open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello,
when I tried starting my simulation it was aborted with the following message:

C:\opentelemac-mascaret\v6p3\scripts\python27>runcode.py telemac2d -c wing64mpi
TEST.cas


Loading Options and Configurations
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

... parsing configuration file: C:\opentelemac-mascaret\v6p3\configs\systel.cfg


Running your CAS file for:
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

+> configuration: wing64mpi
+> root: C:\opentelemac-mascaret\v6p3
+> version v6p3
+> options: parallel mpi


~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~



Running TEST.cas with telemac2d under C:\opentelemac-mascaret\v6p3\scripts\pytho
n27
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

... reading module dictionary
... running in English with TEST.cas
copying: telemac2d.dico C:\opentelemac-mascaret\v6p3\scripts\python27\TEST.
cas_2014-02-04-09h59min05s\T2DDICO
copying: condInit.ser C:\opentelemac-mascaret\v6p3\scripts\python27\TEST.ca
s_2014-02-04-09h59min05s\T2DPRE
copying: Netz ohne PH.slf C:\opentelemac-mascaret\v6p3\scripts\python27\TES
T.cas_2014-02-04-09h59min05s\T2DGEO
re-writing: C:\opentelemac-mascaret\v6p3\scripts\python27\TEST.cas_2014-02-04-
09h59min05s\T2DCAS
copying: Boundaries_ohne PH.cli C:\opentelemac-mascaret\v6p3\scripts\python
27\TEST.cas_2014-02-04-09h59min05s\T2DCLI
partitioning: T2DPRE
+> C:\opentelemac-mascaret\v6p3\builds\wing64mpi\bin\partel.exe < PARTEL.PA
R >> partel_T2DPRE.log
partitioning: T2DGEO
+> C:\opentelemac-mascaret\v6p3\builds\wing64mpi\bin\partel.exe < PARTEL.PA
R >> partel_T2DGEO.log


Running your simulation :
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

C:\opentelemac-mascaret\mpich2\bin\mpiexec.exe -wdir C:\opentelemac-mascaret\v6p
3\scripts\python27\TEST.cas_2014-02-04-09h59min05s -n 7 C:\opentelemac-mascaret\
v6p3\scripts\python27\TEST.cas_2014-02-04-09h59min05s\out_telemac2d.exe


MASTER PROCESSOR NUMBER 0 OF THE GROUP OF 7
EXECUTABLE FILE: C:\opentelemac-mascaret\v6p3\scripts\python27\TEST.cas_2014-02
-04-09h59min05s\A.EXE
BARRIER PASSED

LISTING OF TELEMAC-2D

---

---

TTTTT EEEEE L EEEEE M M AAAAA CCCCC
T E L E MM MM A A C
T EEE L EEE M M M AAAAA C
T E L E M M A A C
T EEEEE LLLLL EEEEE M M A A CCCCC

2D VERSION 6.3 FORTRAN 90
WITH SEVERAL TRACERS
COUPLED WITH SISYPHE AND TOMAWAC

********************************************
* LECDON: *
* AFTER CALLING DAMOCLES *
* CHECKING OF DATA READ *
* IN THE STEERING FILE *
********************************************

EXITING LECDON. NAME OF THE STUDY:
TEST

OPENING FILES FOR TELEMAC2D

*****************************
* MEMORY ORGANIZATION *
*****************************

READGEO1: TITLE= newSelafin

NUMBER OF ELEMENTS: 9207
NUMBER OF POINTS: 4735
MXPTEL (BIEF) : MAXIMUM NUMBER OF ELEMENTS AROUND A POI
NT: 8
MAXIMUM NUMBER OF POINTS AROUND A POINT
: 9
SEGBOR (BIEF) : NUMBER OF BOUNDARY SEGMENTS = 261
INCLUDING THOSE DUE TO DOMAIN DECOMPOSITION
CORRXY (BIEF):NO MODIFICATION OF COORDINATES

