[Debian-med-packaging] Bug#848410: python-biom-format: FTBFS randomly (failing tests)

Daniel McDonald wasade at gmail.com
Tue Dec 20 07:42:33 UTC 2016


Try a different seed? Same issue: it's a stochastic test. Other option is
bump the number of iterations it performs.

Best,
Daniel

On Dec 19, 2016 23:32, "Andreas Tille" <andreas at fam-tille.de> wrote:

> Hi Daniel,
>
> thanks for the helpful hint.  I've uploaded a package with the suggested
> fix and I hope this random error will be solved now.  However, there seems
> to be another issue which happens on i386 architecture only:
>
>
> I: pybuild base:184: cd /«BUILDDIR»/python-biom-
> format-2.1.5+dfsg/.pybuild/pythonX.Y_2.7/build; python2.7 -m nose tests
> ............................................................
> ............................................................
> ............................................................
> ............................................................
> ...........S.............
> ----------------------------------------------------------------------
> Ran 265 tests in 2.759s
>
> OK (SKIP=1)
>         pybuild --test --test-nose -i python{version} -p 3.5
> I: pybuild pybuild:212: cp -r /«BUILDDIR»/python-biom-format-2.1.5+dfsg/examples
> /«BUILDDIR»/python-biom-format-2.1.5+dfsg/tests /«BUILDDIR»/python-biom-
> format-2.1.5+dfsg/.pybuild/pythonX.Y_3.5/build
> I: pybuild base:184: cd /«BUILDDIR»/python-biom-
> format-2.1.5+dfsg/.pybuild/pythonX.Y_3.5/build; python3.5 -m nose tests
> ............................................................
> ............................................................
> ..........................F.................................
> ............................................................
> ...........S.............
> ======================================================================
> FAIL: test_subsample (tests.test_table.SparseTableTests)
> ----------------------------------------------------------------------
> Traceback (most recent call last):
>   File "/«BUILDDIR»/python-biom-format-2.1.5+dfsg/.pybuild/
> pythonX.Y_3.5/build/tests/test_table.py", line 2263, in test_subsample
>     (3, 1, 1)})
> AssertionError: Items in the second set but not the first:
> (3, 0, 0)
>
> ----------------------------------------------------------------------
> Ran 265 tests in 3.098s
>
> FAILED (SKIP=1, failures=1)
>
>
> You can find the full build log here:
>
>    https://buildd.debian.org/status/fetch.php?pkg=python-
> biom-format&arch=i386&ver=2.1.5%2Bdfsg-6&stamp=1482186822
>
> Any idea what might be wrong here?
>
> Kind regards
>
>       Andreas.
>
> On Tue, Dec 20, 2016 at 05:25:10AM +1100, Daniel McDonald wrote:
> > Thanks, Greg.
> >
> > Andreas, the particular test that is failing as it is stochastic. It is
> > likely possible to just set a random seed at entry into the test here:
> >
> > https://github.com/biocore/biom-format/blob/master/tests/
> test_util.py#L48
> >
> > Specifically, doing something like "np.random.seed(123)"
> >
> > Best,
> > Daniel
> >
> > On Tue, Dec 20, 2016 at 5:15 AM, Greg Caporaso <gregcaporaso at gmail.com>
> > wrote:
> >
> > > Hi Andreas, I'm cc'ing Daniel McDonald on this message, who is the
> primary
> > > maintainer of the biom-format package these days. Daniel, could you
> follow
> > > up with Andreas?
> > >
> > > On Sat, Dec 17, 2016 at 3:53 PM, Andreas Tille <andreas at an3as.eu>
> wrote:
> > >
> > >> Hi Greg,
> > >>
> > >> the Debian packaged biom-format received a bug report which leaves me
> > >> clueless how to track it down.  Is there possibly some way to make the
> > >> test suite more verbose to find out what part might cause the random
