Well, it depends what you want to do.
If you just need the data, an option is to split the grib2 file into
separate messages, then run cnvgrib on each message. Split can be done with
wgrib2, or you can write a simple program to do it: each message starts
with the string 'GRIB' and ends with '7777'.
There are other utilities to convert grib2 to grib1 format: Python package
pygrib2 available at https://pypi.python.org/pypi/pygrib2 contains utility
cnvgrib2to1 that would work for you.
If you want to track the cause of cnvgrib2 failure, try to run it via gdb.
The debugger will catch program crashes. I tested cnvgrib on a corrupted
files, it completes without an error message or simply prints some codes
without any indication of an error. You could try different grib2 file (be
sure it contains more than one message), though I doubt the result will be
different. The files on NOMADS site are fine.
George
On Wed, Apr 20, 2016 at 4:30 AM, Michelle Serino <mmserino@xxxxxxxx> wrote:
> I downloaded the newest cnvgrib (1.4.1) and its libraries from
> http://www.nco.ncep.noaa.gov/pmb/codes/GRIB2/, and it was compiled on the
> same operating system for which I sent you the info, CentOS 6.7. Does what
> crash? When I run cnvgrib to convert the file, it seems to run just fine,
> and I do not get any error messages. Would modifying either of the
> libraries be helpful in fixing this?
>
> I will have to talk with my advisor to determine if we really want to use
> these .grb2 files and take the next steps for that.
>
> Michelle M. Serino
> M.S. Candidate
> Texas A&M University
> Atmospheric Sciences Dept.
> Eller O&M, Suite 1017
> College Station, TX 77843
> (484) 769-0949
>
> On Tue, Apr 19, 2016 at 11:18 PM, George Trojan <george.trojan@xxxxxxxx>
> wrote:
>
>> And where cnvgrib comes from? Was it compiled on this or same system? Is
>> it the newest version, i.e. 1.4.1? Does it crash after completing the first
>> message in the file?
>>
>> My only suggestion is to build your own executable.This is not a trivial
>> task, the makefiles in the source retrieved from NCEP site would need to be
>> updated. There are dependencies on w3lib and g2lib, those libraries would
>> have to be build first. I do know that version 1.4.1 does run on Centos6.
>>
>> Let me know you want to go this route, I might be able to dig out
>> makefiles I used to compile it.
>>
>> George
>>
>>
>> On 20/04/16 12:55 AM, Michelle Serino wrote:
>>
>> That's good. Here's the info for my computer:
>>
>> CentOS release 6.7 (Final)
>>
>>
>> LSB_VERSION=base-4.0-amd64:base-4.0-noarch:core-4.0-amd64:core-4.0-noarch:graphics-4.0-amd64:graphics-4.0-noarch:printing-4.0-amd64:printing-4.0-noarch
>>
>> CentOS release 6.7 (Final)
>>
>> CentOS release 6.7 (Final)
>>
>> Michelle M. Serino
>> M.S. Candidate
>> Texas A&M University
>> Atmospheric Sciences Dept.
>> Eller O&M, Suite 1017
>> College Station, TX 77843
>> (484) 769-0949
>>
>> On Tue, Apr 19, 2016 at 8:41 PM, George Trojan <george.trojan@xxxxxxxx>
>> wrote:
>>
>>> So grib2 file looks OK. Something wrong with cnvgrib. As I said before,
>>> I have had problems with cnvgrib in the past. If I recall correctly it had
>>> to do with default integer sizes 4 vs 8 bytes. What version of operating
>>> system are you using?
>>>
>>> On 20/04/16 12:11 AM, Michelle Serino wrote:
>>>
>>> Sorry about that. On the .grb2 file, I get: 492053148 21637412
>>> ruc2anl_130_20090605_2200_000.grb2.
>>>
>>> Thanks, too, for the cksum info.
>>>
>>> Michelle M. Serino
>>> M.S. Candidate
>>> Texas A&M University
>>> Atmospheric Sciences Dept.
