Message 348384 - Python tracker

➜

This issue tracker has been migrated to GitHub, and is currently read-only.
For more information, see the GitHub FAQs in the Python's Developer Guide.

Author	santagada
Recipients	docs@python, neologix, owenlin, santagada, sbt, vstinner
Date	2019-07-24.12:30:27
SpamBayes Score	-1.0
Marked as misclassified	Yes
Message-id	<1563971427.93.0.743102544084.issue19809@roundup.psfhosted.org>
In-reply-to

Content
This is still the case on windows as the pipes created to talk to the process might be inherited by two or more simultaneous CreateProcess calls. I've found a suggested solution to this: https://devblogs.microsoft.com/oldnewthing/20111216-00/?p=8873 By only inheriting the stdout/err/in handles and them supporting close_fds for windows. Would more users be interested in a proper patch for this? For us now we have a lock around Popen.__init__ but that obviously doesn't suport subinterpreters and other calls to CreateProcess that might happen.

This is still the case on windows as the pipes created to talk to the process might be inherited by two or more simultaneous CreateProcess calls.

I've found a suggested solution to this:

https://devblogs.microsoft.com/oldnewthing/20111216-00/?p=8873

By only inheriting the stdout/err/in handles and them supporting close_fds for windows.

Would more users be interested in a proper patch for this? For us now we have a lock around Popen.__init__ but that obviously doesn't suport subinterpreters and other calls to CreateProcess that might happen.

History
Date	User	Action	Args
2019-07-24 12:30:27	santagada	set	recipients: + santagada, vstinner, neologix, docs@python, sbt, owenlin
2019-07-24 12:30:27	santagada	set	messageid: <1563971427.93.0.743102544084.issue19809@roundup.psfhosted.org>
2019-07-24 12:30:27	santagada	link	issue19809 messages
2019-07-24 12:30:27	santagada	create