Bastien, How can you uninstall MPI in a safe way? If I uninstall MPI, some related ( *important *) packages will be broken.
Best, Yaakov On Monday, April 4, 2016 at 6:19:34 PM UTC+2, Bastien Lauras wrote: > > Bruno, > > The first time I ran mpirun --version, it said : > > mpirun (Open MPI) 1.6.5 > > Report bugs to http://www.open-mpi.org/community/help/ > > > So I uninstalled openmpi and mpich, and then re-installed mpich. Now, > mpirun --version gives me : > > HYDRA build details: > Version: 3.0.4 > Release Date: Wed Apr 24 10:08:10 CDT 2013 > CC: cc -D_FORTIFY_SOURCE=2 -g -O2 > -fstack-protector --param=ssp-buffer-size=4 -Wformat > -Werror=format-security -Wl,-Bsymbolic-functions -Wl,-z,relro > CXX: c++ -D_FORTIFY_SOURCE=2 -g -O2 > -fstack-protector --param=ssp-buffer-size=4 -Wformat > -Werror=format-security -Wl,-Bsymbolic-functions -Wl,-z,relro > F77: gfortran -g -O2 > -Wl,-Bsymbolic-functions -Wl,-z,relro > F90: gfortran -Wl,-Bsymbolic-functions > -Wl,-z,relro > Configure options: '--disable-option-checking' > '--prefix=/usr' '--build=x86_64-linux-gnu' '--includedir=${prefix}/include' > '--mandir=${prefix}/share/man' '--infodir=${prefix}/share/info' > '--sysconfdir=/etc' '--localstatedir=/var' > '--libdir=${prefix}/lib/x86_64-linux-gnu' > '--libexecdir=${prefix}/lib/x86_64-linux-gnu' '--disable-maintainer-mode' > '--disable-dependency-tracking' '--enable-shared' '--enable-fc' > '--disable-rpath' '--disable-wrapper-rpath' '--sysconfdir=/etc/mpich' > '--libdir=/usr/lib/x86_64-linux-gnu' '--includedir=/usr/include/mpich' > '--docdir=/usr/share/doc/mpich' '--with-hwloc-prefix=system' > '--enable-checkpointing' '--with-hydra-ckpointlib=blcr' > 'build_alias=x86_64-linux-gnu' 'MPICHLIB_CFLAGS=-g -O2 -fstack-protector > --param=ssp-buffer-size=4 -Wformat -Werror=format-security' > 'MPICHLIB_CXXFLAGS=-g -O2 -fstack-protector --param=ssp-buffer-size=4 > -Wformat -Werror=format-security' 'MPICHLIB_FFLAGS=-g -O2' > 'MPICHLIB_FCFLAGS=-g -O2' 'CFLAGS=-g -O2 -fstack-protector > --param=ssp-buffer-size=4 -Wformat -Werror=format-security -g -O2 > -fstack-protector --param=ssp-buffer-size=4 -Wformat > -Werror=format-security -O2' 'LDFLAGS=-Wl,-Bsymbolic-functions -Wl,-z,relro > ' 'CPPFLAGS=-D_FORTIFY_SOURCE=2 -I/build/buildd/mpich-3.0.4/src/mpl/include > -I/build/buildd/mpich-3.0.4/src/mpl/include > -I/build/buildd/mpich-3.0.4/src/openpa/src > -I/build/buildd/mpich-3.0.4/src/openpa/src > -I/build/buildd/mpich-3.0.4/src/mpi/romio/include' 'CXXFLAGS=-g -O2 > -fstack-protector --param=ssp-buffer-size=4 -Wformat > -Werror=format-security -g -O2 -fstack-protector --param=ssp-buffer-size=4 > -Wformat -Werror=format-security -O2' 'F77=gfortran' 'FFLAGS=-g -O2 -g -O2 > -O2' 'FC=gfortran' '--cache-file=/dev/null' '--srcdir=.' 'CC=cc' 'LIBS=-lrt > -lcr -lpthread ' > Process Manager: pmi > Launchers available: ssh rsh fork slurm ll lsf sge > manual persist > Topology libraries available: hwloc > Resource management kernels available: user slurm ll lsf sge pbs > cobalt > Checkpointing libraries available: blcr > Demux engines available: poll select > > > I tried to run the step-17 tutorial, I had the same output than before > (like : Number of degrees of freedom: 570 (by partition: 570) ) > Should I recompile Deal.II now? > > Thanks for your help. > > Bastien > > > On