STINNER Victor added the comment:
I created a new repository just for this issue:
http://hg.python.org/sandbox/issue21645
I added a lot of debug. It looks like a race condition: the SIGCHLD signal is
handled in a thread B (a C thread, not a Python), whereas the asyncio event
loop is running
STINNER Victor added the comment:
I'm running tests on the buildbot AMD64 FreeBSD 9.0 custom:
http://buildbot.python.org/all/builders/AMD64%20FreeBSD%209.0%20custom/builds/27/steps/test/logs/stdio
--
___
Python tracker rep...@bugs.python.org
STINNER Victor added the comment:
(1) the thread B gets the SIGCHLD signal: it writes a byte into the self pipe
of the event loop
C signal_handler: sig_num=20, thread=34468857856
trip_signal(20): write()
(2) in the main thread, the event loop is awaken by the write in the pipe...
but
Roundup Robot added the comment:
New changeset a0e6a370755f by Victor Stinner in branch 'default':
Issue #21645: Add debug code to analyze a failure on FreeBSD 9
http://hg.python.org/cpython/rev/a0e6a370755f
--
___
Python tracker
Roundup Robot added the comment:
New changeset dbf991650441 by Victor Stinner in branch 'default':
Issue #21645: test_asyncio, log debug trace into sys.__stderr__, not in
http://hg.python.org/cpython/rev/dbf991650441
--
___
Python tracker
STINNER Victor added the comment:
Buildbot where the issue occurs:
http://buildbot.python.org/all/builders/AMD64%20FreeBSD%209.0%203.x/
--
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue21645
Roundup Robot added the comment:
New changeset 69d474dab479 by Victor Stinner in branch 'default':
Issue #21645: asyncio: add a watchdog in test_read_all_from_pipe_reader() for
http://hg.python.org/cpython/rev/69d474dab479
--
nosy: +python-dev
___
Changes by Giampaolo Rodola' g.rod...@gmail.com:
--
components: +Asyncio
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue21645
___
___
New submission from STINNER Victor:
http://buildbot.python.org/all/builders/AMD64%20FreeBSD%209.0%203.4/builds/191/steps/test/logs/stdio
[ 8/389] test_asyncio
Timeout (1:00:00)!
Thread 0x000801407400 (most recent call first):
File
Guido van Rossum added the comment:
Maybe see Tulip issue 168? That test was added to support that. Maybe the fix
doesn't work for this platform?
--
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue21645
10 matches
Mail list logo