This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.
On Mon, 10 Feb 2003, HOETH, BRIAN R. (JSC-ZS) (LM) wrote: > Robb, > > Thanks for all the info. I've been really busy and am just now getting > around to trying to install everything. Unfortunately, I've run into a snag > immediately when trying to install LDM version 5.2.2? First a little > background for what was on the machine before I even tried anything: > > We are currently running LDM on OS of HP-UX 11.11. We had LDM version 5.1.3 > installed on our HP under OS HP-UX 10.20 and then upgraded the OS to HP-UX > 11.11 and didn't bother reinstalling LDM after the OS upgrade because the > LDM binaries appeared to be forward compatible since LDM still worked after > the OS upgraded. Brian, I apologize for the slow response, I attended the AMS all last week and yesterday was a holiday. Just cathing up on my email. One can't build the LDM on HP-UX 11.11, HP changed the rpc libraries to use different names, etc. The code actually compiles but it doesn't run. We support HP-UX 11.11 with the HP-UX 10.20 binaries since they are upwards compatibile. To obtain the binaries: % ftp www.unidata.ucar.edu login as anonymous % cd pub/binary/hpux_10.20-hp9000 % get ldm-5.2.binary.tar.Z There are detailed install instructions at: http://www.unidata.ucar.edu/packages/ldm/ldmBinaryInstallList.html Robb... > > Anyways, that should be enough background I hope? I went to install LDM > version 5.2.2 and ran into a snag upon doing the 'make install' step? I've > attached the output from my attempted 'make install'. > > Any idea what's going on? > > I would just install the binary version of 5.2.2, but I don't see a binary > version of 5.2.2? The latest binary version I see is 5.2 and this is for > HP-UX 10.20? > > Thanks, > Brian > > > -----Original Message----- > From: Robb Kambic [mailto:address@hidden] > Sent: Thursday, January 23, 2003 4:07 PM > To: HOETH, BRIAN R. (JSC-ZS) (LM) > Cc: decoders > Subject: RE: Decoders release 2.4.0 > > > On Thu, 23 Jan 2003, HOETH, BRIAN R. (JSC-ZS) (LM) wrote: > > > Robb, > > > > I have been plowing through all of the online documentation on > > my.unidata.ucar.edu and I guess I'm slightly overwhelmed? What are the > > basic steps I should take to install decoders.tar.Z? > > Brian, > > Here's a url to the old web site, the info is more current there. Haven't > had time to update new site my.unidata yet > > http://www.unidata.ucar.edu/packages/decoders/ > > > > > > I guess I need to install netCDF and netCDF perl in order for the > > decoders.tar.Z package to work properly? I am also wanting to upgrade my > > There are 3 packages that need to be install for the decoders packages. > - udunits > - netcdf > - netcdf perl > > I would install them in the above order. > > > > version of LDM to the latest version. Is there a specific order in which > I > > I probably install the LDM first. > > http://www.unidata.ucar.edu/packages/ldm > > I would look at these tags on the page: > > # LDM Preinstallation Requirements > # LDM Binary Installation Check List > # LDM Source Installation Check List > # LDM Setup > > I would install the binary if it's possible. > > > should install things (i.e. LDM first, then netCDF, then netCDF perl, then > > decoders.tar.Z) or does it matter? Also, where do netCDF, netCDF perl, > and > > decoders.tar.Z typically get installed? We have a user account "ldm" > where > > our LDM stuff is installed under /home/ldm, is this the ideal place for > > netCDF, netCDF perl, and decoders.tar.Z to be installed as well or does > it > > matter? > > > A common practise is to install the packages in the LDM home dir. Some > users also make a decoders and util dirs under LDM home where they place > all the binaries. Then ~ldm/decoders ~ldm/util are added to the PATH > variable so all the ldm programs can find them. > > Robb... > > > > > Thanks, > > --------------- > > Brian Hoeth > > Spaceflight Meteorology Group > > Johnson Space Center > > Ph: 281-483-3246 > > Ops: 281-483-1051 > > > > > > > > -----Original Message----- > > From: Robb Kambic [mailto:address@hidden] > > Sent: Friday, December 06, 2002 4:40 PM > > To: address@hidden > > Cc: decoders > > Subject: Decoders release 2.4.0 > > > > > > Hiya, > > > > I'm pleased to announce the beta release of Decoders 2.4.0. This > > release includes many changes as well as a new program, 'gribtocdl'. > > Most of the changes are to add parameters and levels for new models > > and grids, such as the AVN model on a 1-degree grid (avn-1.0x1.0.cdl). > > The new 'gribtocdl' program creates a CDL file from GRIB files. Much > > of the work was contributed by Arild Burud and Terje Reite from the > > Norwegian Service Centre for Climate Modeling at the Norwegian > > Meteorological Institute. I wish to thank these