Static IP addressing is, at best, a nightmare for network administrators. Managing IP addresses manually for hundreds or thousands of devices is a tedious task that opens the door to a wide variety of possible mistakes. Imagine how you’d feel if you were asked to keep your local telephone book accurate and up-to-date using an Excel spreadsheet. Now imagine how much easier your job would be if the telephone book was stored in a database and everyone had the ability to update their own information. You’d still have some work to do managing the database, but once all the information was entered, the day-to-day management tasks would be reduced tremendously.
Dynamic Host Configuration Protocol (DHCP) provides you with the ability to manage IP addresses dynamically. This Daily Drill Down will introduce you to DHCP and show you how to configure it on your NetWare 5.1 server to manage IP address assignments on your network.
What is DHCP?
In a TCP/IP environment, each device must have its own IP address. If the network is small, you might be able to manually manage IP addresses with little difficulty. You’d assign each device an address that it would use indefinitely. As the network administrator, you’d make all additions and changes. This scheme is commonly called static addressing, and it works well for small, stable environments.
Before writing off static addressing as a nightmare from the past, you should realize that there are situations that require static IP addresses. Servers, host systems, and IP-based printers are normally configured with static addresses to provide a reliable means for clients to contact them. Using DHCP to manage these types of devices would be as confusing as the post office changing ZIP codes every day. Once in a while a letter would reach the correct destination, but more often than not the letter would get lost.
As the network grows in size, it’s plain to see that a static addressing scheme has the potential to become an administrative nightmare. To ease the management burden, an addressing scheme was created that automatically assigns IP addresses to requesting devices. This type of management scheme is called dynamic addressing, and it is used by the majority of networks in existence today.
To configure dynamic IP addressing, the network administrator must install client and server software that will communicate using DHCP. The administrator then defines a range of addresses that the server can dynamically assign to the clients. When a client comes online, it will broadcast a request for an IP address. The DHCP server will receive the broadcast message and respond to the client with an available IP address. If multiple DHCP servers respond to the broadcast, the client accepts the nearest server’s address and declines all other offers. The beauty of DHCP is evident when you move a device to a new subnet. Instead of you having to manually change the IP address, the device will dynamically receive a new address that is appropriate for that network segment.
Because DHCP uses broadcast messages to locate an IP address provider, the client requesting an IP address must be on the same subnet as the DHCP server. This limitation exists because routers typically do not forward broadcast messages. To overcome this problem, you must install relay agents on all network segments that pass DHCP traffic. A relay agent is software that allows DHCP broadcasts to be forwarded to remote segments. The Novell relay agent is BOOTPFWD.NLM. This NLM runs on a NetWare file server and can be configured to pass DHCP traffic throughout a network.
DNS/DHCP Management Console
The DHCP management tool for NetWare is called DNS/DHCP Management Console. This Java-based utility can be run from the Tools menu in NetWare Administrator, or it can be launched from the Run command on a Windows workstation.
During server installation, DNS/DHCP Management Console is not installed. If you’re going to implement DHCP, you must install the utility on an administrative Windows workstation. The Management Console software requires 8.5 MB of free disk space, as well as a minimum of 48 MB of RAM. Novell recommends that the computer have at least 64 MB of RAM.
To install DNS/DHCP Management Console, you must log on to your NDS tree as a user with Admin rights to the tree. Although Management Console is not installed during server installation, the files needed to install it are located at SYS:\PUBLIC\DNSDHCP\SETUP.EXE. To begin the installation, execute the Setup program.
After you accept the license agreement, Setup asks if you want to change the destination directory. Once you’ve selected the destination directory, click Next to begin copying the files and continue the installation.
When Setup finishes copying the files, it will ask if you want to copy the NetWare Administrator snap-in files for DNS/DHCP Management Console. This is the default setting, but the choice is optional. After making your decision, click Next to continue the installation.
The following screen asks you to select the directory of the NetWare Administrator utility. Using the Browse button, navigate to SYS:\PUBLIC\WIN32, unless you’ve placed the files in another location. Click Next when you’ve navigated to the correct location.
I was prompted to overwrite a read-only file, and according to the instructions, I elected to do so. This apparently is a normal part of the installation routine, so you should elect to overwrite the existing files.
The next screen provides you with the opportunity to view the readme file. This choice is optional, but regardless of the decision you make, the installation is complete. You’ll notice that the installation routine has created a handy shortcut for DNS/DHCP Management Console on the desktop for your convenience.
Before launching Management Console, you must extend the schema to include the DHCP objects. To do this, load DNIPINST.NLM on the server. After logging in as Admin, you’ll see a screen that allows you to set the context for the DHCP objects. You can accept the default settings if they’re appropriate for your network, or you can modify them to suit your needs.
