Exit code

Applications and commands in Unix return an exit status—a numerical code indicating success or failure. By convention, a zero exit code means success, while any non-zero code signals an error. Understanding these codes is crucial for writing robust shell scripts and ensuring predictable automation.

Analogy: The Food Delivery Driver

Consider a food delivery driver verifying an address before dispatch. If the address is correct, the driver proceeds. If it’s wrong, they return with an error report. This mirrors how scripts and commands operate: they check conditions, perform tasks, and then report their status.

How Exit Codes Work in Unix

Every Unix-based command finishes with an exit code:

• 0: Success
• Non-zero: An error occurred

Common Exit Codes Table

Certain exit codes are reserved by the shell or operating system. Defining your own codes (above 2) helps callers distinguish between different failure modes.

Custom Exit Codes in Your Script

Below is a template that checks for a configuration file and terminates with meaningful exit statuses.

#!/usr/bin/env bash

export CONF_FILE="/var/tmp/file.conf"

terminate() {
    local message="$1"
    local code="${2:-1}"
    echo "${message}" >&2
    exit "${code}"
}

echo "Starting script execution"
echo "Sourcing configuration file"
if [[ ! -f "${CONF_FILE}" ]]; then
    terminate "Configuration file not found: ${CONF_FILE}" 2
fi

exit 0

References

• GNU Bash Manual: Exit Status
• Linux Shell Exit Codes