|
[Sponsors] |
November 12, 2021, 15:34 |
SBATCH decomposePar error
|
#1 |
Member
cal
Join Date: Feb 2020
Location: nowhere
Posts: 65
Rep Power: 6 |
Hi,
This is my first run with server. I've a good looking test slurm file down below. Meshing goes well but I can't decompose the case. Other than that, all other lines work fine. I mean, if I decompose the case on my PC and upload to the server everything runs smoothly. Code:
decomposePar: error while loading shared libraries: libmetis.so: cannot open shared object file: No such file or directory I'm open to any suggestions. Files are attached. Kind regards, Said. |
|
November 12, 2021, 16:24 |
|
#2 |
Member
Join Date: Aug 2017
Location: Algeria
Posts: 98
Rep Power: 9 |
What is the output of:
Code:
module spider openfoam
__________________
"When in doubt, use brute force." -- Ken Thompson |
|
November 12, 2021, 16:52 |
|
#3 |
Member
cal
Join Date: Feb 2020
Location: nowhere
Posts: 65
Rep Power: 6 |
Hi s1291,
Output is: Code:
cmdTrace.c(713):ERROR:104: 'spider' is an unrecognized subcommand cmdModule.c(411):ERROR:104: 'spider' is an unrecognized subcommand Said. |
|
November 12, 2021, 17:05 |
|
#4 |
Member
Join Date: Aug 2017
Location: Algeria
Posts: 98
Rep Power: 9 |
My guess is that you're using Tcl modules instead of Lmod. nevertheless, what is the output of:
Code:
module avail 2>&1 | grep -i openfoam Code:
module avail 2>&1 | grep -i metis
__________________
"When in doubt, use brute force." -- Ken Thompson |
|
November 12, 2021, 17:13 |
|
#5 |
Member
cal
Join Date: Feb 2020
Location: nowhere
Posts: 65
Rep Power: 6 |
openfoam exist on the list "module avail" but metis not on the list. Also there is no output for both lines, just returns nothing. If I type "module avail" I can see the "centos7.3/app/openfoam/v1912" module.
Edit: Code:
-bash-4.2$ module avail 2>&1 | grep -i openfoam centos7.3/app/openfoam/8 centos7.3/app/openfoam/v1912 -bash-4.2$ module avail 2>&1 | grep -i metis -bash-4.2$ |
|
November 12, 2021, 17:18 |
|
#6 |
Member
Join Date: Aug 2017
Location: Algeria
Posts: 98
Rep Power: 9 |
Another question:
__________________
"When in doubt, use brute force." -- Ken Thompson |
|
November 12, 2021, 17:33 |
|
#7 |
Member
cal
Join Date: Feb 2020
Location: nowhere
Posts: 65
Rep Power: 6 |
I copied that part from server's example slurm file, now I tried to run with switch them but nothing changed. I'm still getting the same error.
|
|
November 13, 2021, 05:55 |
|
#8 |
Member
cal
Join Date: Feb 2020
Location: nowhere
Posts: 65
Rep Power: 6 |
Hi,
I solved the previous problem with sourcing everything in the slurm file but now I can't use restore0Dir command. It is not working. Code:
/tmp/slurmd/job7894762/slurm_script: line 235: restore0Dir: command not found Kind regards, Said. |
|
November 13, 2021, 07:55 |
|
#9 |
Member
Join Date: Aug 2017
Location: Algeria
Posts: 98
Rep Power: 9 |
That command is not necessary. But you can use it after sourcing the RunFunctions script:
Code:
source ${WM_PROJECT_DIR}/bin/tools/RunFunctions
__________________
"When in doubt, use brute force." -- Ken Thompson |
|
November 13, 2021, 08:51 |
|
#10 | |
Member
cal
Join Date: Feb 2020
Location: nowhere
Posts: 65
Rep Power: 6 |
Quote:
Code:
source /truba/sw/centos7.3/app/openfoam/OpenFOAM-v1912/bin/tools/RunFunctions Kind regards, Said. |
||
November 13, 2021, 16:07 |
|
#11 |
Member
cal
Join Date: Feb 2020
Location: nowhere
Posts: 65
Rep Power: 6 |
Hi again,
I thought the problem was solved but changeDictionary can not import boundary conditions to processor files correctly. I searched for the reason for this but couldn't find a solution. Code:
[4] --> FOAM FATAL ERROR: [4] cannot be called for a calculatedFvPatchField on patch H_spheres1 of field U in file "/truba/home/tbag100/openfoam_run/simulation_1/processor4/0/U" You are probably trying to solve for a field with a default boundary condition. [4] [4] From function Foam::tmp<Foam::Field<Type> > Foam::calculatedFvPatchField<Type>::gradientInternalCoeffs() const [with Type = Foam::Vector<double>] [4] in file fields/fvPatchFields/basic/calculated/calculatedFvPatchField.C at line 188. [4] FOAM parallel run aborting Code:
ideasUnvToFoam Mesh_Simulation_1.unv > log.ideasUnvToFoam 2>&1 transformPoints -scale "(0.001)" > log.transformPoints 2>&1 createPatch -overwrite > log.createPatch 2>&1 decomposePar > log.decomposePar 2>&1 source /truba/sw/centos7.3/app/openfoam/OpenFOAM-v1912/bin/tools/RunFunctions restore0Dir -processor mpirun changeDictionary -parallel > log.changeDictionary 2>&1 mpirun buoyantPimpleFoam -parallel > log.buoyantPimpleFoam 2>&1 reconstructPar > log.reconstructPar 2>&1 Edit: I figured it out how to fix it. I just used restore0Dir and changeDictionary (serial) before the decomposition. Right now everythings works fine for me. Kind regards, Said. Last edited by saidc.; November 15, 2021 at 03:32. |
|
Tags |
decompose, decomposepar, sbatch, slurm |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
[OpenFOAM.org] compile error in dynamicMesh and thermophysicalModels libraries | NickG | OpenFOAM Installation | 3 | December 30, 2019 01:21 |
[blockMesh] blockMesh with double grading. | spwater | OpenFOAM Meshing & Mesh Conversion | 92 | January 12, 2019 10:00 |
[swak4Foam] GroovyBC the dynamic cousin of funkySetFields that lives on the suburb of the mesh | gschaider | OpenFOAM Community Contributions | 300 | October 29, 2014 19:00 |
OpenFOAM without MPI | kokizzu | OpenFOAM Installation | 4 | May 26, 2014 10:17 |
DecomposePar links against liblamso0 with OpenMPI | jens_klostermann | OpenFOAM Bugs | 11 | June 28, 2007 18:51 |