×

Notice

The forum is in read only mode.

Problem in Ubuntu 9.10 with SalomeMeca

  • Borja AP
  • Topic Author
  • Offline
  • Junior Member
  • Junior Member
More
15 years 3 months ago #3767 by Borja AP
Problem in Ubuntu 9.10 with SalomeMeca was created by Borja AP
Hi all,

I recently installed Salome Meca 2009 GPL, I started trying to solve a linear elastic problem but I'm getting an error message as in this post:

caelinux.com/CMS/index.php?option=com_jo...t=6&limitstart=0

I noticed there were some libs missing so I tried to fix it adding some paths in the "profile.sh" script. which I got from looking for the libraries in the terminal as "locate library".

I got this in the terminal
[code:1]borja@borja-laptop:~/SALOME-MECA-2009.1-GPL$ locate libstdc++.so
/home/borja/GiDx32/lib/libstdc++.so.6
/home/borja/SALOME-MECA-2009.1-GPL/SALOME/SALOME4/Prerequis/debianForSalome-v1_2_0/lib/libstdc++.so.5
/home/borja/SALOME-MECA-2009.1-GPL/SALOME/SALOME4/Prerequis/debianForSalome-v1_2_0/lib/libstdc++.so.5.0.7
/home/borja/SALOME-MECA-2009.1-GPL/SALOME/SALOME4/Prerequis/debianForSalome-v1_2_0/lib/gcc-lib/i486-linux/3.3.5/libstdc++.so
/home/borja/scilab-5.1.1/lib/thirdparty/libstdc++.so
/home/borja/scilab-5.1.1/lib/thirdparty/libstdc++.so.6
/home/borja/scilab-5.1.1/lib/thirdparty/libstdc++.so.6.0.10
/opt/google-earth/libstdc++.so.6
/usr/lib/libstdc++.so.6
/usr/lib/libstdc++.so.6.0.13
/usr/lib/gcc/i486-linux-gnu/4.4/libstdc++.so
/usr/local/matlab/sys/os/glnx86/libstdc++.so.5
/usr/local/matlab/sys/os/glnx86/libstdc++.so.5.0.3[/code:1]

My "profile.sh" which is in "/home/borja/SALOME-MECA-2009.1-GPL/aster/STA9.4" looks like this after adding the LD_LIBRARY_PATH.

[code:1]# profile for Code_Aster version 9.4

PATH=$PATH:$ASTER_ROOT/outils
export PATH

LD_LIBRARY_PATH=/local01/assire/SALOME-MECA-2009.1-BUILD/Salome_4.1.4/SALOME4/Prerequis/OmniORB-4.1.0/lib:$LD_LIBRARY_PATH
LD_LIBRARY_PATH=/local01/assire/SALOME-MECA-2009.1-BUILD/Salome_4.1.4/SALOME4/Prerequis/Scotch-4.0/bin:$LD_LIBRARY_PATH
LD_LIBRARY_PATH=/home/borja/SALOME-MECA-2009.1-GPL/aster/public/mumps-4.7.3/libseq:/home/borja/SALOME-MECA-2009.1-GPL/aster/public/mumps-4.7.3/PORD/lib:/home/borja/SALOME-MECA-2009.1-GPL/aster/public/mumps-4.7.3/lib:$LD_LIBRARY_PATH
LD_LIBRARY_PATH=/local01/assire/SALOME-MECA-2009.1-BUILD/Salome_4.1.4/SALOME4/Prerequis/Med-2.3.5/lib:$LD_LIBRARY_PATH
LD_LIBRARY_PATH=/local01/assire/SALOME-MECA-2009.1-BUILD/Salome_4.1.4/SALOME4/Prerequis/Hdf5-1.6.5/lib:$LD_LIBRARY_PATH
#LD_LIBRARY_PATH=/home/borja/SALOME-MECA-2009.1-GPL/aster/public/Python-2.4.4/lib/python2.4:$LD_LIBRARY_PATH
LD_LIBRARY_PATH=/home/borja/SALOME-MECA-2009.1-GPL/aster/public/lib:$LD_LIBRARY_PATH
LD_LIBRARY_PATH=/usr/lib/X11:/usr/lib:/lib:/usr/X11R6/lib:$LD_LIBRARY_PATH
LD_LIBRARY_PATH=/home/borja/SALOME-MECA-2009.1-GPL/SALOME/SALOME4/Prerequis/debianForSalome-v1_2_0/lib:$LD_LIBRARY_PATH
export LD_LIBRARY_PATH

#?OPT_ENV?


