open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi All,

As described in a recent forum post to the Tomawac group, I have been attempting a 3-way coupling for a large coastal domain, with tides, waves and sediments including bed-roughness feedback, using v7p2r0 on HPC Wales.

A run of, for example, 12 hr duration ends correctly and the solution files for T2D, Tom & Sis are fine.
However, if I try again with Sisyphe now initialised with the 12hr solution file, as follows:

COMPUTATION CONTINUED = YES
PREVIOUS SEDIMENTOLOGICAL COMPUTATION FILE = NorthWales_sis_12h.slf

the run fails with the error below. I have changed nothing else (apart from giving a new results file name). It would seem that a conflict arises between Sisyphe and the global wave results file (WACRBI), in much the same way as described in my Tomawac post.

Does anyone know whether it should be possible in a 3-way coupled run to initialise Sisyphe in the way described above?

Thanks, Alan D.

$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$

Error with 1 processor (same error occurs with 16 processors)

=========================================================================
TEMPS : 0.0000 SECONDES 0IEME ITERATION

---

_____________
runcode::main:
:
|runCode: Fail to run
|mpirun /space02/users/a.g.davies/v7p2/NorthWales_3waycoup/NorthWales_t2d.cas_2018-01-16-11h43min30s/out_bord_agd_NEW
|~~~~~~~~~~~~~~~~~~
|forrtl: severe (257): formatted I/O to unit open for unformatted transfers, unit 17, file /space02/users/a.g.davies/v7p2/NorthWales_3waycoup/NorthWales_t2d.cas_2018-01-16-11h43min30s/WACRBI

Hello Alan

Can you run it in serial mode with keyword DEBUGGER=1 in both Telemac-2d and sishyphe steering files.

kind regards

Riadh

Dear Riadh,

Sorry I should have set DEBUGGER = 1 previously. Below I give first the error message with DEBUGGER = 1 in T2D and Sis and then, second, with DEBUGGER = 1 additionally in Tomawac.

Incidentally, this type of error arises both in v7p2r0 and also v7p2r3 which we have also tested.

Regards, Alan

Error in serial mode (1 processor) with DEBUGGER = 1 in T2D & Sis:

$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
BACK FROM CORFON
CALLING CONDIN
BACK FROM CONDIN
CCT SYNCHRONISATION AVANT
CCT SYNCHRONISATION APRES
CALLING PROPIN
BACK FROM PROPIN
CALLING FRICTION_CHOICE
BACK FROM FRICTION_CHOICE

================================================================================
ITERATION 0 TIME: 0.0000 S
CALLING PRERES_TELEMAC2D
BACK FROM PRERES_TELEMAC2D
CALLING DESIMP
BACK FROM DESIMP
CALLING HPROPA
BACK FROM HPROPA
CALLING DE CHPCONV
BACK FROM CHPCONV
CALLING BILAN
BACK FROM BILAN
TELEMAC2D COUPLED WITH:
TOMAWAC,SISYPHE

TELEMAC-2D: INTERNAL COUPLING WITH TOMAWAC
PREMIER APPEL DE TOMAWAC
INBIEF (BIEF): NOT A VECTOR MACHINE (ACCORDING TO YOUR DATA)
STRCHE (BIEF): NO MODIFICATION OF FRICTION