> > >> failure which you can see in the buil logs linked from the bug page
> > >>
> > >>    https://bugs.debian.org/848410
> > >>
> > >> Thanks for any idea to track this down
> > >>
> > >>       Andreas.
> > >>
> > >> On Sat, Dec 17, 2016 at 01:34:49PM +0100, Santiago Vila wrote:
> > >> > Package: src:python-biom-format
> > >> > Version: 2.1.5+dfsg-5
> > >> > Severity: serious
> > >> >
> > >> > Dear maintainer:
> > >> >
> > >> > I tried to build this package in stretch with "dpkg-buildpackage -A"
> > >> > (which is what the "Arch: all" autobuilder would do to build it)
> > >> > but it failed:
> > >> >
> > >> > ------------------------------------------------------------
> > >> --------------------
> > >> > [...]
> > >> >  debian/rules build-indep
> > >> > dh build-indep --with python2,python3,bash-completion,sphinxdoc
> > >> --buildsystem=pybuild
> > >> >    dh_testdir -i -O--buildsystem=pybuild
> > >> >    dh_update_autotools_config -i -O--buildsystem=pybuild
> > >> >    dh_auto_configure -i -O--buildsystem=pybuild
> > >> >       pybuild --configure --test-nose -i python{version} -p 2.7
> > >> > I: pybuild base:184: python2.7 setup.py config
> > >> > running config
> > >> >       pybuild --configure --test-nose -i python{version} -p 3.5
> > >> > I: pybuild base:184: python3.5 setup.py config
> > >> > running config
> > >> >    debian/rules override_dh_auto_build
> > >> > make[1]: Entering directory '/<<BUILDDIR>>/python-biom-for
> > >> mat-2.1.5+dfsg'
> > >> >
> > >> > [... snipped ...]
> > >> >
> > >> > ImportError: No module named _filter
> > >> > /<<BUILDDIR>>/python-biom-format-2.1.5+dfsg/doc/
> documentation/table_objects.rst:1:
> > >> WARNING: autodoc: failed to import module u'biom.table'; the following
> > >> exception was raised:
> > >> > Traceback (most recent call last):
> > >> >   File "/usr/lib/python2.7/dist-packages/sphinx/ext/autodoc.py",
> line
> > >> 526, in import_object
> > >> >     __import__(self.modname)
> > >> >   File "/<<BUILDDIR>>/python-biom-format-2.1.5+dfsg/biom/__init_
> _.py",
> > >> line 53, in <module>
> > >> >     from .table import Table
> > >> >   File "/<<BUILDDIR>>/python-biom-format-2.1.5+dfsg/biom/table.py",
> > >> line 195, in <module>
> > >> >     from ._filter import _filter
> > >> > ImportError: No module named _filter
> > >> > looking for now-outdated files... none found
> > >> > pickling environment... done
> > >> > checking consistency... done
> > >> > writing... biom.1 { documentation/index documentation/biom_format
> > >> documentation/format_versions/biom-1.0 documentation/format_versions/
> biom-2.0
> > >> documentation/format_versions/biom-2.1 documentation/quick_usage_
> examples
> > >> documentation/table_objects documentation/biom_conversion
> > >> documentation/adding_metadata documentation/summarizing_biom_tables
> > >> BIOM_LICENSE }
> > >> > build succeeded, 3 warnings.
> > >> > make[1]: Leaving directory '/<<BUILDDIR>>/python-biom-for
> > >> mat-2.1.5+dfsg'
> > >> >    dh_auto_test -i -O--buildsystem=pybuild
> > >> >       pybuild --test --test-nose -i python{version} -p 2.7
> > >> > I: pybuild pybuild:212: cp -r /<<BUILDDIR>>/python-biom-
> format-2.1.5+dfsg/examples
> > >> /<<BUILDDIR>>/python-biom-format-2.1.5+dfsg/tests
> > >> /<<BUILDDIR>>/python-biom-format-2.1.5+dfsg/.pybuild/
> pythonX.Y_2.7/build
> > >> > I: pybuild base:184: cd /<<BUILDDIR>>/python-biom-form
> > >> at-2.1.5+dfsg/.pybuild/pythonX.Y_2.7/build; python2.7 -m nose tests
> > >> > ............................................................
> > >> ............................................................
> > >> ............................................................
