Chapter 1: UNIX Operating System: Mind the Gap

cedar_1982

Chapter 1: UNIX Operating System: Mind the Gap
Overview
Sun Certified System Administrator for the Solaris 10 Operating System. Sounds great, I want this certification. But by the way, what do they mean by Solaris 10 Operating System?
If this sounds like you, welcome to this chapter. Otherwise, you can quickly browse through this chapter, or if you are already familiar with UNIX, you can skip this chapter altogether. In other words, this chapter is meant for folks【人们】 who are relatively new to the UNIX world and want to come up to speed quickly to fill the gap before they start this journey toward their Solaris System Administration certification. If you have bought this book, you cannot possibly be a perfect stranger to an operating system. When you work on any computer, you actually interact with the operating system such as Windows, Mac, Linux, or Solaris.
So the central question in this chapter to think about is: what is Solaris, and how do we use it? In search of an answer to this question, we will explore three thought streams: what is an operating system (OS), how is Solaris OS related to UNIX, and how do you interact with Solaris?
Understanding the Operating System
Your computer is made of hardware and software components. Some examples of the hardware components include the central processor unit (CPU), which processes the instructions coded in a software program; the physical memory, which stores the instructions and data of a currently running program; and a hard disk, which stores the programs that are not currently running. You can look at a computer as a processor. It takes some input data from the user, processes it, and produces results in terms of output data. You use some input device to feed the input data to the computer, and the computer uses some output device to display the results or the output data. Examples of input and output (I/O) devices connected to a computer include monitor, keyboard, disk, and printer.
Applications (the software programs that do something useful for users) such as the Netscape browser and Microsoft Word are examples of software components. Without applications computers would be of little use for their users. The applications (software components) need hardware components to run. For example, an application needs the CPU to process the instructions, as well as memory and a hard disk to store instructions and data. The relationship between computer applications and the computer hardware gives rise to the following three issues:
Each application requires access to resources such as CPU and I/O devices. The access must be managed.
The resources themselves must be managed.
If multiple applications are running concurrently, they may try to access a resource at the same time. Who would break the tie?
These issues apply to all applications. Therefore, it is not efficient to handle them separately within each application, which is possible in principle but would be a duplication of efforts. Consequently, these issues are handled for all applications in one software system called the operating system. As shown in
Figure 1-1
, all the applications access the computer resources through the operating system. This is a very high level view: each layer may have sublayers.
Figure 1-1: Relationship among applications, operating system, and computer hardware
An operating system performs the following types of functions:
Control program. The operating system is a control program that controls the I/O devices and the execution of the user's programs (and applications) to ensure the correct operation of the computer. For example, if two programs are trying to access the hard disk at the same time, the OS would decide in which order they would have access, and it manages the usage of the CPU by multiple programs executing concurrently.
Resource manager. The operating system also works as a resource manager. It allocates the hardware and software resources needed by a running program. CPU, I/O devices, and files are some examples of resources.
Mediator. The operating system acts as a mediator (agent) between the users (or applications) and the computer hardware.
So, the operating system is a software system that controls and manages computer resources and acts as an agent between the user and the computer hardware, as well as between the applications and the computer hardware. There are a number of operating systems available in the market—for example, Windows, which is a family of operating systems from Microsoft, and UNIX, which is a family of UNIX-based operating systems, such as Linux and Solaris, from multiple vendors. If you know the history of UNIX, you will know not only that UNIX was the first operating system but that it has also been a trail blazer【传播者】 in the field of operating systems for providing state of the art features. Other operating systems such as Windows have adopted the features first proposed by and implemented in UNIX.
Evolution of the UNIX Operating System
Solaris is a flavor【味道】 of UNIX—that is, it is one of many members of a family of operating systems called UNIX. As a Solaris system administrator, you need to understand the history of UNIX—its origin, how it evolved, and where it is now. The most important point to remember in exploring the history of UNIX is that UNIX is not an operating system that was built by one company with a wonderful marketing department.
So, repeat: UNIX, pronounced as yoo-niks, is not the name of one operating system; it refers to a family of operating systems.
