27.1. /dev

The /dev directory contains entries for the physical devices that may or may not be present in the hardware. [1] The hard drive partitions containing the mounted filesystem(s) have entries in /dev, as a simple df shows.

bash$ df Filesystem 1k-blocks Used Available Use% Mounted on /dev/hda6 495876 222748 247527 48% / /dev/hda1 50755 3887 44248 9% /boot /dev/hda8 367013 13262 334803 4% /home /dev/hda5 1714416 1123624 503704 70% /usr

Among other things, the /dev directory also contains loopback devices, such as /dev/loop0. A loopback device is a gimmick that allows an ordinary file to be accessed as if it were a block device. [2] This enables mounting an entire filesystem within a single large file. See Example 16-8 and Example 16-7.

A few of the pseudo-devices in /dev have other specialized uses, such as /dev/null, /dev/zero, /dev/urandom, /dev/sda1, /dev/udp, and /dev/tcp.

For instance:

To mount a USB flash drive, append the following line to /etc/fstab. [3]

1 /dev/sda1 /mnt/flashdrive auto noauto,user,noatime 0 0

(See also Example A-24.)

Checking whether a disk is in the CD-burner (soft-linked to /dev/hdc):

1 head -1 /dev/hdc 2  3  4 # head: cannot open '/dev/hdc' for reading: No medium found 5 # (No disc in the drive.) 6  7 # head: error reading '/dev/hdc': Input/output error 8 # (There is a disk in the drive, but it can't be read; 9 #+ possibly it's an unrecorded CDR blank.) 10  11 # Stream of characters and assorted gibberish 12 # (There is a pre-recorded disk in the drive, 13 #+ and this is raw output -- a stream of ASCII and binary data.) 14 # Here we see the wisdom of using 'head' to limit the output 15 #+ to manageable proportions, rather than 'cat' or something similar. 16  17  18 # Now, it's just a matter of checking/parsing the output and taking 19 #+ appropriate action.

When executing a command on a /dev/tcp/$host/$port pseudo-device file, Bash opens a TCP connection to the associated socket.

Getting the time from nist.gov:

 bash$ cat </dev/tcp/time.nist.gov/13
 53082 04-03-18 04:26:54 68 0 0 502.3 UTC(NIST) *
 	      

[Mark contributed the above example.]

Downloading a URL:

 bash$ exec 5<>/dev/tcp/www.net.cn/80
 bash$ echo -e "GET / HTTP/1.0\n" >&5
 bash$ cat <&5
 	      

[Thanks, Mark and Mihai Maties.]

   1 #!/bin/bash
   2 # dev-tcp.sh: /dev/tcp redirection to check Internet connection.
   3 
   4 # Script by Troy Engel.
   5 # Used with permission.
   6  
   7 TCP_HOST=www.dns-diy.com   # A known spam-friendly ISP.
   8 TCP_PORT=80                # Port 80 is http.
   9   
  10 # Try to connect. (Somewhat similar to a 'ping' . . .) 
  11 echo "HEAD / HTTP/1.0" >/dev/tcp/${TCP_HOST}/${TCP_PORT}
  12 MYEXIT=$?
  13 
  14 : <<EXPLANATION
  15 If bash was compiled with --enable-net-redirections, it has the capability of
  16 using a special character device for both TCP and UDP redirections. These
  17 redirections are used identically as STDIN/STDOUT/STDERR. The device entries
  18 are 30,36 for /dev/tcp:
  19 
  20   mknod /dev/tcp c 30 36
  21 
  22 >From the bash reference:
  23 /dev/tcp/host/port
  24     If host is a valid hostname or Internet address, and port is an integer
  25 port number or service name, Bash attempts to open a TCP connection to the
  26 corresponding socket.
  27 EXPLANATION
  28 
  29    
  30 if [ "X$MYEXIT" = "X0" ]; then
  31   echo "Connection successful. Exit code: $MYEXIT"
  32 else
  33   echo "Connection unsuccessful. Exit code: $MYEXIT"
  34 fi
  35 
  36 exit $MYEXIT