[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

20000613: Gempak - making gifs




Mike,
Chris' problem turned out to be that he had edited $GEMTBL/colors/coltbl.tbl and
added more than 32 colors into the table. As it turned out, if you use NTL which
uses the color tables in $NAWIPS/tables/nawips.clr, then the problem was 
bypassed.
When he ran the gf driver without ntl, then the $GEMTBL/colors files are used.
See : http://www.unidata.ucar.edu/glimpse/gempak/3600

In general, if you are not using ntl, which makes sure you have all the colors 
you will need),
then you run the risk of the driver not being able to obtain all the colors it 
needs for
the graphic if you are sending to a display with browsers, answerbook etc 
running on it.

Steps to check for your problem.

1) make sure you don't have any hung massage queue that are causing trouble 
(ipcs will show these).

2) Make sure you have your DISPLAY environmental variable set so that the 
driver can contact the
   X display.

3) Check to see if you have a file or directory called "gf" in your woring 
directory. If so,
   gempak will try to execute that as the driver instead of $GEMEXE/gf.

4) Make sure you can get all the colors you need by launching ntl.


Have you previously had success and now have a problem, or have you just 
started to use the gf driver?
Remember, when running from a cron that you don't have the DISPLAY variable set 
in the environment
until you define it in the script. Also make sure you can run other X programs 
like xterm to
the display (Sun has magic cookies, and you must be allowed to draw to the 
display).


Steve Chiswell
Unidata User Support



>From: Michael Voss <address@hidden>
>Organization: UCAR/Unidata
>Keywords: 200006131834.e5DIYuT15191

>Steve,
>
>I'm trying to make some gifs with device=gf in gempak. I get an error:
>*** TERMINATING  gf
>*** Received signal 11 SIGSEGV
>
>I looked in the support archives and followed a thread you had going with Chri
> s Hennon who seemed to have a similar problem. If I follow the conclusion of 
> that correctly, you ended up copying over an executable built on your machine
> :
>
>>> I copied over my solaris executable of $GEMEXE/gf (moved the one built
>>> on your machine to $GEMEXE/gf.old. All appears to work fine using
>>> or copy. You will want to verify this on your own of course.
>
>...and this cleared up at least this portion of his problem.
>
>I'm running Solaris 2.6 with gempak patched through pl15. Could you make a sug
> gestion as to how I might proceed in my touble shooting efforts?
>
>Thank You,
>
>Mike 
>
>
>--------------------------
>Mike Voss                                
>Department of Meteorology              
>San Jose State University                        
>One Washington Square                             
>San Jose, CA 95192-0104   
>             
>408.924.5204 voice
>408.924.5191 fax   
>