NOTE: The decoders mailing list is no longer active. The list archives are made available for historical reasons.
On Wed, 22 Dec 2004, Francois Vandenberghe wrote:
Robb, I installed the 3.0.7 decoders on my computer, the gridtocdl and gridump work fine but the gridtonc crashed at the very beginning. I suspect the problem is that my NETCDF library is compiled with pgf90 and I used the defaut compilation option (gcc) with the decoder. I need to fix this problem first before I can test the tiling. I'll let you know my progress (if any) and/or my problems (certainly).
francois, good news, yesterday i downloaded a couple different files and decoded to netcdf, it worked w/o errors. i'll attach the cdl used. i agree the netcdf and udunits libraries need to be rebuilt. a word of advice on building on linux, do % setenv CC gcc before running configure to get the compiler flags correct. robb...
Thanks, Francois On Tue, 21 Dec 2004, Robb Kambic wrote: > francois, > > with the test file, i discovered that level 245 was missing. it was > recently added. a new decoders release was made decoders-3.0.7.tar.Z. > it's available in the ftp directory. > > after building the release, run gribtocdl again to create a new cdl, then > run gribtonc. i don't know if this will solve your problem but it's a step > in the right direction. > > robb... > > > > > > On Tue, 21 Dec 2004, Francois Vandenberghe wrote: > > > > > Setting back to 255 brings the previous problem of the La1 error message: > > > > Dec 21 19:22:53 gribtonc.exe[9804]: 379, eta.t00z.awip21806.nc nav. > > mismatch La1: 33.011 != 19.9433 > > > > as I specify the south west corner at 19.9433N while the tile I am reading > > begins at 33.011N > > > > I have attached the tile file (it's binary but not very big) in case your > > connection would allow this transfer. > > > > Thanks again, > > > > Francois > > > > > > On Tue, 21 Dec 2004, Robb Kambic wrote: > > > > > On Tue, 21 Dec 2004, Francois Vandenberghe wrote: > > > > > > > > > > > Robb, > > > > > > > > Thank for the quick response. I have made the modifications you suggested, > > > > ie setting La1/Lo1 to the south west corner of the full ETA218 grid > > > > and setting the grid_number to the list of all the tiles (there are 54, > > > > see map at http://www.emc.ncep.noaa.gov/mmb/research/tiles.218.html). > > > > I ran into a new problem: it seems that the GRIB file carries the "255" > > > > code for all the tiles and I got the following error messages: > > > > > > > > Dec 21 15:54:04 gribtonc.exe[28513]: 374 grid 255 not in grid_number list > > > > in eta.t00z.awip21806.nc > > > francois, > > > > > > sometimes in stitched together grid files the grid numbers are not used, > > > only 255 is used. so replace with grid_number = 255; > > > > > > > > I have attached my CDL file for your information and put two tiled > > > > files (tile 23 and tile 32) at: > > > > ftp://ftp.rap.ucar.edu:incoming/irap/vandenb/eta.t00z.awip21806.23 > > > > ftp://ftp.rap.ucar.edu:incoming/irap/vandenb/eta.t00z.awip21806.32 > > > > > > i get connection timed out errors on retrieval or chdir errors > > > > > > GET ftp://ftp.rap.ucar.edu:incoming/irap/vandenb/eta.t00z.awip21806.23 > > > > eta.t00z.awip21806.23 > > > <HTML> > > > <HEAD><TITLE>An Error Occurred</TITLE></HEAD> > > > <BODY> > > > <H1>An Error Occurred</h1> > > > 404 Can't chdir to irap > > > </BODY> > > > </HTML> > > > > > > > > > > > Other files can be found under the naming eta.tAAz.awip218FF.TT > > > > (where AA is the analysis time, FF the forecast lead time and > > > > TT the tile number) at: > > > > ftp://ftpprd.ncep.noaa.gov/pub/data/nccf/com/eta/prod/eta.20041221/ > > > > > > i get: > > > > > > Can't build data connection: Connection timed out > > > > > > > > > i'm working at home and my connection doesn't permit tranfering of large > > > data files. if the above solution doesn't work, i look at it tomorrow. > > > > > > robb... > > > > > > > > Thanks for the support, > > > > > > > > Francois > > > > > > > > > > > > > > > > On Tue, 21 Dec 2004, Robb Kambic wrote: > > > > > > > > > On Mon, 20 Dec 2004, Unidata Support wrote: > > > > > > > > > > > > > > > > > ------- Forwarded Message > > > > > > > > > > > > >To: support-decoders@xxxxxxxxxxxxxxxx > > > > > > >From: "Vandenberghe" <vandenb@xxxxxxxx> > > > > > > >Subject: netCDF Decoders - Decoding/Tiling ETA218 (12km) with gribtonc > > > > > > >Organization: ? > > > > > > >Keywords: 200412201842.iBKIgRcV018844 netCDF decoders gribtonc > > > > > > > > > > > > --=_76dbc2476f6428fae144519a869f88c1 > > > > > > Content-Type: text/plain; charset="ISO-8859-1" > > > > > > Content-Transfer-Encoding: 7bit > > > > > > > > > > > > Institution: NCAR > > > > > > Package Version: gribtonc > > > > > > Operating System: LINUX > > > > > > Hardware Information: Intel Pentium > > > > > > Inquiry: Hi, > > > > > > > > > > > > I am trying to use gribtonc to convert and tile the ETA 218 tiled > > > > > > products available from: > > > > > > > > > > > > ftpprd.ncep.noaa.gov:/pub/data/nccf/com/eta/prod/ > > > > > > > > > > Francois, > > > > > > > > > > i went to the site, there were no files available. > > > > > > > > > > > > I have attached my CDL file. The program works fine with one tile, but > > > > > > when I try to use it with several tiles, then I got the following error > > > > > > message: > > > > > > > > > > > > Dec 20 18:29:49 gribtonc.exe[29974]: 1, eta.t00z.awip21806.nc nav. > > > > > > mismatch La1: 33.011 != 19.9433 > > > > > > > > > > > I didn't receive the cdl but here' a few comments. the La1 and Lo1 should > > > > > be the south west corner of the full ETA218 domain, sound like you coded > > > > > ti correctly. the grib_number should be all the grid tiles, for ecmwf it > > > > > is: grid_number = 1,2,3,4,5,6,7,8,9,10,11,12 ; > > > > > > > > > > > Here 33.011 is the latitude of the south west corner of the tile, while > > > > > > 19.9433 is the south west corner of the full ETA218 domain. One tile > > > > > > having the dimensions 69x72 while the full domain have the dimensions > > > > > > 491x303. > > > > > > > > > > > > I was able to tile other products such as ECMWF tiled data which were > > > > > > not on a projection map. So I was wondering if the projection is the > > > > > > problem. > > > > > > > > > > all files should have the same projection > > > > > > > > > > robb... > > > > > > > > > > > > > > > > > > > > > > Thanks in advance for any help, > > > > > > > > > > > > Francois > > > > > >
============================================================================== Robb Kambic Unidata Program Center Software Engineer III Univ. Corp for Atmospheric Research rkambic@xxxxxxxxxxxxxxxx WWW: http://www.unidata.ucar.edu/ ==============================================================================
decoders archives: