WRF_ndown.exe

Atmosphere
, , ,
1

Hi everyone,

Got stuck for WRF again. This time, I am trying to run ndown.exe for 1-way nesting. I encountered this situation, the script ran, but unsuccessfully. Upon checking error logs, the qsub did not generate any error message (ie. rsl.out. or run_ndown.e). It only generated run_ndown.o. Here is the message I only got, and it’s really out of my scope! This situation also happens, when I tried to run real.exe for another case. please have a look…

running with 6 mpi ranks
 starting wrf task            0  of            6
 starting wrf task            1  of            6
 starting wrf task            2  of            6
 starting wrf task            3  of            6
 starting wrf task            4  of            6
 starting wrf task            5  of            6
--------------------------------------------------------------------------
Primary job  terminated normally, but 1 process returned
a non-zero exit code. Per user-direction, the job has been aborted.
--------------------------------------------------------------------------
--------------------------------------------------------------------------
mpirun detected that one or more processes exited with non-zero status, thus causing
the job to be terminated. The first process to do so was:

  Process name: [[35649,1],0]
  Exit code:    29
--------------------------------------------------------------------------

real    0m5.686s
user    0m0.127s
sys     0m0.547s
ndown.exe is finished.

======================================================================================
                  Resource Usage on 2025-03-24 07:05:55:
   Job Id:             137468687.gadi-pbs
   Project:            v46
   Exit Status:        0
   Service Units:      0.04
   NCPUs Requested:    6                      NCPUs Used: 6
                                           CPU Time Used: 00:00:10
   Memory Requested:   32.0GB                Memory Used: 1.69GB
   Walltime requested: 03:00:00            Walltime Used: 00:00:09
   JobFS requested:    100.0MB                JobFS used: 0B
======================================================================================
1 Like
2

Hi @ChristalX. Just to let you know this is out of scope for ACCESS-NRI support, but I hope someone in the community can assist.

I guess you know about it already, but there is a WRF support forum run by NCAR

That may be another option, to either search for tips on debugging WRF or register and create a topic to ask for assistance.

And please do feel free to continue to ask your questions here, especially if there are local WRF users who can assist each other.

1 Like
3

Hi @ChristalX, it’s been a good few years since I have run WRF (I think I was last using v3.7.1!) but I will try to help if I can.

Did real.exe produce input and boundary conditions for your outer nest (wrfinput_d01 and wrfbdy_d01)? Can you check the real.exe log file carefully, and also try opening those files to see if they can be read?

This job is failing very quickly into its walltime, my suspicion is that it is reading incomplete or corrupt data.

When you say real.exefails similarly for another case (i.e., doesn’t produce the outer nest input and boundary data?), are you using the same met_em.d0* data for both cases? It might be that metgrid didn’t complete successfully and the problems are being picked up downstream in both cases.

1 Like
4

Also just checking you are running ndown as per the user guide instructions: