HI, Dear MEEP users and Steven,
I need your help on using openmpi 1.4.1 and meep-mpi (meep -1.1.1). When I was
running a test on several nodes, if there is some hdf5 output, it will show
errors like this:
-----------
creating output file "./eps-000000.00.h5"...
HDF5-DIAG: Error detected in HDF5 library version: 1.6.10 MPI-process 0. Back
trace follows.
#000: H5F.c line 1992 in H5Fcreate(): unable to create file
major(04): File interface
minor(17): Unable to open file
#001: H5F.c line 1749 in H5F_open(): unable to open file
major(04): File interface
minor(17): Unable to open file
#002: H5FD.c line 1033 in H5FD_open(): open failed
major(22): Virtual File Layer
minor(29): Unable to initialize object
#003: H5FDmpio.c line 802 in H5FD_mpio_open(): MPI_File_open failed
major(03): Internal HDF5 error
minor(78): Some MPI function failed
#004: H5FDmpio.c line 802 in H5FD_mpio_open(): MPI_ERR_ACCESS: invalid access
mode
major(03): Internal HDF5 error
minor(79): MPI Error String
meep: error on line 450 of h5file.cpp: error opening HDF5 output file
--------------------------------------------------------------------------
MPI_ABORT was invoked on rank 0 in communicator MPI_COMM_WORLD
with errorcode 1.
NOTE: invoking MPI_ABORT causes Open MPI to kill all MPI processes.
You may or may not see output from other processes, depending on
exactly when Open MPI kills them.
--------------------------------------------------------------------------
--------------------------------------------------------------------------
mpirun has exited due to process rank 0 with PID 13410 on
node compute-0-23 exiting without calling "finalize". This may
have caused other processes in the application to be
terminated by signals sent by mpirun (as reported here).
--------------------------------------------------------------------------
In an early post
"http://www.mail-archive.com/[email protected]/msg02928.html";, it
is said that openmpi will works for the similar errors. However, if there is no
hdf5 output, the simulation goes well on my nodes.
For hdf5, it is hdf5-1.6.10, configured with --enable-parallel CC=mpicc
--prefix=/opt/hdf5. And it works fine on another cluster.
What's the reason for the above errors. And the solution...
MANY THANKS!!
Best,
Jiangjun
_______________________________________________
meep-discuss mailing list
[email protected]
http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss
