[Met_help] MET installation problem

[John Halley Gotway]

Litta,

Great, I'm glad you got it compiled.  Thanks for sending that log file.  It makes it very easy to diagnose the problem.

When the MET tools are run, some of them use some data files that are distributed with the MET package.  At compile time, the code remembers the path where it was built, and uses that path the find
those data files it needs at run time.  So if you were to compile MET in on directory and then move it to a different directory, you'd see these types of error messages.  The error messages are just
saying that it can't find those data files it's looking for.

Here's something I'd like you to try... try setting the environment variable "MET_BASE" to the full path for the top-level METv2.0 directory.  Here's how you'd do it using the C-shell:
setenv MET_BASE /full/path/to/METv2.0

And then try rerunning the test scripts.  If that works, then just put the MET_BASE definition into your shell configuration file (.cshrc for C-shell).

Thanks,
John

Litta A J wrote:
> Dear sir,
> 
> Thank you very much for your kind information... I have got all the
> executables after the successful compilation.
> 
> thunder>ls
> ascii2nc*       grid_stat*      mode_analysis*  pcp_combine*
> stat_analysis*
> gen_poly_mask*  mode*           pb2nc*          point_stat*
> wavelet_stat*
> thunder>
> 
> But it was not properly creating output files. test run was not successful.
> I am getting some errors.. I am attaching the test_all.log and out folder
> along with this mail.. I am waiting for your valuable suggestion
> 
> with best regards,
> Litta
> On Tue, Feb 23, 2010 at 8:45 PM, John Halley Gotway <johnhg at ucar.edu> wrote:
> 
>> Litta,
>>
>> It looks like your compiler is having a lot of problems with include files.
>>  For example, it's finding the definition for the "abs" (absolute value
>> function) in two different locations.  This looks
>> like more of a problem with your compiler than with the MET code itself.  I
>> looked at the include directories on our IBM and found the file "stdlib.h",
>> but I did not find the file "cstdlib".  Those
>> are the two files on your system which contain the conflicting definitions
>> of "abs".
>>
>> Unfortunately, there's not much I can do to help.  You'll need to find
>> someone to help you resolve these issues with your compiler.  And it'd be
>> good to compile some simple C++ code using that
>> compiler to make sure the compiler is working as expected.
>>
>> Sorry I can't be of more help.
>>
>> John
>>
>>
>>
>>
> 
> 
>