[Nek5000-users] Shutdown after "Failed in HMHOLTZ"

[nek5000-users at lists.mcs.anl.gov]

Hi Johs,

It depends on your compiler.  For my favorite, pg, I set the flag
in makenek:

G="-Ktrap=fp -mcmodel=medium"

The -Ktrap=fp flag causes the code to stop at the first inf or NaN.

I'm not 100% certain what happens in a parallel environment
when only one of the cores encounters the NaN...

At least that's a start... perhaps some others have other ideas.

Paul

On Thu, 17 Feb 2011, nek5000-users at lists.mcs.anl.gov wrote:

> Neks,
>
> Is there a simple way to make Nekton quit if the simulation blows up?
> I know Nek will quit if the CFL gets too high, but it doesn't seem to
> want to quit something get messed up in the Helmholtz or Pressure
> iterations (resulting in NaNs).
>
> I am submitting my jobs to a cluster, and I may not be able to babysit
> the simulation all the time (particularly if my turn in the queue
> comes up late at night).  Therefore, hours can get wasted.
>
> Any help would be appreciated!
>
> -- 
> Josh Camp
>
> "All that is necessary for the triumph of evil is that good men do
> nothing" -- Edmund Burke
> _______________________________________________
> Nek5000-users mailing list
> Nek5000-users at lists.mcs.anl.gov
> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users
>