CAELinux 2013: development and testing

amd64 only means it's built for 64 Bit architectrue.
The Name comes from the fact, that the original specification was from AMD.
There are no compability issues using it on an Intel processor.

Greetings back from Austria

I installed Code_Aster 11.3-MPI and Salome-Meca 2013.1 with these instructions (www.code-aster.org/forum2/viewtopic.php?id=18691) on Ubuntu 12.04
However when i run the short test list with (/opt/aster113/bin/as_run astout.export) the following result is returned in NOOK (/opt/aster113/testing/resu_test/)

forma10a (18/42) <F>_ERROR 1.11 0.37 1.48 5.53
forma01c (19/42) <F>_ERROR 1.09 0.30 1.39 4.39
forma40b (22/42) <F>_ERROR 1.11 0.39 1.50 1.63
forma01a (23/42) <F>_ERROR 1.02 0.40 1.42 2.56
forma10b (25/42) <F>_ERROR 1.17 0.27 1.44 2.08
forma01b (27/42) <F>_ERROR 1.04 0.33 1.37 1.51
forma02c (42/42) <F>_ERROR 1.11 0.37 1.48 1.65
forma11b (35/42) <F>_ERROR 46.93 0.66 47.59 48.31

When i look in the .mess files of these tests i can see that too little memory is assigned.

Is this normal behaviour?

#

---

!

---

!
! <F> <JEVEUX1_71> !
! !
! La mémoire totale de 256.00 Mo allouée à l'étude est insuffisante, il est nécessaire !
! de disposer d'au moins 331.57 Mo uniquement pour démarrer l'exécution. !
! !
! !
! Cette erreur est fatale. Le code s'arrête. !
!

---

!


La mémoire demandée au lancement est sous estimée, elle est de 256.00 Mo.
Le pic mémoire utilisée est de 331.58 Mo.

End of the Code_Aster execution - MPI exits with errors

EXECUTION_CODE_ASTER_EXIT_0145-2556=6

<I>_EXIT_CODE = 1

Hello,

I added the repos successfully and downloaded and installed code aster 11.3 mpi. When I attempt to run a simulation, I get exit code 255 before it starts. I cannot find any information as to what this error code means. Any Guidance is greatly appreciated.

FYI: Im on ubuntu 12.04. I have existing installations of MPI code aster that work fine.

EDIT: OOPS I made a mistake when adding this to my ASTK list of version. It works fine now!!!

Has anyone tried to run a code aster simulation using the PETSc solver in the build from the repos?

On the code aster forums, TdS says that PETSc 3.2p7 is supported by code aster, but the current build in the repos uses 3.4.

Actually, I cant even get PETSc to start solving. I get the error message shown in the attached screenshot.

EDIT: I cannot attach the screenshot for some reason, but you can see it here:
www.code-aster.org/forum2/viewtopic.php?id=18919

(No screenshot is attached)

However i experience problems when i try to solve large linear elastic simulations with Petsc on Code-Aster11.3-MPI (installed from the sourceforce repo).
The following error shows up when running the simulation on 1 core:
[METIS Fatal Error] ***Memory allocation failed for CompressGraph: gdata. Requested size: -1029597756 bytes

I tried to compile an parallel version of 11.3 with Petsc 3.2-p7 with the build script in this topic.
However when the compilation was finished i could not run the forma1a.export test sum.

(edit: the reply and reply to topic button are confusing.)

When I try to access the repository I get the following error(s). I tried to add a ticket at sourceforge, but could not.