David Edelsohn added the comment:
There now are two zLinux buildbots: zlinux (running SUSE) and zwheezy (running
Debian). zlinux (running on SUSE) has the libc problem causing the timezone
error. A second buildbot was added, not converting or upgrading the existing
buildbot.
I still would
STINNER Victor added the comment:
The bug still occurs on the System Z buildbot:
http://buildbot.python.org/all/builders/System%20Z%20Linux%203.x/builds/2375/steps/test/logs/stdio
==
FAIL: test_list_command_verbose
Antoine Pitrou added the comment:
Any other ideas for a reliable method to restore the correct timezone after
running a test?
No. The best would be for you to investigate. Perhaps contact some glibc guys.
--
___
Python tracker
David Edelsohn added the comment:
It doesn't fail on the Debian system. The Debian system will be successful
after the test_gdb patch is installed.
--
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue20220
David Edelsohn added the comment:
Any other ideas for a reliable method to restore the correct timezone after
running a test?
--
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue20220
___
Serhiy Storchaka added the comment:
May be issue6478 is related.
--
nosy: +berker.peksag
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue20220
___
Antoine Pitrou added the comment:
May be issue6478 is related.
I'm not sure. issue6478 looks like a Python stdlib bug, while this issue looks
like a libc problem.
--
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue20220
Antoine Pitrou added the comment:
David, could you try the patch in msg230703 and see if it fixes the problem on
your buildbot?
--
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue20220
___
David Edelsohn added the comment:
Unfortunately, the patch does not fix the failures of running test_imaplib
before test_tarfile or test_datetime.
--
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue20220
Antoine Pitrou added the comment:
I can reproduce under Linux using:
$ TZ='America/New_York' ./python -m test -v test_imaplib test_tarfile
test_datetime
--
nosy: +David.Edelsohn, pitrou
___
Python tracker rep...@bugs.python.org
Antoine Pitrou added the comment:
We seem to be bitten by the following bit of glibc (?) oddity:
os.environ['TZ'] = 'America/New_York'
time.tzset()
time.daylight, time.timezone, time.altzone, time.tzname
(1, 18000, 14400, ('EST', 'EDT'))
os.environ['TZ'] = 'STD-1DST'
time.tzset()
Antoine Pitrou added the comment:
First unsetting TZ seems to fix it:
diff --git a/Lib/test/support/__init__.py b/Lib/test/support/__init__.py
--- a/Lib/test/support/__init__.py
+++ b/Lib/test/support/__init__.py
@@ -1507,11 +1507,11 @@ def run_with_tz(tz):
try:
New submission from Serhiy Storchaka:
TarFile.list() outputs a time exactly 6 hours back from expected.
http://buildbot.python.org/all/builders/System%20Z%20Linux%203.x/builds/1023/steps/test/logs/stdio
==
FAIL:
13 matches
Mail list logo