IT 244: Introduction to Linux/Unix

IT 244: Introduction to Linux/Unix
Class 9

Today's Topics

Reading Assignment

The reading assignment for this week is chapter 4 of Sobell, The Filesystem.

Homework 5

I have posted homework 5 here.

It is due this coming Sunday at 11:59 PM.

Quiz 2

Let's look at the answers to Quiz 2.

Mid-term

The mid-term exam for this course will be held on Tuesday, October 24th.

It will consist of 25 questions like those on the quizzes.

You will have the entire class period to work on the exam.

60% of the questions will come from the Ungraded Class Quizzes.

The last class before the exam, Thursday, October 19th, will be a review session.

You will only be responsible for the material in the Class Notes for that class on the exam.

The Mid-term is a closed book exam.

Tips and Examples

Testing Your Class Exercises

The Class Exercises offer you two ways to test your scripts
The first method is to run the script using bash ...
but only look at the error messages
Do do this for the script ex9.sh you would enter the following on the command line
```
bash ex9.sh  > /dev/null
```
bash ex9.sh runs the script ex5.sh
But we don't want to see the normal output of the script
We want to see the error messages
/dev/null throws away the normal output ...
leaving only the error messages
These error messages tell you if one of the lines in your script ...
is not working properly
Most Class Exercises if done properly ...
will not produce error messages
If you run the above command and see nothing ...
it means that your script contains not Linux errors
But it doesn't mean that your script does everything it should
Why not?
Here is a script file with nothing in it
```
$ cat empty.sh 
```
When I run this script looking for errors I get nothing
```
$ bash empty.sh > /dev/null
```
But I also get nothing if I run it normally
```
$ bash empty.sh
```
To check whether you Class Exercise script does what it is supposed to do ...
you need to run my test script ...
which you will find in ~ghoffman/it244_test

A Subtle Pathname Problem

Several semesters ago a student came to me with a problem ...
that took me a few minutes to unravel
It all had to do with the absence of one character in a pathname
The student was trying to go to his it244 directory using the command
```
cd ~/it244
```
Once there he tried to create a file using the touch command ...
but he got an error message saying he didn't have permission
After staring at his prompt for a few minutes ...
I realized what was going on
The student thought he had entered
```
cd  ~/it244
```
which would have put him in his it244 directory
What he really typed was
```
cd ~it244
```
which took him to the home directory of the it244 account, /home/it244
This account was created by Prof. Bo Sheng ...
who also teaches this course
It is a dummy account that he uses for teaching and testing ...
like my it244gh account
The only account that can create a file in this directory ...
is the account for which the home directory was created
Here again a single character was the source of the problem

Review

Hidden Filenames

A file whose filename begins with a period, . , is a "hidden" or "invisible" file
ls does not display these files unless you use the -a option
These files are used to configure your Unix environment

Startup Files

Files like .login and .bash_profile ...
which appear in your home directory ...
are startup files
They contain Unix commands ...
that are run before you get the first shell prompt
These commands customize your Unix environment
Among other things, you can set Unix variables ...
which can help you with your work

The . and .. Directory Entries

Every directory has at least two entries . and ..
When a new directory is created ...
these are the first two entries
. stands for the current directory
. . stands for the parent directory of your current directory
. . is the directory immediately above your current location
. is most often used in two circumstances
- To run a program in your current directory
- To move or copy a file to your current directory

Pathnames

Every file has a pathname which is used to access the file
A pathname has two components
- A path to reach the file
- The name of the file
The path is a list of directories that you must go through ...
to reach the file you want
A pathname is like an address on a letter ...
a name and directions to get there
The name of the file is always at the end of a pathname
When the slash, / appears between names in a pathname ...
it is used to separate a directory name ...
from what comes after it
When a / is the first character in a pathname ...
it stands for the root directory
There are two types of pathnames
- Absolute
- Relative

