no ip subnet-zero command

When would you ever want to use the "no ip subnet-zero command" given that it takes away from your allowable address range?

take an example of class C /26 (i.e 192). When ip subnet-zero is removed, we have a subnet of 64 & 128 only.
But when subnet-zero is enabled we can have a subnets of 0, 64, 128 & 192.

No I understand how it works... I just want to know why you would ever want to do that? What is the point of making hosts 1-62 and 191-254 not usable? It just seems like a waste.

Hi,

As far as I know there were issue with those subnets in classful networks (where netmask information were not exchanged by the routing protocols).

To give an example, how would you differentiate network address 10.0.0.0/8 from 10.0.0.0/10 ? Same for the broadcast of the last /10, how would you differentiate 10.255.255.255/8 from 10.255.255.255/10?

To give an example, how would you differentiate network address 10.0.0.0/8 from 10.0.0.0/10 ? Same for the broadcast of the last /10, how would you differentiate 10.255.255.255/8 from 10.255.255.255/10?

Well Subnet Zero does indeed get used:

www.cisco.com/en/US/tech/tk648/tk361/tec...186a0080093f18.shtml

I'm not exactly sure how that is managed I assume it has something to do with the fact that while the hosts ranges over lap the subnet masks are respectively different.



So at this point I'm wondering when is it okay and not ok to use the zero subnet?

hi
you should use "no ip sub.." while you have network devices that cannt communicate with devices which are on zero subnets , usually old routers. something in the way they read the binaries of the host in the zero subnet will not let them forward or reply to your request

hope that helps

Roy