open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi, I'm looking at this for PL

SELAFIN INPUT NAME <INPUT_NAME>:
INPUT: T2DGEO

BOUNDARY CONDITIONS FILE NAME :
INPUT: SERAFIN
FILE DOES NOT EXIST: SERAFIN



PLANTE: PROGRAM STOPPED AFTER AN ERROR
STOP 2

Backtrace for this error:
+ function plante_ (0x42413E)
+ in the main program
from file homere_partel.f
+ /lib64/libc.so.6(__libc_start_main+0xfd) [0x3a8ce1ed5d]

Here is the error.
+> /gpfs/stfc/local/HCP012/jjd66/cxh71-jjd66/v7p1r0/builds/encore/bin/partel < PARTEL.PAR >> partel_T2DGEO.log
/bin/sh: line 1: 11862 Killed /gpfs/stfc/local/HCP012/jjd66/cxh71-jjd66/v7p1r0/builds/encore/bin/partel < PARTEL.PAR >> partel_T2DGEO.log
runPartition:
|runPARTEL: Could not split your file T2DGEO (runcode=137) with the error as follows:
|
|... The following command failed for the reason above (or below)
|/gpfs/stfc/local/HCP012/jjd66/cxh71-jjd66/v7p1r0/builds/encore/bin/partel < PARTEL.PAR >> partel_T2DGEO.log
|
| You may have forgotten to compile PARTEL with the appropriate compiler directive
| (add -DHAVE_MPI to your cmd_obj in your configuration file).
|
|Here is the log:
|


Here is the config line
cmd_obj: /gpfs/stfc/local/apps/gcc/gcc/4.9.1/bin/gfortran -c -O3 -fconvert=big-endian -frecord-marker=4 -DHAVE_MPI <mods> <incs> <f95name>

If I run the test command again

/gpfs/stfc/local/HCP012/jjd66/cxh71-jjd66/v7p1r0/builds/encore/bin/partel < PARTEL.PAR

It appears to work until the following is output

"BEGIN PARTITIONING WITH METIS
Killed"

I'm using metis 5.1.0 as follows.

module load gcc/4.9.1 openmpi-gcc cmake/2.8.12 python/2.7.8
wget glaros.dtc.umn.edu/gkhome/fetch/sw/metis/metis-5.1.0.tar.gz
gunzip metis-5.1.0.tar.gz
tar -xvf metis-5.1.0.tar
rm -f metis-5.1.0.tar
cd ~/metis-5.1.0
vi include/metis.h # to be 64 bit
make config
make

Then adding this into the config file

libs_partel: /gpfs/stfc/local/HCP012/jjd66/cxh71-jjd66/metis-5.1.0/build/Linux-x86_64/libmetis/libmetis.a

Hi yugi,

Unfortunately i am going on holidays for a couple of weeks but chardacre took over the troubleshooting.

Thanks for all the help!

Best regards
PL

Hi,

Could you post the file partel.par.
It seems like you are missing the boundary condition file.

Hope it helps.

Many thanks for your help

-bash-4.1$ cat PARTEL.PAR
T2DGEO
SERAFIN
T2DCLI
16
1


T2DGEO
SERAFIN

I missed the other messages.

The error seems to come from metis.
Could you try reinstalling with metis in 32bit to see if there is a difference?

Hope it helps

Thanks again, 32 bit got me a bit further thank you.

The following errors are seen (attached for complete output)

PRERES: MAXAt line 132 of file /gpfs/stfc/local/HCP012/jjd66/cxh71-jjd66/v7p1r0/sources/utils/gretel/homere_gretel.f (unit = 5, file = 'stdin')
Fortran runtime error: End of file
Backtrace for this error:
+ in the main program
from file homere_gretel.f
+ /lib64/libc.so.6(__libc_start_main+0xfd) [0x379881ed1d]

---

Then

... merging separated result files

+> cas.cas
collecting: T2DRES
runRecollection:
|runGRETEL: Could not split your file T2DRES (runcode=2) with the error as follows:
|
|... The following command failed for the reason above (or below)
|/gpfs/stfc/local/HCP012/jjd66/cxh71-jjd66/v7p1r0/builds/encore/bin/gretel < gretel_T2DRES.par >> gretel_T2DRES.log

Ah this file has been copied from windows, and contains ^M

cat -v randvillkor.cli
2 2 2 0.000 0.000 0.000 0.000 2 0.000 0.000 0.000 27 1 # ^M
2 2 2 0.000 0.000 0.000 0.000 2 0.000 0.000 0.000 72 2 # ^M
2 2 2 0.000 0.000 0.000 0.000 2 0.000 0.000 0.000 148 3 # ^M
2 2 2 0.000 0.000 0.000 0.000 2 0.000 0.000 0.000 241 4 # ^M
2 2 2 0.000 0.000 0.000 0.000 2 0.000 0.000 0.000 337 5 # ^M

that didn't change the errors