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

20010404: Garp - level widget problem

Luis,

For my testing purposes, is this under Digital Unix? 

Steve Chiswell
Unidata User SUpport




>From: LUIS M FARFAN <address@hidden>
>Organization: UCAR/Unidata
>Keywords: 200104041950.f34Jo3L06721

>
>--Boundary_(ID_wsv72YLIr10c+yUyEDIY3A)
>Content-type: text/plain; charset="us-ascii" ; format="flowed"
>
>I want to report a problem that has been occurring in our model 
>displays with garp.
>This happens when the model level (lev1) is searched and garp is 
>terminated with
>a core dump.
>
>Initially, the "model plan view" looks normal and I have attached a 
>gif file with this
>message to show its appearance. In this example, there is only one 
>file in the directory
>$GEMDATA/hds and this file is called 2001040412_eta211.gem. The application
>of the program "gdinfo" indicates that there is data at several 
>pressure, height and other
>levels. As a second attachment, there is a copy of the output from 
>gdinfo on this file.
>In addition, we can use the garp function/widget info to open a list 
>of the file content.
>By the way, this eta file comes from the ldm/idd data-stream.
>
>The problem occurs when the widget called "lev1" is clicked. After 
>this is done, it takes
>garp a few seconds to terminate its execution with a:
>       Floating exception (core dumped)
>If a valid level is typed (manually entered into the space next to 
>lev1), we can get the
>display of any field. So, the problem is reduced to have garp tell us 
>the available levels
>for a given coordinate system. Whenever this is attempted, there is a 
>termination and
>generation of a core file.
>
>I found, in your archive section of the unidata/gempak web-site, 
>comments from other
>users regarding similar problems but none of them helped me to 
>identify and fix my
>problem. I looked into the "core" file via dbx, limited the content 
>of the hds directory
>to a single file, etc.
>Do you have any comments or suggestions about this problem?
>
>Thank you,
>
>Luis M Farfan.
>
>