address test_time.py failures under Redhat 6.2

[7172c766-024f-43cc-8cbc-e13efd4e927c]

BPO	762934
Nosy	@smontanaro, @warsaw, @brettcannon, @kbkaiser
Files	configure.in.patch.kbk: Catches RH Linux 6.2 "tzset" Error

^{Note: these values reflect the state of the issue at the time it was migrated and might not reflect the current state.}

Show more details

GitHub fields: ```python assignee = 'https://github.com/brettcannon' closed_at = created_at = labels = ['library'] title = 'address test_time.py failures under Redhat 6.2' updated_at = user = 'https://bugs.python.org/zenzen' ``` bugs.python.org fields: ```python activity = actor = 'brett.cannon' assignee = 'brett.cannon' closed = True closed_date = None closer = None components = ['Library (Lib)'] creation = creator = 'zenzen' dependencies = [] files = ['5424'] hgrepos = [] issue_num = 762934 keywords = ['patch'] message_count = 15.0 messages = ['44143', '44144', '44145', '44146', '44147', '44148', '44149', '44150', '44151', '44152', '44153', '44154', '44155', '44156', '44157'] nosy_count = 7.0 nosy_names = ['skip.montanaro', 'barry', 'nnorwitz', 'brett.cannon', 'kbk', 'zenzen', 'furie'] pr_nums = [] priority = 'normal' resolution = 'accepted' stage = None status = 'closed' superseder = None type = None url = 'https://bugs.python.org/issue762934' versions = ['Python 2.3'] ```

[7172c766-024f-43cc-8cbc-e13efd4e927c]

A mangled version of this patch is also in bug: http://www.python.org/sf/728051

Looks like tzset(3) is broken under Redhat 6.2 in a way that wasn't being detected by configure. This patch adds stricter tzset(3) checking to configure.in.

[d21744ff-f396-4c71-955e-7dbd2e886779]

Logged In: YES user_id=33168

Stuart, what boxes did you test this on? How confident are you that this won't break some other platform? I'm asking to try to determine if this should go into 2.3final (we only have 1 release candidate before release) or if this should wait for 2.3.1. Thanks for all your work on tzset.

[7172c766-024f-43cc-8cbc-e13efd4e927c]

Logged In: YES user_id=46639

This patch has only been tested under OS X.

I'm confident that it won't break other platforms.

I have no real way of proving that it addresses the problem it is supposed to solve, however, as I don't have access to a box that fails the tzset test in test_time.py.

The only reported platform that this is failing on is at http://www.python.org/sf/728051, so we could just flag this test as expected to fail on that platform, if someone knows how to do that.

[brettcannon]

Logged In: YES user_id=357491

We can say it should fail under Linux, but we can't specify Red Hat 6.2.

I am keeping an eye on this patch for bug bpo-763153, but I have to wait until the OP applies it and tests it.

Looking at the patch, beyond not realizing that the X-mas time was GMT initially and the unneeded variable assignments, the patch looks fine to me (might want to comment that tzset does not return a value; rather non-standard) although I am no autoconf expert and I am assuming it just compiles this C code and any problems it just says it fails.

Neal, what OS are you running? If it is non-OS X (sorry, that's what I am running as well) can you give the patch a try?

[warsaw]

Logged In: YES user_id=12800

This patch didn't break RH9.

[brettcannon]

Logged In: YES user_id=357491

Well, it didn't work for two people for Red Hat 6.2 . Perhaps being more explicit for the test? To give that a shot, I am uploading a patch that tests explicitly for AEDT as the daylight-savings timezone. I snagged the code mostly from Modules/timemodule.c .

Now this is untested so there could be syntax problems. I can't get a proper version of Autoconf to run on my system so I can run autoreconf. Hopefully this will deal with the problem.

[brettcannon]

Logged In: YES user_id=357491

Quick update: I got autoreconf to work and it seems to work for me. I also tested the C code in isolation and had no problems. So I now I just need other people to apply the patch and say whether it works.

[smontanaro]

Logged In: YES user_id=44345

Works for me (Mac OS X)

[d21744ff-f396-4c71-955e-7dbd2e886779]

Logged In: YES user_id=33168

See this mail for a possible fix. http://mail.python.org/pipermail/python-dev/2003-July/037116.html

Stuart, is that correct? It fixes the problem on RedHat 6.2 and doesn't break on RedHat 9. The change is to define YEAR as 365 24 3600, instead of adding 6 hours.

[brettcannon]

Logged In: YES user_id=357491

Well, it is looking like tzset_AEST is not working as a solution.
Hopefully Neal's patch will do the trick.

