search

plesager / ece3-postproc

Suite of processing tools for EC-Earth3 output
5 stars 8 forks source link

NEMO monitoring crash #47

Closed plesager closed 5 years ago

plesager commented 5 years ago

Got this error message from monitor_ocean.sh of hiresclim2:

[17:23:33] monitor_ocean.sh[476]> ncks -3 -h -A -v tot_area_ice_north tmp_north.nc -o atun_1870_time-series_ocean.tmp

nco_err_exit(): ERROR Short NCO-generated message (usually name of function that triggered error): nco_put_vara() nco_err_exit(): ERROR Error code is -57. Translation into English with nc_strerror(-57) is "NetCDF: Start+count exceeds dimension bound" ERROR: nco_put_vara() failed to nc_put_vara() variable "nav_lat" nco_err_exit(): ERROR NCO will now exit with system call exit(EXIT_FAILURE)

Not sure with which version of EC-Earth output this started, and if this is specific to ECMWF/cca platform. But I have the error with both CMIP6 and vanilla control output files of EC-Earth (essentially a version that will be 3.3.1.1).

oloapinivad commented 5 years ago

Which version of NCO are you using? I am currently using the following modules nco/4.6.7 netcdf python/2.7.12-01 cdo/1.9.6. I am monitoring a historical simulation with r6903 which is basically the same as 3.3.1.1. and the timeseries subpackage works smooth.

plesager commented 5 years ago

I tried both older nco/4.3.7 and newest nco/4.7.8. Maybe I should try 4.6.7.... I have cdo/1.9.6 but that does not matter here I think.

plesager commented 5 years ago

Ok, I figured it out. nco/4.7.8 became the default on ECMWF/cca in June 2019 and that was creating the problem. Switching back to 4.6.7 fixes it. Thanks @oloapinivad

plesager commented 5 years ago

Committed fix 6699891