[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: THREDDS GRIB-1 Collections

I think i have this fixed.

The newest release candidate is here if youd like to test:



On 5/22/2013 10:02 AM, Kevin Manross wrote:

Yeah, that looks to be it.


On 5/22/13 10:02 AM, John Caron wrote:
On 5/22/2013 9:02 AM, Kevin Manross wrote:

Hello John,

I'll bet that you are pretty swamped with things.  I wanted to check
on the status of the GRIB-1 Collections with regard to the "must be
unique within group" error I encountered when trying to set up a
timePartition and dateFormat.

Can you let me know where that stands and if there is any timeline
for resolution?  This will help me in my planning.

Many thanks!


-- Kevin Manross NCAR/CISL/Data Support Section Phone: (303)-497-1218 Email:address@hidden <mailto:address@hidden> Web:http://rda.ucar.edu

Hi Kevin:

I think I have this fixed. Is  this the same as the message below?


Hi John,

Sorry to be such a pain in the backside with all these

Conversion from GRIB1 to netcdf was done with netcdfAll-4.3.jar.
Version stuff follows:

Manifest-Version: 1.0
Implementation-Vendor: UCAR/Unidata
Implementation-Title: NetCDF-Java Library
Implementation-Version: 4.3.15
Implementation-Vendor-Id: edu.ucar
Built-On: 20121218.1413
Built-By: caron
Build-Jdk: 1.7.0_05

I then dumped the netCDF file headers using ncdump from the (assumed)
C library on yellowstone (netcdf library version 4.2 of Aug  3 2012

The Pressure_surface dump was merely a grep of
Pressure_surface:Grib1_TableVersion of all the resulting .cdl header files

The bleeding edge thredds war file was offered to me by Lansing to
test the timePartition fixes for GRIB1

I'm attaching the first two files of that dump list.  They're gzipped
for email only.  We normally don't zip them (though I don't suspect
the compressing to be a factor).

Let me know if you need any more info.  Thanks!


On 5/2/13 4:27 PM, John Caron wrote:
Hi Kevin:

On 5/2/2013 4:12 PM, Kevin Manross wrote:


I know there is work being preformed on the GRIB1 in THREDDS, so I do not know if this already being covered or not. I came across the error in the subject line when I expanded my TDS catalog config to cover the extent of my dataset. This caused an Error 500 when trying to view Best Timeseries data via any of the TDS access types/services.

I'll spare the gory details unless they are needed.  I have isolated
this error to a period spanning two months in 1999 (Aug & Sep).

converting the GRIB1 files to netCDF and dumping the netCDF headers,

using what program ?

only variation I see with the Pressure_surface variable is that some
Grib1_TableVersion = 1 and the rest use version 2:

fnl_19990814_00_00.cdl: Pressure_surface:Grib1_TableVersion = 2 ;
fnl_19990814_12_00.cdl: Pressure_surface:Grib1_TableVersion = 1 ;
fnl_19990814_18_00.cdl: Pressure_surface:Grib1_TableVersion = 1 ;
fnl_19990815_00_00.cdl: Pressure_surface:Grib1_TableVersion = 2 ;
fnl_19990815_06_00.cdl: Pressure_surface:Grib1_TableVersion = 2 ;

what is this a dump from?

maybe send me two of the original grib files, one from each version

I would be glad to dig into this further if it provides any help.

I am currently using THREDDS Data Server [Version 4.3.17-SNAPSHOT -

wow, thats recent, did you build it yourself ??

-- Kevin Manross NCAR/CISL/Data Support Section Phone: (303)-497-1218 Email:address@hidden <mailto:address@hidden> Web:http://rda.ucar.edu

-- Kevin Manross NCAR/CISL/Data Support Section Phone: (303)-497-1218 Email:address@hidden <mailto:address@hidden> Web:http://rda.ucar.edu

NOTE: All email exchanges with Unidata User Support are recorded in the Unidata inquiry tracking system and then made publicly available through the web. If you do not want to have your interactions made available in this way, you must let us know in each email you send to us.