Welcome, Guest
Username: Password: Remember me

TOPIC: Error run after installation v7p2

Error run after installation v7p2 7 years 3 months ago #27505

  • yugi
  • yugi's Avatar
  • OFFLINE
  • openTELEMAC Guru
  • Posts: 851
  • Thank you received: 244
OKay then i would suggest running:
compileTELEMAC.py -c wing64mpi --clean
Removing the executabale and rerunning telemac2d.py ...
There are 10 types of people in the world: those who understand binary, and those who don't.
The administrator has disabled public write access.
The following user(s) said Thank You: Yunhao Song

Error run after installation v7p2 7 years 3 months ago #27509

  • Yunhao Song
  • Yunhao Song's Avatar
  • OFFLINE
  • Senior Boarder
  • Posts: 118
  • Thank you received: 9
Hi Yoann,

After running some tests finally I am assured that the problem has been solved!

Thank you so much for your expertise and patience!

Best regards,

Yunhao
The administrator has disabled public write access.

Error run after installation v7p2 7 years 3 months ago #27517

  • yugi
  • yugi's Avatar
  • OFFLINE
  • openTELEMAC Guru
  • Posts: 851
  • Thank you received: 244
Cool.
There are 10 types of people in the world: those who understand binary, and those who don't.
The administrator has disabled public write access.

Error run after installation v7p2 6 years 1 month ago #31757

  • Yunhao Song
  • Yunhao Song's Avatar
  • OFFLINE
  • Senior Boarder
  • Posts: 118
  • Thank you received: 9
Dear Yugi,

I allow myself to continue this post as the same problem occurred when running examples in parallel on Cluster:
MASTER PROCESSOR NUMBER 0 OF THE GROUP OF 1
P_INIT: FILE PARAL IS INCONSISTENT WITH MPI PARAMETERS
MPI NCSIZE = 1
PARAL NCSIZE = 16

After the warning message the example kept running and could be completed, but the time costed was close to that of serial mode, so I suppose the pb comes from my mpi settings. Could you please help me check the config file enclosed?

Besides, is there a simple case to check the mpi on Cluster? Just like you posted at #27502.

Thank you very much!
Yunhao
Attachments:
The administrator has disabled public write access.

Error run after installation v7p2 6 years 1 month ago #31758

  • yugi
  • yugi's Avatar
  • OFFLINE
  • openTELEMAC Guru
  • Posts: 851
  • Thank you received: 244
This message means that the number of cores detected by MPI was 1.
So i guess the error comes from your mpi_cmdexec.
Maybe use mpirun instead of srun.
There are 10 types of people in the world: those who understand binary, and those who don't.
The administrator has disabled public write access.

Error run after installation v7p2 6 years 1 month ago #31759

  • Yunhao Song
  • Yunhao Song's Avatar
  • OFFLINE
  • Senior Boarder
  • Posts: 118
  • Thank you received: 9
Do you mean changing the mpi_cmdexec into mpirun -wdir <wdir> -n <ncsize> <exename>? Just FYI, slurm was used to submit the job and before launching the example some modules were loaded as below.
module load anaconda/2.7.13
module load openmpi/2.1.5gcc5.3.1
The administrator has disabled public write access.

Error run after installation v7p2 6 years 1 month ago #31760

  • yugi
  • yugi's Avatar
  • OFFLINE
  • openTELEMAC Guru
  • Posts: 851
  • Thank you received: 244
Yes something like that is there any info on your cluster on how mpi applications should be launch ?
I am afraid that what srun -n <ncsize> <exe> is doing is launching ncsize time one <exename>
There are 10 types of people in the world: those who understand binary, and those who don't.
The administrator has disabled public write access.

Error run after installation v7p2 6 years 1 month ago #31761

  • Yunhao Song
  • Yunhao Song's Avatar
  • OFFLINE
  • Senior Boarder
  • Posts: 118
  • Thank you received: 9
Yes here is a mpi test case:
[simon@n0255 ~]$ which mpicc
/home/software/mpich_3.2/bin/mpicc
[simon@n0255 ~]$ mpicc -o mpitest mpitest.c
[simon@n0255 ~]$ srun --mpi=pmi2 -p hpib -n 4 mpitest
Hello world: rank 0 of 4 running on n0317
Hello world: rank 1 of 4 running on n0317
Hello world: rank 2 of 4 running on n0317
Hello world: rank 3 of 4 running on n0317

Please find the code enclosed.
Attachments:
The administrator has disabled public write access.

Error run after installation v7p2 6 years 1 month ago #31762

  • yugi
  • yugi's Avatar
  • OFFLINE
  • openTELEMAC Guru
  • Posts: 851
  • Thank you received: 244
You do not have the --mpi option in your srun could that be it ?
There are 10 types of people in the world: those who understand binary, and those who don't.
The administrator has disabled public write access.

Error run after installation v7p2 6 years 1 month ago #31770

  • Yunhao Song
  • Yunhao Song's Avatar
  • OFFLINE
  • Senior Boarder
  • Posts: 118
  • Thank you received: 9
Dear Yugi,

Thanks for your help, the mpi_cmdexec: mpiexec -wdir <wdir> -n <ncsize> <exename> worked.

Best regards,
The administrator has disabled public write access.
Moderators: borisb

The open TELEMAC-MASCARET template for Joomla!2.5, the HTML 4 version.