Re: [thredds] featureCollection - Not reading GRIB-2 file

On 12/19/2011 5:42 AM, Comiskey, Glenn wrote:
Thanks for the reply John.

Further to my posting and your reply though I have subsequently noted
that another dataset I've published appears to have separate reference
times for each record.

I have attached details in the file icec.doc, which I've successfully
published and which presents over 45000 time slices (see MHT file).

         <featureCollection featureType="FMRC" name="Global ice cover
proportion" harvest="true" path="icecon">
             <collection
spec="//192.168.2.121/data/cfsr/grib2/flx/cfsr.icecon.gdas.*grb2"/>
         </featureCollection>

I' assuming there is some difference between the ICEC and SALTY data
files, but I can't readily see why one publishes successfully and the
other doesn't - albeit I get the impression neither should.

Hi Glenn:

It looks to me like the SALTY data will be wrong if using FMRC with 4.2. If possible, you should check the time coordinates carefully against what they are supposed to be. This will be fixed with 4.3, although Im not yet sure if we will have the 2D time coordinates, but perhaps just the "best" dataset. Does that matter to you?

So for anyone else, CDM 4.2 assumes that GRIB files have a common reference time in any one file. Please check your datasets!

John



  • 2011 messages navigation, sorted by:
    1. Thread
    2. Subject
    3. Author
    4. Date
    5. ↑ Table Of Contents
  • Search the thredds archives: