runtime error ....

General scientific issues regarding ROMS

Moderators: arango, robertson

Post Reply
Message
Author
mashinde
Posts: 135
Joined: Mon Jun 22, 2009 3:46 pm
Location: Indian Institute of Tropical Meteorology, Pune, INDIA

runtime error ....

#1 Unread post by mashinde »

Dear frds,

I have following error....

This error comes when I use bulk forcing.........

forrtl: error (65): floating invalid
Image PC Routine Line Source
oceanG 000000000069E78C time_string_ 490 get_date.f90
oceanG 00000000006947EF get_2dfld_ 456 get_2dfld.f90
oceanG 00000000005B200C get_data_ 73 get_data.f90
oceanG 000000000044C2F9 initial_ 217 initial.f90
oceanG 000000000044A524 ocean_control_mod 107 ocean_control.f90
oceanG 000000000044A18E MAIN__ 97 master.f90
oceanG 0000000000449FFC Unknown Unknown Unknown
libc.so.6 000000316BA1D994 Unknown Unknown Unknown
oceanG 0000000000449F09 Unknown Unknown Unknown
--------------------------------------------------------------------------
mpirun noticed that process rank 0 with PID 7118 on node compute-0-21.local exited on signal 6 (Aborted).
--------------------------------------------------------------------------
forrtl: error (78): process killed (SIGTERM)
forrtl: error (78): process killed (SIGTERM)
Image PC Routine Line Source
mca_btl_openib.so 00002AB65678DED0 Unknown Unknown Unknown

----------------------------------------------------------------------------------

is this error due to cluster problem or model problem. Is the problem occurs due to miscommunication in the cluster processors.
I did not understand why this error occurs when I use bulk forcing with bulk parametrization key is ON in .h file.

without bulk i.e. with momentum and heat fluxes it runs okay.




Thanks !

User avatar
kate
Posts: 4088
Joined: Wed Jul 02, 2003 5:29 pm
Location: CFOS/UAF, USA

Re: runtime error ....

#2 Unread post by kate »

The routine time_string is relatively simple and has nothing to do with parallel processing.
Do you know what your time is? Did you send it something that is too big for the format:

DDDDD HH:MM:SS

i.e., would overflow past 99999 days? Or perhaps check the time in your winds file?

mashinde
Posts: 135
Joined: Mon Jun 22, 2009 3:46 pm
Location: Indian Institute of Tropical Meteorology, Pune, INDIA

Re: runtime error ....

#3 Unread post by mashinde »

Hi kate,

thanks for the reply. I am doing climatological run. my u-wind nc file looks like.

dimensions:
xi_rho = 940 ;
eta_rho = 424 ;
xi_psi = 939 ;
eta_psi = 423 ;
xi_u = 939 ;
eta_u = 424 ;
xi_v = 940 ;
eta_v = 423 ;
wind_time = UNLIMITED ; // (12 currently)
variables:
double wind_time(wind_time) ;
wind_time:long_name = "atmospheric forcing time" ;
wind_time:units = "days" ;
wind_time:cycle_length = 360. ;
double Uwind(wind_time, eta_rho, xi_rho) ;
Uwind:long_name = "surface u-wind component" ;
Uwind:units = "m/s" ;

// global attributes:
:type = "ROMS uwnd forcing file" ;
:title = "Rmed16" ;
:history = "surface u-wind forcing file" ;



and same netcdf attributes for other parameters.



thanking you

User avatar
kate
Posts: 4088
Joined: Wed Jul 02, 2003 5:29 pm
Location: CFOS/UAF, USA

Re: runtime error ....

#4 Unread post by kate »

But what are those twelve time values? Numerically?

mashinde
Posts: 135
Joined: Mon Jun 22, 2009 3:46 pm
Location: Indian Institute of Tropical Meteorology, Pune, INDIA

Re: runtime error ....

#5 Unread post by mashinde »

hi kate,

Thanks !

these are the monthly climatology from COADS data.
when I print time from the nc file it gives me like.....

nc=netcdf('rmed16_grd_etopo5_e_uwnd_n.nc');
>> time=nc{'wind_time'}(:);
>> time

time =

15
45
75
105
135
165
195
225
255
285
315
345

Total 12 time elements
so these are the days..........time in days(climatology)

or is there any way how can i set time in nc file.



Thanking you

mashinde
Posts: 135
Joined: Mon Jun 22, 2009 3:46 pm
Location: Indian Institute of Tropical Meteorology, Pune, INDIA

Re: runtime error ....

#6 Unread post by mashinde »

Hi Kate,

I think I solved the problem........model run with bulk forcing...

Thanks for the advices.

Post Reply