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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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.
Mike, The problem appears to be the -g flag in the dcgrib actions. This points to /gemtbl in your pqact.conf, but that directory doesn't exist. This might have been a link at one time since I recall you had a really long path name that exceeded the line length. The idea is to define the GEMTBL directory using the -g flag so that the decoder can find all those tables under $GEMTBL/grid, so probably -g /export/meaculpa/nawips/gempak5.4/tables would work, or create your link from /export/meaculpa/nawips/gempak5.4/tables to /gemtbl. This also appears in the -s and -p options in the dchrly and other decoders. So either you made the link previously to save the typing- or you needed it to save a line length problem. At any rate, the quickest solution is to reinstate the link that someone removed- but you might want to jest put that whole path in the pqact.conf entries. Steve On Tue, 1 Feb 2000, Mike Trexler wrote: > Steve, > > Thanks for the advice on compiling nawips. I'm sure that you would have > no trouble believing that we were running SC3.0.x on the unity machines. > But, we now have access to the new compilers so everything look like > it'll work out. > > New question....Suddenly all the gempak products stopped writing to the > right place and I can't figure out why. I read the dcgrib.log which indicated > that it couldn't write to the right place. So, then I changed the pqact file > to explicitly define GEMHDS, but still got the same results. Here's what > the log says: > > Feb 01 22:38:32 dcgrib[9787]: Starting Up > Feb 01 22:38:32 dcgrib[9787]: Error creating file: > -1/newdata/gempak/realtime/00020118_meso21 > Feb 01 22:38:32 dcgrib[9787]: can't open file > /newdata/gempak/Realtime/00020118_meso212.gem > Feb 01 22:38:32 dcgrib[9787]: Exiting > Feb 01 22:38:32 dcgrib[9787]: 1 WMO msgs, 1 GRIBs decoded, 0 written > Feb 01 22:38:33 dcgrib[9788]: Starting Up > Feb 01 22:38:33 dcgrib[9788]: Error creating file: > -1/newdata/gempak/realtime/00020118_meso21 > Feb 01 22:38:33 dcgrib[9788]: can't open file > /newdata/gempak/Realtime/00020118_meso212.gem > > I also checked to make sure all the permissions on the directory were correct > and > nothing had changed. Any ideas? > > Also this is not just happening with the model data, but also with the > surface and > upperair > files. What's funny though is that there are no entries in the log files for > the past > 2 days at all, but I know that the data is coming in because the text > products (and grib > > model datasets) are being written fine. Can't understand what's up andwhy it > suddenly > started happening!?!?!? > > Thanks for your help! > Mike > > > > > Steve Chiswell wrote: > > > Mike, > > > > Was the distribution on the 2.5.1 system compiled from scratch on that > > platform, > > or was it copied over from the 2.7 system? > > > > What version of the C & fortrans compilers are the 2 systems running? > > We are running SC5.0 here. > > > > It could be that the motif interface library was modified between the 2 > > Solaris releases. The oldest system we have here is 2.6. > > > > Steve > > > > On Thu, 20 Jan 2000, Mike Trexler wrote: > > > > > Hey Steve, > > > > > > Hope you had a good trip. I downloaded the latest gempak distrib (through > > > patch 15) and installed it on a couple of Suns running Solaris 2.7. > > > Everything > > > seems to work nicely in all respects. Now I also installed this same > > > distribution > > > on a Sun running Solaris 2.5.1. I was less sucessful here. The problem > > > is with > > > Garp. At first you think everything is ok, because it starts up fine and > > > you can > > > plot surface data, nids/nowrad/satellite data all fine. The models part > > > is where > > > things get weird. Garp will crash and shut down if you open up any of > > > the model > > > related widgets (cross sections, plan proj, profs...) after having > > > already open one > > > of these widgets once. Like say I looked at eta 500mb hghts and pressed > > > "display and close", then when I open up the plan proj widget the next > > > time it > > > will crash. However, if you put up the 500mb hghts and choose "display" > > > rather > > > than also closing the window you can add as many other things and clear as > > > many times as you like, so long as you never close the first window. > > > > > > Any ideas? > > > > > > Thanks > > > Mike > > > > > > -- > > > __________________________________________________ > > > ************************************************** > > > ****** C. Michael Trexler ****** > > > ****** N C State Univ. - Box 8208 ****** > > > ****** Dept. of Mar/Earth/Atmos. Sciences ****** > > > ****** Research III Rm. 100 ****** > > > ****** Raleigh, NC 27695-8208 ****** > > > ****** (919) 515-1447 Phone ****** > > > ****** (919) 515-1683 Fax ****** > > > ************************************************** > > > -------------------------------------------------- > > > ****** address@hidden ****** > > > __________________________________________________ > > > > > > > > > > > > > > -- > __________________________________________________ > ************************************************** > ****** C. Michael Trexler ****** > ****** N C State Univ. - Box 8208 ****** > ****** Dept. of Mar/Earth/Atmos. Sciences ****** > ****** Research III Rm. 100 ****** > ****** Raleigh, NC 27695-8208 ****** > ****** (919) 515-1447 Phone ****** > ****** (919) 515-1683 Fax ****** > ************************************************** > -------------------------------------------------- > ****** address@hidden ****** > __________________________________________________ > > > >