For anyone else on the list who is interested:
There definitely was a bug in the system that was causing Open MPI to
not forcibly terminate all processes when one called abort. However,
we also have a backup mechanism that should catch things even if our
primary method fails.
MPI processe
Thank you.
DM
On Fri, 26 Jun 2009, Ralph Castain wrote:
Man, was this a PITA to chase down. Finally found it, though. Fixed on trunk
as of r21549
Thanks!
Ralph
So something else is wrong.
On Jun 25, 2009, at 3:19 PM, Mostyn Lewis wrote:
Just local machine - direct from the command line w
On Fri, Jun 26, 2009 at 6:25 PM, Ralph Castain wrote:
> Not sure I understand the problem. Your server will have the ability to recv
this is what i wanted. feel free to add to the FAQ. In the TODO list
is still remove the step of copying of the URI from one machine to
another, but that can be done
On Jun 25, 2009, at 12:06 PM, Robert Jackson wrote:
When using OpenMPI and nwchem standalone (mpirun --byslot --mca btl
self,sm,tcp --mca btl_base_verbose 30 --mca btl_tcp_if_exclude
lo,eth1 $NWCHEM h2o.nw > & h2o.nwo.$$) the job runs fine.
When running the same job via the PBSPro scheduler
For the web archives, the user posted a similar question on the
OpenFabrics list and had their question answered by someone from QLogic.
On Jun 26, 2009, at 9:46 PM, Nifty Tom Mitchell wrote:
On Thu, Jun 25, 2009 at 10:29:39AM -0700, D'Auria, Raffaella wrote:
>
>Dear All,
>I have bee