Absolute Pathnames

The top of the filesystem is a directory called the root directory
The root directory is represented by a single slash character, /
It can stand alone ...
or appear as the first character before a directory name
An absolute path is a list of directories ...
starting with the root directory ...
and ending with the directory that contains the file
When you add the filename to the end of an absolute path ...
you have an absolute pathname

Tilde ~ in Pathnames

There is one form of absolute path that is very short
This is the tilde character, ~
Tilde stands for your home directory
This means you can use tilde, ~ ...
anywhere you would normally use a path to your home directory
When you put a tilde in front of a Unix username ...
it stands for the home directory of that account
~ always means an absolute path

Relative Pathnames

Absolute pathnames are useful because you can use them anywhere
But they are long and easy to mistype
For most purposes, it is easier to use relative pathnames
In a relative pathname, the path starts from your current directory
In an absolute pathname, the path starts from the root, /
While all absolute pathnames start with a slash, /, or a tilde, ~ ...
relative pathnames never do
As far as Unix is concerned it makes no difference ...
whether you use and absolute or relative pathname
There are four types of relative pathnames:
- When the file is in your current directory
- When the file is in a subdirectory of your current directory
- When the file is in a directory that is above your current directory
- When the file is in a directory that is neither above or below the current directory

Relative Pathnames in Your Current Directory

A relative pathname of a file or directory ...
inside your current directory ...
is simply the name of that file or directory

Relative Pathnames in a Subdirectory

Things get a little more complicated ...
when you are dealing with a file in a subdirectory
Here, you must list every directory ...
between your current directory ...
and the file you want
You must use a slash, / ...
to separate the name of each directory ...
from what comes after it

Relative Pathnames above the Current Directory

When the file or directory is above the current directory ...
you can't list the directory names
Instead, you have to use the special . . entry in each directory
Use one . . for each directory up the chain in the path
Use a slash, /, between each ..

Relative Pathnames Neither above Nor below the Current Directory

What if the file is neither above nor below?
Here you have to go up to a common ancestor ...
and then down to the directory that holds what you want
The path starts with one or more . .
You keep going up ...
until you get to a directory ...
that is both an ancestor of your current directory ...
and the file you are trying to reach
Once you get to the common ancestor ...
you go down to the directory that holds the file

New Material

Access Permissions

All Unix files and directories have access permissions
The access permissions allow the owner of a file or directory ...
to decide who gets to do what with the file or directory
By default, the owner of a file or directory ...
is the account that created it
Every file, directory or device on a Unix filesystem
has three types of permissions
- Read
- Write
- Execute
If you have read permission on a file ...
you can look at the data in the file
You can run cat, more or less on these files
If you only have read permission on a file ..,
you cannot change it
To change a file you need write permission
To run a program or script file ...
you must have execute permission
Each of the three types of permissions ...
are set either on or off ...
to three classes of users
- The owner
- The group
- Every other Unix account
Every file or directory has an owner ...
and a group assigned to it
The account that created the file is usually the owner
A group is a collection of Unix accounts
Every account also has a default group ...
that is assigned to every file or directory ...
that account creates
This default group is created when the account is created
Only a system administrator can add users to a group
Every file or directory is assigned to a group ...
though the owner can change this to another group
The last class of users is any account ...
that is not the owner or a member of the group
Unix calls this class of users "other"

Viewing Access Permissions

To view the permissions of a file or directory use ls -l

$ ls -l
total 5
-rw------- 1 it244gh libuuid 316 2011-09-20 21:32 dead.letter
lrwxrwxrwx 1 it244gh libuuid  34 2011-09-06 13:21 it244 -> /courses/it244/s12/ghoffman/it244gh
drwx------ 2 it244gh libuuid 512 2011-09-07 15:03 mail
drwxr-xr-x 2 it244gh libuuid 512 2011-09-25 15:48 test
-rw-r--r-- 1 it244gh libuuid  15 2011-09-20 16:18 test.txt