When the schema extension is finished, you should be able to launch DNS/DHCP Management Console. The initial screen you’ll see is the DNS Service screen. To manage DHCP, click the DHCP Service tab. You should be presented with a screen that looks similar to Figure A.
|You’ll first see the DNS Service screen when you start DNS/DHCP Management Console.|
When you first configure DHCP, you must start by creating the DHCP Server object. To do this, click the Create button, which is the second button from the left on the toolbar. For some reason, the button shows a picture of a box. After clicking the Create button, a pop-up window containing the different types of DHCP records that you can create will be displayed. To create the DHCP Server object, you should select DHCP Server.
Another pop-up window will appear, allowing you to configure the DHCP server. Using the Browse button, locate the container that holds the Server object. When the server name appears in the left pane of the window, click OK, and the Server object will be created. Figure B shows what the Server object looks like after it has been created.
|The Server object appears in the lower-left pane.|
One of the options that you can configure on the DHCP server is Ping Enabled. When you turn on this option, the server will ping the requesting device to ensure that it can communicate with that device. To enable this option, click the DHCP Server icon, click the Options tab, and scroll down the page until you see the Ping Enabled option. Click the check box to select it, and then click the Save button.
Now that you’ve configured your DHCP server, it’s time to create a subnet. To do this, highlight the Our Network object and click Create. Select Subnet from the pop-up window and click OK.
A pop-up window will be displayed so that you can enter the subnet parameters, as shown in Figure C. For easy recognition, you might want to structure the subnet name to read <server name>subnet. My subnet was named RHYTHM01SUBNET. Once you’ve entered the appropriate parameters for your network, you can click the Create button. After refreshing the tree, your subnet object should appear.
|You can enter subnet information here.|
Now that you’ve created the Subnet object, you must create a subnet address range. The specified range will contain the available IP addresses that can be provided to requesting devices. To specify an address range, you should highlight the Subnet object that you just configured and click Create. Select Subnet Address Range from the pop-up window and click OK.
You’ll then see a configuration window similar to the one shown in Figure D. I used the name ADDRESSRANGE and then selected the appropriate range of addresses for my network. Remember that unless you’re going to use relay agents, your DHCP server’s address must be within the IP address range that you specify. After entering the appropriate information, click the Create button.
|You must also enter the subnet range.|
Now that you’ve defined the subnet address range, you must exclude the file server’s IP address. As we mentioned earlier, all file servers should have a static IP address. To exclude an IP address from the subnet’s specified address range, select the Subnet object (RHYTHM01SUBNET in my case) and then click the Create button. Select IP Address from the pop-up window, click OK, and a configuration window will appear.
Simply enter the IP address that you’re excluding from the address range and select the default assignment type of Exclusion. After clicking Create, refresh the tree and verify that all of the objects that you’ve created are displayed. The exclusion will be displayed as another object. Figure E shows the configuration that was made on my test network.
|Here’s what my configuration looks like.|
If all the settings are correct and you can see the objects that have been created, your DHCP server configuration is complete. You can exit DNS/DHCP Management Console and move on to configuring the file server and workstation for DHCP.
Configuring the file server for DHCP
Configuring the file server will prove to be one of the easiest steps that you’ll perform when implementing DHCP. From the server console screen, load DHCPSRVR. Voila! DHCP services have been started on the file server.
To ensure that DHCP services will be started when the server is rebooted, add the command LOAD DHCPSRVR to the AUTOEXEC.NCF file. Be sure to put this command after the MOUNT ALL command.
Configuring the workstation for DHCP
Now that you’ve configured the NDS and the file server for DHCP services, you can turn your attention to the final piece in the puzzle: the workstation. My network has an NT 4.0 workstation, so those of you running Windows 9x may find that your screens differ slightly.
To configure the workstation, right-click the Network Neighborhood icon and select Properties. You should click the Protocols tab, highlight TCP/IP, and click Properties. Select the option to obtain an IP address from a DHCP server and click OK. Answer yes to the question that asks if you really want to enable DHCP. Click OK again to accept the changes, and then reboot the workstation to allow the changes to take effect.
After rebooting, verify that the workstation has an IP address that’s within the address range you specified earlier. Since you were able to log on, it’s a relatively safe bet that DHCP has been configured correctly. To prove that everything worked normally, however, you can run WNTIPCFG (Windows 9x users should use WINIPCFG) to verify the IP address of the workstation.
In this Daily Drill Down, I introduced you to DHCP and showed you how to configure it on your NetWare 5.1 server to manage IP address assignments on your network. Unless you manage an ultra-small network like my test network, which is made up of two servers and a workstation, configuring DHCP for your network will be well worth the initial investment of time. You’ll not only make IP address management more efficient, you’ll also ensure that future growth will be easier. Whenever you add a workstation to the network with DHCP enabled, it will receive an IP address automatically. How cool!
Steve Pittsley is a desktop analyst for a Milwaukee hospital. He enjoys playing drums, bowling, and most sports.The authors and editors have taken care in preparation of the content contained herein but make no expressed or implied warranty of any kind and assume no responsibility for errors or omissions. No liability is assumed for any damages. Always have a verified backup before making any changes.