[Met_help] Cygwin build and runtime error for MET v1.0

Dr. John Huddleston John.M.Huddleston at noaa.gov
Wed Jan 16 03:23:34 MST 2008


Good morning

I ran gdb and stepped through the Cygwin build of MET v1.0. You can see 
the results below. It does indeed fail at the engine.set function call.  
During the build there were two include files that had warnings on a 
deprecated use of interger and character* Maybe that has something to do 
with it.

John Huddleston

Engine::set (this=0x5889e0, fcst_wd=@0x22c6e0, obs_wd=@0x22c610)
    at engine.cc:246
246        for(j=0; j<max_singles; j++) {
(gdb) print j
$23 = 2280697
(gdb) s
247           fcst_single[j].clear();
(gdb) print j
$24 = 2280697
(gdb) print max_singles
$25 = 500


John Halley Gotway wrote:
> John,
>
> Sounds like quite an effort!  I haven't tried to port MET to cygwin yet myself.  But I could give it a shot if there are enough users who would be interested in that.
>
> I would like to hear about what type of issues you've turned up.  I looked through the attached PDF, and unfortunately there isn't anything obvious in there.  All MODE is doing there is reading
> through several grib records in the input grib file and dumping out header information about them (when -v >= 3).  I don't see any warning or error messages in there.  When you ran that MODE command,
> what happens next?  Does it just hang or is there some error message that wasn't included at the end of the PDF?  The last status message from MODE says that it's defining the objects in the forecast
> and observation fields.
>
> When I run that same command here (with -v 1), here's the messages that are printed to the screen:
>
> ------------------------------------
> ../bin/mode ../data/sample_fcst/2005080700/wrfprs_ruc13_12.tm00_G212 ../data/sample_fcst/2005080712/wrfprs_ruc13_00.tm00_G212 config/WrfModeConfig_RH -outdir ../out/mode -v 1
>
> Forecast File: ../data/sample_fcst/2005080700/wrfprs_ruc13_12.tm00_G212
> Observation File: ../data/sample_fcst/2005080712/wrfprs_ruc13_00.tm00_G212
> Match Config File: config/WrfModeConfig_RH
> Merge Config File: config/WrfModeConfig_RH
> Forecast Field: RH at ISBL_500
> Observation Field: RH at ISBL_500
> Identifying objects in the forecast and observation fields...
> Computing contingency table statistics...
> Identified: 8 forecast objects and 12 observation objects.
> Performing merging (no merging) in the forecast field.
> Performing merging (no merging) in the observation field.
> Remaining: 8 forecast objects and 12 observation objects.
> Performing matching between the forecast and observation fields.
> Creating Contingency Table Statistics file: ../out/mode/mode_RH_ISBL_500_vs_RH_ISBL_500_120000L_20050807_120000V_000000A_cts.txt
> Creating Fcst-Obs Object Statistics file: ../out/mode/mode_RH_ISBL_500_vs_RH_ISBL_500_120000L_20050807_120000V_000000A_obj.txt
> Creating Object NetCDF file: ../out/mode/mode_RH_ISBL_500_vs_RH_ISBL_500_120000L_20050807_120000V_000000A_obj.nc
> Loading forecast raw color table: /pd6/score/MET/MET_releases/METv1.0/data/colortables/mode_raw.ctable
> Loading observation raw color table: /pd6/score/MET/MET_releases/METv1.0/data/colortables/mode_raw.ctable
> Creating postscript file: ../out/mode/mode_RH_ISBL_500_vs_RH_ISBL_500_120000L_20050807_120000V_000000A.ps
> ------------------------------------
>
> Thanks,
> John Halley Gotway
>
> Dr. John Huddleston wrote:
>   
>> Good morning,
>>
>> I finally got a chance to build the METv1.0 system under Windows using
>> Cygwin. Attached is an output of the run of the third sample from the
>> test_mode.sh script. This was run using the GNU debugger "gdb" command
>> and I added the "-v 5" option on the tail end of the run under gdb.
>>
>> It took many hours to get this far. The default compiler for Cygwin is
>> 3.4.4 which includes g77 but not gfortran.  So, I built GCC 4.2.0. That
>> alone took six hours. Then I had to rebuild NetCDF. I used 3.6.1 because
