Re: [OMPI users] exporting PATH ineffective for finding MPI

2014-03-14 Thread Ralph Castain
On Mar 14, 2014, at 6:58 PM, Ross Boylan wrote: > On 3/14/2014 5:42 PM, Ralph Castain wrote: >> Why not skip some of these problems and configure OMPI >> --enable-orterun-prefix-by-default? > Cool. Thanks for the pointer. If MPI is compiled with that option, can it > be overridden by the com

Re: [OMPI users] exporting PATH ineffective for finding MPI

2014-03-14 Thread Ross Boylan
On 3/14/2014 5:42 PM, Ralph Castain wrote: Why not skip some of these problems and configure OMPI --enable-orterun-prefix-by-default? Cool. Thanks for the pointer. If MPI is compiled with that option, can it be overridden by the command line? That is, will orterun --prefix /a/b cause /a/b or

Re: [OMPI users] exporting PATH ineffective for finding MPI

2014-03-14 Thread Ralph Castain
Why not skip some of these problems and configure OMPI --enable-orterun-prefix-by-default? On Mar 14, 2014, at 5:14 PM, Ross Boylan wrote: > I used this script to launch mpi: > > R_PROFILE_USER=~/KHC/sunbelt/Rmpiprofile > LD_LIBRARY_PATH=/home/ross/install/lib:$LD_LIBRARY_PATH > PATH=/home/

[OMPI users] exporting PATH ineffective for finding MPI

2014-03-14 Thread Ross Boylan
I used this script to launch mpi: R_PROFILE_USER=~/KHC/sunbelt/Rmpiprofile LD_LIBRARY_PATH=/home/ross/install/lib:$LD_LIBRARY_PATH PATH=/home/ross/install/bin:/home/ross/install/lib64/R/bin:$PATH orterun -x R_PROFILE_USER -x LD_LIBRARY_PATH \ -x PATH -hostfile ~/KHC/sunbelt/hosts -np 18 R --n

Re: [OMPI users] SIGSEV when running OMPI Java binding

2014-03-14 Thread Saliya Ekanayake
This is really great news!! I'll test the trunk on our cluster. Thank you, Saliya On Fri, Mar 14, 2014 at 4:44 PM, Jeff Squyres (jsquyres) wrote: > We just fixed the segv (see > https://svn.open-mpi.org/trac/ompi/changeset/31073, if you care). > > The issue was an errant large array on the sta

Re: [OMPI users] SIGSEV when running OMPI Java binding

2014-03-14 Thread Jeff Squyres (jsquyres)
We just fixed the segv (see https://svn.open-mpi.org/trac/ompi/changeset/31073, if you care). The issue was an errant large array on the stack in debug builds, which would cause JVMs to run out of stack space. The fix is on the SVN trunk now; it will be on the v1.7 branch shortly. On Mar 11,

Re: [OMPI users] ctrl+c to abort a job with openmpi-1.7.5rc2

2014-03-14 Thread Ralph Castain
Think I got this fixed on the trunk - see https://svn.open-mpi.org/trac/ompi/ticket/4393 On Mar 13, 2014, at 8:51 PM, tmish...@jcity.maeda.co.jp wrote: > > > I happened to misspell a hostname, then it hanged. > > [mishima@manage ~]$ mpirun -np 6 -host node05,nod06 > ~/mis/openmpi/demos/mypro

Re: [OMPI users] Question about '--mca btl tcp,self'

2014-03-14 Thread Jeff Squyres (jsquyres)
On Mar 14, 2014, at 10:11 AM, Ralph Castain wrote: >> 1. If specified '--mca btl tcp,self', which interface application will run >> on, use GigE adaper OR use the OpenFabrics interface in IP over IB mode >> (just like a high performance GigE adapter) ? > > Both - ip over ib looks just like

Re: [OMPI users] Question about '--mca btl tcp,self'

2014-03-14 Thread Ralph Castain
On Mar 14, 2014, at 6:50 AM, Jianyu Liu wrote: > Dear Sir/Madam, > > 1. If specified '--mca btl tcp,self', which interface application will run > on, use GigE adaper OR use the OpenFabrics interface in IP over IB mode > (just like a high performance GigE adapter) ? Both - ip over ib look

[OMPI users] Question about '--mca btl tcp,self'

2014-03-14 Thread Jianyu Liu
Dear Sir/Madam, 1. If specified '--mca btl tcp,self', which interface application will run on, use GigE adaper OR use the OpenFabrics interface in IP over IB mode (just like a high performance GigE adapter) ? 2. If '--mca btl tcp,self' forced application to use GigE adapter, how to forc