The build of python-cassandra-driver fails on s390x (a big endian platform) with ... running install_scripts =============================================================================== The optional C extensions are not supported on big-endian systems. =============================================================================== + mkdir -p /builddir/build/BUILDROOT/python-cassandra-driver-3.7.0-1.fc26.s390x/usr/lib64/python2.7/site-packages + mv '/builddir/build/BUILDROOT/python-cassandra-driver-3.7.0-1.fc26.s390x/usr/lib/python2.7/site-packages/*' /builddir/build/BUILDROOT/python-cassandra-driver-3.7.0-1.fc26.s390x/usr/lib64/python2.7/site-packages mv: cannot stat '/builddir/build/BUILDROOT/python-cassandra-driver-3.7.0-1.fc26.s390x/usr/lib/python2.7/site-packages/*': No such file or directory RPM build errors: ... for full logs please see http://s390.koji.fedoraproject.org/koji/taskinfo?taskID=2359154 The fix-up part in the %install section (http://pkgs.fedoraproject.org/cgit/rpms/python-cassandra-driver.git/tree/python-cassandra-driver.spec#n85) doesn't seem to be correct. Version-Release number of selected component (if applicable): python-cassandra-driver-3.7.0-1.fc26
Hello. Thank you for fill the bug. I investigated the problem and I found same behavior on ppc64 platform. The problem comes from workaround in specfile in the %install section which moves files from /usr/lib to /usr/lib64 on big endian platforms but it looks that this workaround is not necessary anymore because files are already installed in the correct folder. Another problem was empty debugfiles.list so I enabled this automatically created subpackage only for specified architectures. Scratch builds now looks good. Could you please check new specfile version and build logs? Then I can push this change and build new version in rawhide. SPEC: https://lbalhar.fedorapeople.org/python-cassandra-driver.spec Builds: http://koji.fedoraproject.org/koji/taskinfo?taskID=15935431 http://s390.koji.fedoraproject.org/koji/taskinfo?taskID=2364706 http://ppc.koji.fedoraproject.org/koji/taskinfo?taskID=3777298 Thank you one more time. Have a nice day. Lumir
Thanks, it builds, then ship it :-) The logs look good, except the failing test, which I suspect is related to a "hard-coded" little endian value. It might be worth to fix it too.
Hello. New version 3.7.0-2 is in rawhide and updates are prepared. I'll leave this bug opened and I'll try to fix mentioned test for big endian systems. Bodhi updates: https://bodhi.fedoraproject.org/updates/FEDORA-2016-d6d0a70bbb https://bodhi.fedoraproject.org/updates/FEDORA-2016-a2fe74417c Have a nice day. Lumir
python-cassandra-driver-3.7.0-2.fc25 has been pushed to the Fedora 25 testing repository. If problems still persist, please make note of it in this bug report. See https://fedoraproject.org/wiki/QA:Updates_Testing for instructions on how to install test updates. You can provide feedback for this update here: https://bodhi.fedoraproject.org/updates/FEDORA-2016-a2fe74417c
python-cassandra-driver-3.7.0-2.fc24 has been pushed to the Fedora 24 testing repository. If problems still persist, please make note of it in this bug report. See https://fedoraproject.org/wiki/QA:Updates_Testing for instructions on how to install test updates. You can provide feedback for this update here: https://bodhi.fedoraproject.org/updates/FEDORA-2016-d6d0a70bbb
python-cassandra-driver-3.7.0-2.fc25 has been pushed to the Fedora 25 stable repository. If problems still persist, please make note of it in this bug report.
Hello. Because I cannot solve this issue by myself, I've created bug report in upstream JIRA portal: https://datastax-oss.atlassian.net/browse/PYTHON-653 Have a nice day. Lumír
python-cassandra-driver-3.7.0-2.fc24 has been pushed to the Fedora 24 stable repository. If problems still persist, please make note of it in this bug report.
I have a response from upstream developers: > I was wondering if by any chance you or one of your colleagues could submit a > fix for this as I'm not sure how easy it will be for us to get access to such > architectures. Thanks But I am not sure where the problem comes from. If is it the problem only in test (hardcoded results from the hash function), we can write separated test for BE platform. But if is it problem in hash function itself, it would be harder to fix it. What do you think? Thanks for help and have a nice day.
This bug appears to have been reported against 'rawhide' during the Fedora 26 development cycle. Changing version to '26'.
I moved this slightly forward but unfortunately, I still cannot enable tests on all platforms. I fixed the issue in tests of Murmur algorithm [0] but other bugs appeared. I am working on it in cooperation with upstream developers. Because there is the open issue in Jira portal and build/install issue was fixed, I think that we can close this bug. [0] https://github.com/datastax/python-driver/pull/762 Have a nice day. Lumír