Hi Spencer, Chris, Jean, I’ve fixed the problem in my case just by doing CC=mpicc CXX=mpic++ cmake \ -DTHIRDPARTY_BUILD_BOOST=OFF \ -DBoost_USE_MULTITHREADED=OFF \ -DNEKTAR_USE_FFTW=ON \ -DNEKTAR_USE_MKL=ON \ -DNEKTAR_USE_MPI=ON \ -DNEKTAR_USE_SYSTEM_BLAS_LAPACK=OFF \ -DTHIRDPARTY_BUILD_FFTW=OFF .. Thanks again, Cheers, Yan On 30 Nov 2014, at 15:36, Chris Cantwell <c.cantwell@imperial.ac.uk<mailto:c.cantwell@imperial.ac.uk>> wrote: Hi Yan, There is now a boost module on cx2 (version 1.57). Since it looks like you have linked against 1.55 (maybe in your home directory?), you would need to rerun make to link against the correct version of the libraries. Cheers, Chris On 30/11/14 13:43, Bao, Yan wrote: Hi Chris, Many thanks for your help! At this moment, I’ve found libfftw3.so.3 in the library, however, I did not find boost in them, please see below: ybao@cx2:~/CX2/nektar++/builds> ldd dist/bin/IncNavierStokesSolver | grep fftw libfftw3.so.3 => /apps/fftw/3.3.3/lib/libfftw3.so.3 (0x00007fffeb32a000) ybao@cx2:~/CX2/nektar++/builds> ldd dist/bin/IncNavierStokesSolver | grep boost libboost_thread.so.1.55.0 => not found libboost_iostreams.so.1.55.0 => not found libboost_date_time.so.1.55.0 => not found libboost_program_options.so.1.55.0 => not found libboost_filesystem.so.1.55.0 => not found libboost_system.so.1.55.0 => not found libboost_thread.so.1.55.0 => not found libboost_iostreams.so.1.55.0 => not found libboost_date_time.so.1.55.0 => not found libboost_program_options.so.1.55.0 => not found libboost_filesystem.so.1.55.0 => not found libboost_system.so.1.55.0 => not found libboost_thread.so.1.55.0 => not found And, when I’m running the case, I got the following errors: MPI: r3i2n0: 0x447c000054609b62: /home/ybao/CX2/nektar++/builds/dist/bin/IncNavierStokesSolver: error while loading shared libraries: libboost_thread.so.1.55.0:MPI: r3i2n0: 0x447c000054609b62: cannot open shared object file: No such file or directory How can I fix them? Thanks again! Regards, Yan On 30 Nov 2014, at 11:24, Chris Cantwell <c.cantwell@imperial.ac.uk<mailto:c.cantwell@imperial.ac.uk> <mailto:c.cantwell@imperial.ac.uk>> wrote: Hi Yan, It looks like there is no longer an fftw/3.3.2-double module on cx2. Try loading the fftw/3.3.3-double module, recompiling and relinking Nektar++ and trying again. You can test if it is finding the library by running on the login node: ldd dist/bin/IncNavierStokesSolver | grep fftw from your build directory and see if you see a line like: libfftw3.so.3 => /apps/fftw/3.3.3/lib/libfftw3.so.3 Cheers, Chris On 29/11/14 23:46, Bao, Yan wrote: Hi Chris, Thanks for your email. Now, the code has been compiled successfully, However, when I qsub my job on cx2, I get the following message: MPI: r3i3n0: 0x27c6000054417657: /home/ybao/CX2/nektar++/builds/dist/bin/IncNavierStokesSolver: error while loading shared libraries: libfftw3.so.3: cannot openMPI: r3i3n0: 0x27c6000054417657: shared object file: No such file or directory MPI: could not run executable (case #4) In fact, I’ve loadedfftw/3.3.2-double, when running my case. Could you please help me to fix this problem? many thanks! Regards, Yan On 29 Nov 2014, at 21:49, Chris Cantwell <c.cantwell@imperial.ac.uk<mailto:c.cantwell@imperial.ac.uk> <mailto:c.cantwell@imperial.ac.uk> <mailto:c.cantwell@imperial.ac.uk>> wrote: Hi Yan, It sounds like you might have exceeded your disk quota when previously compiling the code, leaving a truncated file. If you are well within your quota, deleting the ExtractMeanModeFromHomo1DFld.cpp.o file should force it to recompile it and resolve the issue. Cheers, Chris On 29/11/14 21:03, Bao, Yan wrote: Dear all, When I’m trying to compile branch “MovingBodies” of nektar++ on cx2, I got some errors as follows: CMakeFiles/ExtractMeanModeFromHomo1DFld.dir/ExtractMeanModeFromHomo1DFld.cpp.o: file not recognized: File truncated make[2]: *** [utilities/PostProcessing/ExtractMeanModeFromHomo1DFld-3.4.0] Error 1 make[1]: *** [utilities/PostProcessing/CMakeFiles/ExtractMeanModeFromHomo1DFld.dir/all] Error 2 make[1]: *** Waiting for unfinished jobs…. However, the compiling process of the “master” branch is very smoothly on cx2, and I did not get any such errors when compiling my branch on victoria/euston nodes. How can I fix it? many thanks. Cheers, Yan On 27 Nov 2014, at 13:57, Kamil Ozden <kamil.ozden.me@gmail.com<mailto:kamil.ozden.me@gmail.com> <mailto:kamil.ozden.me@gmail.com> <mailto:kamil.ozden.me@gmail.com> <mailto:kamil.ozden.me@gmail.com>> wrote: Dear Dr. Moxey, I got in contact with the system administrator. He told that it is impossible to recompile blas and lapack on the system with -fPIC option. Alternatively, he copied the Blas and Lapack libraries from