wrote: »

You're almost there, you just haven't figured out the why. Subnet zero and the broadcast subnet are only considerations when you actually subnet. 192.168.7.0/24 is not a subnetted range, it is a native class C allocation, so if that's what you want, no further subnetting is needed. 192.168.7.128/25 *is* a subnet that falls outside of the classful definitions. The reason it's a broadcast subnet is because it contains the broadcast address (192.168.7.255) for the entire Class C allocation (192.168.7.0/24). This is where the problem used to be - the router didn't know how to differentiate between a broadcast destined for the entire Class C and one destined just for that subnet. This is why for years the idea was to not use the broadcast subnet for assignment to hosts. That restriction went away awhile ago

Yep, that's exactly what I thought. So in other words, /8, /16 and /24 subnets are not subnet zeros. Not sure why this site is going sooooo slow today >_< Thanks for the help dude! This cleared up my question.

MrBrian

Here's a little more to help you along.. 192.168.7.0 with a /24 is not subnetted because that's the classfull subnet mask for a class C address.. So, you have the whole 4th octet, 8 bits, for host addresses. Now, once you start borrowing bits from the 4th octet, they will be used for the subnet. So if you make it a /25 you're taking 1 bit.. a /28 you're taking 4 bits.. a /30 you're taking 6 bits, etc.

The "zero" subnet, is the subnet number with all subnet bits set to 0.. the "broadcast" subnet has all subnet bits 1'd. It's kind of funny to use the example in this scenario, a /25, because there's only 1 bit for the subnet, giving you two possible subnets (2^1=2.. and they are the 0 and 128 subnets). So the "zero" subnet is 192.168.7.0 /25, and the "broadcast" subnet is 192.168.7.128 /25

It's maybe a little easier to see this when you have multiple subnet bits.. say you used a /28 for that network.. there's 4 bits for subnetting so 2^4=16 subnets with that.. the zero subnet will be 192.168.7.0 /25.. and the broadcast subnet would be the 16th subnet.. 192.168.7.240 /28

alan2308

altjx wrote: »

Yep, that's exactly what I thought. So in other words, /8, /16 and /24 subnets are not subnet zeros. Not sure why this site is going sooooo slow today >_< Thanks for the help dude! This cleared up my question.

If you're subnetting a Class A into multiple /16's or multiple /24's, or a Class B into multiple /24's, there is still going to be a subnet zero. If you stay at a classful boundary and did not subnet, then there will be no subnet zero.

In other words, if you're subnetting, then there will be a subnet zero regardless of what mask you're using.

urvi

Forsaken_GA wrote: »

You're almost there, you just haven't figured out the why.

Subnet zero and the broadcast subnet are only considerations when you actually subnet.

192.168.7.0/24 is not a subnetted range, it is a native class C allocation, so if that's what you want, no further subnetting is needed.

192.168.7.128/25 *is* a subnet that falls outside of the classful definitions. The reason it's a broadcast subnet is because it contains the broadcast address (192.168.7.255) for the entire Class C allocation (192.168.7.0/24). This is where the problem used to be - the router didn't know how to differentiate between a broadcast destined for the entire Class C and one destined just for that subnet.

This is why for years the idea was to not use the broadcast subnet for assignment to hosts. That restriction went away awhile ago

Hi
Can you plz tell me,how come nowadays the router knows how to differentiate between a broadcast destined for the entire Class C and one destined just for that subnet only?

WilyOne

urvi wrote: »

Can you plz tell me,how come nowadays the router knows how to differentiate between a broadcast destined for the entire Class C and one destined just for that subnet only?

First of all, Google and cisco.com are your friends.

I was curious how this topic was coming up now; I've heard of it, but it's mostly historical. I found the following doc which talks about it in depth:
Subnet Zero and the All-Ones Subnet  [IP Addressing Services] - Cisco Systems

The direct answer to your question is: starting with IOS v12.0, Cisco routers will allow the "subnet zero" and the "all ones" subnet by default.

DoubleNNs

Saying "google and cisco.com is your friend" in response to their question is essentially implying there is no need for this forum to exist.

Urvi revived a thread created over 2 years ago, meaning they already did some searching for their answer.

Also they know that both subnet zero and the all ones subnet is allowed. But they don't know how the router is able to differentiate between a broadcast meant for the class broadcast address vs for the broadcast subnet.

I'm pretty sure the broadcast message includes the subnet mask in it. So using the broadcast address, it knows which network it was intended for - whether for the overarching class or the subnet.

WilyOne

DoubleNNs wrote: »

Saying "google and cisco.com is your friend" in response to their question is essentially implying there is no need for this forum to exist.

Upon seeing this thread I got curious, and used Google to find the Cisco article talking about the issue. I did not search for "broadcast subnet" I searched for "all ones subnet", since I had heard of this topic before. A forum such as this adds something raw Google searching does not: experience and insight. The point I was trying to make is knowing how to find an answer is often more valuable than the answer itself.

urvi

Thanks DoubleNNs. I got my answer
And @just plain ol Kev from Cisco Forum (aka Wilyone) you are back again here!!
If I had found the answer googling it I wouldn't have posted here. 2ndly,did you find the answer that DoubleNNS said by googling the net?
I didn't!

theodoxa

There are two different concepts: Broadcast Address and Broadcast Subnet. The Broadcast Subnet idea goes back to "Subnet Zero." It used to be MANY MANY years ago (more than 12 years ago) that you could not use the first or last subnet when subnetting a classful network. The first subnet was called the "All Zeroes Subnet" or "Subnet Zero". I presume the last subnet was what they are calling the "Broadcast Subnet". Back in 2001, we were instructed that even though it was the default behavior of current routers to allow subnet zero that for the CCNA exam we should assume it wasn't allowed. This has since changed as Odom's book if I remember correctly stated that you should assume these subnets are usable unless the no ip subnet-zero command is configured or you are using a classful routing protocol (RIP Version 1).

WilyOne

urvi wrote: »

And @just plain ol Kev from Cisco Forum (aka Wilyone) you are back again here!

Huh?

/10char