A Brief History of UNIX
The roots of UNIX lie in Comprehensive Time Sharing System (CTSS) developed by F. Corbato at MIT in the early 1960s. Continuing along the principles of CTSS, a multiuser and multitasking system was designed under the Multiplexed Information and Computing Service (MULTICS) project started by General Electric and AT&T Bell Labs. Because the project fell behind schedule, AT&T pulled out in 1969, but the work of building a system along these lines continued, and during the early 1970s a system called Uniplexed Information and Computing Service (UNICS) was developed at Bell Labs in New Jersey. Some important events in the history of UNIX subsequent to these early developments are listed in
Table 1-1
.
Table 1-1: Some events in the history of UNIX
Year
UNIX Event
1969
Emergence【出现】of UNIX at AT&T Bell Labs for a Digital PDP minicomputer as an alternative to GE's multiuser mainframe running MIT's Multics,
1970–1972
System refinement and many new features added. Confined to AT&T sites in New Jersey. Demand for UNIX started growing, AT&T, reluctant, to go into business outside its arena of telephony and telegraphy, gave away UNIX to universities and research sites for a nominal fee with no support and no bug fixes. This gave birth to the UNIX community.
1973
Sixteen UNIX installations by February,
1974
First UNIX users meeting at Columbia University's College of Physicians and Surgeons in May 1974, attended by 24 people from 12 institutions.
1975
Second UNIX users meeting attended by 40 people from 20 institutions
1978
First Berkeley Software Distribution (BSD) version released.
1982
SUN Microsystems Inc. released first UNIX workstation based on the BSD flavor of UNIX.
1983
BSD 4.2 included TCP/IP implementation. AT&T releases UNIX System V.
1989
SPARCstation 1 introduced.
UNIX is the oldest operating system that pioneered several OS concepts used by other operating systems such as Microsoft DOS and Microsoft Windows. Originally, Bell Labs distributed UNIX along with the source code so that, anybody could modify and customize the OS to meet specific needs. This gave rise to several flavors of UNIX, some of which are listed in
Table 1-2
along with the hardware architectures they support.
Table 1-2: Examples of UNIX flavors
UNIX Flavor
Vendor
Chip Architecture
Solaris
Sun Microsystems
SPARC, x86
HP-UX
Hewlett-Packard
IA-64, and HP PA-RISC
Irix
Silicon Graphics
MIPS
AIX
IBM
PowerPC
UNIXWare
SCO/Caldera
Xeon, Pentium, x86
Linux
Freeware
Alpha, SPARC, Power PC, and Intel
Accordingly, when people say "UNIX," they are really referring to the common functionality of a family of operating systems — that is, multiple flavors of UNIX.
A high-level view of a general architecture of the UNIX OS is presented in
Figure 1-2
. The heart of the OS is the kernel that contains the programs to manage and control resources. It's the kernel that interfaces with the computer hardware. The users (or the applications) communicate with the kernel through a software program called the shell. The shell interprets the commands from the user for the kernel.
Figure 1-2: Structure of UNIX
Some important characteristics of UNIX with their advantages and disadvantages are listed in
Table 1-3
.
Table 1-3: Characteristics of UNIX
UNIX Characteristics
Advantages
Disadvantages
UNIX is built from small components; each component does one job and does it well.
Simplicity. Freedom to combine these components to create new and powerful tools
Makes it difficult to get a unified grasp of the entire system.
Support for virtually all network protocols.
Makes UNIX the premier networking system.
Too sophisticated【 复杂的】 for a common user.
Many flavors of UNIX.
Multiple vendors, competition, more choices for customers.
Confusing because of lack of a unified system.
Wide variety of software available for UNIX.
Increases usability.
System becomes bigger in size and more complex. Troubleshooting becomes more involving.
Open standard, open source.
Freedom to innovate. Makes the system better on a regular basis.
More varieties, less commercial/customer support.
UNIX supports running multiple tasks and processes at the same time. Not only multiple user programs, but also some programs in the background running all the time.
Makes the system efficient and powerful.
Makes the system more cumbersome【不方便的】for most users. Troubleshooting becomes more involving.
Support for multiple users. Multiple users can log on to the same machine and do their work.
Enhances efficiency and usability.
Makes the system more cumbersome for most users. Troubleshooting becomes more involving.
These days, other operating systems such as Windows also offer support for network protocols, multiple processes, and multiple users, but UNIX has been the trail blazer in these areas. The underlying philosophy of UNIX may be outlined as follows:
·         Modularity (independence). UNIX is composed of small, independent components (referred to as programs, utilities, or tools). Every component is designed to do one thing well. This makes UNIX very modular and efficient.
