Open Source

SolutionBase: Configuring Red Hat Enterprise Linux

Installing Linux is only half the battle. Once you get it installed, you've got to configure it to play nice in a Windows environment. Brien Posey shows you what you need to do.

Making the move from Windows to Linux can be a big step. Once you've finished the installation process, there is still quite a bit of configuration that you will have to do before your server is ready to use. In this article, I will discuss the configuration process.

Post-installation housekeeping

When the installation process first completes, you are asked to reboot your server. When the server reboots, Linux greets you with a Welcome screen informing you that there are some additional configuration tasks that you must complete before you can actually use Red Hat Enterprise Linux.

Click Next and Linux will display the software's license agreement. Click Yes to accept the license agreement and then click Next. At this point, you will be asked to set the system's date and time. The process for doing so is pretty straight forward, but if you prefer, you can have the computer's date and time set through the use of the Network Time Protocol, rather than setting it manually.

If you want to use the Network Time Protocol to set the system's date and time, you must simply select the screen's Network Time Protocol tab and then select the Enable Network Time Protocol check box. After doing so, just add your network time server to the list of servers and click Next.

After you set the system's date and time, the next screen that you will see asks you to select your display resolution. My experience has been that Linux should automatically detect your video adapter and your monitor. However, you may not be initially allowed to take full advantage of your video adapter's full capabilities. For example, the system that I am using while writing this article is equipped with a PCI Express video card from ATI that contains 256 MB of RAM. Even so, Linux only allowed me to set my display to a maximum resolution of 800 x 600. I will show you how to correct this later on.

After selecting your display resolution, you will come to a screen that might be a bit foreign to those of you who have previously administered only Windows systems. This screen asks you to enter your Red Hat login information.

What this screen is asking for is for the username and password that you or your company set up on Red Hat's Web site prior to purchasing Enterprise Linux. You can complete the configuration process without entering the account information, but I advise going ahead and entering your account information. The reason is because when you purchase Enterprise Linux, you are also purchasing support and software maintenance services. In order to qualify for these services, you must link an account to your deployment.

After entering your Red Hat Network credentials, you are asked to activate your subscription. The activation process is designed to prove that you own a legitimate Red Hat Enterprise Linux license. The activation process is a little bit different from the way that Microsoft makes you activate Windows though. The Windows activation process transmits your product key to Microsoft and returns an activation code. If you have worked through the Linux installation process then you know that there is no product key for Enterprise Linux.

Rather than basing your server's activation on a product key, Red Hat bases the activation on a subscription number. Your subscription number is e-mailed to you at the time of purchase. Therefore, to activate Linux, you must enter your subscription number and a name that you want to assign to the server that you are setting up. You also have the option of including your server's hardware profile and package list with the activation

Enter your subscription number and a name for the system and click Next. There will now be a brief delay while your activation information is transmitted to Red Hat. At this point, you will see a screen asking you to create a system user account. A system user in Linux is the same thing as a regular user in Windows. It's just a basic user account that does not have administrative privileges.

This particular screen is deceptively simple. You could just enter a username, a full name, and a password, click Next, and be done with the user creation process. The problem is that in doing so, you are creating a local user account that is only good on the server that you are setting up. That's fine if you have a tiny organization and this is going to be your first and only server. If you already have a network in place though, then you probably want to configure Linux to use your existing user accounts.

Because I am writing this article series from the standpoint of guiding Windows administrators through their first Linux deployment, I will assume that you have an existing Windows network and that your existing user accounts are Active Directory based. For now, don't enter any system user account information. I will address this topic at length later in the article. For now, just click Next to continue.

The next screen that you will see allows you to set up your sound card. Normally, Linux should automatically detect your sound card. You can confirm that your sound card is working by clicking the Play Test Sound button. Of course, if you are setting up a server, it's usually no big deal if the machine doesn't have sound.

Click Next and you will see a screen asking you if you have any additional CDs that you would like to install. You can use this screen to install things such as documentation disks, third party plug-ins, and even applications. Of course installing additional CDs is optional. If you don't have any CDs to install, just click Next.

At this point you should receive a message indicating that the initial setup and configuration is complete. Click Next one last time to launch Linux.

Fixing the screen resolution

During the initial configuration process, I explained that when you initially set your display resolution, the choices that Linux gives you might not reflect your video adapter's true capabilities. The reason for this is because Linux attempts to give you display resolutions that are safe for your video adapter and monitor combination. Linux will give you especially low resolution choices if you have an unknown monitor type.

To fix the problem, select the Run Application command from the Actions menu. Now enter the following command:

system-config-display

(This command was previously redhat-config-display in some earlier Enterprise Linux versions). After a brief delay, Linux will display the Display Settings properties sheet.

The properties sheet's Settings tab is selected by default. This tab looks like the screen that you have already seen, which allowed you to select your display resolution and color depth. The Settings tab will still contain the same limited set of choices that you had earlier though.

