Development server does not shutdown cleanly

[rmboggs]

I am able to run the development server for testing but when I hit Ctrl-C to stop it, the error below is thrown. It is occurring in versions 1.2.1, 1.2.2, and 1.2.3. It's not preventing usage that I can see but it is a bit of an eyesore.

Thanks,
Ryan

$: python manage.py runserver
Validating models...
0 errors found

Django version 1.2.1, using settings 'mayham.settings'
Development server is running at http://127.0.0.1:8000/
Quit the server with CONTROL-C.
^CUnhandled exception in thread started by <function inner_run at 0x872cc8b4>
Traceback (most recent call last):
  File "/usr/local/lib/python2.6/site-packages/django/core/management/commands/runserver.py", line 60, in inner_run
    run(addr, int(port), handler)
  File "/usr/local/lib/python2.6/site-packages/django/core/servers/basehttp.py", line 721, in run
    httpd.serve_forever()
  File "/usr/local/lib/python2.6/SocketServer.py", line 224, in serve_forever
    r, w, e = select.select([self], [], [], poll_interval)
select.error: (4, 'Interrupted system call')

[rmboggs]

Sorry, should have specified. This occurred on OpenBSD current using either Python 2.6.5 and 2.6.6.

[mariarchi]

It's not a bug.
By pressing ctrl+c while the system call is executed (select.select) you're indeed interrupting it, which django tells you.

[rmboggs@…]

Yes, I am interrupting it because django is telling me to hit CONTROL-C to quit the development server. No offense intended but it seems like a bug to me if an error occurs while following django's directions.

If I am understanding right, this error is just an eyesore. CONTROL-C quits the dev server like it is suppose to but just displays that error. It works normally for Linux systems but on OpenBSD, the error appears.

[Russell Keith-Magee]

Yes - it is just an eyesore. And yes, it should be cleaned up.

[rmboggs]

The attached patch catches this specific issue but raises anything else. It worked during my tests on OpenBSD and should be good across the board. Please commit patch if acceptable. If not, please let me know what else is needed so I can make the appropriate corrections.

[Julien Phalip]

I think the attached patch is a bit better. Could you test it under OpenBSD?

[rmboggs]

Replying to julien:

I think the attached patch is a bit better. Could you test it under OpenBSD?

Yeah, I'll give it a try today...

[rmboggs]

I just tried the patch you provided on the latest snapshot of OpenBSD and it is not catching the exception. It's unfortunate because that patch did look cleaner than the one I provided.

Is there anything else that I can do to clean up the patch I provided to get it committed?

[Julien Phalip]

Could you track down which exception is actually raised, if any, on OpenBSD? Maybe try placing a breakpoint in a plain try/except statement using pdb (​http://docs.python.org/release/2.5.2/lib/module-pdb.html).

[rmboggs]

Well, the exception that is raised is in the original description of this ticket. It looks like it originates in Python's SocketServer.py. I'll start digging around there unless there is a better suggestion...

[Aymeric Augustin]

Change UI/UX from NULL to False.

[Aymeric Augustin]

Change Easy pickings from NULL to False.

[Unai Zalakain]

Could you please try out with Django 1.4 or newer? Only to confirm that a patch is still needed.

[Markus Amalthea Magnuson]

I can't reproduce this in Django 1.5.5 running on OpenBSD 5.5.

Doing Ctrl-C just gets be back to prompt, no traceback, just as in e.g. Mac OS.