generic network address xxx.xxx.xxx.0

[u856100]

Guru's,

Can anyone explain just exactly what the generic network address is?

I have a B class network (147.89.etc)
subnet mask 255.255.255.128
that contains about 10 nodes.

When I ping the broadcast address, all 10 node packets are returned fine. When I ping the generic network address 147.89.124.0 I receive two ip addresses. Can anyone explain what the replying IP's are?

I'm a bit new to HP UX networking, so any help would be appreciated.

thanks in advance to all replies!

John

[Jeff Schussele · Accepted solution]

Hi,

If you are pinging a subnet addresss of xxx.xxx.xxx.0 then you have a class C address.
But your subnet mask of xxx.xxx.xxx.128 indicates that the network has been "divided" into 2 subnets each capable of 126 hosts.
Remember that the lowest IP is the network address - xxx.xxx.xxx.0 & xxx.xxx.xxx.128 in your case. The highest IP xxx.xxx.xxx.127 & xxx.xxx.xxx.255 will be the broadcast IPs.
Since you didn't specify what the actual host IP was, my guess would be is that your host is > xxx.xxx.xxx.128 & using the xxx.xxx.xxx.0 as the network address is giving the "double IP". If this is the case you should ping the xxx.xxx.xxx.128 address as your network address.

HTH,
Jeff

[u856100]

Hi Jeff,

Thanks for replying, appreciate it.

- I understand what you have said, but would just like to clear up a couple of things for my own understanding.

- My host IP is 147.89.124.147. I thought that because the first octet starts in a the range 128-191 it is a B class network (is this wrong?)

- If you were to subdivide a network into three, what does this make the subnet mask 255.255.255.??? (170?)
(Splitting into four will give 255.255.255.63 255.255.255.127 255.255.255.191 255.255.255.255????0

You were right to assume my subnet is xxx.xxx.xxx.128 and pinging this address gives me a packet response from all connected nodes.
I assumed I was a memeber of the wrong subnet.

Once again, Thanks very much for you help Jeff!!

john

[Jeff Schussele]

Hi (again),

Yes, you're correct if you have official assigned IPs from the registering org in your geographic area - your org has a Class B network with the max possible hosts of (2 to the power 16) - 2. If you have a private network - no connection to the outside world - then I'd consider them class C solely due the subnetting & masks you are employing.

Regardless what class you are the subnetting & masking rules remain. You essentially have split the last octect (256 addresses & 254 poss hosts) into two subnets.
The lowest IP is net address & the highest is broadcast.

Splitting a net into 3 is a "sticky-wicket" as 256 is not divisible by three. You generally want to subdivide nets by powers of 2. But you could divide the net into 4 subnets with a xxx.xxx.xxx.192 subnet mask yielding net IPs of ~.0 ~.64 ~.128 & ~.192 and broadcast IPs of ~.63 ~.127 ~.191 & ~.255.
This would give max hosts of 62 in each subnet.

Not sure if splitting into 3 is possible or even allowed as I've never done it or seen it done. But if it's possible you're in the right range - the last octet would be 170/171 & yield 86/85 addresses and the max hosts would be that value minus 2.

Rgds,
Jeff

[Darrell Allen]

Hi John,

I hope this doesn't add confusion but I want to add-on to the good explanations Jeff has given.

Your IP address is for a class B network address for which the default subnet mask is 255.255.0.0. Your subnet mask (255.255.255.128) says to use an additional 9 bits of the IP address for subnetting. The remaining 7 bits of the address are used as the host portion of the IP address.

For your example (class B with mask 255.255.255.128) you have:
510 subnets
126 host addresses per subnet

The number of subnets = (2 to the power of "the number of subnet bits") - 2
So, in your case, (2 to the power of 9) - 2
or 512 - 2 = 510

The number of hosts per subnet = (2 to the power of "the number of host bits") - 2
Again, in your case, (2 to the power of 7) - 2
or 128 - 2 = 126

Why "- 2" for the number of subnets? Some routers require at least one "1" subnet bit and one "0" subnet bit. That is, no all-"1"s nor all-"0"s subnet portions of the IP address are allowed.

Why "- 2" for the number of hosts per subnet? A host portion of all-"0"s is the subnet address and all-"1"s is the broadcast address.

Can you subnet into 3 subnets? No. The smallest number of subnets you can have is 2. Next would be 6, then 14. Remember, (2 to the power of "the number of subnet bits") - 2 = number of subnets. For 1 subnet bit, that equals 0 subnets. Again, that's an issue with routers.

Lastly, 255.255.255.170 or 255.255.255.171 would be invalid subnet masks. In the old days you could get away with such a mask. Now, the standard is to allocate subnet bits contiguously. Since 170 decimal = 10101010 binary, you see the "on" bits are not contiguous.

Darrell

[Ron Kinner]

Once upon a time the broadcast address was all 0's. Nowadays it is all 1's. Some systems remember this and will reply to an all 0 as if it were an all 1. Most UNIX devices do. Microsoft doesn't. The reason you only hear from two devices is that they are probably UNIX based while the others which didn't reply this time are probably Window boxes.

Ron