Monday, April 4, 2016 at 10:31:16 AM UTC-5, Bruno Turcksin wrote: >> >> Bastien, >> >> Can you try mpirun --version it should say MPICH However, I strongly >> advise you to have either openmpi or mpich installed but not both. Having >> both installed will lead to hard to understand bug. >> >> Best, >> >> Bruno >> >> On Monday, April 4, 2016 at 11:23:12 AM UTC-4, Bastien Lauras wrote: >>> >>> And here is what I have in my *bin* folder : >>> >>> >>> <https://lh3.googleusercontent.com/-Ow-PmTNx3IM/VwKG22nnZ-I/AAAAAAAAAAQ/b3d8YRndBMkEYzxnCC6DMUvQuN4rpzh1g/s1600/bin_001.png> >>> >>> >>> >>> On Monday, April 4, 2016 at 9:51:19 AM UTC-5, Bastien Lauras wrote: >>>> >>>> Hi, Thanks for answering! >>>> >>>> Here are the outputs : >>>> >>>> bastien@PC-Bastien:~$ which mpirun >>>> /usr/bin/mpirun >>>> bastien@PC-Bastien:~$ which mpic++ >>>> /usr/bin/mpic++ >>>> >>>> >>>> bastien@PC-Bastien:~/build$ grep MPI detailed.log >>>> # CMAKE_CXX_COMPILER: GNU 4.8.4 on platform Linux x86_64 >>>> # CMAKE_C_COMPILER: /usr/bin/cc >>>> # CMAKE_Fortran_COMPILER: /usr/bin/gfortran >>>> # DEAL_II_WITH_MPI set up with external dependencies >>>> # MPI_VERSION = 3.0 >>>> # MPI_C_COMPILER = /usr/bin/mpicc >>>> # MPI_CXX_COMPILER = /usr/bin/mpicxx >>>> # MPI_Fortran_COMPILER = /usr/bin/mpif90 >>>> # MPI_CXX_FLAGS = -fstack-protector >>>> # MPI_LINKER_FLAGS = -Wl,-Bsymbolic-functions -Wl,-z,relro >>>> # MPI_INCLUDE_DIRS = /usr/include/mpich >>>> # MPI_USER_INCLUDE_DIRS = /usr/include/mpich >>>> # MPI_LIBRARIES = >>>> /usr/lib/x86_64-linux-gnu/libmpichcxx.so;/usr/lib/x86_64-linux-gnu/libmpichf90.so;/usr/lib/x86_64-linux-gnu/libmpich.so;/usr/lib/x86_64-linux-gnu/libopa.so;/usr/lib/x86_64-linux-gnu/libmpl.so;rt;/usr/lib/libcr.so;pthread >>>> >>>> >>>> I probably have more than one MPI library installed. I had quite a hard >>>> time to install it. >>>> Should I uninstall everything (including Deal.II) and install >>>> everything again? >>>> >>>> Many thanks, >>>> Bastien >>>> >>>> On Saturday, April 2, 2016 at 5:25:42 AM UTC-5, Timo Heister wrote: >>>>> >>>>> Do you have more than one MPI library installed? Can you post the >>>>> output of "which mpirun" and "which mpic++" (or whatever your compiler >>>>> is called)? Also give us the output of "grep MPI detailed.log" in your >>>>> deal.II build directory. >>>>> >>>>> On Fri, Apr 1, 2016 at 11:40 PM, Bastien Lauras <laur...@umn.edu> >>>>> wrote: >>>>> > Moreover, if I display the number of threads I seem to have by >>>>> adding this >>>>> > line in my run method : >>>>> > >>>>> > void TopLevel<dim>::run () >>>>> > { >>>>> > pcout << "Number of threads : " << MultithreadInfo::n_threads() >>>>> << >>>>> > std::endl; >>>>> > >>>>> > >>>>> > I have the following answer (times the number of thrads I wanted to >>>>> run the >>>>> > program on: >>>>> > >>>>> > Number of threads : 1 >>>>> > >>>>> > >>>>> > So, why is this value false? It is the same thing if I display the >>>>> value of >>>>> > n_mpi_processes. >>>>> > >>>>> > Many thanks for your answer beforehand! >>>>> > >>>>> > On Friday, April 1, 2016 at 3:34:57 PM UTC-5, Bastien Lauras wrote: >>>>> >> >>>>> >> Hi! >>>>> >> >>>>> >> I had quite a few problems to run step-17 and step-18 of the >>>>> tutorial. I >>>>> >> think I have installed everything needed (BLAS, Lapack, MPI, petsc, >>>>> metis). >>>>> >> Though, when I run the "make test" commmand after the "make >>>>> install" and >>>>> >> "cmake ..." in build directory of Deal.II, I get the following >>>>> error while >>>>> >> running test 4 (mpi.debug) : >>>>> >> >>>>> >> [100%] Built target mpi.debug >>>>> >> mpi.debug: RUN failed. Output: >>>>> >> Hi from 0/1 >>>>> >> Hi from 0/1 >>>>> >> ERROR: process does not see nproc=2! >>>>> >> ERROR: process does not see nproc=2! >>>>> >> >>>>> -------------------------------------------------------------------------- >>>>> >>>>> >> mpiexec noticed that the job aborted, but has no info as to the >>>>> process >>>>> >> that caused that situation. >>>>> >> >>>>> -------------------------------------------------------------------------- >>>>> >>>>> >> >>>>> >> >>>>> >> mpi.debug: ****** RUN failed ******* >>>>> >> >>>>> >> =============================== OUTPUT END >>>>> >> =============================== >>>>> >> Expected stage PASSED - aborting >>>>> >> CMake Error at >>>>> /home/bastien/dealii-8.4.0/cmake/scripts/run_test.cmake:140 >>>>> >> (MESSAGE): >>>>> >> *** abort >>>>> >> >>>>> >> >>>>> >> Moreover, when I try to run step-17, the "cmake ." command runs >>>>> without >>>>> >> problem, as the "make", but when I type "mpirun -np 4 ./step-17", >>>>> the code >>>>> >> seems to be executed 4 times on 4 processes. I mean, every process >>>>> works on >>>>> >> every degree of freedom and so on. In fact, I get the following >>>>> output : >>>>> >> >>>>> >> mpirun -np 4 ./step-17 >>>>> >> Cycle 0: >>>>> >> Cycle 0: >>>>> >> Number of active cells: 64 >>>>> >> Number of active cells: 64 >>>>> >> Cycle 0: >>>>> >> Number of degrees of freedom: 162 (by partition: Number of >>>>> degrees of >>>>> >> freedom: 162 (by partition: 162) >>>>> >> 162) >>>>> >> Number of active cells: 64 >>>>> >> Cycle 0: >>>>> >> Number of degrees of freedom: 162 (by partition: 162) >>>>> >> Number of active cells: 64 >>>>> >> Number of degrees of freedom: 162 (by partition: 162) >>>>> >> Solver converged in 9 iterations. >>>>> >> Solver converged in 9 iterations. >>>>> >> Cycle 1: >>>>> >> Cycle 1: >>>>> >> Solver converged in 9 iterations. >>>>> >> Number of active cells: 124 >>>>> >> >>>>> >> [.......] >>>>> >> >>>>> >> >>>>> >> It's not the one I should have, it seems that every process execute >>>>> the >>>>> >> code, and n_mpi_processes = 1, and this_mpi_process is always >>>>> equal to 0. >>>>> >> What can be the problem? Are both issues related? >>>>> >> >>>>> >> Many thanks for your help. >>>>> >> >>>>> >> Bastien >>>>> > >>>>> > -- >>>>> > The deal.II project is located at http://www.dealii.org/ >>>>> > For mailing list/forum options, see >>>>> > https://groups.google.com/d/forum/dealii?hl=en >>>>> > --- >>>>> > You received this message because you are subscribed to the Google >>>>> Groups >>>>> > "deal.II User Group" group. >>>>> > To unsubscribe from this group and stop receiving emails from it, >>>>> send an >>>>> > email to dealii+un...@googlegroups.com. >>>>> > For more options, visit https://groups.google.com/d/optout. >>>>> >>>>> >>>>> >>>>> -- >>>>> Timo Heister >>>>> http://www.math.clemson.edu/~heister/ >>>>> >>>> -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to dealii+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
