opengeosys data explorer for linux using gmsh

Mailing List Archives ogs-users
1

Hello

I’m considering to use OpenGeoSys in a bioremediation project. At the moment I’m testing its geometry generation capabilities. I understand that it uses Gmsh to generate meshes?
As a first step I’m using the example files in http://www.opengeosys.org/resources/examples , specifically the file GROUNDWATER_FLOW.tar.gz.

The version of gmsh i’m using: 2.9.3
The version of linux: Fedora 21 (4.0.4-202.fc21.x86_64 GNU/Linux)

I downloaded two versions of the ogs data explorer for linux from: https://www.opengeosys.org/ogs-5/ and http://docs.opengeosys.org/download#latest

The name of the first version is : ogs-6.0.0 Beta-Linux-2.6.32-431.el6.x86_64-x64
The name of the second version is: ogs-6.0.1-398-g4b6c26d-Linux-2.6.32-504.8.1.el6.x86_64-x64

with both versions I have tried to open the file “q_quad.gli” included in the example package mentioned previously and the results seems to be the same.

The output after reading the file is:

info: GeoLib::readGLIFile(): open stream from file /home/zerpiko/GROUNDWATER_FLOW/q_quad/q_quad.gli.
info: GeoLib::readGLIFile(): done.
info: GeoLib::readGLIFile(): read points from stream.
info: GeoLib::readGLIFile(): ok, 5 points read.
info: Geometry “q_quad.gli” built. 5 points added.
info: GeoLib::readGLIFile(): read polylines from stream.
info: GeoLib::readGLIFile(): ok, 2 polylines read.
info: GeoLib::readGLIFile(): tag #SURFACE not found.
info: 2 polylines added.

It seems that the file is read correctly, except for the line “info: GeoLib::readGLIFile(): tag #SURFACE not found.”. What is the reason for this?
Then after specifying the path to gmsh I go to Tools->Mesh Generation and in the new window I transfer the geometry file q_quad.gli from “Available information” to “Employed information”.
In the second tab “Advanced” I leave the default options (Adaptive meshing) and untick the “Delete GMSH geo-file after generating mesh” so that I can check this file by myself. After accepting these
options the terminal output is:

info: Start meshing …
info: GEOObjects::getStationVec() - No entry found with name “q_quad.gli”.
Info : Running ‘/opt/gmsh-2.9.3-Linux/bin/gmsh -2 -algo meshadapt /home/zerpiko/GROUNDWATER_FLOW/q_quad/tmp_gmsh.geo -o /home/zerpiko/GROUNDWATER_FLOW/q_quad/tmp_gmsh.msh’ [Gmsh 2.9.3, 1 node, max. 1 thread]
Info : Started on Wed Jun 10 13:30:39 2015
Info : Reading ‘/home/zerpiko/GROUNDWATER_FLOW/q_quad/tmp_gmsh.geo’…
Info : Done reading ‘/home/zerpiko/GROUNDWATER_FLOW/q_quad/tmp_gmsh.geo’
Info : Meshing 1D…
Info : Done meshing 1D (1e-06 s)
Info : Meshing 2D…
Info : Done meshing 2D (9.53674e-07 s)
Info : 0 vertices 0 elements
Info : Writing ‘/home/zerpiko/GROUNDWATER_FLOW/q_quad/tmp_gmsh.msh’…
Info : Done writing ‘/home/zerpiko/GROUNDWATER_FLOW/q_quad/tmp_gmsh.msh’
Info : Stopped on Wed Jun 10 13:30:39 2015
info: GMSHInterface::isGMSHMeshFile(): Found GMSH mesh file version: 2.2 0 8.

It seems like Data Explorer is not finding the correct file? And then, the generation of the .geo file is incorrect, in this case is empty, in others with different .gli files it has incomplete content.

Do you have any suggestions?

Regards
Javier

2

Dear Javier,

take a look at the gli-file you are working with. There is no definition
of a surface or even a closed line loop gmsh could make a mesh from.
This is, why you don't get any output.

