Issue 9978: Better wait for slow machine in test_os (_kill_with_event) (original) (raw)

Created on 2010-09-28 23:13 by pitrou, last changed 2022-04-11 14:57 by admin. This issue is now closed.

Files
File name Uploaded Description Edit
py3k_fix_test_os.patch ocean-city,2010-09-29 14:57 review
Messages (9)
msg117561 - (view) Author: Antoine Pitrou (pitrou) * (Python committer) Date: 2010-09-28 23:13
This has started showing up since r85073. ====================================================================== ERROR: test_CTRL_BREAK_EVENT (test.test_os.Win32KillTests) ---------------------------------------------------------------------- Traceback (most recent call last): File "D:\cygwin\home\db3l\buildarea\3.x.bolen-windows\build\lib\test\test_os.py", line 1068, in test_CTRL_BREAK_EVENT self._kill_with_event(signal.CTRL_BREAK_EVENT, "CTRL_BREAK_EVENT") File "D:\cygwin\home\db3l\buildarea\3.x.bolen-windows\build\lib\test\test_os.py", line 1039, in _kill_with_event os.kill(proc.pid, event) WindowsError: [Error 5] Access is denied http://www.python.org/dev/buildbot/builders/x86%20XP-4%203.x
msg117590 - (view) Author: Hirokazu Yamamoto (ocean-city) * (Python committer) Date: 2010-09-29 10:21
I cannot reproduce this in my environment... E:\python-dev\py3k>py3k -m test.regrtest test_os [1/1] test_os [33503 refs] 1 test OK. [85140 refs]
msg117615 - (view) Author: Hirokazu Yamamoto (ocean-city) * (Python committer) Date: 2010-09-29 14:57
I created the patch to fix this issue. WINFUNCTYPE is not member of ctypes.wintypes, but ctypes.
msg117616 - (view) Author: Brian Curtin (brian.curtin) * (Python committer) Date: 2010-09-29 14:59
I just checked in the importing changes to r85109 and r85110 (I saw your branch checkins happening at the same time, sorry).
msg117617 - (view) Author: Brian Curtin (brian.curtin) * (Python committer) Date: 2010-09-29 15:03
Do you think we need the other parts of that patch? It seems like the only real issue here was the importing.
msg117624 - (view) Author: Hirokazu Yamamoto (ocean-city) * (Python committer) Date: 2010-09-29 15:48
Other parts are needed for my environment. If machine is really slow, there is possibility of subprocess being killed before execution reaches "wintypes.WINFUNCTYPE" line.
msg118008 - (view) Author: Brian Curtin (brian.curtin) * (Python committer) Date: 2010-10-05 13:23
In that case, the patch seems alright to me.
msg118170 - (view) Author: Hirokazu Yamamoto (ocean-city) * (Python committer) Date: 2010-10-08 09:47
Thank you. Committed in r85315(py3k), r85318(release27-maint).
msg118171 - (view) Author: Hirokazu Yamamoto (ocean-city) * (Python committer) Date: 2010-10-08 09:52
I modified the patch slightly to avoid potential hang.
History
Date User Action Args
2022-04-11 14:57:07 admin set github: 54187
2010-10-08 09:52:43 ocean-city set messages: +
2010-10-08 09:47:31 ocean-city set status: open -> closedmessages: +
2010-10-05 13:23:14 brian.curtin set messages: + versions: + Python 2.7
2010-10-05 04:50:25 ocean-city set title: test_os failures on XP-4 buildbot -> Better wait for slow machine in test_os (_kill_with_event)
2010-09-29 15:48:20 ocean-city set messages: +
2010-09-29 15:03:39 brian.curtin set resolution: fixedmessages: + stage: needs patch -> commit review
2010-09-29 14:59:51 brian.curtin set messages: +
2010-09-29 14:57:36 ocean-city set files: + py3k_fix_test_os.patchkeywords: + patchmessages: +
2010-09-29 10:21:57 ocean-city set messages: +
2010-09-28 23:13:58 pitrou create