CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   OpenFOAM Native Meshers: snappyHexMesh and Others (https://www.cfd-online.com/Forums/openfoam-meshing-snappyhexmesh/)
-   -   HelyxOS Doesn't load my stl geometry (https://www.cfd-online.com/Forums/openfoam-meshing-snappyhexmesh/119610-helyxos-doesnt-load-my-stl-geometry.html)

pitpat231 June 20, 2013 09:47

HelyxOS Doesn't load my stl geometry
 
pls help me out, i installed HelyxOS using the procedures i saw on the Engys website. my problem now is that my helyx does not load (show) my stl geometry. pls i need help.

chegdan June 20, 2013 10:51

More information is helpful in order to solve this
  1. Is your STL in binary or ASCII format?
    HELYX-OS only accepts ASCII formatted STL files. You can convert between a binary STL (.obj) and an ASCII (.stl) via the command
    Code:

    surfaceConvert inputFile.obj outputFile.stl
  2. What is the decimal separator in your STL file, a "," or a "."? What are your language settings? i.e. what is your locale on the machine?
    If you have a language that uses "," as a decimal separator but your .stl file uses a "." instead then this could be a cause but HELYX-OS will provide an error to let you know this is the issue.
  3. Any error messages with the install? Does it complain about a VTK error? If so then its a problem with your libmpi.so discussed here
  4. What system are you using and how did you install OpenFOAM (via .deb, precompiled source, you compiled OpenFOAM, etc.)?
    In general, problems that users encounter with the precompiled versions of OpenFOAM and are linked to the libmpi.so issue in (3)

pitpat231 June 21, 2013 09:59

thanks, i can now load stl geometry, but now i encounter the following problem wen i run a case in helyx. i use a intel corei7 personal laptop.



MPI_ABORT was invoked on rank 3 in communicator MPI_COMM_WORLD
with errorcode 1.
NOTE: invoking MPI_ABORT causes Open MPI to kill all MPI processes.
You may or may not see output from other processes, depending on
exactly when Open MPI kills them.
mpirun has exited due to process rank 3 with PID 3555 on
node chijioke-W240HU-W250HUQ exiting without calling "finalize". This may
have caused other processes in the application to be
terminated by signals sent by mpirun (as reported here).
[chijioke-W240HU-W250HUQ:03550] 1 more process has sent help message help-mpi-api.txt / mpi-abort
[chijioke-W240HU-W250HUQ:03550] Set MCA parameter "orte_base_help_aggregate" to 0 to see all help / error messages




pls any help on how i can solve this?????
thanks in advance.

chegdan June 22, 2013 18:27

We need to do a few test to see if there is a problem with your MPI install

1: See if we are pointing to your MPI Install
What is the result of the command

Code:

echo $MPI_ARCH_PATH
when you run that in a bash terminal? If nothing comes up then we have a problem. If something comes up in the terminal, let me know what it is if you make it to the end of this post and its still not clear what the problem is. Continue to step 2.

2: Can you run a case in parallel without HELYX-OS?

Assuming you have setup your $FOAM_RUN directory correctly and you are capable of running on 4 processors then we can test on some tutorials. To test this please copy the following into your terminal one-by-one

Code:

cd $FOAM_RUN/tutorials/multiphase/interFoam/laminar/damBreak
blockMesh >blockMesh.log 2>&1
cp 0/alpha1.org 0/alpha1
decomposePar >decomposePar.log 2>&1
mpirun -np 4 interFoam -parallel > interFoam.log 2>&1

and see if you are able to run a parallel case without HELYX-OS.

If you get an error on the first command saying that there is no such directory, then you need to go to 3. If you have an MPI_ABORT error when the last command is run, then its something with your MPI and not with HELYX-OS.

3: setting up your $FOAM_RUN correctly

copy the tutorials to your $FOAM_RUN directory with

Code:

mkdir -p $FOAM_RUN
cd $FOAM_RUN
cp -r $WM_PROJECT_DIR/tutorials $FOAM_RUN

and then start over with the previous steps with running the test case in parallel (part 2).

