[torqueusers] mpiexec jobs got stuck
tbaer at utk.edu
Tue May 12 13:16:11 MDT 2009
On Tue, 2009-05-12 at 15:10 -0400, Abhishek Gupta wrote:
> Some of our jobs run for a while and then got stuck in between for ever.
> I tried various node numbers and as soon as I go beyond 20nodes, it
> shows this behaviour. Can anyone suggest something if you ever face the
> same problem?
Which MPI and mpiexec implementations are you using?
Troy Baer, HPC System Administrator
National Institute for Computational Sciences, University of Tennessee
More information about the torqueusers