How to Keep Long‑Running Linux Jobs Alive with nohup

Why use nohup?

When a program runs for hours or days, closing the terminal or losing network connectivity normally sends a HUP (hang‑up) signal that terminates the process. The nohup command (short for “no hangup”) tells the shell to ignore this signal, allowing the program to continue running.

Basic syntax

$ nohup command arguments

or $ nohup options Use nohup --help for help and nohup --version to see the version.

Running a program with nohup

Execute a command with nohup so it keeps running after the shell exits: $ nohup my_program The standard output and error are written to a file named nohup.out in the current directory (or home directory if not writable).

Redirecting output

To store output in a custom location and filename, redirect it with >:

$ nohup ./myScript.sh > ~/output/myOutput.txt

Background execution

Append & to run the job in the background. The process will still be detached from the terminal. Use fg to bring it back to the foreground if needed.

Running multiple commands

Chain several commands with && inside a bash -c string, and redirect the combined output:

$ nohup bash -c 'mkdir files && ping -c 1 baidu.com && ls' > output.txt

Stopping a background process

Find the process ID (PID) with ps or pgrep, then terminate it with kill -9 PID:

$ ps aux | grep myScript.sh

$ kill -9 14942