>>> Eller O&M, Suite 1017
>>> College Station, TX 77843
>>> (484) 769-0949
>>>
>>> On Tue, Apr 19, 2016 at 7:10 PM, George Trojan - NOAA Federal <
>>> george.trojan@xxxxxxxx> wrote:
>>>
>>>> By the way, I asked for checksum on the GRIB2 file. We know the
>>>> converted one is too short.
>>>>
>>>> On Tue, Apr 19, 2016 at 11:07 PM, George Trojan - NOAA Federal <
>>>> <george.trojan@xxxxxxxx>george.trojan@xxxxxxxx> wrote:
>>>>
>>>>> Checksum value and file size. Google should be your friend:
>>>>> https://en.wikipedia.org/wiki/Cksum
>>>>>
>>>>> I did have in the past issues with cnvgrib built on Centos 6 and
>>>>> recent Fedora's. Before going into this, let's eliminate simple causes.
>>>>>
>>>>> On Tue, Apr 19, 2016 at 10:53 PM, Michelle Serino <
>>>>> <mmserino@xxxxxxxx>mmserino@xxxxxxxx> wrote:
>>>>>
>>>>>> I get:
>>>>>>
>>>>>> 942363691 228074 ruc2anl_130_20090605_2200_000.grb.
>>>>>>
>>>>>>
>>>>>> What do those numbers represent? I am not familiar with the cksum
>>>>>> command.
>>>>>>
>>>>>> Michelle M. Serino
>>>>>> M.S. Candidate
>>>>>> Texas A&M University
>>>>>> Atmospheric Sciences Dept.
>>>>>> Eller O&M, Suite 1017
>>>>>> College Station, TX 77843
>>>>>> (484) 769-0949 <%28484%29%20769-0949>
>>>>>>
>>>>>> On Tue, Apr 19, 2016 at 6:49 PM, George Trojan - NOAA Federal <
>>>>>> <george.trojan@xxxxxxxx>george.trojan@xxxxxxxx> wrote:
>>>>>>
>>>>>>> Either the input file is corrupted, or your version of cnvgrib is
>>>>>>> bad. Try this to eliminate the first possibility:
>>>>>>>
>>>>>>> $ cksum ruc2anl_130_20090605_2200_000.grb2
>>>>>>> 492053148 21637412 ruc2anl_130_20090605_2200_000.grb2
>>>>>>>
>>>>>>> The numbers 492053148 21637412 should be the same.
>>>>>>>
>>>>>>>
>>>>>>> On Tue, Apr 19, 2016 at 10:42 PM, Michelle Serino <
>>>>>>> <mmserino@xxxxxxxx>mmserino@xxxxxxxx> wrote:
>>>>>>>
>>>>>>>> When I do 'wgrib ruc2anl_130_20090605_2200_000.grb,' I get,
>>>>>>>> '1:0:d=09060522:HGT:kpds5=7:kpds6=100:kpds7=1000:TR=1:P1=0:P2=0:TimeU=1:1000
>>>>>>>> mb:anl:NAve=0,' and that is the only output. I am not sure
>>>>>>>> exactly how to read that output and if that indicates that there is
>>>>>>>> only
>>>>>>>> one field.
>>>>>>>>
>>>>>>>> I got the length of the .grb file as only 228074 bytes: -rwxr-xr-x
>>>>>>>> 1 mserino 30490 228074 Apr 7 17:39
>>>>>>>> ruc2anl_130_20090605_2200_000.grb
>>>>>>>>
>>>>>>>> Michelle M. Serino
>>>>>>>> M.S. Candidate
>>>>>>>> Texas A&M University
>>>>>>>> Atmospheric Sciences Dept.
>>>>>>>> Eller O&M, Suite 1017
>>>>>>>> College Station, TX 77843
>>>>>>>> (484) 769-0949 <%28484%29%20769-0949>
>>>>>>>>
>>>>>>>> On Tue, Apr 19, 2016 at 6:00 PM, George Trojan - NOAA Federal <
>>>>>>>> <george.trojan@xxxxxxxx>george.trojan@xxxxxxxx> wrote:
>>>>>>>>
>>>>>>>>> Does wgrib list only one field? I just ran convert on the same
>>>>>>>>> file. There should be 317 data fields. What is the length of .grb
>>>>>>>>> file? I
>>>>>>>>> got 65403248 bytes.
