Loading shared library in open-mpi/ mpi-run

mpi pbs torque
piyush_sao picture piyush_sao · Jul 6, 2012 · Viewed 13k times · Source

I'm trying to run my program using torque scheduler using mpi run. Though in my pbs file I load all the library by 

export LD_LIBRARY_PATH=/path/to/library

yet it gives error i.e. 

error while loading shared libraries: libarmadillo.so.3: 
cannot open shared object file: No such file or directory.

I guess error lies in variable LD_LIBRARY_PATH not set in all the nodes. How would I make it work?

Answer

Hristo 'away' Iliev picture Hristo 'away' Iliev · Jul 6, 2012

LD_LIBRARY_PATH is not exported automatically to MPI processes, spawned by mpirun. You should use

mpirun -x LD_LIBRARY_PATH ...

to push the value of LD_LIBRARY_PATH. Also make sure that the specified path exists on all nodes in the cluster and that libarmadillo.so.3 is available everywhere.

Related questions

fork: retry: Resource temporarily unavailable

I tried installing Intel MPI Benchmark on my computer and I got this error: fork: retry: Resource temporarily unavailable Then I received this error again when I ran ls and top command. What is causing this error? Configuration of my …

Read More
What is "Signal 15 received"

What might cause a C, MPI program using a library called SUNDIALS/CVODE (a numerical ODE solver) running on a Gentoo Linux cluster to give me repeated Signal 15 received.? Is that code being issued by MPI, Sundials, Linux, C or …

Read More
fatal error: mpi.h: No such file or directory #include <mpi.h>

when I compile my script with only #include <mpi.h> it tells me that there is no such file or directory. But when i include the path to mpi.h as #include "/usr/include/mpi/mpi.h" (the …

Read More