STINNER Victor added the comment: Again, the problem is that the exception exits from the function which owns the last reference to the Popen object. If the Popen is left alive when you exit the function, you create a zombi process, you can leave open pipes, etc.
It's unclear to me if CTRL+c sends SIGTERM to all processes or only a few of them (only the parent process?). Even if SIGTERM is sent to all processes, a child process can decide to completly ignore it, or can block in a deadlock or whatever. Giving a few seconds to the child process to wait until it ends is not easy because it's hard to choose an arbitrary timeout. If the timeout is too low, you kill the child process (SIGKILL) before it flushes files. If the timeout is too long, the parent process is blocked too long when the child process is really blocked. I suggest to keep the current behaviour by default. If you really want to give time to the child process, I suggest to add a *new* optional parameter . For example ctrlc_timeout=5.0 to send SIGTERM and then wait 5 seconds. I don't know if the parent process must always send a SIGTERM to the child process or not. One signal or two don't have the same behaviour. It's possible to explicitly send a SIGTERM to only one process using the kill command. ---------- _______________________________________ Python tracker <rep...@bugs.python.org> <http://bugs.python.org/issue25942> _______________________________________ _______________________________________________ Python-bugs-list mailing list Unsubscribe: https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