·         Interrelation (interconnection). The components are designed to be able to connect to each other. For example, the output of one component may become input into another component. This makes UNIX very powerful.
·         Evolution (innovation). UNIX is designed to facilitate building new components. This keeps UNIX evolving【进化的】and multiplying【繁殖】.
This book is about Solaris which is the current name of a flavor of UNIX that became the reason for founding a company called SUN (Stanford University Network) Microsystems, and it still remains one of its two flagship【旗舰】product lines, the other being Java.
A Brief History of Solaris
Except for the recent rise in the popularity of Linux, no other flavor of UNIX has been as popular as Solaris from SUN Microsystems, founded in February 1982 by Bill Joy (among others), who was previously involved in the development of BSD 4.1. It was the improved version of BSD 4.1 that was released by SUN as SunOS in 1983.
In the early 1980s, AT&T saw a future in UNIX; it released System III in 1983, followed by System V. From there on, for a long time, the evolution of UNIX was for the most part carried on by the interaction between BSD and System V. In 1984, AT&T released System V release 2, followed by release 3 in 1987. In 1988, AT&T shocked the UNIX community by purchasing a percentage of Sun Microsystems. This triggered efforts for merging the distribution of BSD and System V. In 1990, AT&T released System V release 4, which was a merger of System V and BSD. So, SunOS 5.x, released in 1992, was based on System V release 4 and as a result was significantly different from SunOS 4.x, which was based entirely on BSD. At this time, AT&T once again exited the UNIX world (it was a distraction for the company from its focus on telecommunications hardware) by selling its UNIX Software Lab to Novell; Novell eventually sold its UNIX (UNIXware) to Santa Cruz Operation in 1995.
SunOS 5.x, based on System 5, was also referred to as Solaris. The rest of the Solaris releases are listed in
Table 1-4
.
Table 1-4: History of Solaris releases
Year
Solaris Release
1992
Solaris (also called SunOS 5.x, Based on System 5.)
1994
Solaris 2.4
1995
Solaris 2.5
1997
Solaris 2.6
1998
Solaris 7
2000
Solaris 8
2002
Solaris 9
2004
Solaris 10
The ability to create new flavors of UNIX offers the freedom to innovate, but it can also cause confusion and incompatibilities. So, to keep some coherence within the multiplicity of flavors, some open standards were needed.
The UNIX Unification: POSIX Standards
The problems posed by variations among several flavors of UNIX motivated efforts to developing standards for UNIX. A standard called Portable Operating System Interface (POSIX) was developed by the Institute of Electrical and Electronics Engineers (IEEE) and the American National Standards Institute (ANSI). The goal of this standard is to define a standard operating system that functions as UNIX.
In fact, POSIX is a set of standards. Each standard defines a particular aspect of the standard operating system. For example, POSIX. 1 defines the standard for an application interface that the OS would provide, whereas POSIX.2 defines the command interface—that is, the shell.
In this age of the Internet (the information age), you hardly ever work on an isolated computer; instead, you work on (or interact with) a system. For example, when you send an email to your friend, you are using the largest system of connected computers on the planet, called the Internet. A system could be as small as your machine with various components, and as large as the Internet. So, it's very important that you understand the system concepts.
Understanding the System Concepts
A system is a combination of independent but related elements functioning together as a unified whole in order to accomplish a set of specific tasks. The term system has Greek and Latin roots and means combination. An element of a system may be a system in itself and is called a subsystem of the system. For example, a country is a system of citizens, cities, and states, and a city is a subsystem of its country. An operating system is another example of a system, and a file system is a subsystem that the operating system manages. Your Solaris system machine is a part of a network system that may contain client and server machines. In the following list we define some general and some UNIX-related system concepts that are referred to throughout this book:
·         Machine. Machine is a very generic term used for any standalone computer or a computer connected to the network.
·         
Host
. A host is a machine connected to a network and having a network (e.g., IP) address.
·         Workstation. The term workstation refers to a machine that has its own storage, CPU, and memory to run the applications. This term originally developed as an alternative to dummy terminals during the era of mainframe computers. The dummy terminals did not have their own resources (storage, memory, and CPU power) to run applications on their own. So, generally speaking, a workstation can be a client or a server. However, vendors such as Microsoft and Sun use the term workstation in a more specific sense which is different from what is described here.
