[issue11954] 3.3 - 'make test' fails

Changes by Stefan Krah : -- status: pending -> closed ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe: http://ma

[issue11954] 3.3 - 'make test' fails

Stefan Krah added the comment: I can confirm that test_distutils runs fine on Fedora both as root and as non-root (rpmbuild is installed). Also, I don't see any test failures when running the whole test suite as root. IMO this means that all remaining issues in this report are fixed. -

[issue11954] 3.3 - 'make test' fails

Nick Coghlan added the comment: I'm not sure what platform this was on, but if it was a Fedora system then test_distutils should also be working as non-root now (see #14443) -- nosy: +ncoghlan ___ Python tracker

[issue11954] 3.3 - 'make test' fails

Stefan Krah added the comment: I just ran into this with test_httpserver/Python2.7, which fails if run as root. If I understand correctly this is the main problem here. Jason, do you agree to change the issue title to reflect this? -- nosy: +skrah versions: +Python 2.7, Python 3.1, Pyth

[issue11954] 3.3 - 'make test' fails

R. David Murray added the comment: test_distutils should not be dependent on the existence of rpm (if it references the system rpm it should skip if it doesn't exist). It is difficult to find someone willing to run a buildbot as the root user, so while we will see about fixing the test suite

[issue11954] 3.3 - 'make test' fails

Jason Vas Dias added the comment: One last niggle : when I do $ DESTDIR=`pwd`/inst make install The configure '--libdir=/usr/lib64' setting I specified is ignored and python installs itself under /usr/lib . I guess I need to raise a different bug on this ? --

[issue11954] 3.3 - 'make test' fails

Jason Vas Dias added the comment: Nope, all tests except rpm dependant test_distutils OK as non-root with the patch I submitted for #11956 - so I guess that's good enough . Please fix the python 'make test' to work when run as root user . Thank You! --

[issue11954] 3.3 - 'make test' fails

Changes by Brian Curtin : -- stage: -> needs patch ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail

[issue11954] 3.3 - 'make test' fails

Jason Vas Dias added the comment: hmm... not sure if "make test" completely succeeds as non-root user yet: "test_subprocess" has been sitting in this state for @ 30mins : [282/354] test_subprocess . this bit of output is from a test of stdout in a different process ... and is getting rathe

[issue11954] 3.3 - 'make test' fails

Jason Vas Dias added the comment: argparse.py needs a similar fix, but I'm not sure where - I raised issue #11955 on this . -- ___ Python tracker ___ ___

[issue11954] 3.3 - 'make test' fails

Jason Vas Dias added the comment: OK, so those test_argparse.py and test_import.py tests succeed if run as a non-root user , because as the super-user one is allowed to override mode 0300 ( lack of 'w' bit for owner ) on a directory , so some tests that check if they are unable to create files w

[issue11954] 3.3 - 'make test' fails

Changes by Brian Curtin : -- components: +Tests -Build nosy: -brian.curtin type: crash -> behavior versions: +Python 3.3 -Python 3.4 ___ Python tracker ___ _

[issue11954] 3.3 - 'make test' fails

Brian Curtin added the comment: > I'm not confident to start using this build until I can pin down why eg > test_argparse and test_import are failing. Feel free to look into the failures in Lib/test/test_argparse.py and Lib/test/test_import.py -- nosy: +brian.curtin _

[issue11954] 3.3 - 'make test' fails

R. David Murray added the comment: A focused bug report would focus on *one* of the test failures (as in the failures from running a single test_x). Python3 does not support Berkeley DB out of the box, you need a third party library to get bdb support. You might be interested to look at

[issue11954] 3.3 - 'make test' fails

Jason Vas Dias added the comment: 'make test' failures after $ /usr/src/cypython/configure --prefix=/usr --libdir=/usr/lib64 --enable-shared $ make -j2 && make test (make test fails) So, to run in "verbose mode", I do : $ LD_LIBRARY_PATH=`pwd` LD_PRELOAD=`pwd`/libpython3.so \ ./python -Wd -E

[issue11954] 3.3 - 'make test' fails

Jason Vas Dias added the comment: OK, so the ' vanilla configure ' build succeeds too - using DB module only gdbm , and with internal libffi: $ make clean $/usr/src/cpython/configure --prefix=/usr --libdir=/usr/lib64 --enable-shared ... $ echo $? 0 $ make -j2 ... $ echo $? 0 But exactly the sa

[issue11954] 3.3 - 'make test' fails

Jason Vas Dias added the comment: In reply to last comment : RE: > a single, focused bug What is this bug if not single and focused ? The "SINGLE FOCUS" of this bug, in case you missed it , is that there appears to be no source release of python that can build and pass its make tests on my pla