4: Report Back
Report back on errors that you are receiving and also with some information about your system like operating system and whether its 32 bit or 64 and if you compiled OpenFOAM or installed it from a .deb package. Also, is there a VtkError folder created in your home directory or where you launched HELYX-OS?

pitpat231 June 24, 2013 10:08

@Chegdan.

for process 1. when i put the command echo $MPI_ARCH_PATH, i got the following; /usr/lib/openmpi

for process 2. ; after inputing the commands one after the other, it all ran well with no errors. no error report at all.

When i initially launch my Helyx the following report shows on my terminal, even though the helyx still opens;

java -Xms128m -Xmx1024m -XX:OnError="/home/chijioke/OpenFOAM/Engys/HelyxOS/v1.0.1/bin/collectInfo.run" -jar /home/chijioke/OpenFOAM/Engys/HelyxOS/v1.0.1/lib/HelyxOS.jar HelyxOS 2>&1 | tee /home/chijioke/OpenFOAM/Engys/HelyxOS/v1.0.1/helyxEE.log

pls wat next??? thanks

chegdan June 24, 2013 10:49

That line

Code:

java -Xms128m -Xmx1024m -XX:OnError="/home/chijioke/OpenFOAM/Engys/HelyxOS/v1.0.1/bin/collectInfo.run" -jar /home/chijioke/OpenFOAM/Engys/HelyxOS/v1.0.1/lib/HelyxOS.jar HelyxOS 2>&1 | tee /home/chijioke/OpenFOAM/Engys/HelyxOS/v1.0.1/helyxEE.log
is not an error so you are fine. Do you have a file in your home directory (or where you launch HELYX-OS) called

Code:

vtkError.txt
When you run a case in parallel and get an error?

atg April 25, 2014 23:03

I am having some issues getting HelyxOS 2.0.1 to run in parallel on ubuntu 12.04.

Code:

echo $MPI_ARCH_PATH
yields /usr/lib/openmpi

Tutorials run fine in parallel outside of Helyx as described by chegdan above. pitzDaily converges just fine on multiple processors.

HelyxOS itself seems to launch fine, load old cases, etc.

I can run checkmesh and blockmesh in Helyx and get an OK.

decomposePar seems to work, but then the meshing run ends with the following appearing in the little terminal window within HelyxOS:

Code:

End.

mpirun:  Symbol  'orte_plm' has different size in shared object, consider re-linking
mpirun:  symbol lookup error: mpirun: undefined symbol: orte_cmd_basename

I'm not sure where to start with this. I noticed my bashrc in home/openFoam-2.2.2/etc had WM_MPLIB set to "OPENMPI" instead of "SYSTEMOPENMPI", so I thought that might be the problem. But after changing it to SYSTEMOPENMPI and re-sourcing in the terminal, restarting helyx and reloading the case, the same message resulted and the case would not run.

Anybody have any thoughts? I must have done something to OF while compiling that Helyx does not like. But OF seems perfectly happy otherwise.

Also, is there any way to capture the text from the little terminal within Helyx? It would have saved me some typing!

Thanks for any suggestions! And sorry to bother on a Friday night!
Karl

Quote:

Originally Posted by pitpat231 (Post 435624)
@Chegdan.

for process 1. when i put the command echo $MPI_ARCH_PATH, i got the following; /usr/lib/openmpi

for process 2. ; after inputing the commands one after the other, it all ran well with no errors. no error report at all.

