Generating files of any size
/dev/zerois a character special device, which infinitely returns the zero byte ( ).
The above command will create a file called junk.datathat is exactly 1MB in size. Let's go
through the parameters: ifstands for – inputfile, ofstands for – outputfile, bsstands for
BYTES for a block, and countstands for the number of blocks of bsspecified to be copied.
dd if=/dev/zero of=junk.data bs=1M count=1
1+0 records in
1+0 records out
1048576 bytes (1.0 MB) copied, 0.00263553 s, 398 MB/s
Comm command demo:
[hadoop@namenode test]$ cat file1.txt file2.txt
1
2
3
4
1
2
4
5
[hadoop@namenode test]$ comm file1.txt file2.txt
1
2
3
4
5
[hadoop@namenode test]$ comm file1.txt file2.txt -1
1
2
4
5
[hadoop@namenode test]$ comm file1.txt file2.txt -2
1
2
3
4
[hadoop@namenode test]$ comm file1.txt file2.txt -3
3
5
[hadoop@namenode test]$ comm file1.txt file2.txt -3 -1
5
-1 removes first column from output
-2 removes the second column
-3 removes the third column
[hadoop@namenode test]$ ll
total 1052
-rw-rw-r--. 1 hadoop hadoop 99 Feb 16 08:34 all_txt_files.txt
-rw-rw-r--. 1 hadoop hadoop 8 Feb 17 03:29 file1.txt
-rw-rw-r--. 1 hadoop hadoop 8 Feb 17 03:29 file2.txt
-rw-rw-r--. 1 hadoop hadoop 1048576 Feb 17 03:20 junk.data
-rw-rw-r--. 1 hadoop hadoop 44 Feb 16 02:06 mul_bank.txt
-rw-rw-r--. 1 hadoop hadoop 30 Feb 16 08:55 num.txt
drwxrwxr-x. 2 hadoop hadoop 4096 Feb 16 03:05 sub
-rw-rw-r--. 1 hadoop hadoop 44 Feb 16 02:11 test.txt
"-"—if it is a regular file.
"d"—if it is a directory
"c"—for a character device
"b"—for a block device
"l"—if it is a symbolic link
"s"—for a socket
"p"—for a pipe
chmod u=rwx g=rw o=r filename
Here:
u =specifies user permissions
g =specifies group permissions
o =specifies others permissions
chmod a+x filename
a statnd for all.
Read,write,and execute permissions have unique octal numbers as follows:
r--=4
-w-=2
--x=1
Touch is a command that can create blank files or modify the timestamp of files if they
already exist.
Symbolic links are just pointers to other files.
ln -s targetfilename(directory) link_name
ln -s test.txt test_link.txt
Counting number of lines, words, and characters in a file
$ wc -l file # count number of lines as follows:
$ wc -w file #count words number
$ head -10 filename
$ tail -10 filename
Head used to get the first n lines of the file.
Tail is used to get the last n lines of the file.