[issue11954] 3.3 - 'make test' fails

R. David Murray added the comment: Just so you know, you aren't likely to get much help using this approach to bug reporting. A single, focused bug report is much more likely to get attention. You might also want to try starting with a "vanilla configure" and see how things go with that firs

[issue11954] 3.3 - 'make test' fails

Jason Vas Dias added the comment: [237/354] test_pyexpat test_ordered_attributes (test.test_pyexpat.SetAttributeTest) ... ok test_specified_attributes (test.test_pyexpat.SetAttributeTest) ... ok test_parse_file (test.test_pyexpat.ParseTest) ... ok test_unicode (test.test_pyexpat.ParseTest) ... o

[issue11954] 3.3 - 'make test' fails

Jason Vas Dias added the comment: test_too_high_from_package (test.test_import.RelativeImportFromImportlibTests) ... test test_import failed -- Traceback (most recent call last): File "/usr/src/cpython/Lib/test/test_import.py", line 545, in test_unwritable_directory '__pycache__', '{}.{}

[issue11954] 3.3 - 'make test' fails

Jason Vas Dias added the comment: test_start_with_double_slash (test.test_httpservers.SimpleHTTPRequestHandlerTestCase) ... /usr/src/cpython/Lib/unittest/case.py:799: BytesWarning: str() on a bytes instance (i, item1, item2)) test test_httpservers failed -- multiple errors occurred ok

[issue11954] 3.3 - 'make test' fails

Jason Vas Dias added the comment: test_successes_one_group_sysargs (test.test_argparse.TestTypeUserDefined) ... test test_argparse failed -- multiple errors occurred ok == FAIL: test_failures_many_groups_listargs (test.test_ar

[issue11954] 3.3 - 'make test' fails

Jason Vas Dias added the comment: [ 16/354] test_argparse ... # all ok up to: test_wb_1 (test.test_argparse.TestFileTypeRepr) ... ok test_failures_many_groups_listargs (test.test_argparse.TestFileTypeW) ... FAIL test_failures_many_groups_sysargs (test.test_argparse.TestFileTypeW) ... FAIL test_f

[issue11954] 3.3 - 'make test' fails

Jason Vas Dias added the comment: OK, so getting out strace shows I need to do : $ LD_LIBRARY_PATH=`pwd` LD_PRELINK=`pwd`/libpython3.3.so.1.0 ./python -Wd -E -bb /usr/src/cpython/Lib/test/regrtest.py -l -v 2>&1 | tee make.test.verbose.log I'll show the failures from the above command when i

[issue11954] 3.3 - 'make test' fails

Jason Vas Dias added the comment: no, 'make test V=1' still fails with 'run in verbose mode for details' . does it mean 'make test verbose=1' ? 'make test mode=verbose' ? -- ___ Python tracker ___

[issue11954] 3.3 - 'make test' fails

Jason Vas Dias added the comment: $ ls -l /usr/lib64/libffi* -rwxr-xr-x 1 root root 36839 May 25 2008 /usr/lib64/libffi-2.00-beta.so -rw-r--r-- 1 root root 193480 Apr 29 13:22 /usr/lib64/libffi.a -rwxr-xr-x 1 root root904 Apr 29 13:22 /usr/lib64/libffi.la lrwxrwxrwx 1 root root 15 Apr

[issue11954] 3.3 - 'make test' fails

Jason Vas Dias added the comment: suspect cause #1 - bad system libffi ? I just built it, but : $ ls -l /usr/lib64/libffi* -rwxr-xr-x 1 root root 36839 May 25 2008 /usr/lib64/libffi-2.00-beta.so -rw-r--r-- 1 root root 201320 Apr 8 03:46 /usr/lib64/libffi.a lrwxrwxrwx 1 root root 15 Apr

[issue11954] 3.3 - 'make test' fails

Jason Vas Dias added the comment: my expat was just built a few weeks ago from expat-2.0.1.tar.gz: $ ls -l /usr/lib64/libexpat* -rw-r--r-- 1 root root 577052 Apr 8 21:52 /usr/lib64/libexpat.a lrwxrwxrwx 1 root root 17 Apr 8 21:52 /usr/lib64/libexpat.so -> libexpat.so.1.5.2 lrwxrwxrwx 1 r

[issue11954] 3.3 - 'make test' fails

Changes by Jason Vas Dias : -- type: -> crash ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.pyth

[issue11954] 3.3 - 'make test' fails

Changes by Jason Vas Dias : -- components: +Build versions: +Python 3.4 ___ Python tracker ___ ___ Python-bugs-list mailing list Unsub

[issue11954] 3.3 - 'make test' fails

New submission from Jason Vas Dias : I do : $ hg clone http://hg.python.org/cpython ( my existing Python-2.7, following upgrade to glibc-2.13, started producing erroneous results - see gnome.org glib bug : https://bugzilla.gnome.org/show_bug.cgi?id=648863 So I tried downloading and build