the directory in the system to another folder in my home directory and recompiled them there with -fPIC option. However, when I tried to reinstall Nektar by showing the path of new Blas and Lapack libraries in my home directory I got the same error. Is there any other alternative way to overcome this problem? Regards, Kamil On 27-11-2014 13:10, David Moxey wrote: Hi Kamil, The issue is related to linking. We generate shared libraries in Nektar++, but here we are trying to link the shared library with a static library. You can only do this if the static library was compiled with the -fPIC option, which generates position-independent code that shared libraries need in order to work at runtime. If you have a shared library, you should use this. Otherwise, you should recompile BLAS/Lapack installation with the -fPIC option. This may be something that your cluster system administrators can help with. Thanks, Dave On 27 Nov 2014, at 10:37, Kamil ÖZDEN <kamil.ozden.me@gmail.com<mailto:kamil.ozden.me@gmail.com> <mailto:kamil.ozden.me@gmail.com> <mailto:kamil.ozden.me@gmail.com> <mailto:kamil.ozden.me@gmail.com>> wrote: Dear All, I'm trying to install Nektar++ 4.0 to a cluster with the options NEKTAR_USE_BLAS_LAPACK ON NEKTAR_USE_SYSTEM_BLAS_LAPACK ON with the path of the libraries as follows: NATIVE_BLAS /truba/sw/centos6.4/lib/blas/netlib-gcc/blas_LINUX.a NATIVE_LAPACK /truba/sw/centos6.4/lib/lapack/netlib-3.5.0-gcc/liblapack.a However, I'm getting the following error: Linking CXX shared library libLibUtilities.so /usr/bin/ld: /truba/sw/centos6.4/lib/lapack/netlib-3.5.0-gcc/liblapack.a(dgbtrf.o): relocation R_X86_64_32 against `.rodata' can not be used when making a shared object; recompile with -fPIC /truba/sw/centos6.4/lib/lapack/netlib-3.5.0-gcc/liblapack.a: could not read symbols: Bad value collect2: ld returned 1 exit status make[2]: *** [library/LibUtilities/libLibUtilities.so.4.0.0] Error 1 make[1]: *** [library/LibUtilities/CMakeFiles/LibUtilities.dir/all] Error 2 make: *** [all] Error 2 What may be the reason for this problem and how can I solve it? Regards, Kamil _______________________________________________ Nektar-users mailing list Nektar-users@imperial.ac.uk<mailto:Nektar-users@imperial.ac.uk> <mailto:Nektar-users@imperial.ac.uk> <mailto:Nektar-users@imperial.ac.uk><mailto:Nektar-users@imperial.ac.uk> https://mailman.ic.ac.uk/mailman/listinfo/nektar-users -- David Moxey (Research Associate) d.moxey@imperial.ac.uk<mailto:d.moxey@imperial.ac.uk> <mailto:d.moxey@imperial.ac.uk> <mailto:d.moxey@imperial.ac.uk><mailto:d.moxey@imperial.ac.uk> | www.imperial.ac.uk/people/d.moxey<http://www.imperial.ac.uk/people/d.moxey> <http://www.imperial.ac.uk/people/d.moxey> <http://www.imperial.ac.uk/people/d.moxey> <http://www.imperial.ac.uk/people/d.moxey> Room 363, Department of Aeronautics, Imperial College London, London, SW7 2AZ, UK. _______________________________________________ Nektar-users mailing list Nektar-users@imperial.ac.uk<mailto:Nektar-users@imperial.ac.uk> <mailto:Nektar-users@imperial.ac.uk> <mailto:Nektar-users@imperial.ac.uk><mailto:Nektar-users@imperial.ac.uk> https://mailman.ic.ac.uk/mailman/listinfo/nektar-users -- Chris Cantwell Imperial College London South Kensington Campus London SW7 2AZ Email:c.cantwell@imperial.ac.uk<mailto:c.cantwell@imperial.ac.uk> <mailto:c.cantwell@imperial.ac.uk><mailto:c.cantwell@imperial.ac.uk> www.imperial.ac.uk/people/c.cantwell<http://www.imperial.ac.uk/people/c.cantwell> <http://www.imperial.ac.uk/people/c.cantwell> <http://www.imperial.ac.uk/people/c.cantwell> _______________________________________________ Nektar-users mailing list Nektar-users@imperial.ac.uk<mailto:Nektar-users@imperial.ac.uk> <mailto:Nektar-users@imperial.ac.uk><mailto:Nektar-users@imperial.ac.uk> https://mailman.ic.ac.uk/mailman/listinfo/nektar-users -- Chris Cantwell Imperial College London South Kensington Campus London SW7 2AZ Email:c.cantwell@imperial.ac.uk<http://imperial.ac.uk> <mailto:c.cantwell@imperial.ac.uk> www.imperial.ac.uk/people/c.cantwell<http://www.imperial.ac.uk/people/c.cantwell> <http://www.imperial.ac.uk/people/c.cantwell> _______________________________________________ Nektar-users mailing list Nektar-users@imperial.ac.uk <mailto:Nektar-users@imperial.ac.uk> https://mailman.ic.ac.uk/mailman/listinfo/nektar-users -- Chris Cantwell Imperial College London South Kensington Campus London SW7 2AZ Email: c.cantwell@imperial.ac.uk<mailto:c.cantwell@imperial.ac.uk> www.imperial.ac.uk/people/c.cantwell<http://www.imperial.ac.uk/people/c.cantwell> _______________________________________________ Nektar-users mailing list Nektar-users@imperial.ac.uk https://mailman.ic.ac.uk/mailman/listinfo/nektar-users