CSCI 2150
Linux Package Installation Lab

The following set of steps takes you through the installation of software using the Red Hat Package Manager and by compiling source code using make. It uses the text-based command line terminal, which due to the overwhelming majority of programs that use a graphical user interface (GUI), may be unfamiliar to most people.

For the most part, there are a few things you need to know about using the terminal interface for Linux.

Depending on how you installed Linux, your installation of Linux may or may not boot straight into a text terminal. If it goes to X-Windows, there are two methods to open the terminal interface.
- The first method, which may be different depending on which distribution of Linux you installed and which desktop you are using, is to open the terminal interface from the desktop menu. In Gnome running on Redhat 9.0, you can select "System Tools" from the main menu then click on "Terminal". This will bring up a terminal window on the desktop.
- The second method is the most reliable, but is also the most difficult to use because it turns the entire desktop into a text screen. Simply press CTRL-ALT-F2. This opens terminal window 2. To return to the desktop, press ALT-F7
In Linux, there are multiple text terminals, typically 6. To switch back and forth between them, press ALT-Fx where 'x' represents 1 through 6.
Each command will be entered one at a time by typing the command then pressing Enter to execute the command. The up arrow and down arrow can be used to scroll through previous commands which can be edited before pressing Enter to execute.
All Linux commands and file names are case sensitive. Be sure to match the uppercase and lowercase format used in all of the commands printed in this document. Be careful to also match the spacing. This is where the majority of errors occur in this lab.
If you are typing the name of a file or a directory, Linux can help you by filling in the rest of the name before you are finished. For example, assume a directory/folder contains the following files:

private .txt
program
myfile.dat

If you are trying to type the name "myfile.dat", simply type 'm', then press the tab key. Linux will fill in the rest of the name for you if there is only one option. If there are more than one possible option, for example if you typed "pr" then tab, Linux will beep at you to tell you it cannot isolate which file you are referring to. Simply type an additional letter, for example "pro", then tab, and Linux will finish the string of characters

Important: Some of the operations you will be performing in this lab require that you are logged in as the superuser with the username "root". If you have not logged into your system as root, type "su" at the command prompt (without the quotes). You will then be asked for the root password. After entering it, you will be logged into the terminal as the superuser. To determine who you are logged into a terminal as, type "whoami" without the quotes.

Make the DOS floppy visible to the Linux operating system

First, we will need to get access to the files to be installed. Virtual PC makes it difficult to access the physical drives of the machine, so what I have done is created an image of a floppy disk with the required files. To make the Linux virtual machine see this file as a floppy diskette, you will need to "capture floppy disk image." To do this, click on the "Floppy" menu item and select "Capture Floppy Disk Image..." This will open a file window which will allow you to select the image you downloaded from the above link. All that this did was to act like you inserted a disk in the floppy disk drive. Linux does not see it yet. To do this, you need to mount the floppy disk.

Linux treats everything including disk drives as subdirectories of "/". Therefore, in order to read a floppy diskette, we need to make a directory for the floppy drive to be visible through. To do this, we must create an empty directory that we can mount to the floppy drive. (Mount simply means to connect or enable the resource to make the it available.)

Your install of Linux has probably already done this for you. Enter the command

ls /mnt/floppy

"ls" is a command to list all of the files and folders in a specific folder, so ls /mnt/floppy tried to list the contents of the folder /mnt/floppy. If the command executed without an error (most likely there would have been no output), that folder exists. If there was an error, something like, "No such file or directory," then follow the following instructions.

Move to the top directory in the directory structure by typing cd / then pressing Enter.
I usually create a subdirectory called mount under which I add the directories that I will be mounting my devices to. To do this, type mkdir mnt then press Enter. If an error like, "mkdir: cannot create directory '/mnt/': File exists," appears, then the /mnt/ folder already exists.
Make the current or active directory the mnt directory by typing cd /mnt then pressing Enter.
Now that you're inside the mnt directory, we can create the directory that we will be connecting the floppy to. Type mkdir floppy then press Enter.

"cd" is a command to change the current directory or folder. It allows you to pass from folder to folder in a system.

Now that we've verified that the directory to connect the floppy to exists, we need to tell Linux to go connect to the floppy. It is important to note that the diskette image you downloaded is of a Microsoft DOS diskette, not a Linux diskette. It will be explained later how to address this.

Place the floppy disk that you received from me in the A-drive.
Type mount -t vfat /dev/fd0 /mnt/floppy and press Enter.
- mount is the command to tell Linux to make the connection between a device and a directory name
- -t vfat tells the mount command that the diskette is of type vfat, i.e., it is a Microsoft DOS diskette.
- /dev/fd0 tells the mount command which device is being connected to the directory. All Linux devices including serial ports, printer ports, CDROMs, and hard drives have names defined under the /dev directory.
- /mnt/floppy is the directory we created to connect the diskette to.
Now type ls /mnt/floppy/ to see if the files appear. You should see two files: faq-6.2-1.noarch.rpm and shed-1.10.tar.gz.

What we want to do now is copy these two files to your root directory so that we can use them for our two installations. Perform the following steps to copy the files from the floppy.

First, make the current or active directory the one that we defined for the floppy. Type cd /mnt/floppy then press Enter.
To see what files are on this disk, type ls -l then press Enter. (The character after the dash is a lowercase L, not a one. The -l makes ls show the details of each of the files.) Once again, a list of the files and folders that are contained in this folder should appear, but this time it will have details such as the last store dates and times, the file sizes, the userid of the person who created the file, and the privileges associated with the file. Your list should look somewhat like the list shown below.

-rwxr-xr-x 1 root root 393115 Oct 30 19:38 faq-6.2-1.noarch.rpm -rwxr-xr-x 1 root root 30869 Oct 30 19:38 shed-1.10.tar.gz
The command "cp" is used to copy files from one location to another. Typing cp faq-6.2-1.noarch.rpm /root/ then pressing Enter will copy the file faq-6.2-1.noarch.rpm to your directory /root. Remember to use Tab when entering the file name. For example, type "faq" then hit Tab. This will make it so that the file name will not have any typos. It will also mean that any type I've created in this set of instructions will not be transferred to your lab.
Repeat the previous step to copy shed-1.10.tar.gz to your /root/ directory by replacing the string faq-6.2-1.noarch.rpm with the shed file name.
Lastly, to release the mounted floppy diskette from the operating system, we must use the umount command. Before we can release it, we must be out of the directory it is mounted to. Type cd / and Enter, then type umount /mnt/floppy then press Enter. This must be done before you eject the diskette. Note: The command is "umount", not "unmount."

At this point, you should have the correct files loaded to your virtual machine's hard drive to continue with the rest of the lab.

Installing Software Using RPM

Although the installation seemed to take forever and it seems like everything you'd ever need is included in your Linux installation, you may want to install other applications or files later on. One of the simplest ways to do this is to use a tool called the Red Hat Package Manager (RPM).

Packages are groups of files including applications, settings, configurations, etc. that the developer or distributor has compiled into a single file for installation. (Not all Linux software comes as a package.) The file that you copied to the /root directory was a package, not of applications, but of HTML files. Specifically, it is a set of HTML files that are Frequently Asked Questions (FAQs) about Linux. Follow these steps to install the files.

Change your current directory to /root/ by typing cd /root the pressing Enter.
Next, use the ls command to verify that the file faq-6.2-1.noarch.rpm was successfully copied to the directory. Unless you've added other files to your /root directory, there should only be two files present, the rpm file and the shed file.

You will be using RPM to install these files. Some basic formats of the RPM command are shown in the table below.

Command	Action performed
rpm -ivh packagename	Install a package
rpm -e applicationname	Uninstall a package
rpm -qa	List all installed packages
rpm -Va	Check that all packages are installed correctly
rpm -qpl packagename	See the names of the files installed by the RPM
rpm -qpi packagename	Get general information about a package
rpm -q -whatrequires filename	Find out which packages rely on a file

Type rpm -qa then press Enter. According to the table above, this will produce a list of all of the packages that have been installed on your machine. There should be quite a few of them.
To install software, the table above indicates that we should use the command rpm -ivh packagename where "packagename" is the name of the package we wish to install. Type rpm -ivh faq-6.2-1.noarch.rpm then press Enter. Remember again to use the Tab key to complete the typing of the file name.
If the installation was successful, you should see output similar to that shown on below.

