Understanding the Nohup Command in Linux

This article provides a concise guide to the nohup command in Linux, explaining its primary purpose, how it prevents process termination during remote sessions, and how to use it effectively. By reading this guide, you will learn how nohup handles terminal hangup signals, manages command output, and ensures that critical scripts continue executing seamlessly even after an SSH connection drops or closes.

The Purpose of Nohup

The term nohup stands for "no hangup." Its primary purpose is to allow processes and scripts to continue running in the background even after the user logs out, closes the terminal, or loses their network connection to a remote server.

When you connect to a remote Linux server via SSH, the shell session acts as the parent process for any commands you run. When the SSH session closes, the operating system sends a SIGHUP (Signal Hangup, signal 1) to all child processes running in that session, which terminates them by default. Prefixing a command with nohup instructs the operating system to intercept and ignore the SIGHUP signal, ensuring the targeted process remains active.

How Nohup Works

Under standard execution, terminal-bound tasks rely directly on the active session's standard input (stdin), standard output (stdout), and standard error (stderr).

When invoked, nohup modifies this behavior:

  1. Signal Immunity: It configures the process to ignore the SIGHUP signal entirely.
  2. Output Redirection: Because the controlling terminal will eventually close, nohup automatically redirects stdout and stderr to a file named nohup.out in the current directory. If write permissions are lacking there, it attempts to write to $HOME/nohup.out.
  3. Detachment: When combined with an ampersand (&), it releases the command line prompt back to the user while keeping the script active in the background.

Standard Usage

To run a script that persists after disconnecting from an SSH session, use the following syntax:

nohup /path/to/script.sh &

The & operator places the command in the background, allowing you to continue using the terminal immediately.

To specify a custom destination for the generated output rather than relying on nohup.out, redirect the streams manually:

nohup /path/to/script.sh > custom_output.log 2>&1 &

In this command, > custom_output.log sends standard output to the specified file, and 2>&1 redirects standard error to the same destination.

Managing Nohup Processes

Because a nohup process is decoupled from your terminal session, you cannot stop it using standard keyboard shortcuts like Ctrl + C once the terminal is closed.

To manage or terminate a running nohup process:

  1. Locate the process ID (PID) using ps or pgrep: 
    ps aux | grep script.sh
  2. Terminate the process using the kill command followed by the PID: 
    kill <PID>

Using nohup is a standard, lightweight method for executing system maintenance, data backups, and long-running data processing tasks on remote servers without the risk of premature interruption caused by unstable network connections or closed SSH windows.