> > >> ............................................................
> > >> ...........S.............
> > >> > ------------------------------------------------------------
> ----------
> > >> > Ran 265 tests in 3.736s
> > >> >
> > >> > OK (SKIP=1)
> > >> >       pybuild --test --test-nose -i python{version} -p 3.5
> > >> > I: pybuild pybuild:212: cp -r /<<BUILDDIR>>/python-biom-
> format-2.1.5+dfsg/examples
> > >> /<<BUILDDIR>>/python-biom-format-2.1.5+dfsg/tests
> > >> /<<BUILDDIR>>/python-biom-format-2.1.5+dfsg/.pybuild/
> pythonX.Y_3.5/build
> > >> > I: pybuild base:184: cd /<<BUILDDIR>>/python-biom-form
> > >> at-2.1.5+dfsg/.pybuild/pythonX.Y_3.5/build; python3.5 -m nose tests
> > >> > ............................................................
> > >> ............................................................
> > >> ............................................................
> > >> ............................................................
> > >> ...........S......F......
> > >> > ============================================================
> ==========
> > >> > FAIL: test_generate_subsamples (tests.test_util.UtilTests)
> > >> > ------------------------------------------------------------
> ----------
> > >> > Traceback (most recent call last):
> > >> >   File "/<<BUILDDIR>>/python-biom-format-2.1.5+dfsg/.pybuild/python
> > >> X.Y_3.5/build/tests/test_util.py", line 58, in
> test_generate_subsamples
> > >> >     (3, 1, 1)})
> > >> > AssertionError: Items in the second set but not the first:
> > >> > (3, 0, 0)
> > >> >
> > >> > ------------------------------------------------------------
> ----------
> > >> > Ran 265 tests in 3.575s
> > >> >
> > >> > FAILED (SKIP=1, failures=1)
> > >> > E: pybuild pybuild:276: test: plugin distutils failed with: exit
> > >> code=1: cd /<<BUILDDIR>>/python-biom-format-2.1.5+dfsg/.pybuild/
> pythonX.Y_3.5/build;
> > >> python3.5 -m nose tests
> > >> > dh_auto_test: pybuild --test --test-nose -i python{version} -p 3.5
> > >> returned exit code 13
> > >> > debian/rules:18: recipe for target 'build-indep' failed
> > >> > make: *** [build-indep] Error 25
> > >> > dpkg-buildpackage: error: debian/rules build-indep gave error exit
> > >> status 2
> > >> > ------------------------------------------------------------
> > >> --------------------
> > >> >
> > >> > The failure happens randomly on the single-CPU KVM/QEMU virtual
> machines
> > >> > I use as autobuilders. The approximate failure rate is about 1%.
> > >> >
> > >> > I attach several build logs (made on different autobuilders).
> > >> >
> > >> > If you could not reproduce this using sbuild on a single CPU virtual
> > >> > machine (as I did), of course trying a lot of times (because of the
> > >> > failure rate above), please do not downgrade or mark as
> unreproducible,
> > >> > I would then consider giving you access to a virtual machine on
> which
> > >> > I can reproduce it so that you can as well. (In such case, please
> > >> > contact me off-list for details).
> > >> >
> > >> > Thanks.
> > >>
> > >>
> > >>
> > >>
> > >>
> > >> > _______________________________________________
> > >> > Debian-med-packaging mailing list
> > >> > Debian-med-packaging at lists.alioth.debian.org
> > >> > http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/debi
> > >> an-med-packaging
> > >>
> > >>
> > >> --
> > >> http://fam-tille.de
> > >>
> > >
> > >
>
> --
> http://fam-tille.de
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.alioth.debian.org/pipermail/debian-med-packaging/attachments/20161219/55aa1c3c/attachment-0001.html>


More information about the Debian-med-packaging mailing list