To change the choices that are made available to you, select the Hardware tab. The Hardware tab allows you to set your monitor type and your display adapter type. Linux generally does a pretty good job of automatically detecting your video card type, but make sure that the correct video card has been detected. If the video card is set wrong, you can change it by clicking the Configure button and selecting a new video card from the list.

Now, look at the monitor type. For the purposes of this article, I am going to assume that Linux is set to an unknown monitor type as it is on my test machine. Click Configure and you will see a long list of monitor brands. If you have a generic monitor or have a model that isn't listed, you can use the Generic CRT Display or Generic LCD Display options.

Just selecting one of these generic monitor types won't do anything to increase your screen resolution though. For that, you need to click the arrow icon to the left of the generic monitor listing. This will cause the generic monitor listing to expand and display a list of display resolutions. Now, just select your monitor's maximum display resolution and click OK. If you go back to the Settings tab, you should now be able to select a higher resolution. Click OK to return to the Linux desktop. You will have to reboot in order for your changes to take effect.

Networking

Earlier in this article, I explained that the server's network configuration is of particular importance if you are bringing a Linux server into a Windows environment. Before I show you how to configure anything, let me just say that there are many different ways to make Linux co-exist with Windows.

For example, a common practice is to continue to allow the Active Directory to store the user account information and to configure Linux to authenticate users through the Active Directory. For the purposes of this article, I am going to show you how to use SAMBA to turn your newly configured Linux server into a file server that is accessible to Windows users.

In case you are not familiar with SAMBA, it is an add-on for Linux that allows the Linux server to function as a Windows compatible file or print server. Windows clients access the SAMBA shares through Server Message Blocks (SMB). Although there are a number of protocols that support the use of Server Message Blocks, SAMBA is only compatible with TCP/IP.

The first step in the configuration process is to install the SAMBA software. To do so, select the System Settings | Add / Remove Applications commands from your Linux Server's Applications menu. When you do, you will see a list of all of the available packages. Scroll down to the Servers section of the package list and select the Windows File Server check box. Click the Update button and follow the prompts to install the necessary packages.

To configure SAMBA, select the Server System Settings | Server Settings | Samba commands from the server's Applications menu. When you do, Linux will open the Samba Server Configuration dialog box. The first thing that you need to do to prepare SAMBA for use is to configure a few basic server settings. To do so, select the Server Settings command from the Preferences menu. When you do, the Server Settings properties sheet will appear.

This properties sheet is divided into two tabs. The Basic tab contains a workgroup field and a Description field. Even though most Windows networks don't use workgroups any more, you must still fill in the Workgroup field. Rather than entering a workgroup name, enter the name of the Windows domain that you would make the server a part of if it were a Windows Server. Windows users will have the illusion that the Samba server is a part of the domain that you specify.

The Description is also important. The description that you enter will be part of what Windows users see when they browse the domain. For example, since I only have one Linux server on my network, I assigned this server the name Linux. I then set the Description field to Samba Server. When I browse my specified domain from a Windows workstation, the Linux server is listed as Samba Server (Linux). As you can see, the name that the Windows users see is based on the Samba description and the server name.

After filling in a workgroup and description, select the properties sheet's Security tab. The first thing that you have to set on this tab is the authentication mode. The default authentication mode is User, which means that authentication is based on a Windows password and a Linux password. It is possible to authenticate users through the Active Directory, but doing so is beyond the scope of this article.

While you are on this tab, it is very important that you set the Encrypt Passwords option to Yes. Windows XP users will not be able to access the Samba shares unless the password is encrypted. The guest account should typically be set to No Guest Account, and the Authentication Server and Kerberos Realm options should be left blank if user authentication is being used. Click OK to accept your settings.

Assuming that you are configuring Samba to use user authentication, then the next thing that you will have to do is to create some Samba users. You can do so by selecting the Samba Users command from the Preferences menu and then clicking the Add User button. The process of creating a Samba user involves mapping a Unix user account to a Windows Unix account. You must begin the process by selecting the Unix account that you want to use from the UNIX Username drop down list.

If you don't have any UNIX accounts set up yet, you can define them by using the Users and Groups option found on the Applications | System Settings menu. Take care to give any user accounts that you create a name that matches that of the Windows account that you will map to it.

After you select the UNIX username from the Create New Samba User dialog box, you must enter the name of the corresponding Windows account. This will usually match the name of the UNIX account that you have selected. Now just enter a Samba password. This password should match the passwords used by the UNIX account and the Windows account. Click OK to define the Samba user.

The last step in the process is to create a Samba share. To do so, click the Add button found on the Samba Server configuration dialog box. When the Create Samba Share dialog box appears, enter the path to the directory that you want to share, followed by a share name and description. I recommend not using spaces in the share name because avoiding spaces will make it easier to manually map to the share from Windows machines.

The Basic tab also contains a permissions option. You can set the share's permissions to either Read Only or to Read / Write. The Create Samba Share properties sheet also contains an Access tab. You can use this tab to select which Samba users have access to the share. After doing so, just click OK and you are in business.

Editor's Picks

Free Newsletters, In your Inbox