================================================================================
TEMPS : 0.0000 SECONDES 0IEME ITERATION

---

_____________
runcode::main:
:
|runCode: Fail to run
|mpirun /space02/users/a.g.davies/v7p2/NorthWales_3waycoup/NorthWales_t2d.cas_2018-01-16-18h00min59s/out_bord_agd_NEW
|~~~~~~~~~~~~~~~~~~
|forrtl: severe (257): formatted I/O to unit open for unformatted transfers, unit 17, file /space02/users/a.g.davies/v7p2/NorthWales_3waycoup/NorthWales_t2d.cas_2018-01-16-18h00min59s/WACRBI
|Image PC Routine Line Source
|out_bord_agd_NEW 0000000000A9A169 Unknown Unknown Unknown
|out_bord_agd_NEW 0000000000AE4D3C Unknown Unknown Unknown
|out_bord_agd_NEW 00000000005E32F3 Unknown Unknown Unknown
|out_bord_agd_NEW 00000000005B514D Unknown Unknown Unknown
|out_bord_agd_NEW 0000000000442402 Unknown Unknown Unknown
|out_bord_agd_NEW 0000000000436890 Unknown Unknown Unknown
|out_bord_agd_NEW 0000000000404B2E Unknown Unknown Unknown
|libc.so.6 000000337621ED1D Unknown Unknown Unknown
|out_bord_agd_NEW 0000000000404A39 Unknown Unknown Unknown
|~~~~~~~~~~~~~~~~~~

$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$

Error with DEBUGGER = 1 additionally in Tom:

$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$

================================================================================
ITERATION 0 TIME: 0.0000 S
CALLING PRERES_TELEMAC2D
BACK FROM PRERES_TELEMAC2D
CALLING DESIMP
BACK FROM DESIMP
CALLING HPROPA
BACK FROM HPROPA
CALLING DE CHPCONV
BACK FROM CHPCONV
CALLING BILAN
BACK FROM BILAN
TELEMAC2D COUPLED WITH:
TOMAWAC,SISYPHE

TELEMAC-2D: INTERNAL COUPLING WITH TOMAWAC
PREMIER APPEL DE TOMAWAC
APPEL DE LECLIM POUR MESH2D
SORTIE DE LECLIM
APPEL DE INBIEF POUR MESH2D
INBIEF (BIEF): NOT A VECTOR MACHINE (ACCORDING TO YOUR DATA)
SORTIE DE INBIEF
APPEL DE LECLIM POUR MESH3D
SORTIE DE LECLIM
APPEL DE INBIEF POUR MESH3D
RETOUR DE INBIEF
APPEL DE FONSTR
STRCHE (BIEF): NO MODIFICATION OF FRICTION

RETOUR DE FONSTR
APPEL DE WRITE_HEADER
RETOUR DE WRITE_HEADER
APPEL DE WRITE_MESH
RETOUR DE WRITE_MESH
APPEL DE CONDIW
RETOUR DE CONDIW
APPEL DE INITAB
RETOUR DE INITAB
APPEL DE IMPR

================================================================================
TEMPS : 0.0000 SECONDES 0IEME ITERATION

---

RETOUR DE IMPR
APPEL DE LIMWAC
RETOUR DE LIMWAC
APPEL DE INIPHY
RETOUR DE INIPHY
APPEL DE ECRETE
RETOUR DE ECRETE
APPEL DE PROXIM
RETOUR DE PROXIM
APPEL DE DUMP2D
RETOUR DE DUMP2D
APPEL DE BIEF_DESIMP
RETOUR DE BIEF_DESIMP
APPEL DE ECRSPE
_____________
runcode::main:
:
|runCode: Fail to run
|mpirun /space02/users/a.g.davies/v7p2/NorthWales_3waycoup/NorthWales_t2d.cas_2018-01-16-18h04min06s/out_bord_agd_NEW
|~~~~~~~~~~~~~~~~~~
|forrtl: severe (257): formatted I/O to unit open for unformatted transfers, unit 17, file /space02/users/a.g.davies/v7p2/NorthWales_3waycoup/NorthWales_t2d.cas_2018-01-16-18h04min06s/WACRBI
|Image PC Routine Line Source
|out_bord_agd_NEW 0000000000A9A169 Unknown Unknown Unknown
|out_bord_agd_NEW 0000000000AE4D3C Unknown Unknown Unknown
|out_bord_agd_NEW 00000000005E32F3 Unknown Unknown Unknown
|out_bord_agd_NEW 00000000005B514D Unknown Unknown Unknown
|out_bord_agd_NEW 0000000000442402 Unknown Unknown Unknown
|out_bord_agd_NEW 0000000000436890 Unknown Unknown Unknown
|out_bord_agd_NEW 0000000000404B2E Unknown Unknown Unknown
|libc.so.6 000000337621ED1D Unknown Unknown Unknown
|out_bord_agd_NEW 0000000000404A39 Unknown Unknown Unknown
|~~~~~~~~~~~~~~~~~~
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$

