Four Powerful kill Command Techniques for Managing Linux Processes

In Linux systems, managing processes is essential for developers and administrators, and when a process hangs, using the kill command with the right signal allows graceful termination and data safety.

1. SIGTERM (15): Graceful termination signal

The default behavior of kill is to send SIGTERM, which asks the process to shut down cleanly, allowing it to perform cleanup. Run the simple command:

kill 123

Example output:

bogon:~ sdcui$ jps
73113 Launcher
72522 Main
73114 PdfviewerApplication
72716 RemoteMavenServer36
73118 Jps
bogon:~ sdcui$ kill 73114

2. SIGQUIT (3): Termination with core dump

SIGQUIT works like SIGTERM but also generates a core dump file that records the process memory state, which is useful for debugging.

kill -3 123

Note: Java processes do not exit on SIGQUIT; they only produce a core dump.

Example:

bogon:~ sdcui$ jps
73248 PdfviewerApplication
73259 Jps
72522 Main
72716 RemoteMavenServer36
73247 Launcher
bogon:~ sdcui$ kill -3 73248

3. SIGKILL (9): Forceful termination

When a process does not respond to SIGTERM or SIGQUIT, SIGKILL is the last resort. It tells the kernel to kill the process immediately, which may cause data loss.

kill -9 123

4. SIGSTOP (19) and SIGCONT (18): Pause and resume

SIGSTOP pauses a process without terminating it. You can later resume the process with SIGCONT.

kill -19 123  # pause process
kill -18 123  # resume process

Summary

Mastering the four key kill techniques—SIGTERM, SIGQUIT, SIGKILL, and SIGSTOP—enables you to handle stuck processes efficiently while avoiding potential data risks; start with SIGTERM or SIGQUIT and resort to SIGKILL only when absolutely necessary.

Reference: https://man7.org/linux/man-pages/man7/signal.7.html