Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

  • Congratulations gkittelson on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

/16 for a LAN with 30 Hosts 1

Status
Not open for further replies.

br0ck

MIS
Apr 25, 2002
467
US
I have a customer that has 4 offices and a CO all are using a /16 subnet for the networks

What is the point of this if there are less that 30 hosts on the LAN?

65K addresses seems like overkill and there are no chances of growth beyond 100 hosts

thanks

 
It doesn't hurt anything to have too large a subnet if you've got the space. That was probably setup by someone who didn't understand the mask or had no idea how big the network could grow.
 
We had a /24 subnet set-up years ago and everyone at the time said we would never need more IP's than that. Of course that wasn't the case. Moving to a /16 subnet was no walk in the park.

While yes, this is overkill in your example, it really doesn't hurt anything.
 
Yes, that's correct picohat. We narrowed our DHCP scope as well to just over what we actually needed even though our subnet is /16.
 
Which private IP class is it? The least common I have seen is B, which by default has a /16 mask. If your customer is paranoid and wants to have a scheme that is hard to guess, the class B is the way to go. If they use RIP version 1 as the routing protocol, they would have no choice but to use a /16, if they use a class B range, because RIP version 1 does not send subnet mask info in the updates.

Burt
 
Using Rip v1 is somewhat insecure from what I understand. As suggested if someone is using a Class B to make the network scheme harder to guess (Paranoid then isn't RIP v1 kinda just giving them all the info... Just a thought point out any flaws n my thoughts please.



Gb0mb

........99.9% User Error........
 
Some routers only support version one. The only reason that I can think that it may be insecure is from the danger of rerouting attacks, in which case MD5 update authentication between hosts would be best. I do not think RIP ver1 supports this, but I would not necessarily deem this as an insecure protocol.

Burt
 
I think the issue with rip one is that since it has no authentication it would be possible to send fake rip packets and mess up the routing of the network.

Glen

Gb0mb

........99.9% User Error........
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top