open TELEMAC-MASCARET The mathematically superior suite of solvers

I agree, there is a problem
But I don't understand how you could obtain only 1 file when splitting in 8.
Could you try to use the running script in parallel but with --split option to only run the preprocessing step and then give us the feedback with the partel log files

Regards

thank you for your help. I think I found the reason:

... partitioning base files (geo, conlim, sections, zones and weirs)
+> /home/hafedh/v8p3r1/builds/debgfopenmpi/bin/partel < partel_T2DGEO.par >> partel_T2DGEO.log

... splitting / copying other input files
partitioning: T2ATMB
+> /home/hafedh/v8p3r1/builds/debgfopenmpi/bin/partel < partel_T2DGEO.par >> partel_T2ATMB.log

for the partition it uses the same file as input (partel_T2DGEO.par) for bith the GEO and Atmospheric files.

In fact, using python3 instead of python2.7, I got a problem with the the configuration file. PARTEL.PAR which is not created in the working directory. So I replaced it by partel_T2DGEO.par and this was working as far as I did not use input data in format file.

[debgfopenmpi]
#
#par_cmdexec: <config>/partel < PARTEL.PAR >> <partel.log>
par_cmdexec: <config>/partel < partel_T2DGEO.par >> <partel.log>

I think some is wrong with my configure file or with python3. How should I do to get:
partitioning: T2ATMB
+> /home/hafedh/v8p3r1/builds/debgfopenmpi/bin/partel < partel_T2ATMB.par >> partel_T2ATMB.log
rather than,
partitioning: T2ATMB
+> /home/hafedh/v8p3r1/builds/debgfopenmpi/bin/partel < partel_T2DGEO.par >> partel_T2ATMB.log

putting this:
par_cmdexec: <config>/partel < <partel.par> >> <partel.log>
the problem is solved.
many thanks