You might have to tell mpirun (or whatever load-manager your cluster 
uses) the max file-size per node that you expect. I've run into issues 
in my cluster where we use a PBS script to submit jobs and we are 
required to tell PBS the upper limits on resources we expect to use so 
that the scheduler can work properly.

cheers
Manoj


Baker D.J. wrote:
> Hello,
> 
> I'm installing meep for one of the users of our HPC cluster, and we're 
> finding that one of the meep (test) examples fails for some reason. This is 
> very difficult -- I'm neither a parallel hdf5 or meep user. I can tell you 
> that I have compiled hdf5 and meep using the same version of (GNU compiled) 
> mpich (1.2.6). Running this test
> 
> mpirun -machinefile machinefile 
> /local/software/meep/0.10.1/gcc-3.4.4/bin/meep-mpi bend-flux.ctl
> 
> we find that hdf5 is failing to open a file for some reason -- or at least it 
> appears that way. We get the trace as shown below as part of the output of 
> the above command. Could anyone in the hdf5 or meep support 
> groups/communities please advise us regarding this issue.
> 
> Thanks & Regards -- David Baker.
> 
> HDF5-DIAG: Error detected in HDF5 library version: 1.6.6 MPI-process 0.  Back 
> trace follows.
>   #000: H5F.c line 2049 in H5Fopen(): unable to open file
>     major(04): File interface
>     minor(17): Unable to open file
>   #001: H5F.c line 1726 in H5F_open(): unable to open file
>     major(04): File interface
>     minor(17): Unable to open file
>   #002: H5FD.c line 1045 in H5FD_open(): open failed
>     major(22): Virtual File Layer
>     minor(29): Unable to initialize object
>   #003: H5FDsec2.c line 367 in H5FD_sec2_open(): unable to open file
>     major(04): File interface
>     minor(17): Unable to open file
>   #004: H5FDsec2.c line 367 in H5FD_sec2_open(): No such file or directory
>     major(03): Internal HDF5 error
>     minor(87): System error message
> meep: error on line 249 of h5file.cpp: error opening HDF5 input file
> [0] MPI Abort by user Aborting program !
> [0] Aborting program!
> p0_26999:  p4_error: : 1
> 
> 
> 
> 
> 
> 
> _______________________________________________
> meep-discuss mailing list
> meep-discuss@ab-initio.mit.edu
> http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss
> 
> 

_______________________________________________
meep-discuss mailing list
meep-discuss@ab-initio.mit.edu
http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss

Reply via email to