rpmbuild - empty RPMS directory

[arnaud gaboury]

On Wed, Sep 16, 2015 at 11:28 AM, Michael Schwendt <mschwendt at gmail.com> wrote:
> On Tue, 15 Sep 2015 18:09:44 +0100, Jonathan Wakely wrote:
>
>> On 15/09/15 13:58 +0200, arnaud gaboury wrote:
>> >I have been working on my first rpmbuild since quite a few days now.
>>
>> As a more gneral comment, why not try something *much* simpler for
>> your first rpmbuild?
>>
>> Try something that doesn't use an alternative compiler, doesn't
>> require lots of environment variables, and doesn't use a new spec file
>> that has never been successfully tested.
>>
>> Maybe just take a working spec file for another project and run
>> rpmbuild, and check the results. That would reveal if you're looking
>> in the wrong place for the output, and you wouldn't waste time looking
>> for problems in the spec file if you know the spec file is already
>> correct.
>
> Also try --short-circuit builds for the individual %build and %install
> sections. "rpmbuild -bi --short-circuit specfile" is extremely helpful
> for debugging problems in %install and %files sections as it skips the
> earlier successful %build stage.

Building rpm step by step gave me the solution:
$ rpmbuild -v -bc SPECS/R.spec
/var/tmp/rpm-tmp.nTfP8D: line 58: export: `-ip': not a valid identifier
/var/tmp/rpm-tmp.nTfP8D: line 58: export: `-ipo': not a valid identifier
/var/tmp/rpm-tmp.nTfP8D: line 58: export: `-opt-mem-layout-trans=3':
not a valid identifier
/var/tmp/rpm-tmp.nTfP8D: line 58: export: `-xHost': not a valid identifier
/var/tmp/rpm-tmp.nTfP8D: line 58: export: `-mavx -fp-model precise
-DMKL_ILP64 -qopenmp-parallel
-I/opt/intel/compilers_and_libraries_2016.0.109/linux/mkl/include':
not a valid identifier

My build flags were not correctly set.
Then I kept on secion by section, with no errors.

$ rpmbuild -v -ba SPECS/R.spec
<----------BUILD---------------->
Wrote: /home/poisonivy/rpmbuild/SRPMS/R-3.2.1-mkl.fc22.src.rpm
Wrote: /home/poisonivy/rpmbuild/RPMS/x86_64/R-3.2.1-mkl.fc22.x86_64.rpm
Wrote: /home/poisonivy/rpmbuild/RPMS/x86_64/R-core-3.2.1-mkl.fc22.x86_64.rpm
Wrote: /home/poisonivy/rpmbuild/RPMS/x86_64/R-core-devel-3.2.1-mkl.fc22.x86_64.rpm
Wrote: /home/poisonivy/rpmbuild/RPMS/x86_64/R-devel-3.2.1-mkl.fc22.x86_64.rpm
Wrote: /home/poisonivy/rpmbuild/RPMS/x86_64/R-java-3.2.1-mkl.fc22.x86_64.rpm
Wrote: /home/poisonivy/rpmbuild/RPMS/x86_64/R-java-devel-3.2.1-mkl.fc22.x86_64.rpm
Wrote: /home/poisonivy/rpmbuild/RPMS/x86_64/libRmath-3.2.1-mkl.fc22.x86_64.rpm
Wrote: /home/poisonivy/rpmbuild/RPMS/x86_64/libRmath-devel-3.2.1-mkl.fc22.x86_64.rpm
Wrote: /home/poisonivy/rpmbuild/RPMS/x86_64/libRmath-static-3.2.1-mkl.fc22.x86_64.rpm
Wrote: /home/poisonivy/rpmbuild/RPMS/x86_64/R-debuginfo-3.2.1-mkl.fc22.x86_64.rpm
Executing(%clean): /bin/sh -e /var/tmp/rpm-tmp.f7bTCJ
+ umask 022
+ cd /home/poisonivy/rpmbuild/BUILD
+ cd R-3.2.1
+ /usr/bin/rm -rf /home/poisonivy/rpmbuild/BUILDROOT/R-3.2.1-mkl.fc22.x86_64
+ exit 0
---------------------------------------------------------------

Thank you for your interest in my issue and your patience. And please
apologize for the "tons" of emails in an open source community mailing
list ;-)
Result is here.

>
> --
> devel mailing list
> devel at lists.fedoraproject.org
> https://admin.fedoraproject.org/mailman/listinfo/devel
> Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct



-- 

google.com/+arnaudgabourygabx