·         
Client
. A client is a machine connected to the network, running a program that makes a request for a service. For example, when you are browsing the web, your machine is acting as a client, and the web browser application (e.g., Netscape or Internet Explorer) is the client program running on your machine. Obviously, a client must have a network address.
·         
Server
. A server is a machine connected to the network, running a program that accepts and serves the requests made by the client machines on the network. A machine connected to the network that is running a web server is an example of a server.
·         Process. A process is a program being executed on the machine, whereas a program is an executable file residing on a disk. In other words, a process is an executing instance of a program, which has a unique numeric identifier called process ID.
·         Daemon. A daemon is a process that often runs continuously as long as the system is up. The daemons are often started when the system is hooted, and they terminate when the system is shut down. It is said that the daemons run in the background, meaning that there is no controlling terminal window for a daemon. It runs independent of a terminal or a login session. The term daemon is derived from the Greek mythology where it means an intermediary between the gods and mankind. This is an accurate description of daemons in UNIX in the sense that UNIX provides services through daemons, which act as the intermediary between services and users.
·         
Kernel
. An operating system has a core, which consists of the defining programs for that operating system, and some utilities around it that can use the core. The core of an operating system is called the kernel. Programs (or utilities) interact with the kernel through system calls.
·         
Shell
. A shell is a wrapper around the kernel; it protects the kernel from the user. In other words, it's a program that lets a user interact with the kernel. When a user issues a command, the command goes to the shell; the shell gets the information from the kernel and provides it to the user. So, the user interacts with the kernel through the shell. The user can put multiple commands and their logic in a file called a shell script, written according to the rules of that shell. That shell script can be executed like any other program. UNIX and Solaris support a number of shells such as C shell, Korn shell, and Bourne shell. When a user logs on to the system, a shell is activated that the user interacts with during the login session.
·         Command and utilities. A command is an instruction that a user writes at a prompt in a terminal window in order to accomplish a task—for example, to see the list of Ales in a directory. A utility is a program that does something useful for a user and is not part of the kernel. A command accomplishes the task by executing a program. For that reason, commands are sometimes also called utilities. In other words, there are a number of programs other than kernel and shell that are part of UNIX. These programs can be run as commands in a shell. The shell interprets the command for the kernel.
·         Multiuser and mutitasking. In the primitive days of computing, a computer could support only one user running one application at a time. However, UNIX has been a multiuser and multitasking operating system, which means it allows multiple users to log on to a machine at the same time and multiple applications to be run on the same machine concurrently, sharing resources such as CPU, memory, and file systems. Currently, most operating systems (even Windows) support multitasking and multiusers, but UNIX was the pioneer in this arena, as in many other areas of the operating systems.
·         Distributed system. A distributed computer system is a collection of autonomous computers networked together that collaborate to accomplish a common task. The Internet is the largest distributed system that supports services such as the World Wide Web and email.
·         Fault tolerance. A distributed system is called fault tolerant if it keeps working even if one or more of its components should fail. (The capability of a system to keep functioning when one or more of its components fails is called fault tolerance.) One common method to implement fault tolerance is through redundancy. That means there are multiple components capable of offering the given functionality; when one component fails, the other component takes over to provide the functionality of the failed component. Consequently, the user does not experience the failure.
An operating system manages resources for the users, and one of those resources is the file system. Users interact with the file system through the operating system. There are two ways to interact with the operating system: GUI and command line. Although substantial【真实的】 work has been done on the GUI front, UNIX traditionally has been a command-based operating system. As a system administrator, you must become familiar with the UNIX commands and know how to use those commands to interact with the file system.
Understanding the UNIX Commands and File Organization
The UNIX commands and the file systems get a fair coverage in this book. To make use of that coverage, you must understand the basic command structure and the general organization of files on a UNIX system.
The UNIX Command Structure
A command is the text that you type after the command prompt on the command console. Issuing a command involves typing the command on the command prompt and pressing RETURN on the computer keyboard. A command has a structure. It is, in general, composed of the following elements:
·         Name. The name of the command specifies the action to be performed. This is the name of a program that: would be executed when you issue the command. The name is always the first element of a command.
·         Option. This is a command modifier in the sense that: it specifies the way in which the action (specified by the command name) would be performed. There may be more than one option in a command. Options usually begin with a hyphen (-). In the OS literature, options are also called switches. Options generally follow the name.