In this case, the mesh was probably created externally (e.g. via stand
alone gmsh or gina). This way, you don't have to define the whole domain
in the gli-file but only those geometric entities of interest, like one
line for a source term, one line for the boundary condition and a point
for the output. OGS will then automatically interpolate (or nearest
neighbour search??? someone help me out here!) between the mesh in the
*.msh file and the geometric entities in the *.gli file. Btw, the
$EPSILON parameter you are seeing in the *.gli file is specifying
"something like" the interpolation distance between the geometric
entities and the mesh nodes (again, someone with a more precise
explanation?).

My personal recommendation regarding mesh generation would be working
with gmsh directly, as long as you work with an artificial geometry. I
would use the ogs data explorer only if you have to work with some kind
of GIS or raster files. Take a look at the rich documentation of gmsh
(links below), you have a lot of options for geometry and mesh creation!
Once you created your mesh, you can import (not open, same ending but
different format!) the *.msh file via ogs data explorer and save it as
an ogs-mesh.

Another approach would be using "gina", which is some kind of an
ogs-gui. Unfortunately, gina is not publicly available, but I'm sure,
someone in this mailing list is going to provide you with a copy, if you
ask for it.

geuz.org/gmsh/ (seems to be down atm)
info gmsh (old docu)
2D Mesh Tutorial using GMSH - OpenFOAMWiki (nice
tutorial for getting started)

Good Luck!

Gerrit

···

Am 10.06.2015 um 15:32 schrieb Jose Javier Muñoz Criollo:

Hello

I'm considering to use OpenGeoSys in a bioremediation project. At the
moment I'm testing its geometry generation capabilities. I understand that
it uses Gmsh to generate meshes?
As a first step I'm using the example files in
http://www.opengeosys.org/resources/examples , specifically the file
GROUNDWATER_FLOW.tar.gz.

The version of gmsh i'm using: 2.9.3
The version of linux: Fedora 21 (4.0.4-202.fc21.x86_64 GNU/Linux)

I downloaded two versions of the ogs data explorer for linux from:
https://www.opengeosys.org/ogs-5/ and
http://docs.opengeosys.org/download#latest

The name of the first version is : ogs-6.0.0
Beta-Linux-2.6.32-431.el6.x86_64-x64
The name of the second version is:
ogs-6.0.1-398-g4b6c26d-Linux-2.6.32-504.8.1.el6.x86_64-x64

with both versions I have tried to open the file "q_quad.gli" included in
the example package mentioned previously and the results seems to be the
same.

The output after reading the file is:

info: GeoLib::readGLIFile(): open stream from file

/home/zerpiko/GROUNDWATER_FLOW/q_quad/q_quad.gli.
info: GeoLib::readGLIFile(): done.
info: GeoLib::readGLIFile(): read points from stream.
info: GeoLib::readGLIFile(): ok, 5 points read.
info: Geometry "q_quad.gli" built. 5 points added.
info: GeoLib::readGLIFile(): read polylines from stream.
info: GeoLib::readGLIFile(): ok, 2 polylines read.
info: GeoLib::readGLIFile(): tag #SURFACE not found.
info: 2 polylines added.

It seems that the file is read correctly, except for the line "info:
GeoLib::readGLIFile(): tag #SURFACE not found.". What is the reason for
this?
Then after specifying the path to gmsh I go to Tools->Mesh Generation and
in the new window I transfer the geometry file q_quad.gli from "Available
information" to "Employed information".
In the second tab "Advanced" I leave the default options (Adaptive meshing)
and untick the "Delete GMSH geo-file after generating mesh" so that I can
check this file by myself. After accepting these
options the terminal output is:

info: Start meshing ...

