Shell by Example: Exit Codes

Shell by Example: Exit Codes POSIX + Bash

Every command returns an exit code (0-255). By convention, 0 means success and non-zero indicates an error. The special variable `$?` holds the last command’s exit code. Successful command returns 0:	Edit `#!/bin/sh true echo "Exit code of 'true': $?"`
	Output: Exit code of 'true': 0

Failed command returns non-zero:	Edit `#!/bin/sh false echo "Exit code of 'false': $?"`
	Output: Exit code of 'false': 1

Common exit codes by convention: `0` - Success `1` - General errors `2` - Misuse of shell command `126` - Command not executable `127` - Command not found `128+N` - Fatal error signal N Check exit code with `if` statement:	Edit `#!/bin/sh if grep -q "root" /etc/passwd; then echo "Found root user (grep returned 0)" else echo "No root user found" fi`
	Output: Found root user (grep returned 0)

Using `$?` explicitly:	Edit `#!/bin/sh ls /nonexistent 2>/dev/null status=$? if [ "$status" -ne 0 ]; then echo "ls failed with exit code: $status" fi`
	Output: ls failed with exit code: 2

The `exit` command sets the script’s exit code:	Edit `#!/bin/sh run_check() { if [ -f "/etc/passwd" ]; then return 0 # Success else return 1 # Failure fi } run_check echo "run_check returned: $?"`
	Output: run_check returned: 0

Return codes in functions use `return`, not `exit`. `exit` would terminate the entire script.	Edit `#!/bin/sh check_file() { [ -f "$1" ] # Return code is the test's result } check_file "/etc/passwd" echo "Check /etc/passwd: $?" check_file "/nonexistent" echo "Check /nonexistent: $?"`
	Output: Check /etc/passwd: 0 Check /nonexistent: 1

Chain commands based on exit codes:	Edit `#!/bin/sh echo "Using && (run if previous succeeded):" true && echo " This runs because true succeeded" false && echo " This won't run" echo "Using \|\| (run if previous failed):" false \|\| echo " This runs because false failed" true \|\| echo " This won't run"`
	Output: Using && (run if previous succeeded): This runs because true succeeded Using \|\| (run if previous failed): This runs because false failed

Combine `&&` and `\|\|` for simple conditionals:	Edit `#!/bin/sh test -d "/tmp" && echo "/tmp exists" \|\| echo "/tmp missing"`
	Output: /tmp exists

Exit codes with pipes - `$?` gives last command’s code:	Edit `#!/bin/sh echo "hello" \| grep -q "world" echo "Pipe exit code: $?"`
	Output: Pipe exit code: 1

Bash Bash provides a `PIPESTATUS` array for all pipe exit codes:	Edit `#!/bin/bash cat /nonexistent 2>/dev/null \| grep "pattern" \| head -n 1 echo "First command: ${PIPESTATUS[0]}" echo "Second command: ${PIPESTATUS[1]}" echo "Third command: ${PIPESTATUS[2]}"`
	Output: First command: 1 Second command: Third command:

Use `set -e` to exit on first error:	Edit `#!/bin/sh echo "Without set -e:" ( false echo " This still runs" ) echo "With set -e:" ( set -e true echo " This runs" # false would cause exit here )`
	Output: Without set -e: This still runs With set -e: This runs

Bash Use `set -o pipefail` with `set -e` for pipe errors. This makes the script exit if any command in a pipeline fails, not just the last one.	Edit `#!/bin/bash set -eo pipefail`

Custom exit codes for different error types:	Edit `#!/bin/sh ERR_INVALID_ARG=1 validate_arg() { case "$1" in start \| stop) return 0 ;; *) return "$ERR_INVALID_ARG" ;; esac } validate_arg "invalid" code=$? case $code in 0) echo "Valid argument" ;; "$ERR_INVALID_ARG") echo "Invalid argument (code $code)" ;; esac`
	Output: Invalid argument (code 1)

Best practice: always check critical commands:	Edit `#!/bin/sh if ! command -v ls >/dev/null 2>&1; then echo "ls command not found" >&2 exit 127 fi echo "Script completed successfully"`
	Output: Script completed successfully

« Here Documents | Traps »