developers on behalf > > of the Unidata community. > > > > Since this is a beta release, I would appreciate help in testing it > > and reporting problems. > > > > A couple of gribtocdl notes: > > > > - Use a representative dataset when creating a CDL file with gribtocdl. If > > the > > dataset is missing a variable, it will also be missing in the CDL. > > > > - Consider the created CDL as a starting point, minor tweaks are usually > > needed. > > > > - The external table file feature for variable name override has not > > been tested much, but needs to be used with gribtocdl as well as > > gribtonc to override variable names. > > > > - Default variable names follow NUWG conventions. > > > > I would have liked to test the code more exhaustively but did not want > > to further delay the release. If problems are found, please send email > > to address@hidden with information necessary to reproduce > > the problem, error messages, and log messages. > > > > Thanks, > > Robb... > > > > > > Included man page from gribtocdl: > > > > User Commands GRIBTOCDL(1) > > > > NAME > > gribtocdl - create netCDF CDL-file from GRIB dataset > > > > SYNOPSIS > > gribtocdl [-v] [-g tabfile] [-q qspecs] gribfiles > > > > DESCRIPTION > > This program decodes GRIB products and generates to stdout > > the corresponding CDL code that can be used in gribtonc(1) > > to convert the GRIB products to netCDF format. The default > > parameter translation is made according to a definition by > > UCAR, but the user may supply specific translations in a > > separate file by the -g tabfile option. The user-supplied > > translations will override the default ones. > > > > OPTIONS > > -v Verbose, reports decoding steps to stderr. > > > > -g tabfile > > Use GRIB-netcdf translation defined in this file. The > > format of this file is described below. > > > > -q qspecs > > Specification for how to expand so-called quasi-regular > > or thinned grids into rectangular grids. This argument > > is a comma-delimited text string that specifies the > > method of interpolation used (currently lin for linear > > or cub for cubic), and the latitude and longitude in- > > crements in degrees for the resulting grid (e.g. > > dlat=2.5,dlon=5.0). > > > > > > FORMAT OF GRIB TO NETCDF TRANSLATION > > The format of the translation table is as follows: Each > > line contains > > GRIB parameter number > > netCDF parameter name > > netCDF long description > > netCDF units > > An optional comment > > > > These are all separated by semicolon (";"). To comment a > > whole line, put any letter or special character (i.e. not a > > digit) in front of it. Trailing spaces are ignored before > > each semicolon, but not after. The GRIB number may have > > spaces in front. Parameters found in the separate file take > > precedence over the default parameter table, those not list- > > ed in the separate file are unchanged. All parameters will > > be named with a level suffix in addition to the name as- > > signed in the separate file. > > > > EXAMPLE > > The command > > gribtocdl -v -g my.tab sst.grib > sst.cdl > > > > will decode the products in sst.grib and generate the > > corresponding CDL code into file sst.cdl. The file my.tab > > contains a specific conversion table to be used when naming > > a parameter from GRIB. Later the GRIB file may be converted > > to netCDF by using gribtonc: > > > > gribtonc -v -g my.tab sst.cdl sst.nc < sst.grib > > > > The file my.tab looks like this: > > Grib;netCDF;long name ;units;comments > > 31 ;CI ;Sea-ice cover ;none ;(0-1) > > 34 ;SSTK ;Sea surface temperature ;K ; > > 151 ;MSL ;Mean sea-level pressure ;Pa ; > > 167 ;T2M ;2 meter temperature ;K ; > > > > SEE ALSO > > gribdump(1), gribtonc(1), netcdf(3), udunits(3) > > > > DIAGNOSTICS > > Error messages and log messages are written to stderr. > > > > > > > > > ============================================================================ > > === > > Robb Kambic Unidata Program Center > > Software Engineer III Univ. Corp for Atmospheric > > Research > > address@hidden WWW: http://www.unidata.ucar.edu/ > > > ============================================================================ > > === > > > > ============================================================================ > === > Robb Kambic Unidata Program Center > Software Engineer III Univ. Corp for Atmospheric > Research > address@hidden WWW: http://www.unidata.ucar.edu/ > ============================================================================ > === > > =============================================================================== Robb Kambic Unidata Program Center Software Engineer III Univ. Corp for Atmospheric Research address@hidden WWW: http://www.unidata.ucar.edu/ ===============================================================================