info: GEOObjects::getStationVec() - No entry found with name "q_quad.gli".
Info : Running '/opt/gmsh-2.9.3-Linux/bin/gmsh -2 -algo meshadapt
/home/zerpiko/GROUNDWATER_FLOW/q_quad/tmp_gmsh.geo -o
/home/zerpiko/GROUNDWATER_FLOW/q_quad/tmp_gmsh.msh' [Gmsh 2.9.3, 1 node,
max. 1 thread]
Info : Started on Wed Jun 10 13:30:39 2015
Info : Reading '/home/zerpiko/GROUNDWATER_FLOW/q_quad/tmp_gmsh.geo'...
Info : Done reading '/home/zerpiko/GROUNDWATER_FLOW/q_quad/tmp_gmsh.geo'
Info : Meshing 1D...
Info : Done meshing 1D (1e-06 s)
Info : Meshing 2D...
Info : Done meshing 2D (9.53674e-07 s)
Info : 0 vertices 0 elements
Info : Writing '/home/zerpiko/GROUNDWATER_FLOW/q_quad/tmp_gmsh.msh'...
Info : Done writing '/home/zerpiko/GROUNDWATER_FLOW/q_quad/tmp_gmsh.msh'
Info : Stopped on Wed Jun 10 13:30:39 2015
info: GMSHInterface::isGMSHMeshFile(): Found GMSH mesh file version: 2.2 0
8.

It seems like Data Explorer is not finding the correct file? And then, the
generation of the .geo file is incorrect, in this case is empty, in others
with different .gli files it has incomplete content.

Do you have any suggestions?

Regards
Javier

--
--------------------------------------------------------------------------------------------
            
Gerrit Laube
Department Hydrogeologie
Department Hydrogeology
        
Helmholtz-Zentrum für Umweltforschung GmbH - UFZ
Helmholtz Centre for Environmental Research GmbH - UFZ
Permoserstraße 15 / 04318 Leipzig / Germany
Telefon +49 341 235 1996
gerrit.laube@ufz.de / www.ufz.de

Sitz der Gesellschaft: Leipzig
Registergericht: Amtsgericht Leipzig, Handelsregister Nr. B 4703
Vorsitzender des Aufsichtsrats: MinDirig Wilfried Kraus
Wissenschaftlicher Geschäftsführer: Prof. Dr. Georg Teutsch
Administrative Geschäftsführerin: Dr. Heike Graßmann
            
--------------------------------------------------------------------------------------------

3

Dear Gerrit

Thanks for your answer. Now it makes more sense. I do have a bit of experience working with Gmsh and is my default mesh tool at the moment. However I got a bit confused when I looked at the ogs geometry file, I thought that you would define the whole geometry in it, then export it to .geo to mesh it with Gmsh. Now, if I understood correctly, I need to have a mesh first (created with Gmsh for example) and then define the physical entities with the gli-file.

Thanks for your help.

Regards
Javier

···

On 10/06/15 16:56, Gerrit Laube wrote:

Dear Javier,

take a look at the gli-file you are working with. There is no definition
of a surface or even a closed line loop gmsh could make a mesh from.
This is, why you don't get any output.

In this case, the mesh was probably created externally (e.g. via stand
alone gmsh or gina). This way, you don't have to define the whole domain
in the gli-file but only those geometric entities of interest, like one
line for a source term, one line for the boundary condition and a point
for the output. OGS will then automatically interpolate (or nearest
neighbour search??? someone help me out here!) between the mesh in the
*.msh file and the geometric entities in the *.gli file. Btw, the
$EPSILON parameter you are seeing in the *.gli file is specifying
"something like" the interpolation distance between the geometric
entities and the mesh nodes (again, someone with a more precise
explanation?).

My personal recommendation regarding mesh generation would be working
with gmsh directly, as long as you work with an artificial geometry. I
would use the ogs data explorer only if you have to work with some kind
of GIS or raster files. Take a look at the rich documentation of gmsh
(links below), you have a lot of options for geometry and mesh creation!
Once you created your mesh, you can import (not open, same ending but
different format!) the *.msh file via ogs data explorer and save it as
an ogs-mesh.

Another approach would be using "gina", which is some kind of an
ogs-gui. Unfortunately, gina is not publicly available, but I'm sure,
someone in this mailing list is going to provide you with a copy, if you
ask for it.

geuz.org/gmsh/ (seems to be down atm)
info gmsh (old docu)
2D Mesh Tutorial using GMSH - OpenFOAMWiki (nice
tutorial for getting started)

Good Luck!