Preparing... ######################################## [100%] 1:faq ######################################## [100%]

Using the Installed Software

So where did it all go? In order to use the software, it would be nice to know where RPM put it. According to the table above, the command rpm -qpl packagename will list all of the files installed by RPM where "packagename" is replaced with the name of the package we installed.

Type rpm -qpl faq-6.2-1.noarch.rpm then press Enter. You should see a list of files fly by, and in that list you should see that the HTML files were installed in a directory /usr/doc/FAQ/html.
Change to this directory by typing cd /usr/doc/FAQ/html and pressing Enter. Typing ls should reveal all of the files that were installed here.
The files that you installed comprise a set of FAQs on Linux topics. Depending on what options you selected for your Linux install, there may be a simple text-based HTML browser called lynx that you can use to view these files. If there is, typing lynx Linux-FAQ.html then pressing Enter will display them. Many of you may not have this browser installed. In this case, use the graphical web browser that you access from your desktop or type cat Linux-FAQ.html to simply dump the HTML file to the screen.

If you have the text-only browser lynx, the following is a brief tutorial on its use.

The up and down arrow keys highlight each link
The right arrow follows that link to its referenced page
The left arrow goes back to the previous page
PgUp and PgDn scroll through the displayed document
Q quits

If you wish, you may call me over to verify that your installation was successful.

If you wish to uninstall the HTML files that you just installed, type rpm -e faq then press Enter. Note that "faq" is the application name, the same name which was displayed when you installed the software.

Compressed Packages

Linux typically delivers groups of files and directories in a package that has been compressed to reduce the overall size of the package. There are actually two steps to restoring these files before they can be used on a Linux system.

To begin with, when a software developer has completed a set of files and is ready to ship them, he or she groups them together using a program referred to as tar. The name tar stands for "tape archive" and was originally used to make it easier to move a large group of files as a whole.

The next thing the software developer does is to compress the "tar-ed" file into a smaller package. The following file extensions are used to distinguish the different types of files.

filename.tar -- this is the format of a file name that has simply been packaged but not compressed.
filename.tar.gz or filename.tgz -- this is the format of a file name that has been packaged and also compressed.

Therefore, in order to restore the files you have downloaded from a developer, you must perform the reverse set of operations. First, you must uncompress the files in order to retrieve the original tar file. Then, you must apply the tar program in order to restore the files and directory structure that the developer created. With the correct switches, later versions of tar can be used to uncompress the package and unpack all of the files and directories.

Move to the /root directory by typing cd /root. (Do not include the period.)
You should have already copied the necessary file to this folder when you mounted the floppy diskette. All we need to do is uncompress it using the command tar -xvzf shed-1.10.tar.gz. (Do not include the period.) Remember to use Tab!
Take a directory using ls -l, and you should see a new directory/folder has been created called "shed-1.10". This contains all of the files we will be using to install the simple application by compiling it.

Using Make

The files you uncompressed in the previous section are for a program called shed. Shed, which stands for "simple hex editor", will allow you to open a file and view its contents in ASCII, hexadecimal, decimal, octal, and binary. (You can download your own copy from the URL http://sourceforge.net/projects/shed.) The following instructions show how to install shed from the downloaded package.

To install the software, move into the directory shed-1.10 by typing cd shed-1.10.
The installation instructions say to run the command ./configure (the "./" tells Linux that the program or script that you are running is in the same directory as the one you are currently in.). This command prepares your files to be compiled into an application for your system.

As a convenience to programmers and the end users who might be compiling software, almost all software development environments come with a method to maintain all of the settings including a list of all of the files needed to create an application. For command line compilers, this operation is usually called "make". When you installed the development packages for Linux, make came as part of that package. Each compilable program contains the parameters for make that in a text file called "Makefile". (Type cat Makefile to type these parameters to the screen.)

To make your shed application, type the command make while you are in the shed-1.10 directory.
Typing ls -1 should reveal a program called shed in your directory.

If you are interested in seeing shed run, type ./shed README to open the text file README and see what the different characters are equal to in ASCII, hex, decimal, octal, and binary.