·         Argument. Arguments usually specify the target of the action. For example, if the command instructs to display the content of a directory, the argument would be the name of the directory whose content is to be displayed. Arguments often follow options in the command. Arguments are also called parameters in the literature.
Following is the syntax of a UNIX command:
   $  [] []
The symbol $ specifies the prompt and is already there on the console. The  specifies a list of options, and the  specifics a list of arguments. The square brackets around  and  indicate that these elements are optional. For example, the following command displays the names of the files in the directory named mydir in the long format.
   ls -l mydir
The command name
ls
specifies that a list of files should be displayed, the option -l specifies that the display should be in the long format, and the argument mydir specifies the name of the directory whose files should be displayed. If the directory name is not specified, the current directory is assumed by default. You type ls -l mydir after the prompt and press RETURN. The command is issued. The important basic points about UNIX commands are summarized below:
·         In a command:
o        There must be a space between any two elements.
o        There must not be any space within the same element.
·         The name of a command is a required element. The other two elements are often optional. If not given, their default values are used.
·         UNIX commands are case sensitive.
When a command is issued, it executes a program that interacts with the kernel through a shell to do its job, as shown in
Figure 1-3
.
Figure 1-3: Relationship among the commands, shells, and kernel

On the Job
Solaris provides both the options GUI and the command console for the user to interact with the OS.
You use the commands to interact with the operating system to perform various tasks. Quite a few of those tasks involve browsing through the files on your system. The files on a UNIX (and therefore Solaris) system are organized into a directory tree, so you should become quite comfortable with the concepts related to the directory tree.
Understanding the Directory Tree
One important task of an operating system is to manage files. The UNIX operating system keeps the files in a hierarchy called the directory tree. You can use a number of commands to manage files in the directory. Before you do that, you should be familiar with the following concepts related to the directory tree:
·         
File
. A file is the smallest container of data. In other words, the digital (computer) data is stored in files. Each file has a name. In UNIX, the file names are case sensitive. For example, myfile and myFile are the names of two different files. File in UNIX is a very general concept. The files are of the following types:
o        Regular file. Such a file contains data. The text files, binary files, and programs are examples of regular files.
o        Special file. I/O devices are also treated as files in UNIX.
o        Links. A link is a mechanism to allow several file names to refer to the same file on the disk.
o        
Directory
. This is a binary file containing the list of other files; it may include other directories as well. Files are organized into directories. Any file exists in some directory, which is a UNIX equivalent of folder in the Windows OS.
·         Storage device. The files are stored on a storage device such as a hard disk, a floppy disk, a CD-ROM, or a tape.
·         
File system
. The way an operating system organizes files on the storage device is called a file system. The operating system uses a file system to store, locate, and retrieve the data in a convenient and efficient fashion. The file system is maintained on a storage device.
·         
Directory
. Files are organized into directories. Any given file exists in some directory. Directory is a UNIX equivalent of folder in the Windows OS.
During OS installation, a file system is created. The layout of the file system is in the form of a hierarchical tree with the root directory (/) as the root of the tree. This hierarchy is called (inverted【转换】) directory tree with root (/) at the top. Part of the hierarchy created during installation is shown in
Figure 1-4
. The root directory (/) contains subdirectories including bin, etc, export, and home. These subdirectories may contain their own subdirectories, and so on. A file exists inside a directory (or a subdirectory).
Figure 1-4: A very small part of the directory hierarchy created during Solaris installation (the directory jkerry is a fictitious directory name that you can create inside the home directory; it's not part of the directories automatically created)
When you log into your system, you are put into your home directory as a starting point. From there on you can move around in the directory tree, described as follows:
·         You are here. When you are browsing the directory tree from the command console, there is a concept of being in a directory called current directory. It is important to know in which directory you currently are. This directory would work as a default directory for some commands. If your command prompt does not indicate where you are, you can find out by issuing the command:
·                pwd
·         
Path
. The path is the address of a file or a directory on the directory tree. No two files in a directory tree may have the same path. The path contains a series of names of directories separated by a forward slash (/) and ending in the name of the directory or the file being addressed. For example, a file
passwd
in the directory etc in the directory tree shown in
Figure 1-4
may be addressed as: /etc/passwd. The path indicates that the file
passwd
exists inside the directory etc, and the directory etc exists inside the root directory /. A path may be absolute or relative.
