20000117: 20000114: GARP flakiness

Tom,

Its possible that if you either downloaded the new distribution
in the same directory as your previous, or if your
old $GEMLIB files were still in place, that make would not build
some of the new library routines.

Glad you have gotten past your previous problems. Sorry that you
had such a hassle.

Steve Chiswell
Unidata User Support


>From: "Thomas L. Mote" <address@hidden>
>Organization: .
>Keywords: 200001171235.FAA20643

>
>Steve,
>
>For the third time in the last few weeks, I have rebuilt 
>GEMPAK. This time did the trick. To be safe, I completely 
>removed the gempak5.4 directory first before untarring the 
>file.
>
>I'm not sure why it failed before. I downloaded the full 
>tar file for pl15. I then untarred it and did a top level 
>make. 
>
>Nevertherless, it worked this time. Thanks.
>
>Tom
>
>
>On Fri, 14 Jan 2000 15:24:51 -0700 Unidata Support 
><address@hidden> wrote:
>
>> 
>> Tom,
>> 
>> Did you download pl15 into a clean directory tree- or was in ontop
>> of your current distribution? The problems you mention sound like
>> those that were patched in the new distribution. Its possible that
>> if you downloaded on top of your current distribution, the libraries
>> in $GEMLIB were not rebuilt first.
>> 
>> Steve Chiswell
>> Unidata User SUpport
>> 
>> 
>> >From: Thomas Mote <address@hidden>
>> >Organization: .
>> >Keywords: 200001141542.IAA07889
>> 
>> >
>> >Steve:
>> >
>> >I've been running into some problems with GARP that I just noticed after I
>> >installed pl15 this morning.
>> >
>> >I had installed pl12 a few weeks ago. I noticed GARP was bombing when I
>> >tried to look at satellite imagery. I noticed something in the e-mail
>> >archives that suggested upgrading to pl14 or 15. I did a clean install of
>> >GEMPAK to pl15 this morning.
>> >
>> >I run into some odd problems, some of which are obviously Y2K related.
>> >First, I can call up a surface map or satellite image without problem. On
>> >the surface maps, the title is:
>> >
>> >FRI /1 0001 00 METAR Obs
>> >
>> >for the 14Z map.
>> >
>> >I can call up a satellite image without problem, but the available times
>> >listed look like:
>> >
>> >1000114/1315
>> >
>> >My file name for this satellite image is: VIS_000114_1315
>> >There is also a strange title, just like the surface map.
>> >
>> >The problem is when I try to overlay surface data on a satellite image.
>> >If I load the satellite image and then load the surface data using a
>> >display, then a close, no problem. However, if I try to use the "Display
>> >& Close" option, it bombs with the following error.
>> >
>> >GEMPAK: [FL 21] /data/gempak/surface/sao
>> >GEMPAK: [SF -2]  File /data/gempak/surface/sao could not be opened.
>> >Segmentation Fault (core dumped)
>> >
>> >Obviously, the data is there or I couldn't have displayed it the first
>> >time.
>> >
>> >I had the same problem with upper-air soundings, but not upper-air maps.
>> >It bombed when using the "Display & Close" option. It also bombed on one
>> >sounding I tried to call up even with the display option. I think that may
>> >have been for a sounding that didn't exist, but it still shouldn't have
>> >crashed. I was able to successfully call up Peachtree City's sounding just
>> >using the display option. 
>> >
>> >I got sidetracked, had to call up and then close netscape. I tehn tried
>> >GARP again to keep working on it and got the following
>> >
>> >cacimbo% garp
>
>> >G A R P - v2.02 starting...
>> >!!!!!Whoa!!!!!!  Free() got a 0x0
>> >Segmentation Fault (core dumped)
>> >cacimbo% 
>> >
>> >I am also having trouble calling up the model output, but that may be due
>> >to the fact I don't have the Garp_Defaults file correct yet. I changed
>> >some of the file nameing from the ldm.
>> >
>> >You probably already know about some of this, but I thought I should write
>> >anyway.
>> >
>> >Tom
>> >
>> >Thomas L. Mote
>> >
>> 
>> ****************************************************************************
>> Unidata User Support                                    UCAR Unidata Program
>> (303)497-8644                                                  P.O. Box 3000
>> address@hidden                                   Boulder, CO 80307
>> ----------------------------------------------------------------------------
>> Unidata WWW Service                        http://www.unidata.ucar.edu/     
>> ****************************************************************************
>
>
>