How to Use the kill Command in Linux

Sometimes applications stop responding or consume too many system resources, and a normal exit no longer works. The kill command terminates these processes by sending them a signal, from a graceful shutdown request to an immediate force kill.

There are several utilities that allow you to terminate errant processes, with kill being the most commonly used.

kill Command #

kill is a shell built-in in most Bourne-derived shells such as Bash and Zsh. The command behavior is slightly different between the shells and the standalone /bin/kill executable.

Use the type command to display all locations on your system containing kill:

type -a kill

kill is a shell builtin
kill is /bin/kill

The output above shows that the shell built-in has priority over the standalone executable, and it is used whenever you type kill. If you want to use the binary, type the full path to the file /bin/kill. In this article, we will use the Bash built-in.

The syntax of the kill command takes the following form:

kill [OPTIONS] [PID]...

The kill command sends a signal to specified processes or process groups, causing them to act according to the signal. When the signal is not specified, it defaults to -15 (-TERM).

The most commonly used signals are:

• 1 (HUP) - Hangup, often used by daemons to reload configuration.
• 9 (KILL) - Kill a process immediately.
• 15 (TERM) - Gracefully stop a process.

To get a list of all available signals, invoke the command with the -l option:

kill -l

 1) SIGHUP       2) SIGINT       3) SIGQUIT      4) SIGILL
 5) SIGTRAP      6) SIGABRT      7) SIGBUS       8) SIGFPE
 9) SIGKILL     10) SIGUSR1     11) SIGSEGV     12) SIGUSR2
13) SIGPIPE     14) SIGALRM     15) SIGTERM

Signals can be specified in three different ways:

1. Using number (e.g., -1 or -s 1).
2. Using the “SIG” prefix (e.g., -SIGHUP or -s SIGHUP).
3. Without the “SIG” prefix (e.g., -HUP or -s HUP).

The following commands are equivalent to one another:

kill -1 PID_NUMBER
kill -SIGHUP PID_NUMBER
kill -HUP PID_NUMBER

The PIDs provided to the kill command can be one of the following:

• If PID is greater than zero, the signal is sent to the process with ID equal to the PID.
• If PID is equal to zero, the signal is sent to all processes in the current process group. In other words, the signal is sent to all processes belonging to the PGID of the shell that invoked the kill command. Use the ps -efj command to view process group IDs (PGIDs).
• If PID is equal to -1, the signal is sent to all processes the caller has permission to signal, except special system processes such as PID 1.
• If PID is less than -1, the signal is sent to all processes in the process group with PGID equal to the absolute value of the PID.

Regular users can send signals to their own processes, but not those that belong to other users, while the root user can send signals to other users’ processes.

Terminating Processes Using the kill Command #

To terminate or kill a process with the kill command, first you need to find the process ID number (PID). You can do this using different commands such as top, ps , pidof and pgrep .

For example, you can use ps to find the PID for a running process:

ps aux | grep firefox

Let’s say the Firefox browser has become unresponsive, and you need to kill the Firefox process. To find the browser PIDs use the pidof command:

pidof firefox

The command will print the IDs of all Firefox processes:

6263 6199 6142 6076

Once you know the PIDs, you can kill all of them by sending the KILL signal:

kill -9 6263 6199 6142 6076

Instead of searching for PIDs and then killing the processes, you can combine the above commands into one:

kill -9 $(pidof firefox)

Reloading Processes Using the kill Command #

Another common use case for kill is to send the HUP signal to daemons that handle it as a reload request.

For example, to reload Nginx , you need to send a signal to the master process. The process ID of the Nginx master process can be found in the nginx.pid file, which is typically located in the /var/run directory.

Use the cat command to find the master PID:

cat /var/run/nginx.pid

30251

Once you find the master PID, reload the Nginx settings by typing:

sudo kill -1 30251

The command above must be run as root or user with sudo privileges.

SIGTERM vs SIGKILL #

When terminating processes, you should first try SIGTERM (-15), which allows the process to clean up and exit gracefully. If the process does not respond, use SIGKILL (-9), which forces immediate termination.

# Try graceful termination first
kill -15 PID

# If process does not stop, force kill
kill -9 PID

SIGKILL cannot be caught or ignored by the process, so it is the strongest termination signal. However, it does not allow the process to save data or clean up, and some process states, such as uninterruptible sleep or zombies waiting to be reaped, may not disappear immediately.

Check if a Process Exists #

Use signal 0 to check whether a PID exists without sending a real signal:

kill -0 PID

Kill Processes by Name #

Instead of finding the PID first, you can kill processes by name using pkill or killall.

To kill all Firefox processes by name:

pkill firefox

Or using killall:

killall firefox

The difference is that pkill matches partial names by default, while killall requires the exact process name.

Killing Background Jobs #

If you have background jobs running in the current shell, you can kill them using the job number instead of the PID:

kill %1

To list all background jobs and their numbers, use the jobs command:

jobs

[1]+  Running    sleep 1000 &
[2]-  Running    sleep 2000 &

Troubleshooting #

kill: (PID): No such process
The process already exited or the PID is wrong. Re-check with ps -p PID or pgrep name before retrying.

kill: (PID): Operation not permitted
You do not own the target process. Re-run with sudo if the process belongs to root or another user.

Process keeps running after kill -15
Some processes ignore SIGTERM or cannot handle it while busy in uninterruptible system calls. Try kill -9 PID to request forced termination, but note that the process cannot clean up and data may be lost.

Killed wrong PID and shell now hangs
If you accidentally killed your shell or a parent process, open a new terminal session and use pgrep to verify the target before retrying.

Quick Reference #

For a printable quick reference, see the kill cheatsheet .

FAQ #

What does kill -9 do?
It sends SIGKILL, which immediately terminates the process. The kernel handles the signal directly, so the process cannot catch, block, or ignore it.

What is the difference between kill and killall?
kill accepts a PID and signals one process. killall accepts a process name and signals every process matching that name.

Why is kill -9 not recommended as the first choice?
SIGKILL skips cleanup. The process cannot flush buffers, save state, or release locks, which may corrupt data or leave stale lock files behind.

Can a regular user kill any process?
No. A user can only signal processes they own. Killing other users’ processes requires sudo or running as root.

How do I kill a process by name instead of PID?
Use pkill name for partial name matches or killall name for an exact match. Both accept the same signal flags as kill.

Conclusion #

The kill command is used to send signals to processes in Linux. Always try SIGTERM (-15) first for graceful termination, and use SIGKILL (-9) only when necessary.

About the authors

Dejan Panovski

Dejan Panovski is the founder of Linuxize, an RHCSA-certified Linux system administrator and DevOps engineer based in Skopje, Macedonia. Author of 800+ Linux tutorials with 20+ years of experience turning complex Linux tasks into clear, reliable guides.

View author page