Standard Error and Logging in Shell Scripts

Introduction to Standard Error and Logging

Hello! In this lesson, we will explore the concept of error handling through standard error redirection and logging in shell scripts. These techniques are essential for writing robust and maintainable scripts. By learning how to capture and log error messages, you can ensure that your scripts run smoothly and any issues are well-documented for further analysis. Let's get started!

Understanding Standard Error

In Unix-like operating systems, there are three standard streams for input and output: Standard Input (stdin) : This is the default source of input for a program, typically represented by file descriptor 0. Standard Output (stdout) : This is the default destination for program output, typically represented by file descriptor 1. Standard Error (stderr) : This is the default destination for error messages and diagnostics, typically represented by file descriptor 2. stderr is specifically used to output error messages separately from standard output (stdout). This separation allows you to redirect errors independently from regular output. For instance, you might want stdout to go to a file and stderr to be logged elsewhere to diagnose issues without cluttering standard output.

Custom Error Message to Console

Sometimes you may want to print a custom error message to stderr instead of stdout . This can be useful for immediate feedback during script execution. You can use the >&2 operator to redirect the echo output to stderr . Here is an example: Shell #!/bin/bash echo "Printing error to stderr" >&2 By using >&2, you ensure the error message is sent to stderr instead of stdout, which helps distinguish it from regular output.

Redirecting Standard Error to a Log File

Shell scripts typically display error messages on the terminal. However, it is often useful to redirect these messages to a log file for later review. We have seen how to redirect the output (stdout) of a command using command >> file_name . We can similarly redirect the output of stderr to a file using the 2> operator. The 2> operator overwrites the contents of the file, while 2>> appends to the file. Consider the following script: Shell #!/bin/bash # Define a log file to capture error messages log_file="error_log.txt" ls /nonexistent_directory 2>> $log_file cat $log_file In this script: We define a log file named error_log.txt to store error messages. The ls /nonexistent_directory command will fail because the directory does not exist. 2>> $log_file redirects the error message from the ls command to error_log.txt . cat $log_file prints the contents of error_log.txt The output of the cat $log_file command is: text ls: cannot access '/nonexistent_directory': No such file or directory Using the 2>> operator, we have directed the error message of the ls command to error_log.txt .

Join the 1M+ learners on CodeSignal

Be a part of our community of 1M+ users who develop and demonstrate their skills on CodeSignal

Conditional Handling of Errors and Logging

To enhance error handling, you can use conditional statements to manage errors and log the results accordingly. In this example, we will also use the tee command. The tee command reads from standard input and writes to both standard output and one or more files simultaneously. This allows you to see the output in the terminal while also saving it to a file. The general syntax of the tee command is:

text

command | tee -a output_file

command | tee -a output_file

The following script attempts to copy a non-existent file and logs an error:

Shell

#!/bin/bash

# Define a log file to capture error messages
log_file="error_log.txt"

# Attempt to copy a non-existent file and log errors
if ! cp nonexistent_file.txt file.txt 2>> $log_file; then
  echo "Error: Failed to copy file" | tee -a $log_file >&2
fi

#!/bin/bash

# Define a log file to capture error messages
log_file="error_log.txt"

# Attempt to copy a non-existent file and log errors
if ! cp nonexistent_file.txt file.txt 2>> $log_file; then
  echo "Error: Failed to copy file" | tee -a $log_file >&2
fi

Let's look at the if statement:

The ! operator inverts the exit status of the cp command. Normally, if cp succeeds, the exit status is 0 (true), and if it fails, the exit status is non-zero (false). The ! operator will make a successful command return false and a failing command return true.
cp nonexistent_file file.txt attempts to copy a non-existent file to a file called file.txt. Since the file nonexistent_file does not exist, this command will fail. The ! operator makes the if statement condition true.
2>> $log_file redirects any error messages from the cp command to the error_log.txt file, appending them to the existing contents.

Inside the if ... fi block:

echo "Error: Failed to copy file" prints the error message "Error: Failed to copy file" to the standard output.
| tee -a $log_file pipes the output of the echo command to tee, which appends (-a option) the message to the error_log.txt file
>&2 prints the message to stderr instead of stdout.

The output of this script to stderr is:

text

Error: Failed to copy file

Error: Failed to copy file

The content of error_log.txt file is:

text

cp: cannot stat 'nonexistent_file.txt': No such file or directory
Error: Failed to copy file

cp: cannot stat 'nonexistent_file.txt': No such file or directory
Error: Failed to copy file

Using if statements, 2>>, tee -a, and >&2 we have created a script that logs the error to error_log.txt and also outputs a message to the stderr.

Introduction to Standard Error and Logging

Understanding Standard Error

Custom Error Message to Console

Redirecting Standard Error to a Log File

Join the 1M+ learners on CodeSignal

Discarding Errors Messages

Conditional Handling of Errors and Logging

Summary and Next Steps