[Wrf-users] problem in parallel computer

[Verdi March]

This is normally caused by problematic nodes or network links.

Try to isolate which nodes or links that cause this problem.

Regards,
Verdi

On Tuesday 17 February 2009, mehran khodamorad wrote:
> Dear User
> I have to work on a parallel system with 16 nodes to execute WRF but
> there was a problem with it. Wrf.exe and ndown.exe  must be run with
> 5 and 9 nodes and i dont know why. when i run wrf.exe or ndown.exe
> with other nodes,for example 16 ,there was a message that is"rank 4
> in job 8  gpslhpc_53869 caused collective abort of all ranks
>   exit status of rank 4: killed by signal 9". could any body help me?
> Many Thanks in advance
> Regards
>
> MEHRAN