Skip to content

Latest commit

 

History

History
518 lines (470 loc) · 11.9 KB

linux-interview-questions-part-4.md

File metadata and controls

518 lines (470 loc) · 11.9 KB
Raw

Unix / Linux Interview Questions

I have used GIT Bash to execute the below scripts.

Difficulty Level: Beginner to Intermediate

All the Best !!!

References

Question List

1. Delete blank lines in file

$ sed '^\s*$'/d' file.txt

To save changes back to file

$ sed -i '^\s*$'/d' file.txt

Also

$ grep -v '^\s*$' file.txt

2. Print number of times each word appears in a file

$ cat file | tr '[:space:]' '[\n*]' | grep -v '^\s*$' | sort | uniq -c | sort -bnr

tr just replaces spaces woth newlines
grep -v '^\s*$' trims out empty lines
sort to prepare as input for uniq
uniq -c to count occurrences
sort -bnr sorts in numeric reverse order while ignoring whitespace

3. List all the files in directory except .txt and .pdf

$ ls -I "*.txt" -I "*.pdf"

If you want to iterate across all the subdirectories:

$ ls -I "*.txt" -I "*.pdf" -R

Also

$ find -not -iname "*.txt"
$ find . ! '(' -name '*.txt' -o -name '*.pdf' ')'

Reference: https://unix.stackexchange.com/questions/47151/how-do-i-list-every-file-in-a-directory-except-those-with-specified-extensions

4. Rename all .jpg files to .jpeg

$ vi image.sh

for x in `ls ./images/`;
do
    a=`echo $x | cut -d'.' -f1`
    mv ./images/$x ./images/$a".jpeg";
done

$ ./image.sh

5. Delete files of size more than 100mb in a folder which are older than 90 days

$ find /folder -size +204800 -mtime +90 -exec rm {} \;

6. Reverse the words in a sentence

$  echo "Have a good day" | awk '{ for(i=0;i<NF;++i){ x=NF-i; printf "%s ",$x;}  }'

Output: day good a Have

7. Given an input file having lines with alphabets and numbers. Print only the distinct alpha-numberic lines.

Input

$ cat input.txt

1234567890
0987654321
ABCDEFGH
123456789X
1234567890
123456789X

Output

123456789X

$ vi script.sh

#!/bin/bash
awk '
BEGIN {
i=0; 
}
{ 
    if (($1 ~ /[0-9]/) && ($1 ~ /[A-Z]/)) { 
        flg=0; 
        for (x = 0; $x < $i; x=$x+1) { 
            if($arr[$x] == $1) {
                flg=1;
                break;
            } 
        }
        if(flg==0) {
            arr[$i] = $1; i=$i+1;
        } 
    } 
}
END { 
    for (x = 0; x < $i; x++)
        print $arr[0]
}' input.txt | uniq

$ ./script.sh

8. Given a comma separated input file with items, category and cost. Display the category and its sum.

Input

$ cat input.txt

item1,category1,200
item2,category2,100
item3,category3,300
item4,category1,400
item5,category2,500
item6,category1,600

Output

category1 - 1200
category2 - 600
category3 - 300

$ vi script.sh

awk '
BEGIN{
x=0;
flag=0
}
{ 
    split($0,arr,",")           # split - arr[1] has item, arr[2] has category, arr[3] has cost
    flag=0;
    for(i=0;i<=x;i++){
        if(arr[2]==cat_arr[i]){
            flag=1;
            sum_arr[i]=sum_arr[i]+arr[3]
        }
    }
    if(flag==0){
        cat_arr[x]=arr[2]
        sum_arr[x]=arr[3]
        x++;
    }
}
END {
    for(i=0;i<x;i++){
        print(cat_arr[i],"-",sum_arr[i])
    }
}' input.txt

$ ./script.sh

9. Output of following command ps -ef | awk '{ print $1}' | sort | uniq | wc -l

Answer 2

$ ps -ef
     UID     PID    PPID  TTY        STIME COMMAND
prabakad   20464       1 ?          Jul 13 /usr/bin/mintty
prabakad    4940   18944 pty2       Jul 14 /usr/bin/bash
prabakad   18944       1 ?          Jul 14 /usr/bin/mintty
prabakad   19944   20464 pty1       Jul 13 /usr/bin/bash
prabakad   11492    4940 pty2       Jul 14 /usr/bin/winpty
prabakad   20708   16868 pty0     20:38:09 /usr/bin/ps
prabakad   16868   13672 pty0       Jul 12 /usr/bin/bash
prabakad   13672       1 ?          Jul 12 /usr/bin/mintty

$ ps -ef | awk '{ print $1}'
UID
prabakad
prabakad
prabakad
prabakad
prabakad
prabakad
prabakad
prabakad
prabakad

$ ps -ef | awk '{ print $1}' | sort | uniq
prabakad
UID

$ ps -ef | awk '{ print $1}' | sort | uniq | wc -l
2

10. Kill process based on user and process.

  • Using grep to search
$ ps -ef | grep "bash" | grep "prabakad"| awk 'BEGIN{print $2}{kill -9 $2}

$ ps -ef | grep "bash" | grep "prabakad"| awk '{print $2}' | xargs kill -9
  • Using ps command with -u and -p
$ ps -ef -u prabakad -p 8552 | tail -n +2 | awk '{print $2}' | xargs kill -9

