File types in Linux

Every object in Linux can be called as files. For example,

Also special files,

To check that in terminal, use file <object name>

Also, we can use ls -ld object name

here starting d means directory. check the table to verify

Filesystem hierarchy

The /home directory carries all users (Bob/Dave/Rahim etc) except the root user.

/opt directory carries all 3rd party software etc.

To mounts software to the device, we use /mnt and use /tmp to keep it temporarily

All external drives (usb disks) etc. are kept on /media file

Once you use this command, you will see your USB disk is on the /dev/sdb1 file system mounted on /media/usb

The basic binary files are within /bin folder

/etc stores most of the configuration files in linux

/lib and /lib64 is where imported libraries are kept

/var is where system writes all logs

Linux package management

Package managers like DPKG , APT works for Ubuntu, also RPM for RHEL (Red Hat Enterprise Linux) , CentOs

What's the difference between Ubuntu, RHEL and Centos?

here, Ubuntu is a Debian based and RHEL and Centos is RPM Based

Talking about RHEL and Centos,

RHEL is paid and Centos is a community version.

But what is this package?

Its a compressed archieve which has all files needed to run a software/application.

Assume that we want to install gimp in our ubuntu system

To download gimp, we will use it's debian package which carries all binaries to run it

But there are some issues! As linux has 100s of distributions, these package managers don't fit with every distribution.

While installing, you will get errors and issues

You will find dependencies and get stuck. Here we can see gimp depends on libgimp2.0

Again, those dependencies can have their own dependencies too!

To solve all of these issues, we have package manager

Types of package managers

DPKG -> Package manager for debian based distributions (example, Ubuntu, LInux mint, Elementry os)

APT -> Newer front end for DPKG system

APT-GET -> Traditional front end for DPKG System

RPM -> Base package manager found in Redhat based distribution like RHEL, CIntos, fedora

YUM -> Frontend for RPM System

DNF -> More feature reached front end for RPM System

RPM (Red Hat Package Manager)

The extension is .rpm

the basic functionalities are

Installation rpm -ivh <package name>

-i means installation, -v means verbose (using or expressed in more words than are needed) and

Uninstall rpm -e <package name>
1. Upgrade rpm -Uvh <package name>
  1. Query rpm -q <package name>
    1. Verification

A lab:

rpm -q will also work

Another task

I have checked the location and the file name then installed using rpm -ivh

Let's use yum to solve this issue and the dependencies.

this time we will use firefox, instead of the firefox-68.6.0-1.el7.centos.x86_64.rpm (Specific version file) we have on the folder. Basically yum will install all dependencies to install firefox (Basically a generalized name)

YUM (Yellowdog Updater Modified) Package manager

how yum installs a package?

yum depends on repositories which can be locally available or online.The information about the repository is saved on the etc folder

Yum still uses RPM package manager to install

For redhat based system, a file called redhat.repo can be created. Also, for unofficial file or other 3rd party, we can create <otherfile>.repo .here nginx.repo

Whys this 3rd party repo?

Many times the official repo (redhat.repo) is not updated and thus we create our new repo and install there.

Let's see how it installs

it firstly runs a transaction and then checks the dependencies.

example:

DPKG (Debian Package Manager) and APT

DPKG is a lower level package manager

Sometimes, it creates issue and therefore, we use higher level package manager which is APT or APT-GET

We can solve it using apt or, apt-get command. apt (Advanced package manager) is much better option than apt-get

APT basically uses DPKG just like how Yum used RPM . Also, all of its repositories are listed in /etc/apt/........

Example:

Solution:

sudo apt install firefox

View sizes of file

Archiving files

To create an compressed file, use tar -cf

tar -tf to see files within the compressed file, tar -xf to see the content of the compressed file, tar -zhf to reduce the size of the compressed file

Example:

Solution:

Task:

Compressing and Archive

To see file sizes use du -sk filename or, ls -lh filename

Archiving

here -c is for creating an archive and -f for name of the file.

again tar -tf used to see contents in the compressed file, -xf to extract the contents of the compressed file, -zcf to compress files to a reduced size

Compression

To reduce the size we do the compression.We can use bzip2, gzip, xz. To uncompress use bunzip2, gunzip, unxz

we can also read files without un-compressing using zcat, bzcat, xzcat

Searching for files and directories

To find a file we can use locate

But it can happen that your desired file is created just recently and you might not find it as the database is not updated. So, update that using updatedb and then use locate or find command

Task:

Solution:

Using GREP to find

here, we are looking for 'second' & 'capital' from the sample.txt file

To search case sensitive, use grep -i, if you don’t know which file to search from use grep -r. It will show file name and the result

For lines that does not match a word , use grep -v

Assume you need exact words like exam here.

To search whole word, use grep -w

But if we don’t need exact words but words associated with it like examples

use grep -vw

to print matching line and one more line below, use -A1

to print one row before the desired sentence, use -B1

Task:

Solution:

Task

Solution:

IO Redirection

There are three data streams created when you launch a Linux Command.

Standard input, output and error can be redirected to text files. To redirect standard output to a file instead of printing, we use > .

Here shell.txt file will have $SHELL in it now. But as we know $SHELL means the location of shell which is in /bin/bash

Basically > overwrites existing content

but if we want to add one more line, use >>

Again, in order to redirect the error message , we need to use 2 followed by forward arrow symbol

Task

Solution

echo "a file in my home directory" > /home/bob/file_with_data.txt

Task:

Solution:

Task

Solution

Redirecting data to a file to be later consumed by another command can be tedious process. So, we overcome this by command line pipes. The pipes are defines using vertical bar symbol (|)

Here sample.txt file has 2 lines.

BY making use of the pipe, the output of the command grep Hello sample.txt becomes the standard input for the less command that follows after the operator (|)

Another way can be tee command. With tee, the standard output is still printed on the screen before overwriting the contents of the file with the same string.

to append a line instead of overwrite use tee -a

Task

we will use sudo grep -ir 172.16.238.197 /etc/ > /home/bob/ip

sudo for administrative purpose, grep to search and use -ir as we don’t case sensitive (-i) and we don’t know which folder we have the string . So, use -r too.

then directory name (/etc/) and then save that answer to /home/bob/ip and pass it by >.

We can also verify that our output has been passed to /home/bob/ip channel

Task

Answer should be echo "a file in my home directory" > /home/bob/file_with_data.txt

because echo is to write and we write “a file……….” and then pass it (>) to /home/bob/file_with_data.txt

Task

Here we check the error first using python3 /home/bob/my_python_test.py

We get this error “python3: can't open file '/home/bob/my_python_test.py': [Errno 2] No such file or directory”

Now, we will pass this error using cat and then the command (ython3 /home/bob/my_python_test.py) which generates error and then 2> and then to the address (/home/bob/py_error.txt) where we want to pass the error

cat python3 /home/bob/my_python_test.py 2> /home/bob/py_error.txt

We can verify if the /home/bob/py_error.txt file can has the error or not by writing this:

cat /home/bob/py_error.txt

We get these error messages :

cat: python3: No such file or directory

cat: /home/bob/my_python_test.py: No such file or directory

Task

Use zcat /usr/share/man/man1/tail.1.gz | tee /home/bob/pipes

zcat because we don’t want to extract

then the file path and then command line pipe (|) and then . With tee, the standard output is still printed on the screen before overwriting the contents of the file with the same string and finally the goal path (/home/bob/pipes)