The character in the first column indicates the type of file
A dash, - , means an ordinary file
The letter d indicates a directory
The letter l(el) indicates a link ...
which we will discuss in the next class
The next three characters indicate the access permissions of the owner
r means the owner has read permission
w means the owner has write (change) permission
x means the owner has execute (run) permission
- means the owner does not have the permission ...
that would normally appear in this column
The next three characters indicate the permissions of the group
The last three characters are the permission of all other accounts
After the permissions is a number ...
that indicates the number of links to the file or directory
The following column is the owner of the file or directory
Next, you will find the group assigned to the file or directory
Following this is the size of the file in bytes
Next is the date and time the file or directory was created or last modified
The last column is the name of the file or directory

`chmod`

When a file is created, is has certain default permissions

$ touch test.txt
$ ls -l test.txt 
-rw-r--r-- 1 it244gh libuuid 0 2012-09-17 14:40 test.txt

To change these permissions ...
you use the chmod (change mode ) command
Only the owner of a file can do this
chmod requires two arguments
- The permissions you want to grant
- The name of the file(s) or directory(s) which will be changed

The format for a call to chmod is

chmod  PERMISSIONS  FILES_OR_DIRECTORIES

The permission can be specified in two ways
- Symbolically
- Numerically
Symbolic permissions use letters ...
and the plus and minus signs
The numeric form uses three digits ...
running from 0 to 7
I will teach the numeric format
The numeric format is initially confusing
But in my experience, it is the better way to go
You are free to read about the symbolic format in the textbook
I will not deduct points for using this format ...
in a test, quiz or homework assignment ...
as long as you use it correctly

Using `chmod` with Numeric Arguments

The numeric permissions format uses three digits
Each digit is a number from 0 to 7
The first digit gives the permissions of the owner
The second digit gives the permissions assigned to the group
The third digit gives the permissions for every other account
Each of these classes of users must be assigned values ...
for read, write and execute permissions
How do you get three pieces of information out of one number?
By adding powers of two
Each digit is the sum of three other numbers
When calculating the number you add
- 4 if you want to give read permission
- 2 if you want to give write permission
- 1 if you want to give execute permission
Notice that all the number are powers of two
If we write these values in binary notation
- 100 represents 4
- 010 represents 2
- 001 represents 1
A single decimal digit from 0 to 7 is represented by 3 binary digits
This is how we get three pieces of information out of one digit
For example, to give full permissions I would add
- 4 for read permission
- 2 for write permission
- 1 for execute permission
So the total, 7...
which is 111 in binary ...
grants all three permissions
Let's look at some other digits
6 in binary is 110
The leftmost digit is 1 indicating read permission
The center digit is 1 indicating write permission
The last digit is 0 indicating that execute permission is not granted
5 in binary is 101
The first digit is 1 so read permission is granted
The second digit is 0 so write permission is not granted
The last digit is 1 so execute permission is granted
This scheme is confusing when you first encounter it
But it becomes easier as you use it

Try to remember this chant

4      2      1
read   write  execute
owner  group  everyone

Repeat this to yourself several times and it should sink in
Remember that you need three of these digits ...
to specify the full permissions for a file or directory
Let's look at some examples

When you create a new file, it will have certain default permissions

$ touch foo.txt

$ ls
foo.txt

$ ls -l
total 0
-rw-r--r-- 1 it244gh libuuid 0 2012-02-09 15:51 foo.txt

The owner can read and write the file, but not execute it
The group and everyone else can only read the file

To make the file unreadable to everyone except the owner

$ ls -l
total 0
-rw-r--r-- 1 it244gh libuuid 0 2012-02-09 15:51 foo.txt

$ chmod 600 foo.txt 

$ ls -l
total 0
-rw------- 1 it244gh libuuid 0 2012-02-09 15:51 foo.txt

To change the file back to its default permissions

$ ls -l
total 0

-rw------- 1 it244gh libuuid 0 2012-02-09 15:51 foo.txt
$ chmod 644 foo.txt 

$ ls -l
total 0
-rw-r--r-- 1 it244gh libuuid 0 2012-02-09 15:51 foo.txt