Coupled three meshes

[stage75]

Hi,

I get tocouple two areas of meshes using the USSATC like:
SRC.1 :

Code: Select all

numsat = -1
namsat = ' '
iwarns = 1
nbcsat = 1

do ii = 1, nbcsat
  if (ii .eq. 1) then

    call defsat(numsat, namsat, 'all[]', ' ', ' ', 'coupled1', iwarns)
  endif
enddo

SRC.2:

Code: Select all

numsat = -1
namsat = ' '
iwarns = 1
nbcsat = 1

do ii = 1, nbcsat
call defsat(numsat, namsat, 'all[]', ' ', ' ', 'coupled2', iwarns)
  endif
enddo

But I can't couple three areas SRC.1, SRC.2 and SRC.3, I modefied USSATC like this but its not work :

SRC.1

Code: Select all

numsat = -1
iwarns = 1
nbcsat = 3

do ii = 1, nbcsat
 numsat = 1

    call defsat(numsat, namsat, 'all[]', ' ', ' ', 'coupled1', iwarns)
  endif
enddo

SRC.2

Code: Select all

numsat = -1
iwarns = 1
nbcsat = 3
do ii = 1, nbcsat
 numsat = 1

    call defsat(numsat, namsat, 'all[]', ' ', ' ', 'couple0', iwarns)

else if (ii .eq. 2) then
numsat = 3

call defsat(numsat, namsat, 'all[]', ' ', ' ', 'couple1', iwarns)
  endif
enddo

and SRC.3

Code: Select all

numsat = -1
iwarns = 1
nbcsat = 3
do ii = 1, nbcsat
 if (ii .eq. 2) then

    numsat = 3

    call defsat(numsat, namsat, 'all[]', ' ', ' ', 'couple2', iwarns)
  endif
enddo

Thanks for help

[stage75]

...please there is someone who can correct me the ; numsat, namsat, iwarns and nbcsat For this coupling three meshes operates correctly

Thanks

[Yvan Fournier]

Hello,

Do you get interlock or a crash ? If you get a crash, you should know what to post by now...
If you get an interlock (i.e. code does not crash but hangs), you may at least get tracebacks when killing the job (using CTRL-C or kill -15, not kill -9), which could help.

Best regards,

Yvan

[stage75]

I get this message :

Code: Select all

/home/itouche/Téléchargements/tmp/installer/ncs-2.0.4/src/base/cs_sat_coupling.c:2076: Erreur fatale.

Au moins 1 couplage Code_Saturne a été défini pour lequel
aucune communication avec une instance de Code_Saturne n'est possible.

I think I can not couple correctly three meshes!

[Yvan Fournier]

Hello,

If you have read http://code-saturne.org/forum/viewtopic.php?f=2&t=7, you should have provided your "listing" file. Otherwise, please read it.

In your sources, I believe I have found a typo, but the "coupling" part of the "listing" would confirm this, so I'll leave the rest of the exercise to you.

Best regards,

Yvan

[stage75]

Thank's Yvan ,

I'll check when the mistake in the coupling and I come back here

[stage75]

...Frankly I tried the coupling between the three areas but don't work!

coupled1: the common surface between area 1 and area 2
couple0 : the common surface between area 2 and area 1
couple1: is the common surface between area 2 and area 3
couple2: is the common surface between area 3 and area 2

Do you have an idea please !

[Yvan Fournier]

Hello,

You still did not read my post or follow the Forum recommendations. If you do not provide enough debugging information, nobody can help you.

Regards,

Yvan

[stage75]

Yes Yvan I read your post and the link . Despite the link I have trouble achieving this coupling of three areas!

attached the listing files

Thanks

[Yvan Fournier]

Hello,

When you get things like this (from second listing, but others are similar) :

Applications accessibles grâce à MPI :
--------------------------------------
0; type : "Code_Saturne 2.0.4"
nom de cas : ""
rang racine : 0 ; nb_rangs : 1

1; type : "Code_Saturne 2.0.4" (this instance)
nom de cas : ""
rang racine : 1 ; nb_rangs : 1

2; type : "Code_Saturne 2.0.4"
nom de cas : ""
rang racine : 2 ; nb_rangs : 1

Couplage Code_Saturne :
id de couplage : 0
nom local : "�.���"
nom d'application distante : ""
numéro local : 1
numéro d'application MPI : 1
rang racine MPI : 1
nombre de rangs MPI : 1

Couplage Code_Saturne 0 : initialisation de la communication MPI ...

You can see that the problem is related to coupling names not being defined (with 2 couplings only, the names are not used, as there can be only one match, so this explains why things worked with 2 couplings).

Now things are easy: for each coupling, you need to define numsat, to be consistent with the name of the case (set namsat = " "), as I am not sure assigning a name is possible with version 2.0 (it is used in 2.x, x >0, and 3.0).

Regards,

Yvan

Also,