Hi,
It looks like make distclean will remove the following file:
orte_hosts.7 and prevent a configure, new make ; make install.
I tried the pre-release versions (openmpi-1.3b1,b2) and one nightly
tarball (openmpi-1.3b2r19907):
For openmpi-1.3b2:
Step 1 :
./configure --prefix=/tmp/
Sorry for delayed response - had to dig into this a little since it
has been so long since I wrote the bproc support code.
The problem here is with how you named your nodes. On bproc clusters,
the backend nodes are normally named with just a number. Our system
therefore expects to see node
I would suggest making sure that the /etc/beowulf/config file has a
"libraries" line for every directory where required shared libraries
(application and mpi) are located.
Also, make sure that the filesystems containing the executables and
shared libraries are accessible from the compute nodes.
Thanks for all your help Ralph and Sean!!
I changed the machinefile to just containing the node numbers. I added
the env variable NODES in my .bash_profile and .bashrc.
As per Sean's suggestion I added the $LD_LIBRARY_PATH (shared lib path
which the openmpi lib directory path) and the $AMBERHOME/li
Can your nodes see the openmpi libraries and executables? I have the
/usr/local and /opt from the master node mounted on the compute nodes,
in addition to having the LD_LIBRARY_PATH defined correctly. In your
case the nodes must be able to see /home/rchaud/openmpi-1.2.6 in order
to get the librar
The errors you are seeing aren't related to using g95 vs. gfortran:
1. The warnings from configure are fairly normal. It's just configure
trying to be responsible and telling you things that you might want to
know (e.g., your system has no support for Fortran INTEGER*16, so OMPI
is not inc
You are absolutely correct; looks like a typo in orte/util/
Makefile.am. Thanks for reporting this!
I fixed it on the trunk in r19936 and have filed a CMR to get it over
to the v1.3 branch.
On Nov 5, 2008, at 8:54 AM, Mehdi Bozzo-Rey wrote:
Hi,
It looks like make distclean will remove t
On some of my larger problems ,
my program stalls and does not continue
(50 or more nodes, 'long' runs >5 hours). My program is set up as a
master-worker
and it seems that the master gets stuck in a write to stdout see gdb
backtrace below (It took all day
to get there on 50 nodes). the functi