MESH: MESH ALLOCATED

*************************************
* END OF MEMORY ORGANIZATION: *
*************************************

INBIEF (BIEF): NOT A VECTOR MACHINE (ACCORDING TO YOUR DATA)

job aborted:
rank: node: exit code[: error message]
0: Sarah-HP: 123
1: Sarah-HP: 123
2: Sarah-HP: 123
3: Sarah-HP: 123
4: Sarah-HP: 123
5: Sarah-HP: 0: process 5 exited without calling finalize
6: Sarah-HP: 123
_____________
runcode::main:
C:\opentelemac-mascaret\v6p3\scripts\python27:
|runCode: Fail to run
|C:\opentelemac-mascaret\mpich2\bin\mpiexec.exe -wdir C:\opentelemac-mascaret
\v6p3\scripts\python27\TEST.cas_2014-02-04-09h59min05s -n 7 C:\opentelemac-masca
ret\v6p3\scripts\python27\TEST.cas_2014-02-04-09h59min05s\out_telemac2d.exe
|~~~~~~~~~~~~~~~~~~
|
|~~~~~~~~~~~~~~~~~~

The PE00006-00005.LOG ends with the message:

INBIEF (BIEF): NOT A VECTOR MACHINE (ACCORDING TO YOUR DATA)
GEOELT: ELEMENT 596 : NEGATIVE DETERMINANT

What's the reason and how can I terminate this error?
I already read somthing in the topic: www.opentelemac.org/index.php/kunena/16-...imitstart=0&start=10

But to be honest I didn't really understand what to do!?

As you can see I'm using the wing64mpi configuration for parallel simulations. The *.log-, *.cfg-, *.cas-, *.slf- and *.cli-file is in the Attachments.

I would appreciate any kind of help.

Cheers

Sarah

Hello,

It looks like you have elements which are not well oriented in your mesh. The 3 points of a triangle must be given in a counterclockwise sense.

Regards,

Jean-Michel Hervouet

Ok,
do I have to fix this in BlueKenue? How could this happen? Did i do anything wrong while creating the mesh or do I need to change values of keywords in Fudaa-Prepro?

Cheers

Sarah

Hello,

At least in Blue Kenue there is a mesh analysis saying how many clockwise elements you have, but it should not give such elements. In your case however you have very large coordinates that challenge single precision, you should probably skip the coordinates so that the origin is in your domain.

Regards,

Jean-Michel Hervouet

Hello jean Michel,
Thank you for your help so far. The Mesh Analyse in BlueKenue says:

ElementCount: 64464
Interior Elements: 63037
Normal Edge Elements: 1426
Over constrained Elements: 1

Element Node ordering:
Counter-Clockwise Elements: 64464
Clockwise Elements: 0

What does the Over constrained Element mean. Is maybe this the reason for the error?

I just changed my ORIGIN COORDINATES to 710089,9612689. Is that what you mean?
Unfortunately there is still the same error after this change.

Greetings

Sarah

Hello,

Overconstrained elements are elements with their 3 points on the boundary (this is not recommended but it can work, problems may be not well posed in that case, e.g. if you prescribe the velocity of the 3 points there is no degree of freedom for ensuring continuity). It is not your problem here.

If Blue Kenue says you have no clockwise element and Telemac finds one, it must be the truncation of single precision, this is why shifting the coordinates so that they are smaller should solve the problem. Then Telemac tells you what is the guilty element and maybe you can find it.

Blue Kenue can also give you a double precision selafin file for your geometry, and then stating GEOMETRY FILE FORMAT = 'SERAFIND' you would have no truncation error.

With best regards,

Jean-Michel

Thank you Jean-Michel,
I found the Element but I don't know what to do with it!?

Can you please tell me how i can create a double precision selafin file from BlueKenue!?

Cheers
Sarah

Hello,

Actually I don't know, this is a question for Martin Serrer, to put in Blue Kenue section...

One colleague tells me that he got such a problem and that the solution, even in single precision, is to shift the coordinates to have the smallest numbers possible.

In the next version I'll swap the incorrectly oriented elements, but if it is due to truncation errors it will probably not solve the problem.

Regards,

Jean-Michel Hervouet

What you mean with the smallest numbers available? Are we talking about the keyword ORIGIN COORDINATES?

Hello,

No, you probably need to change your coordinates even in Blue Kenue to get a correct single precision mesh. But well, I am certainly not an expert here, others may object...

JMH