Tony, Im not able to run it in parallel on OF7.
So i decide to use OF7 (with CfMesh) to mesh my case and then switch to OF8 and calculate it :).
Thank you for your help and best regards Kai Am 10.01.21 um 16:49 schrieb Tony Ladd via users:
KaiThat means your case directory is mostly OK. Exactly what command did you use to run the executable. By serial mode I actually meant a single processor. For examplesimpleFoamBut then its surprising if it uses multiple cores. But it may be using multithreading by default.On a multicore node something like simpleFoam -parallel might use a default number of cores (probably all of them).For a proper parallel job you need to decompose the problem first with decomposePar. One possible source of error is the decomposeParDict file. If you don't get a proper decomposition that can be a problem. There are examples online.My typical run script would be something like decomposePar mpirun -np 4 simpleFoam -parallel 2>&1 | tee log reconstructParYou can check the decomposition with paraview (in the individual processor dirs)Tony On 1/10/21 5:29 AM, Kahnbein Kai via users wrote:[External Email] Hey Tony, it works without the -parallel flag, all four cpu's are at 100% and running fine. Best regards Kai Am 05.01.21 um 20:36 schrieb Tony Ladd via users:Just run the executable without mpirun and the -parallel flag. On 1/2/21 11:39 PM, Kahnbein Kai via users wrote:*[External Email]* Ok, sorry, what do you mean with the "serial version" ? Best regards Kai Am 31.12.20 um 16:25 schrieb tladd via users:I did not see the whole email chain before. The problem is not that it cannot find the MPI directories. I think this INIT error comes when the program cannot start for some reason. For example a missing input file. Does the serial version work. On 12/31/20 6:33 AM, Kahnbein Kai via users wrote:*[External Email]* I compared the /etc/bashrc files of both versions of OF (v7 and v8) and i dont found any difference. Here are the lines (i thought related to openmpi) of both files: OpenFOAM v7: Line 86 till 89: #- MPI implementation: # WM_MPLIB = SYSTEMOPENMPI | OPENMPI | SYSTEMMPI | MPICH | MPICH-GM | HPMPI # | MPI | FJMPI | QSMPI | SGIMPI | INTELMPI export WM_MPLIB=SYSTEMOPENMPI Line 169 till 174: # Source user setup files for optional packages # ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ _foamSource `$WM_PROJECT_DIR/bin/foamEtcFile config.sh/mpi` _foamSource `$WM_PROJECT_DIR/bin/foamEtcFile config.sh/paraview` _foamSource `$WM_PROJECT_DIR/bin/foamEtcFile config.sh/ensight` _foamSource `$WM_PROJECT_DIR/bin/foamEtcFile config.sh/gperftools` OpenFOAM v8: Line 86 till 89: #- MPI implementation: # WM_MPLIB = SYSTEMOPENMPI | OPENMPI | SYSTEMMPI | MPICH | MPICH-GM | HPMPI # | MPI | FJMPI | QSMPI | SGIMPI | INTELMPI export WM_MPLIB=SYSTEMOPENMPI Line 169 till 174: # Source user setup files for optional packages # ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ _foamSource `$WM_PROJECT_DIR/bin/foamEtcFile config.sh/mpi` _foamSource `$WM_PROJECT_DIR/bin/foamEtcFile config.sh/paraview` _foamSource `$WM_PROJECT_DIR/bin/foamEtcFile config.sh/ensight` _foamSource `$WM_PROJECT_DIR/bin/foamEtcFile config.sh/gperftools` Are you think these are the right lines ? I wish you a healthy start into the new year, Kai Am 30.12.20 um 15:25 schrieb tladd via users:Probably because OF cannot find your mpi installation. Once you set your OF environment, where is it looking for mpicc? Note the OF environment overrides your .bashrc once you source the OF bashrc. That takes its settings from the src/etc directory in the OF source code. On 12/29/20 10:23 AM, Kahnbein Kai via users wrote:[External Email]Thank you for this hint. I installed OpenFOAM v8 (the newest) on mycomputer and it works ... At the Version v7 i get still this mpi error .... I dont know why ... I wish you a healthy start into the new year :) Am 