Launching a subprocess without waiting around for the result?

John Fabiani jfabiani at
Fri Sep 19 07:27:17 CEST 2008

erikcw wrote:

> On Sep 18, 3:33 pm, Ben Finney <bignose+hates-s... at>
> wrote:
>> erikcw <erikwickst... at> writes:
>> > I have a cgi script where users are uploading large files for
>> > processing. I want to launch a subprocess to process the file so the
>> > user doesn't have to wait for the page to load.
>> For "how do I deal with subprocesses from Python", the (new in Python
>> 2.4) 'subprocess' module is the default go-to answer
>> <URL:>, replacing a
>> rather fragmented set of modules before it.
>> > What is the correct way to launch subprocess without waiting for the
>> > result to return?
>> Creating an instance of 'subprocess.Popen' will launch the process and
>> return the Popen instance. You then have the option of polling it or
>> waiting for it to complete.
>> --
>> \     “To stay young requires unceasing cultivation of the ability to |
>> `\                   unlearn old falsehoods.” —Robert Anson Heinlein |
>> _o__)                                                                  |
>> Ben Finney
> So if I create a Popen object and then just ignore the object and exit
> the program the subproccess will finish it's work and then exit itself
> cleanly?
Just so happens that I ran into the same problem recently.  I started with
exec(), then os.system(), next os.popen(), and last os.spawn().
This is what I discovered on a windows platform.  The exec() replaced the
current running process. os.system did not start a completely new process. 
os.popen() created a new process but did not open a command box to display
any print statements.  Lastly os.spawn() worked - it created a new process
and opened a command box to display any print statements I needed.


More information about the Python-list mailing list