[Mesa-users] Failed with STOP op_load

[Francis Timmes]

> I added "write(*,*) ios" just before stop statement; cd mesa-r10108/kap/; ./clean; ./mk; ./export

yes. 

assuming you are running the test case from star/test_suite/radiative_levitation
be sure to do a ./clean ; ./mk in the test directory to pick up the changes made 
to kap/private/op_load.f. as a check, i added such write statements to kap/private/op_load.f 
and the executable dutifully reported my writes of the  ios variable:

 initial setting ios           0
 ios after open           0
 reading OP cache file /Users/fxt/mesa_work/OP4STARS_1.3/mono/op_mono_cache.bin
 done reading OP cache file
 ios after close           0

you may also want to try using the full pathname for op_mono_data_path and 
op_mono_data_path instead of what appears to be a relative pathname.

fxt




> On Feb 7, 2018, at 9:21 AM, Kuldeep Verma via Mesa-users <mesa-users at lists.mesastar.org> wrote:
> 
> Dear All,
> 
> I recently installed mesa-r10108 on a Linux cluster successfully. I ran few test_suite cases without any problem (specifically, example_solar_model and example_astero). The run failed for the test_suite case radiative_levitation with the following error message:
> 
>                                          version_number       10108
>  read inlist_radiative_levitation
>  extras_controls set pointer to set_op_mono_factors
>  eosDT_use_linear_interp_for_X T
> load saved model sdb.mod
> 
>  reading OP cache file OP4STARS_1.3/mono/op_mono_cache.bin
>  done reading OP cache file
>  failed in reading cache file
> 
> I have attached the inlist with the email (you need to download the op mono opacity from https://sourceforge.net/projects/mesa/files/ if you want to run it). The following are the only two lines that I changed in the inlist:
>          op_mono_data_path = 'OP4STARS_1.3/mono'
>          op_mono_data_cache_filename = 'OP4STARS_1.3/mono/op_mono_cache.bin'
> 
> Let me add that I can run the radiative_levitation test case with mesa-r10108 on my PC. It fails only on the cluster. The following is the piece of code from mesa-r10108/kap/private/op_load.f where it fails
> 
>       ios = 0
>       open(1,file=trim(cache_filename),action='read',
>      >         status='old',iostat=ios,form='unformatted')
>       if (ios == 0) then
>          write(*,*) 'reading OP cache file ' // trim(cache_filename)
>          read(1,iostat=ios) ntotv,dv,dv1,umesh,
>      >      ite1,ite2,ite3,jn1,jn2,jne3,umin,umax,ntotp,nc,nf,int,epatom,oplnck, ne1p,
>      >      ne2p,fionp,np,kp1,kp2,kp3,npp,mx,yy1,yy2,nx,yx
>          write(*,*) 'done reading OP cache file'
>          close(1)
>          if (ios == 0) then
>             have_loaded_op = .true.
>             CALL IMESH(UMESH,NTOTV)
>             goto 1001
>          end if
>          write(*,*) 'failed in reading cache file'
>          stop 'op_load'
>       end if
> 
> I tried printing the variable ios without any success (I added "write(*,*) ios" just before stop statement; cd mesa-r10108/kap/; ./clean; ./mk; ./export). Can someone help me printout this variable, please? Any insight on what may be going on would be highly appreciated.
> 
> Best wishes,
> Kuldeep 
> 
>   
> <inlist_radiative_levitation>_______________________________________________
> mesa-users at lists.mesastar.org
> https://lists.mesastar.org/mailman/listinfo/mesa-users
>