# additionnal paths for Python modules
#PYTHONPATH=/home/borja/SALOME-MECA-2009.1-GPL/aster/public/Python-2.4.4/lib/python2.4/site-packages:$PYTHONPATH
PYTHONPATH=/home/borja/SALOME-MECA-2009.1-GPL/aster/public/Python-2.4.4/lib/python2.4/site-packages:$PYTHONPATH
export PYTHONPATH


PYTHONPATH=/home/borja/SALOME-MECA-2009.1-GPL/SALOME/SALOME4/Prerequis/OmniORB-4.1.0/lib/python2.4/site-packages:$PYTHONPATH
export PYTHONPATH[/code:1]

I didn't experience any change when trying tu run the loadcase, so I edited the "aster_profile.sh" script also, adding the path to my libstdc++5.so

It looks like this
[code:1]# $Id: aster_profile.sh 3255 2008-04-10 17:13:17Z courtois $
#
# aster_profile.sh : initialize the environment for astk services
# (sh, ksh, bash syntax)
#
# If variables are depending on Code_Aster version, use ENV_SH in
# the corresponding 'config.txt' file.
#
# $ASTER_VERSION is now obsolete.
#
ASTER_ROOT=/home/borja/SALOME-MECA-2009.1-GPL/aster
export ASTER_ROOT

LD_LIBRARY_PATH=/home/borja/SALOME-MECA-2009.1-GPL/SALOME/SALOME4/Prerequis/debianForSalome-v1_2_0/lib:$LD_LIBRARY_PATH
export LD_LIBRARY_PATH

# this should not be usefull...
PYTHONHOME=/home/borja/SALOME-MECA-2009.1-GPL/SALOME/SALOME4/Prerequis/Python-2.4.4
export PYTHONHOME

LD_LIBRARY_PATH=/home/borja/SALOME-MECA-2009.1-GPL/SALOME/SALOME4/Prerequis/OmniORB-4.1.0/lib:$LD_LIBRARY_PATH
export LD_LIBRARY_PATH

PATH=/home/borja/SALOME-MECA-2009.1-GPL/SALOME/SALOME4/Prerequis/OmniORB-4.1.0/bin:$PATH
export PATH

PYTHONPATH=/home/borja/SALOME-MECA-2009.1-GPL/SALOME/SALOME4/Prerequis/OmniORB-4.1.0/lib/python2.4/site-packages:$PYTHONPATH
export PYTHONPATH[/code:1]

But anyway, everytime I try to run the loadcase, I'am not able to see the results.

I've attached here the 4 flasheur files for more information.

I will appreciate your comments.

Cheers.<br /><br />Post edited by: Borja AP, at: 2010/01/23 14:13
Attachments:
More
15 years 3 months ago #3770 by kwou
Hoi Borja

can you upload your flesheur files?
Makes it all much easier to understand .... ;-)

Kind regards - kees

Interest: structural mechanics, solar energy (picture at 'my location' shows too little pv panels)

--
kind regards - kees
  • Borja AP
  • Topic Author
  • Offline
  • Junior Member
  • Junior Member
More
15 years 3 months ago #3771 by Borja AP
Replied by Borja AP on topic Re:Problem in Ubuntu 9.10 with SalomeMeca
Kees Wouters wrote:

Hoi Borja

can you upload your flesheur files?
Makes it all much easier to understand .... ;-)

Kind regards - kees


Sorry, I thought they were attached, here they are.

