The rtofs_ncoda_acspo_sst_nc executable (and possibly others) sometimes segfaults when reading a corrupt input files. The subsequent err_chk causes job jrtofs_global_ncoda_qc to abort. Please update the rtofs_ncoda source codes and/or the calling scripts to avoid such failures.
The logic might be updated to print a warning and continue if a corrupt file is found. At the end, if any files were skipped, the mail.py utility can be used to email the list of bad files to the developers and SPA team for further investigation.
DanIredell-NOAA
commented
1 year ago
bugzilla 1330 - http://www2.spa.ncep.noaa.gov/bugzilla/show_bug.cgi