open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi,
I am learning to run a telemac2d program. Thank for PL in the forum, the MPI problem was solved and move on one step. The project was compiled and linked, but the exe file failed to run. I can’t figure out what’s wrong. (Attachment 1)
There are two versions of init_compo.f. The one I used is from Telelmac v7p3 (4k , Attachment 2). The other one is much longer (79k, Attachment 3). I got from my colleague, but it can’t be compiled due to the definition of one or more variables conflict with other subroutines or main program.
Anybody can tell me what might be wrong and find a way out? Thanks in advance.
Eeswcy

Since you are a new user, it would be advisable to start from the latest TELEMAC version that has many fixes.
In your case, the error could originate from mixing subroutines from different telemac versions. You must always take the subroutines from sources and modify them all again if you install another version.
In the latest version, most commonly user modifiable subroutines have a separate user version ('user_' prefix) for modifications only, thus it is less prone to errors.
Please try to use the latest version and post the entire error log. Useat the end of your command to output stdout and stderr logs to a file.
Best Regards,
Costas

Thanks, Coasta. The latest version I have is v8p0. I will check all the subroutines to make sure they are from the same version.
Best regards,
Eeswcy

Hi,
Costas, I run the telemac2d using v8p0. All the f files are from v8v0 except “init_compo_sisyphe.f”. I can’t find it in my v8p0 and got it from somewhere else. The result seems about the same. An exe file “out_init_compo.exe” is produced (attachment 1) but can’t run (attachment 2). The details are in attached file “output_v8p0_1.txt”. Environment variable setting is also attached for information.
I appreciate it if anyone gives suggestions directing me what to do next. So far I am quite confused with Telemac and always get lost..
Thank you in advance.
Eeswcy

Hi,
The source files is named 'init_compo.f' and you should be able to find it unless it got deleted. If you are certain that the one you have is not different from the one in the source, set DEBUGGER = 1 in your steering file, run the computation again and post the output.
Costas

Please post your steering files as well. It appears that in the OPTION FOR TIDAL BOUNDARY CONDITIONS you should give one value for every open boundary, 5 in your case. For example:

Hi,
Costas, I add “DEBUGGER = 1”in the steering file ldyt2d.cas, It is attached with the results of both v7p3 and v8p0. It seems at least something wrong with MPI as indicated in the last line:
|~~~~~~~~~~~~~~~~~~
|application called MPI_Abort(MPI_COMM_WORLD, 2) - process 0
|~~~~~~~~~~~~~~~~~~
Thank you.
Eeswcy

Hello Eeswcy,

I think you should read the advice written by Costas again, it should solve your error message.

Be aware that the error message to read and that may help you, is not always the last line but can be a few lines above.

Hope this helps,

Chi-Tuan

Hello again,

Just 2 remarks about your steering file:
- COEFFICIENT TO CALIBRATE TIDAL VELOCITIES =3. this value is quite big. It is expected to have a value around 1. Needing 3 should be investigated more.
- MINOR CONSTITUENTS INFERENCE =true, this value is not recommended to be turn on to true, but perhaps you choose it on purpose?

Hope this helps,

Chi-Tuan