Attachment flasheur_files-c92c58c6f67c2eaf4e58865633a1bf16.zip not found



Thank you in advance for your help.
Attachments:
More
15 years 3 months ago #3774 by kwou
Hoi Borja

the error messages says that you may not reorient the the group ''ee_local_hyp_1''. I guess it has to do with the double use of this group in the command below where you both apply a boundary condition and a load to this plane:

CHAR=AFFE_CHAR_MECA(MODELE=MODE,
DDL_IMPO=(
_F(GROUP_MA='ee_local_hyp_2',
DX=0,
DY=0,
DZ=0,),
_F(GROUP_MA='ee_local_hyp_1',
DX=0,
DY=0,
DZ=0,),),
PRES_REP=(
_F(GROUP_MA='ee_local_hyp_1',
PRES=1000,),),);


Maybe the PRES_REP needs to be applied to another plane, otherwise you have to omit the DDL_IMPO on this group.

The relevant part in the error file is (xxxxx.o&lt;jobnumber&gt;, line 356 and further):

MAIL=MODI_MAILLAGE(reuse = MAIL,
MAILLAGE=MAIL,
INFO=1,
ORIE_PEAU_3D=_F(GROUP_MA('ee_local_hyp_1',)), );

TRAITEMENT DU GROUP_MA: ee_local DE 6 MAILLES
...
! &lt;EXCEPTION&gt; &lt;PREPOST4_97&gt; ! Erreur Utilisateur
! La maille de peau : M455 ne peut pas être réorientée. !
! Car elle est inserrée entre 2 mailles &quot;support&quot; placées de
part et d'autre : M19007 et M26639. !

Also be careful with using names that are too long (more than 8 characters) because they are truncated (line 269 and further):

Le nom de groupe numéro 1 de la famille FAM_-4_ee_local_hyp_2 !
! est trop long. Il sera tronqué à 8 caractères. !
! Le groupe &quot;ee_local_hyp_2&quot; est renommé en &quot;ee_local&quot;.

This may cause the two DDL_IMPO groups to be effectively the same.


Kind regards - kees<br /><br />Post edited by: Kees Wouters, at: 2010/01/24 17:19

Interest: structural mechanics, solar energy (picture at 'my location' shows too little pv panels)

--
kind regards - kees
  • Borja AP
  • Topic Author
  • Offline
  • Junior Member
  • Junior Member
More
15 years 3 months ago #3779 by Borja AP
Replied by Borja AP on topic Re:Problem in Ubuntu 9.10 with SalomeMeca
Thank you Kees for your help.

I understand the mistake is on my model, so as I am a very beginner with Salome-Meca I followed this youtube tutorial:



but in the video the Loadcase is finnishing with a green &quot;ok&quot; mark over the Loadcase icon, not as on my computer. The Post-Process module is not activating automatically neither, so it seems something is going wrong.

I've checked the flasheur files and I think there's something with fortran?

May you take a look? I've attached the flasheur files.

Thank you very much!

Attachment flasheur_files-39fcd16895d7f12a364cce39e87ab459.zip not found

Attachments:
More
15 years 3 months ago #3781 by kwou
Hoi Borja

The flasheur (err, mess) files end with OK, so normally this should be alright. There is only one message in the xxxx.o&lt;jobname&gt; file that hints towards a file that cannot be accessed after the IMPR_RESU command,

IMPR_RESU(INFO=1,....);

MEDouvrir.c [53] : Impossible d'accéder aux fichier :
MEDouvrir.c [54] : nom = &quot;fort.80&quot;

but I guess this is a kind of warning, not an error?
What happens when you click on the POST_PRO module? Does it show the FIELDS part (with displacements and two stresses) or is this not available?

Because, although ASTER doesnot show a green 'okee', it may have complete the run, maybe with a warning.

Sorry, I cannot be of more help at the moment.

Kind regards - kees

Interest: structural mechanics, solar energy (picture at 'my location' shows too little pv panels)

--
kind regards - kees
Moderators: catux
Time to create page: 0.160 seconds
Powered by Kunena Forum