Windows-only bug

[pombredanne]

I did hit this bug today: http://bugs.python.org/issue14252
which is also documented here I think:
http://stackoverflow.com/questions/5573257/windowserror-error-5-access-is-denied-when-trying-to-kill-a-subprocess-pytho
I will come with a patch to support the same approach patched in the main Python 2.7 regardless of the Python version

[apmorton]

This issue is caused by https://github.com/nickstenning/honcho/blob/master/honcho/process.py#L141

from the python docs: On Windows kill() is an alias for terminate()

The terminate function in the ProcessManager class basically kills each process twice when running on windows.

I think what happens is essentially a race condition with the return code checking.

If I change the function to only issue SIGALRM if not ON_WINDOWS and do nothing otherwise, there are no exceptions and the code still works properly.

On a slightly unrelated note, this issue only occurs on some programs run for me. specifically the django runserver command. Which also happens to not output anything over stdout until I try and terminate it, and then it dumps everything at once, which could also be related, since that would cause the delay in termination and hence trigger the race.

I do not know if the same issue with runserver occurs on linux regarding the output not being sent immediately.

[nickstenning]

Closing in favour of tracking this issue in #39.