>> I remember someone saying that 3.6.2 had problems with METv0.9. I tried
>> many combinations of 4.2.0 and 3.4.4 compilers.
>>
>> I'll try and summarize all the failures and tar them up in a new email;
>> however, I was hoping someone could take a look at the PDF file and
>> maybe they can see something obvious.
>>
>> Thank you
>>
>> John
>>
>> Dr. John M. Huddleston, PE
>> Software Applications Development Team Leader
>> Systems Integration Branch
>> NCEP Central Operations
>> 301-763-8000 x7136
>>
>> Barbara Brown wrote:
>>     
>>> Dear MET beta users,
>>>
>>> I apologize if you have already received the attached message, but we
>>> wanted to be sure that all of you received this notice about Version 1
>>> of the Model Evaluation Tools.
>>>
>>> Many of you provided useful comments on MET that were taken into
>>> account as we put together the current release - we greatly appreciate
>>> your feedback, which led to significant improvements to MET. We hope
>>> you'll continue to let us know about problems you encounter and
>>> improvements and enhancements that you would like to have included in
>>> future releases.
>>>
>>> Best regards from the
>>> MET development team
>>> met_help at ucar.edu
>>>
>>>
>>> -------- Original Message --------
>>> Subject:     DTC announces release of Version 1 of Model Evaluation
>>> Tools (MET)
>>> Date:     Mon, 07 Jan 2008 13:48:02 -0700
>>> From:     Barbara Brown <bgb at ucar.edu>
>>> Organization:     National Center for Atmospheric Research / RAL
>>> To:     wrf-news at ucar.edu
>>>
>>>
>>>
>>> Dear WRF Users,
>>>
>>> The MET Development Team at the Developmental Testbed Center (DTC)
>>> proudly announce the formal release of their verification software,
>>> the Model Evaluation Tools (MET) version 1.0, available on the MET
>>> website:  http://www.dtcenter.org/met/users
>>>
>>> MET is designed to be a highly-configurable, state-of-the-art suite of
>>> verification tools. MET was developed using output from the Weather
>>> Research and Forecasting (WRF) modeling system may but has been
>>> applied to the output of other modeling systems as well. MET provides
>>> a variety of verification techniques for gridded numerical weather
>>> forecasts, including:
>>>
>>>  * Standard verification scores comparing gridded model data to
>>>    point-based observations
>>>  * Standard verification scores comparing gridded model data to
>>>    gridded analyses
>>>  * Object-based verification method comparing gridded model data to
>>>    gridded observations
>>>
>>> Additional verification techniques and analysis tools will be
>>> supported for community use in the future, depending on interest and
>>> availability of resources.  Please visit: 
>>> http://www.dtcenter.org/met/users for more information on MET,
>>> including download and installation instructions.
>>>
>>> We thank those of you who volunteered as beta testers.  Many of you
>>> provided valuable feedback that provided us with ideas on how to
>>> improve MET.  Changes since the beta version include bug fixes, two
>>> new tools (the VSDB-analysis tool and the MODE-analysis tool), the
>>> inclusion of derived quantities, and a variety of other improvements.
>>>
>>> Thanks to the U.S. Air Force Weather Agency for their support of this
>>> work.  Thanks also to the National Oceanic and Atmospheric
>>> Administration for their support of the Developmental Testbed Center.
>>>
>>> Sincerely,
>>> The MET Development Team
>>> Barbara Brown, Lacey Holland, John Halley Gotway, Randy Bullock, Eric
>>> Gilleland, and David Ahijevych
>>> met_help at ucar.edu
>>>
>>>
>>>
>>>       
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> Met_help mailing list
>> Met_help at mailman.ucar.edu
>> http://mailman.ucar.edu/mailman/listinfo/met_help
>>     
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/met_help/attachments/20080116/e2fac6c6/attachment.html


More information about the Met_help mailing list