o        Absolute path. The path that starts from the root directory is called the absolute path. Each file or directory may have only one absolute path. For example, the absolute path for the directory password is /etc/password, as shown in
Figure 1-4
. A directory or a file may be accessed from anywhere on the tree using its absolute path. The absolute path always starts with /.
o        Relative path. The path that starts from the current directory (where you currently are on the command prompt) is called the relative path. The relative path of a file depends on where you are trying to access it. For example, if you are in the root directory (/), the relative path to home is home, and to jkerry is home/jkerry. The double dots (..) take you one level up in the tree. For example, the relative path to the
passwd
file from the jkerry directory is ../../etc/passwd. The relative path never starts with /.

On the Job
The absolute path of a file or a directory always starts with a forward slash (/), and a relative path never starts with the forward slash.
Now that you understand the basic command structure and the concepts related to the directory tree, you can roll up your sleeves【袖子】to engage【着手】 in some hands-on experience with your Solaris machine.
Breaking the Ice with Your Solaris Machine
In this section, you will try some commands to break the ice with your Solaris machine. We are assuming that you have installed Solaris on your machine following the instructions given in
Chapter 2
or that you have access to a machine that has Solaris installed on it.
In order to use the Solaris system, you need to log into it.
Logging In and Logging Out
A standard login session (or work session) is the interval starting with the time you log in to the system and ending with the time you log out. The multiuser environment of Solaris requires that you type in your login name and password as your identification each time you want to log into the system. So you, the system administrator, need to create a login name and a password in order to enable a user to use the system.
Before you log in to the system, your screen should display the following:
    console login:
Type in the login name at the prompt and press RETURN. For example, assume the hostname of your machine is dem, and your login name is jkerry; you would type jkerry at the prompt:
   dem console login: jkerry
When you press RETURN, the system requests your password as follows:
   dem console login: jkerry   Password:
Type in your password at the prompt and press RETURN. Note that, as a security feature, the system does not display your password on the screen as you type it.

On the Job
The key referred to as RETURN in this book may be labeled ENTER on some keyboards, and vice versa.
When you initially log in to the system, you actually log into a shell that is automatically started for you. Your command prompt will indicate the shell, and this shell is called your login shell. The Bourne shell is the default shell, but it can be changed. The availability of some commands or programs may depend upon the login shell you are using.
When you have finished, you are done with your session and are ready to exit the system. You can log out by typing exit on the command prompt:
   exit
When you press RETURN, the system once again displays the login prompt:
    console login:
The appearance of the login prompt indicates that you have successfully logged out and the system is ready for a user to log in.
Now that you know how to log in and log out of the system, you can try a few commands after logging in.
Getting Started with Issuing Commands
When you log into your machine, you will be put into your home directory, which is the directory created when your account was created. You can, if you are a typical user, create directories and files under your home directory. That is, your directory tree starts from your home directory. A command prompt will appear that will depend on the shell that was started when you logged in, as shown in
Table 1-5
. Of course, these prompts are by default and can be changed by the system administrator—that is, you.
Table 1-5: Command prompts displayed by different shells for an ordinary user and a superuser
Shell User
C
Bourne, Korn
Ordinary user
%
$
Superuser
#
#
In this section, we will assume that you are logged into the Bourne shell, and hence your command prompt is $. When you see this command prompt, it means the system is waiting for a command from you. Issuing a command means typing the command at the command prompt and pressing RETURN. The command is not issued until you press RETURN. This means you can retype (or correct) your command before pressing RETURN. Let's try a few commands.