28.12.20 um 19:16 schrieb Benson Muite via users:Have you tried reinstalling OpenFOAM? If you are mostly working in a desktop, there are pre-compiled versions available:https://urldefense.proofpoint.com/v2/url?u=https-3A__openfoam.com_download_&d=DwIDaQ&c=sJ6xIWYx-zLMB3EPkvcnVg&r=kgFAU2BfgKe7cozjrP7uWDPH6xt6LAmYVlQPwQuK7ek&m=9YEwGLzNfCD1pAUuvNpqStsbpagtNfIzEt6wL6f3_7I&s=bZFAwh79J3ZL1Ut9Jt4qj-kBCubrvjsLNhq51hnAwXk&e=If you are using a pre-compiled version, do also consider reportingthe error to the packager. It seems unlikely to be an MPI error, more likely something with OpenFOAM and/or the setup. On 12/28/20 6:25 PM, Kahnbein Kai via users wrote:Good morning,im trying to fix this error by myself and i have a little update.The ompi version i use is the: Code: kai@Kai-Desktop:~/Dokumente$ mpirun --version mpirun (Open MPI) 4.0.3 If i create a *.c file, with the following content: Code: #include <mpi.h> #include <stdio.h> int main(int argc, char** argv) { // Initialize the MPI environment MPI_Init(NULL, NULL); // Get the number of processes int world_size; MPI_Comm_size(MPI_COMM_WORLD, &world_size); // Get the rank of the process int world_rank; MPI_Comm_rank(MPI_COMM_WORLD, &world_rank); // Get the name of the processor char processor_name[MPI_MAX_PROCESSOR_NAME]; int name_len; MPI_Get_processor_name(processor_name, &name_len); // Print off a hello world message printf("Hello world from processor %s, rank %d out of %d processors\n", processor_name, world_rank, world_size); // Finalize the MPI environment. MPI_Finalize(); } After i compile it and execute it: Code:kai@Kai-Desktop:~/Dokumente$ mpirun -np 4 ./hello_world -parallel Hello world from processor Kai-Desktop, rank 0 out of 4 processors Hello world from processor Kai-Desktop, rank 1 out of 4 processors Hello world from processor Kai-Desktop, rank 2 out of 4 processors Hello world from processor Kai-Desktop, rank 3 out of 4 processorsIn conclusion mpi works on my computer, or not ? Why are OpenFoam dosent work with it. ? Best regards Kai Am 27.12.20 um 15:03 schrieb Kahnbein Kai via users:Hello, im trying to ran an Openfoam simulation on multiple cores. It worked on my system in the past. I didnt consciously changed anything especially on mpi related things. The only thing i have changed, i updated my Ubuntu version from 18.04 to 20.04. As always i trying to execute the following line:kai@Kai-Desktop:~/OpenFOAM/kai-7/run/tutorials_of/multiphase/interFoam/laminar/damBreak_stl_II/damBreak$mpirun -np 4 interFoam -parallelIn the past, the simulation started and worked fine. But now thismassage will appear:--------------------------------------------------------------------------It looks like MPI_INIT failed for some reason; your parallel process is likely to abort. There are many reasons that a parallel process can fail during MPI_INIT; some of which are due to configuration or environment problems. This failure appears to be an internal failure; here's someadditional information (which may only be relevant to an Open MPIdeveloper): ompi_mpi_init: ompi_rte_init failed --> Returned "(null)" (-43) instead of "Success" (0)----------------------------------------------------------------------------------------------------------------------------------------------------It looks like MPI_INIT failed for some reason; your parallel process is likely to abort. There are many reasons that a parallel process can fail during MPI_INIT; some of which are due to configuration or environment problems. This failure appears to be an internal failure; here's someadditional information (which may only be relevant to an Open MPIdeveloper): ompi_mpi_init: ompi_rte_init failed --> Returned "(null)" (-43) instead of "Success" (0)----------------------------------------------------------------------------------------------------------------------------------------------------It