[buildd-tools-devel] running mpi in sbuild (petsc testing)

[Johannes Ring]

I guess this is the same problem I have when building DOLFIN in pbuilder,
which I solve by adding "export OMPI_MCA_plm_rsh_agent=/bin/false" in
debian/rules.

Johannes

On Fri, Oct 9, 2015 at 5:24 AM Drew Parsons <dparsons at debian.org> wrote:

> The new packages for PETSc 3.6.2 are all but ready,
> http://anonscm.debian.org/cgit/debian-science/packages/petsc.git/.
>
> The build provides some simple testing.  The tests look successful when
> building at the command line.  But for building packages I prefer to
> use sbuild.
>
> In the sbuild environment the tests fail.  I gather it means that MPI
> (openMPI) cannot run in sbuild, or that I need to change something to
> let it run.
>
> Does anyone have experience running mpi tests in sbuild?
>
> Drew.
>
> p.s. an example of the test failure is:
>
> mpicxx -o ex19.o -c -g -O2 -fstack-protector-strong -Wformat
> -Werror=format-security   -fPIC  -D_FORTIFY_SOURCE=2  -I/<<PKGBUILDDIR>>/include
> -I/<<PKGBUILDDIR>>/x86_64-linux-gnu-debug/include
> -I/usr/include/suitesparse -I/usr/include -I/usr/include/superlu
> -I/usr/include/scotch -I/usr/lib/openmpi/include
> -I/usr/lib/openmpi/include/openmpi  -D_FORTIFY_SOURCE=2  `pwd`/ex19.c
> mpicxx -Wl,-z,relro -Wl,--no-as-needed -g -O2 -fstack-protector-strong
> -Wformat -Werror=format-security   -o ex19  ex19.o
> -L/<<PKGBUILDDIR>>/x86_64-linux-gnu-debug/lib
> -L/<<PKGBUILDDIR>>/x86_64-linux-gnu-debug/lib  -lpetsc_real -ldmumps
> -lzmumps -lsmumps -lcmumps -lmumps_common -lpord -lumfpack -lamd -lcholmod
> -lklu -L/usr/lib -lHYPRE_utilities -lHYPRE_struct_mv -lHYPRE_struct_ls
> -lHYPRE_sstruct_mv -lHYPRE_sstruct_ls -lHYPRE_IJ_mv -lHYPRE_parcsr_ls
> -lsuperlu -lscalapack-openmpi -llapack -lblas -lssl -lcrypto -lptesmumps
> -lptscotch -lptscotcherr -lfftw3 -lfftw3_mpi -lm -L/usr/lib/openmpi/lib
> -L/usr/lib/gcc/x86_64-linux-gnu/5 -L/usr/lib/x86_64-linux-gnu
> -L/lib/x86_64-linux-gnu -lmpi_f90 -lmpi_f77 -lgfortran -lm -lgfortran -lm
> -lquadmath -lm -lmpi_cxx -lstdc++ -L/usr/lib/openmpi/lib
> -L/usr/lib/gcc/x86_64-linux-gnu/5 -L/usr/lib/x86_64-linux-gnu
> -L/lib/x86_64-linux-gnu -L/usr/lib/x86_64-linux-gnu -ldl -lmpi -lhwloc
> -lgcc_s -lpthread -ldl
> /bin/rm -f ex19.o
> Possible error running C/C++ src/snes/examples/tutorials/ex19 with 1 MPI
> process
> See http://www.mcs.anl.gov/petsc/documentation/faq.html
> [schumann:13424] [[INVALID],INVALID] ORTE_ERROR_LOG: Not found in file
> ess_hnp_module.c at line 170
> --------------------------------------------------------------------------
> It looks like orte_init failed for some reason; your parallel process is
> likely to abort.  There are many reasons that a parallel process can
> fail during orte_init; some of which are due to configuration or
> environment problems.  This failure appears to be an internal failure;
> here's some additional information (which may only be relevant to an
> Open MPI developer):
>
>   orte_plm_base_select failed
>   --> Returned value Not found (-13) instead of ORTE_SUCCESS
> --------------------------------------------------------------------------
> [schumann:13424] [[INVALID],INVALID] ORTE_ERROR_LOG: Not found in file
> runtime/orte_init.c at line 128
> --------------------------------------------------------------------------
> It looks like orte_init failed for some reason; your parallel process is
> likely to abort.  There are many reasons that a parallel process can
> fail during orte_init; some of which are due to configuration or
> environment problems.  This failure appears to be an internal failure;
> here's some additional information (which may only be relevant to an
> Open MPI developer):
>
>   orte_ess_set_name failed
>   --> Returned value Not found (-13) instead of ORTE_SUCCESS
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.alioth.debian.org/pipermail/buildd-tools-devel/attachments/20151009/5832ae57/attachment.html>