Re: [python-win32] daemonized child

2009-09-28 Thread Greg Ewing
Elias Fotinis wrote: From: "Antoine Martin" I had forgotten about the pipes (doh), I guess I should start the child with stdin=stdout=stderr=None then? That won't necessarily close the OS-level file descriptors, though. If you want that, you need to do something like for i in xrange(3):

Re: [python-win32] daemonized child

2009-09-28 Thread Elias Fotinis
From: "Antoine Martin" Hah, that's easier than I thought! Sure is. The hard part is terminating a bunch of processes as group (either manually with PIDs or using job objects). I had forgotten about the pipes (doh), I guess I should start the child with stdin=stdout=stderr=None then? Yes,

Re: [python-win32] daemonized child

2009-09-27 Thread Antoine Martin
Elias Fotinis wrote: > From: "Antoine Martin" >> What is the best way to ensure that a child started with >> subprocess.Popen does not get killed when its parent terminates on win32? >> On Linux I can daemonize the child with fork()s and dup2()s >> But what about windows? > > You don't have to do

Re: [python-win32] daemonized child

2009-09-27 Thread Elias Fotinis
From: "Antoine Martin" What is the best way to ensure that a child started with subprocess.Popen does not get killed when its parent terminates on win32? On Linux I can daemonize the child with fork()s and dup2()s But what about windows? You don't have to do anything. The lifetime of a child pr