#17924: Race condition in pexpect
-------------------------------------+-------------------------------------
Reporter: Vincent.Neri | Owner:
Type: defect | Status: needs_work
Priority: major | Milestone: sage-6.8
Component: interfaces | Resolution:
Keywords: docker | Merged in:
Authors: Jeroen Demeyer, | Reviewers: Vincent Neri
Volker Braun | Work issues:
Report Upstream: N/A | Commit:
Branch: | 6a759cd704efd7d7b2109b8c58e2621074ea3b07
u/jdemeyer/failure_in_doctest_framework_when_running_in_docker| Stopgaps:
Dependencies: #17686 |
-------------------------------------+-------------------------------------
Comment (by jdemeyer):
Replying to [comment:42 vbraun]:
> there is actually no !SystemExit caught.
Wrong...
Here is what really happens:
1. `pexpect` calls `os.fork()`, creating 2 Python processes.
1. In the child process, `pexpect` creates a `pty` and redirects
stdin/stdout/stderr to this `pty`. As a consequence, anything `print`ed by
the child process is not seen.
1. The child process receives a `SIGHUP`, which (due to the interrupt code
in Sage) raises `SystemExit`.
1. The child process catches `SystemExit` and runs the code which is
really meant for the parent process (this is the bug that #17686 fixes).
1. The child process puts something in the queue, filling it.
1. The parent process also tries to put something in the queue, which
fails.
--
Ticket URL: <http://trac.sagemath.org/ticket/17924#comment:43>
Sage <http://www.sagemath.org>
Sage: Creating a Viable Open Source Alternative to Magma, Maple, Mathematica,
and MATLAB
--
You received this message because you are subscribed to the Google Groups
"sage-trac" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/sage-trac.
For more options, visit https://groups.google.com/d/optout.
