The exit status in shell scripting

Like in the C programs, the exit command terminates a script. It can also return a value which is read by the script’s parent process.

The exit status is also known as return status or exit code. By convention, a succesfull command returns 0, while an unsuccesfull command returns an integer between 1 and 255.

When a script ends with exit without any parameter, or exit $? or no exit at all, the script’s exit status is the same as the last command’s (in the script)  exit code:

The following script does not end with exit (notice that the exit lines are commented):

#!/bin/bash
FIRST_COMMAND
...
LAST_COMMAND
#exit
#exit $?
#This script will exit with the status of the last command!

The $? parameter gives the script the exit status of the last command executed in the function.

We have this script:

#!/bin/bash
echo "Hello world"
exit $? # This will return 0 because the command is valid.
gcc # Wrong command
echo $? # This will return a non-zero value because the command is not used propperly
useradd
exit 12 # This will return 12 to the shell
# to check this, type "echo $?" after the script shell terminates execution

In the next script, I will use ! to negate commands and show you what happens when we use commands combined with pipes:

#!/bin/bash
true #builtin true command
echo "the exit status of true is $?" # 0 : success
!true ## the ! negates the true command and the exit code
echo "the exit status of !true is $?' # 1 : error
### Using pipes: ! combined with pipe inverts the exit status returned
ls | X #invalid command
echo $? # 1 : error
!ls | X # NOT invalid command
echo $? # 0
# the exit code of this script is 0 ;
#type "echo $?" after the script finishes execution to check it

We have even more interesting shell scripting articles!