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

20010828: GDRADR fails



Tom,

The error messages you are seeing below indeicate that some of the
GEMPAK environmental variables are not correctly configured, or
you have not sourced the Gemenviron file in your script. Remember
that sourcing Gemenviron defines the environmnetal variables
as well as adds the GEMPAK $GEMEXE directory to your current path.

The [IP -10] message indicates that the $GEMPARM and $GEMPDF directories
are not found. The [FL -1] errors indicate that the $GEMTBL/config directory 
is not being found. The most likely cases for both these errors are that the
environmental variables that GEMPAK is referring to are either not set in the
script environment, or you might be pointing to an old GEMPAK tree and not the
current distribution.

Steve Chiswell
Unidata User Support


>From: address@hidden
>Organization: UCAR/Unidata
>Keywords: 200108281830.f7SIUn100950

>
>Steve:
>
>We've been trying to use GDRADR on a RedHat 7 system 
>(cacimbo.ggy.uga.edu). We had some success and generated mosaics using a 
>script to rund GDRADR, following exactly what you suggested in the
>documentation. However, I had to reboot the system for maintenance
>and found that when we tried to restart the script that starts GDRADR,
>it was failing to produce output files. 
>
>We then checked to make sure all processes were dead and did a gpend and 
>checked the ipcs. After we were sure things were clean, we restarted 
>GDRADR. Instead of giving us a list of options, we get...
>
>[ldm@cacimbo ~/mosaic]$ gdradr
> [IP -10] GDRADR
> [FL -1] datatype.tbl
> [GDRADR 2] $RAD/NIDS/%SITE%/%PROD%
> [FL -1] splpat.tbl
> [FL -1] combo.tbl
> [GEMPLT -1]
>waiting 300 seconds<ctrl-c to continue>
>
>(You'll notice this was user ldm, but we also ran as user gempak with 
>the same problems. I have checked all the file/directory permissions and 
>they are fine. Both ldm and gempak are in the same group and all 
>directories/files are group writable.)
>
>We're not sure where to look next.
>
>Thanks.
>
>Tom
>
>Thomas L. Mote
>University of Georgia
>
>