>>>>>>>>>
>>>>>>>>> George
>>>>>>>>>
>>>>>>>>> On Tue, Apr 19, 2016 at 9:31 PM, Michelle Serino <
>>>>>>>>> <mmserino@xxxxxxxx>mmserino@xxxxxxxx> wrote:
>>>>>>>>>
>>>>>>>>>> Thank you everyone for getting back to me.
>>>>>>>>>>
>>>>>>>>>> My apologies for not including the location of my data. I
>>>>>>>>>> downloaded it from <http://nomads.ncdc.noaa.gov/data/rucanl/>
>>>>>>>>>> http://nomads.ncdc.noaa.gov/data/rucanl/ for the appropriate
>>>>>>>>>> date and time. Let me respond to everyone's suggestions.
>>>>>>>>>>
>>>>>>>>>> 1. When I run:
>>>>>>>>>>
>>>>>>>>>> od -c ruc2anl_130_20090605_2200_000.grb | less
>>>>>>>>>>
>>>>>>>>>> I get output:
>>>>>>>>>>
>>>>>>>>>> 0000000 051107 041111 075003 000752 000000 001034 064407 100202
>>>>>>>>>>
>>>>>>>>>> 0000020 062007 164003 003011 013005 000400 000000 000001 000000
>>>>>>>>>>
>>>>>>>>>> 0000040 000025 000400 000000 000052 001777 141401 050401 037400
>>>>>>>>>>
>>>>>>>>>> 0000060 100631 135354 100410 014163 032000 000351 164464 040000,
>>>>>>>>>> etc.
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> 2. When I run:
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> wgrib ruc2anl_130_20090605_2200_000.grb
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> I get output:
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> 1:0:d=09060522:HGT:kpds5=7:kpds6=100:kpds7=1000:TR=1:P1=0:P2=0:TimeU=1:1000
>>>>>>>>>> mb:anl:NAve=0
>>>>>>>>>>
>>>>>>>>>> 3. When I run:
>>>>>>>>>>
>>>>>>>>>> wgrib -v ruc2anl_130_20090605_2200_000.grb
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> I get similar output:
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> 1:0:D=2009060522:HGT:1000 mb:kpds=7,100,1000:anl:winds in grid
>>>>>>>>>> direction:"Geopotential height [gpm]
>>>>>>>>>>
>>>>>>>>>> 4. George, you mentioned that "It is possible that the program
>>>>>>>>>> does not know about GRIB2 fields, specific to the office that
>>>>>>>>>> created the
>>>>>>>>>> file." By program, are you referring to the program I am using to
>>>>>>>>>> process
>>>>>>>>>> these data? Or are you referring to another program? I am not able
>>>>>>>>>> to
>>>>>>>>>> test the wgrib2 command at the moment because I do not have the
>>>>>>>>>> permissions
>>>>>>>>>> to install it (my advisor will have to do that).
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> It appears that wgrib at least recognizes the *.grb file. Are
>>>>>>>>>> the lists of the grids as they should be? Perhaps something small is
>>>>>>>>>> getting lost in the conversion process that my processing code (using
>>>>>>>>>> Warning Decision Support System - Integrated Information algorithms)
>>>>>>>>>> cannot
>>>>>>>>>> reconcile?
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> Thank you for the continued help!
>>>>>>>>>>
>>>>>>>>>> Michelle
>>>>>>>>>>
>>>>>>>>>> Michelle M. Serino
>>>>>>>>>> M.S. Candidate
>>>>>>>>>> Texas A&M University
>>>>>>>>>> Atmospheric Sciences Dept.
