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

Problem : running out of Class C Address

Status
Not open for further replies.

fcf777

Technical User
Sep 15, 2002
4
0
0
CA
Our network uses the 10.10.10.0 mask 255.255.255.0 subnet. A DHCP server assigns IP addresses dynamically. However, there are also servers, routers and switches on the network with static addresses.

My problem is I want to use a class B network mask (10.10.0.0 mask 255.255.0.0) to expand the address range but it will take a lot of work to change the addresses of servers, routers and switches. Can anyone suggest the best possible approach to solving this problem? Thanks in advance..
 
How many computers do you have on the network? Do you really need to change to a class , or can you just up the subnet to a 255.255.x.0 config (i.e. 255.255.248.0)? If nothing else, can you split the subnet in half? Jordan MCP, A+
Assistant Network Administrator
 
Sorry I really meant upping the subnet to a 255.255.248.0 config, for example, to expand the address range to : 10.10.1.0 to 10.10.8.255. This will be simple for DHCP assigned addresses.

I'm not inclined to change addresses/masks on Active Directory DC's, firewalls, routers and switches. What if my workstations masks change network masks but not the servers/routers/switches - will they still see each other? For example, will a PC with address 10.10.2.2 mask 255.255.248.0 still be able to use a default gateway 10.10.10.1 mask 255.255.255.0 or telnet to a server with address 10.10.10.222 mask 255.255.255.0 ?
 
It sounds as if what you really want to do is create a new subnet or two. Rather than increase your netmask size, create a routed network, with subnets off of each interface on your router(s). Then you set up routing and the only thing that you have to change is to have routes defined on your router and dhcp scopes for each of the subnets.

I'm not sure off-hand how you define the scopes appropriate for your subnets other than having connections to each of the subnets from your dhcp server and having it issue addresses based on interface. Maybe someone else out there knows a fancier way to do it.

In my experience, subnets larger than a class 'c' get quickly bogged down by the number of broadcast and unicast packets on the network and really slow down. Subnetting and routing solve these problems. But you have to create an architecture that is friendly to routing. If your Internet connection and all of your servers go out the same interface on the router and there is not any peer-to-peer traffic, then you really haven't solved your congestion problems, you've simply made them more complex.

I assume that you have more than one server with that many clients, so you need to subnet based on which clients need access to which servers most of the time. Then traffic is mostly localized (maybe not even having to go through the router). Also, move your internet connection to a separate interface on the router. This solution will require you to modify much more than simply the router configuration, but it will provide far superior performance than simply increasing the netmask size.

pansophic
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top