Hello Alan

hmmm, this will not help us a lot.
I tried with validation case LITTORAL in examples/sisyphe by adding a previous cmputation file and it works well.
- The question is maybe, is there any problem with your sisyphe result file. try to view it and check if everything is OK
- if it is OK, add subroutine sisyphe.F in your fortran file and add a print at line 670, to view the index of SISPRE, which is the logic unit of the previous computation file.
- it seems to be the logic unit number 17 who causes the issues, just verify that it is not the one corresponding to the previous computation file.

If you do not realize any advances, please send me your model (if possible). I will try to check it up.

with my kind regards

Riadh

Hello Riadh,

Thanks for your response.

The Sisyphe results file seems fine when I look through it in BlueKenue.

So I have attempted the suggested print in sisyphe.F, both at line 670 and also earlier. However there is no indication from this that the run even enters sisyphe.F.

So I have tried the same kind of check in wac.F. Just before the first call to ecrspe.f, I find that:
WACRES= 4
WACSPE= 17
WACLEO= 5
So the apparent conflict involving Sisyphe on unit 17 may be originating in Tomawac through WACSPE??

It would be great if you have some (simple!) suggestion how to get around this problem.

Thanks again, Alan

Hello Alan

this was a bug that was fixed since.
You have 2 choices
1- give a name for the 1D specta file (keyword 1D SPECTRA RESULTS FILE =...)
2- in subroutine ecrspe.f before line 248 add the following if statement
IF(WAC_FILES(NSPE)%NAME.NE.'')THEN

If you need this result file, you need to use the first option otherwise use the second one.

I hope that this helps.

with my kind regards

Riadh

Thanks very much for this steer, Riadh.

I'll try what you have suggested (probably option 2) and let you know the outcome.

I re-ran my intial coupled run without a continuation file and found that SISPRE was on unit 5. I noticed however that a file fort.17 was opened in the working directory, but that this remained empty (0KB) during the run.

Hopefully your fix will work!

Kind regards, Alan

Hello again Riadh,

Can you guide me past this error? I need to define an explicit type to allow this IF statement in ecrspe.f

Thanks, Alan

t2dfort.f(755): error #6404: This name does not have a type, and must have an explicit type. [WAC_FILES]
IF(WAC_FILES(NSPE)%NAME.NE.'')THEN

---

^
t2dfort.f(840): remark #8291: Recommended relationship between field width 'W' and the number of fractional digits 'D' in this edit descriptor is 'W>=D+7'.
WRITE(NSPE,'(100(E10.4,2X))') FREQ(JF),

---

^
t2dfort.f(862): remark #8291: Recommended relationship between field width 'W' and the number of fractional digits 'D' in this edit descriptor is 'W>=D+7'.
WRITE(NSPE,'(100(E10.4,2X))') FREQ(JF),

---

^
t2dfort.f(683): error #6321: An unterminated block exists.
IF(DEBRES) THEN

Hi Riadh,

I should add to my previous post that your Choice 1 DOES work and allows Sisyphe to be initialised from a previous solution. So many thanks again for that suggestion! There is still one outstanding issue about continued computations however, which I will refer to in a further post.

All the best, Alan

Hello Alan
Glad to see that it is working.
Sorry I forget to tell you to add the declaration of wac_files.
Actually, the local name of this file is TISPEF, so what you can do is just change location of line 264 to line 247. This should fix everything properly without any further changes.

with my best regards

Riadh