When i initially launch my Helyx the following report shows on my terminal, even though the helyx still opens;

java -Xms128m -Xmx1024m -XX:OnError="/home/chijioke/OpenFOAM/Engys/HelyxOS/v1.0.1/bin/collectInfo.run" -jar /home/chijioke/OpenFOAM/Engys/HelyxOS/v1.0.1/lib/HelyxOS.jar HelyxOS 2>&1 | tee /home/chijioke/OpenFOAM/Engys/HelyxOS/v1.0.1/helyxEE.log

pls wat next??? thanks


chegdan April 27, 2014 21:51

Is there a way you have multiple installs of OpenFOAM 2.2.2 on your machine and HELYX-OS is pointing to one that it shouldn't? You may have one version sourced in your .bashrc (say $HOME/OpenFOAM/OpenFOAM-2.2.2) and HELYX-OS is pointing to another one in say /opt/openfoam222.

atg April 27, 2014 22:09

Quote:

Originally Posted by chegdan (Post 488551)
Is there a way you have multiple installs of OpenFOAM 2.2.2 on your machine and HELYX-OS is pointing to one that it shouldn't? You may have one version sourced in your .bashrc (say $HOME/OpenFOAM/OpenFOAM-2.2.2) and HELYX-OS is pointing to another one in say /opt/openfoam222.

Thanks chegdan - I do have 2.3.0 and 2.2.2 installed, but they are both in my home directory. Nothing in /opt. It is a fresh ubuntu install but the home folder is old. I did discard my old openfoam directories and install both 2.2.2 and 2.3.0 from freshly extracted tar files. I could easily discard 2.3.0 or both and build again. I may just try that and see what happens, but I wish I understood the problem better so I can avoid repeating it.

Karl

westwindpower January 29, 2016 10:00

mpi error in HelyxOS
 
Hi,
I am having a similar error as mentioned in this thread at the CREATE mesh level. I am following the work shop pipe tutorial that Daniel COmbest. (Very nice introduction. really liking it :) ) I have run into a problem when I am trying to build the mesh.

And this comes from the VTK error file in the HelyxOS root directory. I have followed the steps specified in this thread and have verified that opempi is installed and at

/usr/lib/openmpi

the damn tutorial runs perfectly fine.

I will lookup how to turn off parallelization to see if it is a parallelization issue.

Appreciate any help on this!
Thanks
William

// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //
Create time

Create mesh for time = 0

Read mesh in = 0.05 s

Overall mesh bounding box : (-0.2 0.25 -0.2) (0.2 2.25 0.2)
Relative tolerance : 1e-06
Absolute matching distance : 2.078460969e-06

Reading refinement surfaces.
[0] [1]
[0]
[0] --> FOAM FATAL IO ERROR:
[0] Illegal level specification for surface pipe : minLevel:2 maxLevel:0 levelIncrement:0
[0]
[0] file: /home/parallels/Documents/helyx/Engys/HELYX-OS/v2.3.1/tutorial/pipeflow/system/snappyHexMeshDict.castellatedMeshControls.refineme ntSurfaces.pipe from line 41 to line 41.
[0]
[0] From function refinementSurfaces::refinementSurfaces(const searchableSurfaces&, const dictionary>&
[0] in file autoHexMesh/refinementSurfaces/refinementSurfaces.C
at line 122.
[0]
FOAM parallel run exiting
[0]
[1]
[1] --> FOAM FATAL IO ERROR: --------------------------------------------------------------------------
MPI_ABORT was invoked on rank 0 in communicator MPI_COMM_WORLD
with errorcode 1.

NOTE: invoking MPI_ABORT causes Open MPI to kill all MPI processes.
You may or may not see output from other processes, depending on
exactly when Open MPI kills them.
--------------------------------------------------------------------------

[1] Illegal level specification for surface pipe : minLevel:2 maxLevel:0 levelIncrement:0
[1]
[1] file: IOstream.castellatedMeshControls.refinementSurface s.pipe from line 0 to line 0.
[1]
[1] From function refinementSurfaces::refinementSurfaces(const searchableSurfaces&, const dictionary>&
[1] in file autoHexMesh/refinementSurfaces/refinementSurfaces.C at line 122.
[1]
FOAM parallel run exiting
[1]
--------------------------------------------------------------------------
mpirun has exited due to process rank 1 with PID 4623 on
node ubuntu exiting improperly. There are two reasons this could occur:

1. this process did not call "init" before exiting, but others in
the job did. This can cause a job to hang indefinitely while it waits
for all processes to call "init". By rule, if one process calls "init",
then ALL processes must call "init" prior to termination.

2. this process called "init", but exited without calling "finalize".
By rule, all processes that call "init" MUST call "finalize" prior to
exiting or it will be considered an "abnormal termination"

This may have caused other processes in the application to be
terminated by signals sent by mpirun (as reported here).
--------------------------------------------------------------------------
[ubuntu:04620] 1 more process has sent help message help-mpi-api.txt / mpi-abort
[ubuntu:04620] Set MCA parameter "orte_base_help_aggregate" to 0 to see all help / error messages

chegdan January 29, 2016 11:32

1 Attachment(s)
William,

The error you are seeing is from snappyHexmesh. See the line:

Code:

Illegal level specification for surface pipe : minLevel:2 maxLevel:0 levelIncrement:0
You have probably specified a min level higher than a max level. You will need to specify ascending numbers for min and max similar to the attached image.

westwindpower January 29, 2016 22:10

You were right. It was in the refinement layer. I had min 2 and max 0.
Thanks for the help!:)


Quote:

Originally Posted by chegdan (Post 582961)
William,

The error you are seeing is from snappyHexmesh. See the line:

Code:

Illegal level specification for surface pipe : minLevel:2 maxLevel:0 levelIncrement:0
You have probably specified a min level higher than a max level. You will need to specify ascending numbers for min and max similar to the attached image.


westwindpower January 30, 2016 10:32

Illegal Turbulence Model type in HelyzOS
 
Hey,
Was successful in building the mesh. Was following the tutorial but when I went to run the solver. I am getting the following fatal foam error in the execution:

I am thinking of trying some different solver settings just to see it run and then double back on the simplefoam solver used in the tutorial.

Selecting incompressible transport model Newtonian
Selecting turbulence model type RASModel
[1]
[1]
[1] --> FOAM FATAL ERROR:
[1] Unknown TurbulenceModel type RASModel

Valid TurbulenceModel types:

3
(
LES
RAS
laminar
)
[1]
[1]
[1] From function TurbulenceModel::New(const alphaField&, const rhoField&, const volVectorField&, const surfaceScalarField&, transportModel&, const word&)
[1] in file /home/openfoam/OpenFOAM/OpenFOAM-3.0.0/src/TurbulenceModels/turbulenceModels/lnInclude/TurbulenceModel.C at line 119.
[1]
FOAM parallel run exiting


chegdan January 31, 2016 16:20

Great that you were successful in meshing. From looking at the error:

Code:

[1]
[1]
[1] --> FOAM FATAL ERROR:
[1] Unknown TurbulenceModel type RASModel

Valid TurbulenceModel types:

3
(
LES
RAS
laminar
)
[1]
[1]
[1] From function TurbulenceModel::New(const alphaField&, const rhoField&, const volVectorField&, const surfaceScalarField&, transportModel&, const word&)
[1] in file /home/openfoam/OpenFOAM/OpenFOAM-3.0.0/src/TurbulenceModels/turbulenceModels/lnInclude/TurbulenceModel.C at line 119..

You are using an version currently too new for HELYX-OS 2.3.1. The issue is that there was a change in the turbulenceProperties dictionary structure from OpenFOAM-2x to OpenFOAM-3X. The HELYX-OS-2.X releases are only applicable to OpenFOAM-2.X, more precisely, HELYX-OS-2.3.1 will only work with the last 2.x series (i.e. 2.4.0). So you can do one of three things.
  1. Only use HELYX-OS for meshing for now as that will work with OpenFOAM-3.x. Then use OpenFOAM-2.4.1, since that is compatible with HELYX-OS v2.3.1
  2. You can revert back to a fully supported version (OpenFOAM-2.4.1) if all that you need is in that version
  3. you can use HELYX-OS with OpenFOAM-3.x and use the custom interface to manually change some of the require dictionaries (e.g. turbulenceProperties) to coincide with the changes necessary for the new version.

There will be an update to the new version, this latest release was a maintenance release and a movement to github so expect a bigger change when a major release is done for HELYX-OS.


All times are GMT -4. The time now is 23:31.