|
[Sponsors] |
September 28, 2011, 07:08 |
Cyclic BC with parallel running
|
#1 |
New Member
Roberto Della Ratta
Join Date: Apr 2011
Posts: 8
Rep Power: 15 |
Hi,
we're using the last version of OpenFoam 2.0.x to do a compressible LES with rhoCentralFoam. We're facing a problem reading a mesh with cyclic BC and decomposed with Scotch (the only one present in 2.0). Actually, in serial running and decomposing the case in 3 processors the solver runs well. Trying with more processors (40 or 60) the code stops when reading the mesh ( I attach the error). With empty BC the case works well even on 40 processors. Thanks for the help Code:
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * // Create time Create mesh for time = 0 [29] #0 Foam::error:printStack(Foam::Ostream&)-------------------------------------------------------------------------- An MPI process has executed an operation involving a call to the "fork()" system call to create a child process. Open MPI is currently operating in a condition that could result in memory corruption or other system errors; your MPI job may hang, crash, or produce silent data corruption. The use of fork() (or system() or other calls that create child processes) is strongly discouraged. The process that invoked fork was: Local host: node023 (PID 5648) MPI_COMM_WORLD rank: 29 If you are *absolutely sure* that your application will successfully and correctly survive a call to fork(), you may disable this warning by setting the mpi_warn_on_fork MCA parameter to 0. -------------------------------------------------------------------------- addr2line failed [29] #1 Foam::sigSegv::sigHandler(int) addr2line failed [29] #2 Uninterpreted: /lib64/libc.so.6 addr2line failed [29] #3 Foam::processorPolyPatch::updateMesh(Foam::PstreamBuffers&) addr2line failed [29] #4 Foam::polyBoundaryMesh::updateMesh()[13] #0 Foam::error::printStack(Foam::Ostream&) addr2line failed [29] #5 Foam::polyMesh::polyMesh(Foam::IOobject const&) addr2line failed [29] #6 Foam::fvMesh::fvMesh(Foam::IOobject const&) addr2line failed [29] #7 Uninterpreted: NS_Compressible [29] #8 __libc_start_main addr2line failed [13] #1 Foam::sigSegv::sigHandler(int) addr2line failed [29] #9 addr2line failed [13] #2 Uninterpreted: Foam::UOPstream::write(char) /lib64/libc.so.6 [node023:05648] *** Process received signal *** [node023:05648] Signal: Segmentation fault (11) [node023:05648] Signal code: (-6) [node023:05648] Failing at address: 0x20700001610 [node023:05648] [ 0] /lib64/libc.so.6 [0x2aaaae5262d0] [node023:05648] [ 1] /lib64/libc.so.6(gsignal+0x35) [0x2aaaae526265] [node023:05648] [ 2] /lib64/libc.so.6 [0x2aaaae5262d0] [node023:05648] [ 3] /home/ioba/data/software/OpenFOAM/OpenFOAM-2.0.x/platforms/linux64Gcc44DPOpt/lib/libOpenFOAM.so(_ZN4Foam18processorPolyPatch10updateMeshERNS_14PstreamBuffersE+0x2da) [0x2aaaad6db47a] [node023:05648] [ 4] /home/ioba/data/software/OpenFOAM/OpenFOAM-2.0.x/platforms/linux64Gcc44DPOpt/lib/libOpenFOAM.so(_ZN4Foam16polyBoundaryMesh10updateMeshEv+0x2b1) [0x2aaaad6e24b1] [node023:05648] [ 5] /home/ioba/data/software/OpenFOAM/OpenFOAM-2.0.x/platforms/linux64Gcc44DPOpt/lib/libOpenFOAM.so(_ZN4Foam8polyMeshC2ERKNS_8IOobjectE+0x10ea) [0x2aaaad73347a] [node023:05648] [ 6] /home/ioba/data/software/OpenFOAM/OpenFOAM-2.0.x/platforms/linux64Gcc44DPOpt/lib/libfiniteVolume.so(_ZN4Foam6fvMeshC1ERKNS_8IOobjectE+0x19) [0x2aaaab069fb9] [node023:05648] [ 7] NS_Compressible [0x420881] [node023:05648] [ 8] /lib64/libc.so.6(__libc_start_main+0xf4) [0x2aaaae513994] [node023:05648] [ 9] NS_Compressible(_ZN4Foam9UOPstream5writeEc+0xa9) [0x41d599] [node023:05648] *** End of error message *** addr2line failed [13] #3 Foam::processorPolyPatch::updateMesh(Foam::PstreamBuffers&) addr2line failed [13] #4 Foam::polyBoundaryMesh::updateMesh() addr2line failed [13] #5 Foam::polyMesh::polyMesh(Foam::IOobject const&) addr2line failed [13] #6 Foam::fvMesh::fvMesh(Foam::IOobject const&)-------------------------------------------------------------------------- mpirun noticed that process rank 29 with PID 5648 on node node023.cm.cluster exited on signal 11 (Segmentation fault). -------------------------------------------------------------------------- [node015:31786] 1 more process has sent help message help-mpi-runtime.txt / mpi_init:warn-fork [node015:31786] Set MCA parameter "orte_base_help_aggregate" to 0 to see all help / error messages |
|
September 29, 2011, 14:19 |
|
#2 |
Senior Member
Alberto Passalacqua
Join Date: Mar 2009
Location: Ames, Iowa, United States
Posts: 1,912
Rep Power: 36 |
Hi Roberto,
you might want to report thins on the mantis bug tracker ( http://www.openfoam.com/mantisbt/view_all_bug_page.php ). They changed the implementation of cyclic BC's in OpenFOAM 2.0. so that it should work without problems in parallels, and this might be a bug. Best,
__________________
Alberto Passalacqua GeekoCFD - A free distribution based on openSUSE 64 bit with CFD tools, including OpenFOAM. Available as in both physical and virtual formats (current status: http://albertopassalacqua.com/?p=1541) OpenQBMM - An open-source implementation of quadrature-based moment methods. To obtain more accurate answers, please specify the version of OpenFOAM you are using. |
|
September 30, 2011, 05:43 |
|
#3 |
New Member
Roberto Della Ratta
Join Date: Apr 2011
Posts: 8
Rep Power: 15 |
Hi Alberto,
Thanks very much.. I'm just submitting the issue.. For the moment I resolve partitioning with simple and keeping the corresponding cyclic faces on the same processor. Best |
|
Tags |
cyclic patches, decomposition |
Thread Tools | Search this Thread |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Parallel refineMesh with Cyclic Boundaries | mchurchf | OpenFOAM | 8 | December 22, 2018 11:11 |
Cyclic patches and parallel postprocessing problems | askjak | OpenFOAM Bugs | 18 | October 27, 2010 03:35 |
Statically Compiling OpenFOAM Issues | herzfeldd | OpenFOAM Installation | 21 | January 6, 2009 09:38 |
Kubuntu uses dash breaks All scripts in tutorials | platopus | OpenFOAM Bugs | 8 | April 15, 2008 07:52 |
Cyclic patch in parallel calculations | didomenico | OpenFOAM Running, Solving & CFD | 4 | March 7, 2007 05:46 |