open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello again,

I think your first problem is that you do not use the keyword you think.
In your steering file, you use the BINARY ATMOSPHERIC DATA FILE (associated to the name of variable in the code T3ATMB as written in the log file), giving I suppose an ASCII file (text file), not the expected format. That may lead to your error.

Anyway, a few advice:
- if you have an error in parallel mode, have a look in the temporary directory to the PE0*.*LOG files which are the output log files for every subdomain of your parallel run, in particular the one(s) with the biggest or smallest size. Usually, there is an error message that you cannot read in the one at the screen you look, if not already written,
- run your simulation in sequential (= --ncsize = 0 or 1), you will not have any error message with partel and you should see error message if it is not a problem with parallel computation,
- use the latest release of TELEMAC because:
- new features have been developed,
- some bugs have been fixed that exist in previous releases,
- in particular if you want to use any coupling with WAQTEL (BIOMAS, EUTRO, MICROPOL, O2, THERMIC, AED2), there is at least one fixed bug for every module, I strongly recommend to ban release before release 8.1 if coupling TELEMAC-2D or TELEMAC-3D with WAQTEL,
- last but not least, we do not assist people using releases other than trunk or the latest (today is v8p1r1). I can see that you still use an old release.

Chi-Tuan

Dear Jacques:

The things that I modeled

- test case with hydro alone, and not dry zones => OK
- test case with hydro and temperature (as active tracer?) => OK
- test case with hydro, temperature and waqtel (heat exchange at the atmosphere=1 and 2) without dry zones => NOK Nan error.

Thank you very much

Dear Juli,

Thank you.
And for the first question (the most important).
Do you use the v8p1r1?

There are lot of fixes for water quality and heat exchange with the atmosphere in v8p1r1:
- for athmosex= 1 (with or without dryzone)
- in temperature balance
- in waqtel

To be sure to do not try to fix a bug already fixed...

Best regards,
Jacques

Dear Jacques
I am in the process to change the version.
Thank you very much

Hello everybody.
I change Telemac version, I am using v8p1r1.
1. I ran Mesh type 1,2 and 3.
2. I ran hydrodynamics alone -> Ok
3. I ran temperature as an active tracer -> Ok
4. I ran temperature with initial conditions modifying the FORTRAN file-> Ok
5. I ran temperature with IC and coupling with WAQTEL and I got this:

********************************************
* LECDON: *
* BEFORE CALLING DAMOCLES *
********************************************

application called MPI_Abort(MPI_COMM_WORLD, 2) - process 8
application called MPI_Abort(MPI_COMM_WORLD, 2) - process 9
application called MPI_Abort(MPI_COMM_WORLD, 2) - process 13
application called MPI_Abort(MPI_COMM_WORLD, 2) - process 3
_____________
runcode::main:
:
|runCode: Fail to run
|mpiexec -wdir /work/jalzate/Rejet_v8/cas_rejet_v8-coupled.cas_2020-06-16-22h05min48s -n 16 /work/jalzate/Rejet_v 8/cas_rejet_v8-coupled.cas_2020-06-16-22h05min48s/out_condim_v8
|~~~~~~~~~~~~~~~~~~
|application called MPI_Abort(MPI_COMM_WORLD, 2) - process 3

My files:
Stearing file.

FORTRAN file.


Even with version 8 I could not do it the coupling.
I am so sorry, any suggestion?
Thank you very much

Hello,

I wonder if you have read the TELEMAC-3D user manual provided with release v8p1r1. Defining the mesh transformation when it is not the classical sigma (=1) + initial conditions are written with the recommended user subroutines (sections 4.1, 4.2.2, 7.1).

Just a reminder, you have all the documentation available in the documentation folder (e.g. documentation/telemac3d/user/telemac3d_user_v8p1.pdf) or on the wiki: wiki.opentelemac.org/doku.php?id=documentation_v8p1r1
In particular, TELEMAC-2D and TELEMAC-3D user manuals have been updated for v8p1r1.

I think I have already told you to change IPLAN.EQ.10.D0 to IPLAN.EQ.10 in another post...

Anyway, have a look at the examples of user_mesh_transf and user_condi3d_trac subroutines in the TELEMAC-3D examples. In particular, do not assign values for planes 1 and NPLAN for ZPLANE%R array.

And if you have a problem with a parallel run, start to run it in sequential!

Chi-Tuan