OpenMPI does not use PATH, at least not by default (or my default).
Node 1:
PATH=/usr/kerberos/bin:/usr/local/bin:/bin:/usr/bin:/home/elvedin/bin
Node 2:
PATH=/home/elvedin/mpi/ompi/bin:/usr/kerberos/bin:/usr/lib/ccache:/usr/local/bin:/bin:/usr/bin
Binary "endian" is in /home/elvedin/mpi/ompi
Since HFS+ is not case sensitive, mpicc and mpiCC are the same file.
So on OS X, you need to use mpic++ to compile C++ code. But it's
worse than that, because the GNU compiler is smart enough to use the C+
+ parser / compiler instead of the C one, so you don't see the error
until link time
On MacOS 10.5.2 standard installation with Open MPI 1.2.3 the
following hello world program in normal C compiles just fine and runs
correctly.
But if I rename the file to mpihello.cpp and I get many undefined
symbols at link time!
This error cropped up while I was working with a much more
> One thing about running programs is that the binaries need to be in the
> same absolute path on all systems. This means if you're running the
> program from /home/me on system1, the program you're running must also
> be in /home/me on all the other systems. OpenMPI will not transfer those
>
http://www.open-mpi.org/software/ompi/v1.2/
Download either the gzip or bzip, extract it, then "./configure" and
"make all install" is pretty simple. The library will go into
/usr/local/lib so you might need to add that path to your linker. You
can do this on all three systems. OpenMPI will ha
I am looking for the basic steps for setup of an MPI cluster on a RHEL or
Fedora system with mpi-1.1.
IBM used to have a tutorial on this but I cannot find a complete one now.
I have 3 white box computers which I would like to setup and run basic programs
on and start working with MPI.
I currentl
Hi george,
thanks for your help... config.logs are attached.
2008/4/3, George Bosilca :
> Patrick,
>
> As far as I know this was working fine few months ago ...
> I will take a look at this next week. Can you send me the 2 config.log
> files on your 2 machines.
>
> Thanks,
> george.
>
>
> O
> Do you mean that you are starting it via ./my_mpi_program?
Yes.
> Uck. :-(
Yes. :)
> What happens if you replace readline with a simple fgets() (or
> equivalent)? That is, I'm curious to see if the problem is with
> OMPI's interaction with readline or our I/O forwarding in general.
Yes, di
I followed your direction and it works fine now. Thank you very much.
Appreciate it.
Prakashan
i01:~ {1005}$ qconf -sp orte
pe_name orte
slots 360
user_listsNONE
xuser_lists NONE
start_proc_args /bin/true
stop_proc_args/bin/true
allocation_rule $ro
Thank you very much for the information. Let me try it now.
Prakashan
-Original Message-
From: users-boun...@open-mpi.org on behalf of Pak Lui
Sent: Thu 4/3/2008 1:37 PM
To: Open MPI Users
Subject: Re: [OMPI users] SGE error: executing task of job 22966 failed:
Hi Prakashan,
I believ
Hi Prakashan,
I believe it might be something from PE setting. Could you try this:
Change this parameter in the 'orte' parallel environment from:
> job_is_first_task TRUE
to:
> job_is_first_task FALSE
If you have this set to true, it would take away an available slot in
your job, so it might p
Patrick,
As far as I know this was working fine few months ago ...
I will take a look at this next week. Can you send me the 2 config.log
files on your 2 machines.
Thanks,
george.
On Apr 2, 2008, at 7:55 PM, Patrick Heckeler wrote:
Hi everybody,
I tried to submit a single 'long double' f
Hi,
I just compiled OpenMPI version 1.2.5 with the option
./configure --prefix=/u/local/mpi/openmpi/1.2.5 --with-openib=/usr/local
--enable-static --disable-shared CC=icc CXX=icpc F77=ifort FC=ifort --with-sge
on a X86_64 machine with Infiniband Interconnect and OFED software and CentOS 5
Jeff Squyres wrote:
can OpenMPI also deal with one of the subnets failing?
ie. will OpenMPI automatically fall back to using the last remaining
working IB port out of a node, or even fallback to GigE if all the IB
fails?
Not in the 1.2 series.
The 1.3 series *may* include "APM" support (a
14 matches
Mail list logo
