Hello,
I am running WRF V4.5 with ERA5 data (3 hour interval) for one year, with two nested domains with resolution 27x9 km, and dimensions:e_we = 252, 319, e_sn = 178, 205.
When I execute metgrid.exe, the program crashes every time after running for a fixed duration. In this example, it runs normally from 2020-08-01 00:00 to 2021-02-14 09:00, but automatically terminates at 2021-02-14 12:00. The specific error message is as follows:
---------------------------------------------------------------------------------------------------------
ERROR: Error in ext_pkg_write_commit
application called MPI_Abort(MPI_COMM_WORLD, 44984576) - process 0
srun: Job step aborted: Waiting up to 62 seconds for job step to finish.
slurmstepd: error: *** STEP 9879151.0 ON m3cl2004 CANCELLED AT 2024-01-17T13:13:21 ***
srun: error: m3cl2004: tasks 0-5: Killed
----------------------------------------------------------------------------------------------------------
Furthermore, we run metgrid.exe from 2021-02-14 00:00 to 2021-02-15 00:00, and the program runs smoothly, which indicate that there are no errors in the intermediate files generated by ungrid.
This issue is similar to the problem (ERROR: Error in ext_pkg_write_commit) encountered by Carolina Bernardo in 2019, but it hasn't been resolved yet.
Can anyone now provide a solution to this problem?
I am running WRF V4.5 with ERA5 data (3 hour interval) for one year, with two nested domains with resolution 27x9 km, and dimensions:e_we = 252, 319, e_sn = 178, 205.
When I execute metgrid.exe, the program crashes every time after running for a fixed duration. In this example, it runs normally from 2020-08-01 00:00 to 2021-02-14 09:00, but automatically terminates at 2021-02-14 12:00. The specific error message is as follows:
---------------------------------------------------------------------------------------------------------
ERROR: Error in ext_pkg_write_commit
application called MPI_Abort(MPI_COMM_WORLD, 44984576) - process 0
srun: Job step aborted: Waiting up to 62 seconds for job step to finish.
slurmstepd: error: *** STEP 9879151.0 ON m3cl2004 CANCELLED AT 2024-01-17T13:13:21 ***
srun: error: m3cl2004: tasks 0-5: Killed
----------------------------------------------------------------------------------------------------------
Furthermore, we run metgrid.exe from 2021-02-14 00:00 to 2021-02-15 00:00, and the program runs smoothly, which indicate that there are no errors in the intermediate files generated by ungrid.
This issue is similar to the problem (ERROR: Error in ext_pkg_write_commit) encountered by Carolina Bernardo in 2019, but it hasn't been resolved yet.
Can anyone now provide a solution to this problem?
Last edited: