You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We are trying to convert several data sets from Grib2 to Grib1. I know, we are still using grib1. :).
We are using these libraries on a Rocky 9 system. We have installed all the libraries required and build the grib-util. We are able to convert most of the data sets we use like GFS and NAM. However, with the RAP 20km we are seeing malloc errors when trying to convert it. It appears there are 7 files within this file that can cause this to happen. One of the files is the Haines Index, which I have attached the grib2 file for. When I run cnvgrib on it, this is what I get:
cnvgrib -g21 haines_index.grb2 haines_index.grb
malloc(): invalid next size (unsorted)
Program received signal SIGABRT: Process abort signal.
Backtrace for this error:
#0 0x7fc8f5c47880 in ??? #1 0x7fc8f5c46a25 in ??? #2 0x7fc8f58eb6ef in ??? #3 0x7fc8f593894c in ??? #4 0x7fc8f58eb645 in ??? #5 0x7fc8f58d57f2 in ??? #6 0x7fc8f58d612f in ??? #7 0x7fc8f59429f6 in ??? #8 0x7fc8f59457eb in ??? #9 0x7fc8f5946911 in ??? #10 0x44fb71 in w3fi72_
at /opt/NCEPLIBS-w3emc/src/w3fi72.f:268 #11 0x408a4a in putgbexn_
at /opt/NCEPLIBS-grib_util/src/cnvgrib/putgbexn.F90:218 #12 0x4090a3 in cnv21_
at /opt/NCEPLIBS-grib_util/src/cnvgrib/cnv21.F90:177 #13 0x40b1c4 in cnvgrib
at /opt/NCEPLIBS-grib_util/src/cnvgrib/cnvgrib.F90:242 #14 0x4024dc in main
at /opt/NCEPLIBS-grib_util/src/cnvgrib/cnvgrib.F90:256
Aborted (core dumped)
We use this in centos 7 without issue, so this could be related to moving to Rocky 9. Maybe something in the build process or something with libraries with respect to Rocky 9.
I looked through your previous issues and didn't see this mentioned.
Any thoughts on what could be happening? We will continue to try and debug things on our side. But I wanted to at least mention it.
We are trying to convert several data sets from Grib2 to Grib1. I know, we are still using grib1. :).
We are using these libraries on a Rocky 9 system. We have installed all the libraries required and build the grib-util. We are able to convert most of the data sets we use like GFS and NAM. However, with the RAP 20km we are seeing malloc errors when trying to convert it. It appears there are 7 files within this file that can cause this to happen. One of the files is the Haines Index, which I have attached the grib2 file for. When I run cnvgrib on it, this is what I get:
haines_index.grb2.tar.gz
cnvgrib -g21 haines_index.grb2 haines_index.grb
malloc(): invalid next size (unsorted)
Program received signal SIGABRT: Process abort signal.
Backtrace for this error:
#0 0x7fc8f5c47880 in ???
#1 0x7fc8f5c46a25 in ???
#2 0x7fc8f58eb6ef in ???
#3 0x7fc8f593894c in ???
#4 0x7fc8f58eb645 in ???
#5 0x7fc8f58d57f2 in ???
#6 0x7fc8f58d612f in ???
#7 0x7fc8f59429f6 in ???
#8 0x7fc8f59457eb in ???
#9 0x7fc8f5946911 in ???
#10 0x44fb71 in w3fi72_
at /opt/NCEPLIBS-w3emc/src/w3fi72.f:268
#11 0x408a4a in putgbexn_
at /opt/NCEPLIBS-grib_util/src/cnvgrib/putgbexn.F90:218
#12 0x4090a3 in cnv21_
at /opt/NCEPLIBS-grib_util/src/cnvgrib/cnv21.F90:177
#13 0x40b1c4 in cnvgrib
at /opt/NCEPLIBS-grib_util/src/cnvgrib/cnvgrib.F90:242
#14 0x4024dc in main
at /opt/NCEPLIBS-grib_util/src/cnvgrib/cnvgrib.F90:256
Aborted (core dumped)
We use this in centos 7 without issue, so this could be related to moving to Rocky 9. Maybe something in the build process or something with libraries with respect to Rocky 9.
I looked through your previous issues and didn't see this mentioned.
Any thoughts on what could be happening? We will continue to try and debug things on our side. But I wanted to at least mention it.
I pulled the data from here: https://tgftp.nws.noaa.gov/SL.us008001/ST.opnl/MT.rap_CY.11/RD.20240808/PT.grid_DF.gr2/
The raw files are named like this: fh.0004_tl.press_gr.us20km
The text was updated successfully, but these errors were encountered: