Scheduled Downtime
On Friday 21 April 2023 @ 5pm MT, this website will be down for maintenance and expected to return online the morning of 24 April 2023 at the latest

ERROR: Error in ext_pkg_write_field

hannahv1711

New member
I am running metgrid on my intermediate files generated using data from MPI-ESM1.2-HR.
However I get the following error when writing the files:
ERROR: Error in ext_pkg_write_field
--------------------------------------------------------------------------
MPI_ABORT was invoked on rank 0 in communicator MPI_COMM_WORLD
with errorcode 66715552.

NOTE: invoking MPI_ABORT causes Open MPI to kill all MPI processes.
You may or may not see output from other processes, depending on
exactly when Open MPI kills them.
--------------------------------------------------------------------------
srun: Job step aborted: Waiting up to 32 seconds for job step to finish.
slurmstepd: error: *** STEP 487760.0 ON b5202 CANCELLED AT 2022-11-28T13:55:31 ***
srun: error: b5202: tasks 0-15: Killed

After searching the forums it seems like that there can be many different reasons for this error but I do not know which one applies to my case

The metgrid log also does not seem to specify the problem, so I was wondering if anyone here might have some ideas?
 

Attachments

  • namelist.wps
    1.1 KB · Views: 3
  • metgrid.log
    18.7 KB · Views: 1
  • metgrid.log.0000
    72 KB · Views: 0
Based on your namelist.wps, I suppose you are using ERA5 for this case. Please let me know if I am wrong.
I am a little perplexed by the following options in your namelist.wps:

constants_name = 'FIXED:1980-01-01_06'
fg_name = 'WRF'

What is "WRF" datafile? How did you produce "'FIXED:1980-01-01_06'"?
 
Based on your namelist.wps, I suppose you are using ERA5 for this case. Please let me know if I am wrong.
I am a little perplexed by the following options in your namelist.wps:

constants_name = 'FIXED:1980-01-01_06'
fg_name = 'WRF'

What is "WRF" datafile? How did you produce "'FIXED:1980-01-01_06'"?
hi Ming Chen

The 'FIXED:1980-01-01_06' file contains the static data i.e. LANDSEA + SOILHGT
The 'WRF' files contains all the 2D/3D parameters needed, this was produced using the python package pywinter which reads the MPI netcdfs.
 
Just a little update - after reading some of the related threads in the forum, I discovered that I had:
1. GHT on a different number of levels than the rest of my 3D data so I downloaded the dataset with the same levels and replaced this in the intermediate file
2. TT on level 1 was missing, instead there was a reference to "TTUMD" which I didn't recognise. I re-made the intermediate files with TT again

After implementing (1), I was still getting the same error message. However after implementing (1) and (2) metgrid now runs without any errors :)
 
Thanks for the update. These two changes you made do make sense. Note that all 3D variables in the intermediate file must be on the same levels.
 
Top