pythonprocesspidkill-processsigkill

How to kill all processes of a Python program?

I am running a Python program service_host.py, which started multiple processes. And when I use ps -ef | grep python

to check the pids, it shows a lot:

congmin  26968 22897  0 Jun20 ?        00:00:00 python service_host.py
congmin  26969 22897  0 Jun20 ?        00:00:00 python service_host.py
congmin  26970 22897  0 Jun20 ?        00:00:00 python service_host.py
congmin  26971 22897  0 Jun20 ?        00:00:00 python service_host.py
congmin  26972 22897  0 Jun20 ?        00:00:00 python service_host.py
congmin  26973 22897  0 Jun20 ?        00:00:00 python service_host.py
congmin  26974 22897  0 Jun20 ?        00:00:00 python service_host.py
congmin  26975 22897  0 Jun20 ?        00:00:00 python service_host.py

What's the best way to kill all these processes at once? I am on Linux and don't want to kill one by one through the process ID. Is there a way to kill them by the Python name 'service_host.py'? I tried this but it didn't kill at all:

import psutil

PROCNAME = "service_host.py"

for proc in psutil.process_iter():
    # check whether the process name matches
    if proc.name() == PROCNAME:
        proc.kill()

Is there a terminal command that can do this job easily?

Solution

  • Using htop instead you’ll be able to filter with F4 the processes service_host instead and have a look at the whole tree with F5. From there you can kill them all at once with F9 ...

    Follow these simple 3 steps:

    1. enter htop
      • optionally press F4 and type service_host to filter the process to kill
      • optionally press F5 to sort processes as a tree. This organizes child processes to its primary process ID (PID)
    2. scroll the list and highlight the process to kill, then press F9 for kill options
    3. highlight 15 SIGTERM (signal terminate) to gracefully request to stop the proccess(es) or 9 SIGKILL (signal kill) to forcefully shut down the process(es) and finally press enter

    I usually go with 9 SIGKILL without worrying too much, but it's up to you and what you're doing within your python script!