WinNT Err: DHCP Specified Range Overlaps Or Already Exists

ID: Q164330

The information in this article applies to:

Microsoft Windows NT Server versions 3.5, 3.51, 4.0

SYMPTOMS

When you create a new DHCP scope, one of the following error messages may be displayed:

The specified range either overlaps with an already existing range or is invalid.

The specified subnet already exists.

CAUSE

One possible cause is documented in the Microsoft Knowledge Base article:

ARTICLE-ID: Q129122
TITLE : DHCP: Excluding an Address Before Lease Expires Gives Error

However, if this is a brand new scope, it is more likely that the scope is subnetted, and the Start Address and End Address overlap two or more subnets.

For example, an Internet Service Provider (ISP) provides a scope of addresses for a customer, from 205.241.139.160 to 205.241.139.223 and a subnet mask of 255.255.255.160. Trying to create a single scope with this starting/ending range, and using the supplied subnet mask prompts the user with the error message shown above.

DHCP Manager performs a calculation on the information the user supplies and detects that the ranges of addresses do not all exist within the same subnet. The user can verify this themselves by performing an AND-ing operation against the Start Address and the Subnet Mask and also against the End Address and the Subnet Mask. Then, the user can compare the two resulting values, if they do not match, then the starting and ending addresses are not part of the same subnet.

NOTE: AND-ing is the process of comparing 2 bits (in this case, each bit of the IP address and subnet mask). If both bits are 1, then the result becomes 1. If either bit of the IP address OR subnet mask are 0, then the result is 0. In the result below, where there are 1 values, you can see that the IP address and subnet mask both have 1 for that particular bit. Otherwise, the result for that bit is 0.

To verify that this is the problem, the user can perform the AND-ing operation using the following steps:

NOTE: Windows NT Calculator, in scientific mode, can be used to convert decimal values to binary equivalent. Each binary octet should consist of eight (8) characters, if the calculator returns a value with only seven (7) or fewer characters, then append the appropriate number of zeroes (0) to the beginning of each octet so that all octets consist of eight (8) characters.

Convert each octet of the Start Address to binary (in our case, it is only the last octet that is different, so we only need to convert the last octet, but for the purposes of this article, all four octets are converted).
```
   This example:  205.241.139.160 = 11001101.11110001.10001011.10100000 
```

Convert each octet of the End Address to binary.


   This example:  205.241.139.223 = 11001101.11110001.10001011.11011111

Convert each octet of the Subnet Mask to binary.


   This example:  255.255.255.160 = 11111111.11111111.11111111.10100000

Perform an AND against the Start Address and the Subnet Mask.


   Start Address:  11001101.11110001.10001011.10100000
   Subnet Mask:    11111111.11111111.11111111.10100000
   --------------  -----------------------------------
   Result #1:      11001101.11110001.10001011.10100000

Perform an AND against the End Address and the Subnet Mask.


   Start Address:  11001101.11110001.10001011.10100000
   Subnet Mask:    11111111.11111111.11111111.11011111
   --------------  -----------------------------------
   Result #2:      11001101.11110001.10001011.10000000

Compare Result #1 and Result #2 - if they do not match, the scope overlaps a sub network.


   Result #1:      11001101.11110001.10001011.10100000
   Result #2:      11001101.11110001.10001011.10000000
   --------------  -----------------------------------
                   same    .same    .same    .different --> different
                                                             subnets

This error message will also be displayed if you attempt to create a scope that has a beginning address that is the same as your network ID (scope ID).

RESOLUTION

In the above example, the ISP provided a range of 64 addresses without indicating that it was a range of addresses for two (2) sub-networks, consisting of 32 addresses for each subnet. All you did then, was to divide the range in two, resulting in the following two scopes:


   This example: Scope #1:
                 Start Address - 205.241.139.160
                 End Address   - 205.241.139.191
                 Subnet Mask   - 255.255.255.160

                 Scope #2:
                 Start Address - 205.241.139.192
                 End Address   - 205.241.139.223
                 Subnet Mask   - 255.255.255.160

If you calculated the subnets for the DHCP scope, type in a scope (Starting and Ending Addresses) for each subnet you calculated, individually. If the subnets were provided by a third party (such as an ISP), ask them to provide a list of how the subnets are divided, including the ranges of host IDs within each subnet. Use this list to create your DHCP scopes.

NOTE: Versions of Windows NT DHCP server prior to Windows NT 4.0 Service Pack 2 (SP2) are not capable of assigning addresses from more than one scope to a given physical subnet. One workaround for this situation is to add additional network interface cards (NICs) to the server, and to address each of the NICs to a given logical IP subnet. This involves additional and otherwise unnecessary hardware, so a new solution was developed and implemented in Windows NT 4.0 SP2. The enhanced DHCP server allows the administrator to create different scopes (ranges of IP addresses), and then to group those scopes together into a superscope. The procedure to create superscopes is described in the following Microsoft Knowledge Base article:

ARTICLE-ID: Q161571
TITLE : Using DHCP "Superscopes" to Serve Multiple Logical Subnets

Keywords : kbenv kberrmsg kbnetwork kbhowto nttcp NTSrv ntui
Version : WinNT:3.5,3.51,4.0
Platform : winnt
Issue type : kbprb