looks like MPI_INIT failed for some reason; your parallel process is likely to abort. There are many reasons that a parallel process can fail during MPI_INIT; some of which are due to configuration or environment problems. This failure appears to be an internal failure; here's someadditional information (which may only be relevant to an Open MPIdeveloper): ompi_mpi_init: ompi_rte_init failed --> Returned "(null)" (-43) instead of "Success" (0)--------------------------------------------------------------------------*** An error occurred in MPI_Init_thread *** on a NULL communicator*** MPI_ERRORS_ARE_FATAL (processes in this communicator will nowabort, *** and potentially your MPI job) [Kai-Desktop:12383] Local abort before MPI_INIT completed completed successfully, but am not able to aggregate error messages, and not able to guarantee that all other processes were killed! *** An error occurred in MPI_Init_thread *** on a NULL communicator*** MPI_ERRORS_ARE_FATAL (processes in this communicator will nowabort, *** and potentially your MPI job) [Kai-Desktop:12384] Local abort before MPI_INIT completed completed successfully, but am not able to aggregate error messages, and not able to guarantee that all other processes were killed! *** An error occurred in MPI_Init_thread *** on a NULL communicator*** MPI_ERRORS_ARE_FATAL (processes in this communicator will nowabort, *** and potentially your MPI job) [Kai-Desktop:12385] Local abort before MPI_INIT completed completed successfully, but am not able to aggregate error messages, and not able to guarantee that all other processes were killed!--------------------------------------------------------------------------Primary job terminated normally, but 1 process returned a non-zero exit code. Per user-direction, the job has been aborted.----------------------------------------------------------------------------------------------------------------------------------------------------It looks like MPI_INIT failed for some reason; your parallel process is likely to abort. There are many reasons that a parallel process can fail during MPI_INIT; some of which are due to configuration or environment problems. This failure appears to be an internal failure; here's someadditional information (which may only be relevant to an Open MPIdeveloper): ompi_mpi_init: ompi_rte_init failed --> Returned "(null)" (-43) instead of "Success" (0)--------------------------------------------------------------------------*** An error occurred in MPI_Init_thread *** on a NULL communicator*** MPI_ERRORS_ARE_FATAL (processes in this communicator will nowabort, *** and potentially your MPI job) [Kai-Desktop:12386] Local abort before MPI_INIT completed completed successfully, but am not able to aggregate error messages, and not able to guarantee that all other processes were killed!--------------------------------------------------------------------------mpirun detected that one or more processes exited with non-zero status, thus causing the job to be terminated. The first process to do so was: Process name: [[32896,1],0] Exit code: 1--------------------------------------------------------------------------I looked into the Systemmonitor, but i didnt have a process withthis name or number. If i execute mpirun --version the consol replys this message: mpirun (Open MPI) 4.0.3 Report bugs tohttps://urldefense.proofpoint.com/v2/url?u=http-3A__www.open-2Dmpi.org_community_help_&d=DwIDaQ&c=sJ6xIWYx-zLMB3EPkvcnVg&r=kgFAU2BfgKe7cozjrP7uWDPH6xt6LAmYVlQPwQuK7ek&m=9YEwGLzNfCD1pAUuvNpqStsbpagtNfIzEt6wL6f3_7I&s=xXp3HlEJc7DzUAnJY0RVVKgKZ9HopKf0UUMePlaCV8w&e=How can I solve this problem ? Best regards Kai-- Tony Ladd Chemical Engineering Department University of Florida Gainesville, Florida 32611-6005 USA Email: tladd-"(AT)"-che.ufl.edu Webhttp://ladd.che.ufl.edu Tel: (352)-392-6509 FAX: (352)-392-9514