Explanation

$ ps -ef
     UID     PID    PPID  TTY        STIME COMMAND
prabakad   19208   18800 pty0     21:46:39 /usr/bin/bash
prabakad   11440   19208 pty0     21:46:43 /usr/bin/ps
prabakad   18800       1 ?        21:46:38 /usr/bin/mintty

$ ps -ef -u prabakad -p 19208
     UID     PID    PPID  TTY        STIME COMMAND
prabakad   19208   18800 pty0     21:46:39 /usr/bin/bash

$ ps -ef | grep "bash" | grep "prabakad"
prabakad   19208   18800 pty0     21:46:39 /usr/bin/bash
prabakad   19484   19208 pty0     21:47:52 /usr/bin/bash

grep command does not give header.
ps command gives header, so use tail command to remove it.
The PIDs are sent as arguments using xargs and killed using kill -9

11. Print contents of a file starting from the nth line.

For example, print the give file starting from the 4th line

Input

$ cat lines.txt
line1
line2
line3
line4
line5
line6
line7
line8
line9
line10
line11
  • Using tail command
$ cat lines.txt | tail -n +4
line4
line5
line6
line7
line8
line9
line10
line11
  • Using sed command
$ sed '1,3d' lines.txt
line4
line5
line6
line7
line8
line9
line10
line11

Additional Information

head command displays first n lines.
tail command displays last n lines. If tail is used with +, displays lines starting from nth line as shown above.
The sed command above deletes the first 3 lines and prints output starting from 4th line.

12. Write a command to calculate the total size of all pdf files in the directory.

$ du -hc *.pdf | tail -1
25K     total

13. Write a command to delete the first and last line in every file in the directory and rename it.

Input

$ ls ./sample
abc_20171607.txt      rocket_20171503.txt   xyz_20171507.txt
mno xyz_20171507.txt  sample1_20171607.txt

$ cat ./sample/abc_20171607.txt
header
line1
line2
line3
footer

Output

$ ls sample
renamed_abc_20171607.log      renamed_sample1_20171607.log
renamed_mno xyz_20171507.log  renamed_xyz_20171507.log
renamed_rocket_20171503.log

$ cat ./sample/renamed_abc_20171607.log
line1
line2
line3
$ vi sample.sh

IFS=$'\n'
for filename in `ls ./sample -I '*.sh'`;                        # ls in sample folder searching file other than .sh
do
    newfilename=`echo $filename | cut -d '.' -f1`
    sed -i -e '1d' -e '$d' ./sample/$filename                   # -i to save to file. '1d' and '$d' delete first and last line
    mv ./sample/$filename ./sample/"renamed"_$newfilename".log"
done

$ ./sample.sh

Additional

If IFS=$'\n' (Internal Field Separator) is missing, file names having space in them are not recognised.

$ ./sample.sh
abc_20171607.txt
mno
xyz_20171507.txt
rocket_20171503.txt
sample1_20171607.txt
xyz_20171507.txt

14. Given a file with records. Print the sum of m where name=aman

Input

$ cat input.txt
t=20
m=10
name=aman
--xx--
t=30
m=20
name=deep
--xx--
t=40
m=30
name=aman
--xx--
t=40
m=300
name=aman
--xx--
t=40
m=10
name=rahul

Output

sum of aman is 340

$ vi script.awk

BEGIN {
	RS="--xx--\n";
	FS="\n";
}
{
	split($3,data,"=")
	if(data[2]=="aman"){
        split($2,mdata,"=")
        sum=sum+mdata[2]
	}
}
END{
	print ("sum of aman is",sum)
}

$ awk -f script.awk input.txt

Reference: http://www.thegeekstuff.com/2010/01/8-powerful-awk-built-in-variables-fs-ofs-rs-ors-nr-nf-filename-fnr/

15. Given a file with numbers. Sort each line.

Input

$ cat input.txt
4 6 8 1 7
2 12 9 6 10
6 1 14 5 7

Output

1 4 6 7 8
2 6 9 10 12
1 5 6 7 14

$ awk ' {split( $0, a, " " ); asort( a ); for( i = 1; i <= length(a); i++ ) printf( "%s ", a[i] ); printf( "\n" ); }' input.txt

Reference: http://www.unix.com/shell-programming-and-scripting/180835-sort-each-row-horizontally-awk-any.html

16. Bash shell script to reverse a string / word.

$ vi reverse.sh

#!/bin/bash
input="$1"
reverse="" 
len=${#input}
for (( i=$len-1; i>=0; i-- ))
do 
	reverse="$reverse${input:$i:1}"
done 
echo "$reverse"

$ ./reverse.sh Hello
olleH

Also

$ echo "Hello" | rev
olleH

$ rev<<<"This is a test"
tset a si sihT

Reference: https://www.cyberciti.biz/faq/how-to-reverse-string-in-unix-shell-script/

17. Reverse last 4 digits of a string.

Input

123456****7890

Output

123456****0987

$ vi rev.sh

input="$1"
reverse=""

len=${#input}

for((i=0;i<$len-4;i++))
do
{
reverse="$reverse${input:$i:1}"
}
done
for((i=len-1;i>=len-4;i--))
do
{
reverse="$reverse${input:$i:1}"
}
done

echo $reverse

$ ./rev.sh 123456****7890