make_macros.mk WHICH CONTAINS APPLICATION-DEPENDENT MAKE DEF

[zhaoxinhua]

Hello,

I have problem with build.bash, when I'm running , I got this error:makefile:238: INCLUDING FILE /public/home/zhaoxinhua/make_macros.mk WHICH CONTAINS APPLICATION-DEPENDENT MAKE DEFINITIONS

And the oceanM file is builded, but there are no input files such as ocean_grd.nc, ocean_ini.nc

Thanks so much!!

[kate]

You say it built oceanM? Then you have successfully compiled ROMS.

The ocean_grd.nc and ocean_ini.nc come from other preprocessing steps, most commonly done in Matlab. To become familiar with ROMS, I suggest you start with some of the test problems like UPWELLING for which you don't need these external files.

[zhaoxinhua]

You say it built oceanM? Then you have successfully compiled ROMS.

The ocean_grd.nc and ocean_ini.nc come from other preprocessing steps, most commonly done in Matlab. To become familiar with ROMS, I suggest you start with some of the test problems like UPWELLING for which you don't need these external files.

I had that error msg when compiling UPWELLING. And the oceanM file is builded, but there are no input files such as ocean_grd.nc, ocean_ini.nc

next use mpirun --mca btl self,sm,tcp -np 2 ./oceanM ocean_upwelling.in > my_upwelling.log
the my_upwelling.log says
Output/Input Files:

Output Restart File: ocean_rst.nc
Output History File: ocean_his.nc
Output Averages File: ocean_avg.nc
Output Diagnostics File: ocean_dia.nc

READ_PHYPAR - could not find input file:

Elapsed CPU time (seconds):


ROMS/TOMS - Output NetCDF summary for Grid 01:

ROMS/TOMS - I/O error ............... exit_flag: 4


ERROR: I/O related problem.

[kate]

READ_PHYPAR - could not find input file:

I hate when this happens. The error is supposed to report the name of the missing file here, but in this case the filename is empty. I don't know why you are getting this problem - I've never gotten it from the UPWELLING case which ought to run without modifications. I assume your ocean_upwelling.in is from the same svn revision as ROMS itself? You didn't delete the name of the varinfo.dat file or anything crazy, did you?

[kate]

I got this error just today and it was due to tabs in the ocean.in file. Where you are at in the execution of read_phypar is where it checks on all the input files. For my run, there's a grid file then four forcing files, then the station locations file. The UPWELLING problem has an internal grid and forcing fields and therefore it shouldn't try to open files for them.

[zhaoxinhua]

I got this error just today and it was due to tabs in the ocean.in file. Where you are at in the execution of read_phypar is where it checks on all the input files. For my run, there's a grid file then four forcing files, then the station locations file. The UPWELLING problem has an internal grid and forcing fields and therefore it shouldn't try to open files for them.

Thanks very much.
There are only upwelling.h, ocean_upwelling.in, build.bash in my 'upwelling' folder.
Then I use the command line "./build.bash -j 2 >1.dat"
and "mpirun --mca btl self,sm,tcp -np 2 ./oceanM ocean_upwelling.in > my_upwelling.log"
the oceanM, 1.dat and my_upwelling.log are created.
I only slightly modify the paths to the files and to the netcdf. In other server I could run the program, but in our server,it could not create any .nc file...

the my_upwelling.log says
Output/Input Files:

Output Restart File: ocean_rst.nc
Output History File: ocean_his.nc
Output Averages File: ocean_avg.nc
Output Diagnostics File: ocean_dia.nc

READ_PHYPAR - could not find input file:

Elapsed CPU time (seconds):


ROMS/TOMS - Output NetCDF summary for Grid 01:

ROMS/TOMS - I/O error ............... exit_flag: 4


ERROR: I/O related problem.

I really don't know where the problem lies, could you help me? many thanks

[hadibh]

Hi,

I have a similar problem to the one posted in this topic. When I run upwelling test I get an error like this:

rm -f -r core *.ipo /home/myusername/roms/Projects/Upwelling/Build /home/myusername/make_macros.mk
makefile:238: INCLUDING FILE /home/myusername/make_macros.mk WHICH CONTAINS APPLICATION-DEPENDENT MAKE DEFINITIONS
/home/myusername/roms/trunk/Compilers/CYGWIN-gfortran.mk:29: *** empty variable name. Stop.

Can anybody help?

[kate]

/home/myusername/roms/trunk/Compilers/CYGWIN-gfortran.mk:29: *** empty variable name. Stop.

Have you looked at line 29 of this file? Mine has:

Code: Select all

FFLAGS := -frepack-arrays

Are there tabs or anything there?

[hadibh]

Hi Kate,

Thanks for your reply.

Line 29 is this: FFLAGS := -frepack-arrays
exactly similar to what you wrote. However, the whole file is in only 6 lines in notepad and they are all consecutively written without any new line in between different commands. Following is a picture of it:

[kate]

There are multiple posts in the archives about downloading ROMS on the Windows side, then trying to compile in CYGWIN and running into trouble. Is that what you did? You need to be consistently in CYGWIN while downloading and compiling and all.

[hadibh]

I think I did everything in Cygwin. However, I am going to repeat installation process and if I failed again I'll consider linux. Thanks for your suggestion.