Find out in which directory you are by issuing the following command:
   $ pwd
Remember, you do not type $; that is a prompt already displayed. It will display the absolute path of the directory in which you are, such as:
   /home/jkerry
You can issue the following command to find out the time on the system:
   $ date
The output of this command will look like the following:
   Fri Jul 08 11:25:34 PST 2005   $
Try the following command now:
   $ Date
The output will look like the following:
   Date: Command not found.   $
This tells you that commands (and file names) in UNIX (and hence in Solaris) are case sensitive. Most commands in Solaris are in lowercase. To see the list of content (names of files and subdirectories) in the /etc directory, issue the following command:
   $ ls /etc
Here you used the absolute path to the directory etc. Assuming you are in your home directory, directly under the /home directory you can issue the following command by using the relative path to the /etc directory:
   $ ls ../../etc
This will obviously generate the same output as the ls /etc command. To display the whole directory tree underneath the etc directory, issue the following command:
   $ ls -aR /etc
This will recursively display the whole subtree of directories underneath /etc with the directory names and file names. You can use the cd command to change the directory:
   $ cd
The argument  specifies the relative or the absolute path of the directory to which you want to go. The cd command without any argument takes you to your home directory. You can see a list of commands recently issued on this terminal by issuing the following command:
   $ history（solaris10中实验失败）
It will display an output that looks like this:
   1 pwd   2 date   3 ls /etc   4 ls ../../etc   5 ls -aR /etc   6 history
Here are some useful tips about the commands:
·         You can issue more than one independent command in one command line by separating them from each other with a semicolon (;), such as:
·                   $ date; pwd
·         To repeat the last command, just issue the following command:
·                   $ !!
·         To repeat any previous command, just issue it by its history number. For example, the following will reissue the command with history number 3:
·                   $ !3
You can find the history number by issuing the history command. Let's try some advanced commands now. Create a file named test.txt by issuing the following command:
   $ touch test.txt
It will create an empty file with name test.txt. You can verify that the file is created by issuing the
ls
command. Now you can write the output of an
ls
command into the test.txt file by issuing the following command:
   $ ls /etc > test.txt
This is called redirecting the output, and the symbol > is called the redirection operator. You can verify that the test.txt file does have the output of the
ls
command by issuing the following command:
   $ more test.txt
This will display the content of the file one page at a time, and you can push the space bar to display the next page. You can also direct the output of a command as an input to another command. This is called piping. As an example consider the following:
   $ ls -l/etc | grep test
This will pipe the output of ls -l command into the
grep
command, which will display only those lines of the output that contain the string test. The vertical bar (|) is called the pipe operator or pipe symbol.
Another very useful command, the wc command, can be used to count the characters, words, or lines in a file. The command has the following syntax:
   $ wc [-c] [-C] [-lw] [-m] []
The argument fileName specifics the name of the file under analysis. The options are described here:
·         -c. Count the bytes.
·         -C. Count characters.
·         -l. Count lines.
·         -m. Same as -C.
·         -w. Count words delimited by white space or new line character.
If no options are specified, -lwc is assumed as default. If you want to count only the unique lines (ignore the duplicates) in a file, issue the following command:
   $ uniq -u  | wc -l
The uniq -u command generates the list of unique lines and pipes it as an input into the wc command, which counts the number of lines and displays the number.
To count the number of files in a directory, you can issue the following command:
   $ ls -l | wc -l
The ls -l will produce one line for each file, and the wc -l command will give you the line counts; hence, you have the count for the number of files in the directory.
There will be situations when you will need help regarding commands, such as when you know the name of a command but you are not sure how to use it. In this case use the
man
command, which will display the so-called man (manual) pages for the command. For example, if you want to know about the
ls
command, issue the following
man
command:
   $ man ls
However, be patient. It takes a while to get used to the man pages; they may look cumbersome and sound confusing to a beginner.
The three most important takeaways from this chapter are the following:
·         An operating system is a software system that controls and manages the computer resources and acts as an agent between the user and the computer hardware, and between the applications and the computer hardware.
·         Solaris is an operating system that is one member of a family of operating systems called UNIX.
·         You interact with any UNIX operating system, including Solaris, through a shell that interprets your commands for the kernel, which is the core of the operating system.
Summary
An operating system is a software system that manages computer resources such as files and CPU, and acts as an agent between a computer user and the computer hardware. Solaris 10 from SUN is the latest version of an operating system that is a member of a family of operating systems called UNIX. Although substantial progress has been made on the GUI front, the UNIX operating system has largely been a command-based operating system. It's important for a UNIX system administrator to be proficient【熟练的】 in using commands. When you log into a UNIX operating system such as Solaris, a shell is started. This is a software system that will interpret the commands from you for the kernel—the core of the operating system. The files in a UNIX operating system are organized into a directory tree, and when you log into the system, you are put into your home directory, which is created at the time your account was created.
An operating system is a software system that must be installed on your computer before it can start helping you to use the machine. In the
next chapter
, we show you how to install the Solaris 10 flavor of UNIX operating system.


本文来自ChinaUnix博客，如果查看原文请点：http://blog.chinaunix.net/u2/64847/showart_549322.html