[kbkaiser]

Logged In: YES user_id=149084

[kbk@float ~/PYSRC]$ ./python
Python 2.3c2 (#15, Jul 24 2003, 11:17:16)
[GCC egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import time   
>>> from os import environ
>>> victoria = 'AEST-10AEDT-11,M10.5.0,M3.5.0'
>>> environ['TZ'] = victoria
>>> time.tzset()
>>> time.tzname
('AEST', 'AEST')
>>>

========================================= Hm!! Try a couple of things to try to see what's going on: \=========================================

>>> victoria2 = 'AEST-10AEDT-11'
>>> environ['TZ'] = victoria2
>>> time.tzset()
>>> time.tzname
('AEST', 'AEDT')
>>>
>>> # try reversing the changeover
...
>>> victoria3 = 'AEST-10AEDT-11,M3.5.0,M10.5.0'
>>> environ['TZ'] = victoria3
>>> time.tzset()
>>> time.tzname
('AEST', 'AEDT')

=================================== ok, debug inittimezone: \=================================== Breakpoint 1, inittimezone (m=0x4014053c) at /home/kbk/PYTHON/python/dist/src/Modules/timemodule.c:608 608t = (time((time_t *)0) / YEAR) * YEAR; (gdb) n 609p = localtime(&t); (gdb) p asctime(localtime(&t)) $14 = 0x4013be00 "Wed Jan 1 16:00:00 2003\n" (gdb) p localtime(&t)->tm_zone $19 = 0x8162b78 "AEST"

[std time on Jan 1!! ...back up a day or so....]

(gdb) p t = t - 84000 $20 = 1041316800 (gdb) p localtime(&t)->tm_zone $21 = 0x8162b90 "AEDT" (gdb) p asctime(localtime(&t)) $22 = 0x4013be00 "Tue Dec 31 17:40:00 2002\n" (gdb)

\=================================== so Linux6.2 thinks AEDT switches to AEST in Jan, and six months forward is still AEST.

xmas2002 is AEDT so config test passes but timemodule uses Jan 1 and flubs when setting tzname.

Need to do the config test later than xmas. \=====================================

*** Apply Patch SF 762934 configure.in.patch.kbk

\====================================== autoreconf && ./configure && make clean && make OPT=-g

\====================================== extract from configure log: .... checking for broken nice()... yes checking for working tzset()... no checking for tv_nsec in struct stat... no checking whether mvwdelch is an expression... yes .... \======================================

[kbk@float ~/PYSRC]$ ./python
Python 2.3c2 (#18, Jul 24 2003, 22:40:09)
[GCC egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> from test import test_time
>>> test_time.test_main()
test_asctime (test.test_time.TimeTestCase) ... ok
test_clock (test.test_time.TimeTestCase) ... ok
test_conversions (test.test_time.TimeTestCase) ... ok
test_data_attributes (test.test_time.TimeTestCase) ... ok
test_sleep (test.test_time.TimeTestCase) ... ok
test_strftime (test.test_time.TimeTestCase) ... ok
test_strptime (test.test_time.TimeTestCase) ... ok
test_tzset (test.test_time.TimeTestCase) ... ok

---

Ran 8 tests in 2.523s

OK

>>

\=================================== make test: \=================================== .... .... test_zlib 227 tests OK. 28 tests skipped: test_aepack test_al test_bsddb185 test_bsddb3 test_cd test_cl test_curses test_email_codecs test_gl test_imgfile test_largefile test_linuxaudiodev test_macfs test_macostools test_nis test_normalization test_ossaudiodev test_pep277 test_plistlib test_scriptpackages test_socket_ssl test_socketserver test_sunaudiodev test_timeout test_unicode_file test_urllibnet test_winreg test_winsound Those skips are all expected on linux2.

[brettcannon]

Logged In: YES user_id=357491

Thanks for the debugging work, Kurt. I will take a good look at this when 2.4 dev starts.

[brettcannon]

Logged In: YES user_id=357491

OK, so Kurt's thinking and debugging all seem good to me. I applied the patch and it worked correctly for me. Can other people test it (not just Red Hat but also other platforms; OS X is covered by my test)?

[6ed54e6b-f33a-4eda-9de5-85805cf4132d]

Logged In: YES user_id=218883

I've tested kbk's patch on fresh installs of both RH 6.1 and 6. 2, and they both pass all (non-skipped) tests.

[brettcannon]

Logged In: YES user_id=357491

Kurt's patch has been checked in both to HEAD and release23- maint.

Thanks, Kurt, for all your hard work.