>>>>>>>>>> Eller O&M, Suite 1017
>>>>>>>>>> College Station, TX 77843
>>>>>>>>>> (484) 769-0949 <%28484%29%20769-0949>
>>>>>>>>>>
>>>>>>>>>> On Mon, Apr 18, 2016 at 1:52 AM, Michael Dross <
>>>>>>>>>> <mdross@xxxxxxxxxxxxxxxxxx>mdross@xxxxxxxxxxxxxxxxxx> wrote:
>>>>>>>>>>
>>>>>>>>>>> If you have wgrib installed do a
>>>>>>>>>>>
>>>>>>>>>>> wgrib -v ruc2anl_130_20090605_2200_000.grb
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> If it converted it correctly you should get a listing of the
>>>>>>>>>>> grids. If not, something is amiss with the cnvgrib you have.
>>>>>>>>>>>
>>>>>>>>>>> - Mike
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> Sent from my iPad
>>>>>>>>>>>
>>>>>>>>>>> On Apr 17, 2016, at 9:11 PM, Michelle Serino <
>>>>>>>>>>> <mmserino@xxxxxxxx>mmserino@xxxxxxxx> wrote:
>>>>>>>>>>>
>>>>>>>>>>> Hi all,
>>>>>>>>>>>
>>>>>>>>>>> I am using NCAR's cnvgrib utility to convert grib2 RUC-Analysis
>>>>>>>>>>> data to grib1. An example of one of these data files is below:
>>>>>>>>>>>
>>>>>>>>>>> ruc2anl_130_20090605_2200_000.grb2
>>>>>>>>>>>
>>>>>>>>>>> When I run the command,
>>>>>>>>>>>
>>>>>>>>>>> cnvgrib
>>>>>>>>>>> -g21 ruc2anl_130_20090605_2200_000.grb2
>>>>>>>>>>> ruc2anl_130_20090605_2200_000.grb
>>>>>>>>>>>
>>>>>>>>>>> the converting process seems to work. I get a file with an
>>>>>>>>>>> extension of "grb." However, while the file type is listed as a
>>>>>>>>>>> data file,
>>>>>>>>>>> it appears green, as an executable file. In addition, the code I
>>>>>>>>>>> am using
>>>>>>>>>>> does not recognize the file as a grb1 file. (The code only
>>>>>>>>>>> recognizes grb1
>>>>>>>>>>> files.) The website to which I referred to use this utility (
>>>>>>>>>>> <http://blog.nssl.noaa.gov/wdssii/>
>>>>>>>>>>> http://blog.nssl.noaa.gov/wdssii/) did not list any additional
>>>>>>>>>>> options to use besides the required '-g21.' Am I missing
>>>>>>>>>>> something? This
>>>>>>>>>>> is my first implementation of the cnvgrib utility, so I apologize
>>>>>>>>>>> if I have
>>>>>>>>>>> made a naive mistake. I appreciate any help.
>>>>>>>>>>>
>>>>>>>>>>> Best regards,
>>>>>>>>>>>
>>>>>>>>>>> Michelle Serino
>>>>>>>>>>>
>>>>>>>>>>> Michelle M. Serino
>>>>>>>>>>> M.S. Candidate
>>>>>>>>>>> Texas A&M University
>>>>>>>>>>> Atmospheric Sciences Dept.
>>>>>>>>>>> Eller O&M, Suite 1017
>>>>>>>>>>> College Station, TX 77843
>>>>>>>>>>> (484) 769-0949 <%28484%29%20769-0949>
>>>>>>>>>>>
>>>>>>>>>>> _______________________________________________
>>>>>>>>>>> ldm-users mailing list
>>>>>>>>>>> <ldm-users@xxxxxxxxxxxxxxxx>ldm-users@xxxxxxxxxxxxxxxx
>>>>>>>>>>> For list information or to unsubscribe, visit:
>>>>>>>>>>> <http://www.unidata.ucar.edu/mailing_lists/>
>>>>>>>>>>> http://www.unidata.ucar.edu/mailing_lists/
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>
>>>>>
>>>>
>>>
>>>
>>
>>
>