[MITgcm-support] strange problems: "No. of procs= 1 not equal to nPx*nPy= 4"

[Dustin Carroll]

Hi Li,

What flags are you setting when you call mpirun to run your model?

Dustin

On Oct 8, 2014, at 3:28 PM, 李志远 <oceanlizy at 163.com> wrote:

> From: Li zhi <oceanlizy at 163.com>
> To: MITgcm-support at mitgcm.org
> Subject : strange problems: "No. of procs=     1 not equal to nPx*nPy=     4"
> 
> 
> hello,all;
>     I have met a strange problem recently : I have been using MIT for half a year without any problems ,but just a few days ago , when I use it , it occured such errors:
> /opt/gridview/pbs/dispatcher/mom_priv/jobs/6009.node200.SC: line 1: !/bin/bash: No such file or directory
> EEBOOT_MINIMAL: No. of procs=     1 not equal to nPx*nPy=     4
> EEDIE: earlier error in multi-proc/thread setting
> PROGRAM MAIN: ends with fatal Error
> ABNORMAL END: PROGRAM MAIN
> 
> It puzzled me so much ,because  I changed nothing ,but it did not work well any more !!!!!    From the error ,it seemed that I did not compile with MPI ,but the fact is , I set the right path in /public/userspace/allenson3/MITgcm/tools/build_options ,  here is the MPI path in  my linux_amd64_pgf90+mpi_xd1 file,and I add /dcfs2/program/mpi/mvapich2/1.8/pgcc.pgf90/bin/ to my PATH variable .
> FC='/dcfs2/program/mpi/mvapich2/1.8/pgcc.pgf90/bin/mpif90'
> CC='/dcfs2/program/mpi/mvapich2/1.8/pgcc.pgf90/bin/mpicc'
> LINK='/dcfs2/program/mpi/mvapich2/1.8/pgcc.pgf90/bin/mpif90'
> 
> MPI='true'
> 
> DEFINES='-DWORDLENGTH=4'
> CPP='/usr/bin/cpp -P -traditional'
> EXTENDED_SRC_FLAG='-Mextend'
> INCLUDEDIRS='/dcfs2/program/mpi/mvapich2/1.8/pgcc.pgf90/include'
> INCLUDES='-I/dcfs2/program/mpi/mvapich2/1.8/pgcc.pgf90/include'
> LIBS='-L/dcfs2/program/mpi/mvapich2/1.8/pgcc.pgf90/lib'
> 
> I compile with :../../../tools/genmake2 -mpi -mods=../code -of=../../../tools/build_options/linux_amd64_pgf90+mpi_xd1 ,  then make depend ,make ,it produced the mitgcmuv file ,so I think there is nothing wrong
> with my MPI setting .but once I run it , it occured the error as above .  I do not think I am a new mitgcm user ,because I have used the model and run some results I think is good . Just as I say , it suddenly did not work ....
> I made a test in the lab_sea directory . below is tha code files :  
> SIZE.h file :
>      &           sNx =  10,
>      &           sNy =   8,
>      &           OLx =   4,
>      &           OLy =   4,
>      &           nSx =   1,
>      &           nSy =   1,
>      &           nPx =   2,
>      &           nPy =   2,
>      &           Nx  = sNx*nSx*nPx,
>      &           Ny  = sNy*nSy*nPy,
>      &           Nr  =  23)
> 
> I guess there maybe something wrong in other place , such as the eedata.mth file as below ??
>  # Example "eedata" file  for multi-threaded test
> #   (copy "eedata.mth" to "eedata" to use it)
> # Lines beginning "#" are comments
> # nTx - No. threads per process in X
> # nTy - No. threads per process in Y
>  &EEPARMS
>  nTx=1,
>  nTy=2,
>  &
> # Note: Some systems use & as the namelist terminator (as shown here).
> #       Other systems use a / character.
> 
> did some one met the same problem before ?
> 
> thank you for your quick reply !
> 
> LI ZHI 
> 
> 
> 
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mitgcm.org/pipermail/mitgcm-support/attachments/20141008/d831025e/attachment.htm>