Gerrit

Am 10.06.2015 um 15:32 schrieb Jose Javier Muñoz Criollo:

Hello

I'm considering to use OpenGeoSys in a bioremediation project. At the
moment I'm testing its geometry generation capabilities. I understand that
it uses Gmsh to generate meshes?
As a first step I'm using the example files in
http://www.opengeosys.org/resources/examples , specifically the file
GROUNDWATER_FLOW.tar.gz.

The version of gmsh i'm using: 2.9.3
The version of linux: Fedora 21 (4.0.4-202.fc21.x86_64 GNU/Linux)

I downloaded two versions of the ogs data explorer for linux from:
https://www.opengeosys.org/ogs-5/ and
http://docs.opengeosys.org/download#latest

The name of the first version is : ogs-6.0.0
Beta-Linux-2.6.32-431.el6.x86_64-x64
The name of the second version is:
ogs-6.0.1-398-g4b6c26d-Linux-2.6.32-504.8.1.el6.x86_64-x64

with both versions I have tried to open the file "q_quad.gli" included in
the example package mentioned previously and the results seems to be the
same.

The output after reading the file is:

info: GeoLib::readGLIFile(): open stream from file

/home/zerpiko/GROUNDWATER_FLOW/q_quad/q_quad.gli.
info: GeoLib::readGLIFile(): done.
info: GeoLib::readGLIFile(): read points from stream.
info: GeoLib::readGLIFile(): ok, 5 points read.
info: Geometry "q_quad.gli" built. 5 points added.
info: GeoLib::readGLIFile(): read polylines from stream.
info: GeoLib::readGLIFile(): ok, 2 polylines read.
info: GeoLib::readGLIFile(): tag #SURFACE not found.
info: 2 polylines added.

It seems that the file is read correctly, except for the line "info:
GeoLib::readGLIFile(): tag #SURFACE not found.". What is the reason for
this?
Then after specifying the path to gmsh I go to Tools->Mesh Generation and
in the new window I transfer the geometry file q_quad.gli from "Available
information" to "Employed information".
In the second tab "Advanced" I leave the default options (Adaptive meshing)
and untick the "Delete GMSH geo-file after generating mesh" so that I can
check this file by myself. After accepting these
options the terminal output is:

info: Start meshing ...

info: GEOObjects::getStationVec() - No entry found with name "q_quad.gli".
Info : Running '/opt/gmsh-2.9.3-Linux/bin/gmsh -2 -algo meshadapt
/home/zerpiko/GROUNDWATER_FLOW/q_quad/tmp_gmsh.geo -o
/home/zerpiko/GROUNDWATER_FLOW/q_quad/tmp_gmsh.msh' [Gmsh 2.9.3, 1 node,
max. 1 thread]
Info : Started on Wed Jun 10 13:30:39 2015
Info : Reading '/home/zerpiko/GROUNDWATER_FLOW/q_quad/tmp_gmsh.geo'...
Info : Done reading '/home/zerpiko/GROUNDWATER_FLOW/q_quad/tmp_gmsh.geo'
Info : Meshing 1D...
Info : Done meshing 1D (1e-06 s)
Info : Meshing 2D...
Info : Done meshing 2D (9.53674e-07 s)
Info : 0 vertices 0 elements
Info : Writing '/home/zerpiko/GROUNDWATER_FLOW/q_quad/tmp_gmsh.msh'...
Info : Done writing '/home/zerpiko/GROUNDWATER_FLOW/q_quad/tmp_gmsh.msh'
Info : Stopped on Wed Jun 10 13:30:39 2015
info: GMSHInterface::isGMSHMeshFile(): Found GMSH mesh file version: 2.2 0
8.

It seems like Data Explorer is not finding the correct file? And then, the
generation of the .geo file is incorrect, in this case is empty, in others
with different .gli files it has incomplete content.

Do you have any suggestions?

Regards
Javier

--
Jose Javier Munoz Criollo
Research Associate
Cardiff University School of Engineering
Geoenvironmental Research Centre
+44 (0)29 2087 0528 (direct line)