Custom Query (2349 matches)

Filters
 
Or
 
  
 
Columns

Show under each result:


Results (22 - 24 of 2349)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Ticket Resolution Summary Owner Reporter
#2510 fixed out of bound in dtadyn.F90 mathiot mathiot
Description

Context

SETTE tests on the trunk failed on UGA cluster. The failed occurs during the mid year merge.

Analysis

After running sette in debug mode, it appears there is an out of bound in dtadyn.F90. The use of jf_ubl and jf_vbl need to be protected by ln_trabbl.

Fix

  • src/OFF/dtadyn.F90

     
    295295      CALL fld_fill( sf_dyn, slf_d, cn_dir, 'dta_dyn_init', 'Data in file', 'namdta_dyn' ) 
    296296      sf_dyn(jf_uwd)%cltype = 'U'   ;   sf_dyn(jf_uwd)%zsgn = -1._wp   
    297297      sf_dyn(jf_vwd)%cltype = 'V'   ;   sf_dyn(jf_vwd)%zsgn = -1._wp   
    298       sf_dyn(jf_ubl)%cltype = 'U'   ;   sf_dyn(jf_ubl)%zsgn =  1._wp   
    299       sf_dyn(jf_vbl)%cltype = 'V'   ;   sf_dyn(jf_vbl)%zsgn =  1._wp   
    300298      ! 
     299      IF( ln_trabbl ) THEN 
     300         sf_dyn(jf_ubl)%cltype = 'U'   ;   sf_dyn(jf_ubl)%zsgn =  1._wp   
     301         sf_dyn(jf_vbl)%cltype = 'V'   ;   sf_dyn(jf_vbl)%zsgn =  1._wp   
     302      END IF 
     303      ! 
    301304      ! Open file for each variable to get his number of dimension 
    302305      DO ifpr = 1, jfld 
    303306         CALL fld_def( sf_dyn(ifpr) ) 

How to avoid

  • short term: for every important merge: run sette with debug mode few time step
  • long term: fix sette to do it automatically
#2504 fixed Minor typo in sbc_blk_init that can lead to incorrect reporting of bulk formulation choice acc acc
Description

Context

A simple cut-and-paste error that means the true value of ln_COARE_3p5 is never reported to ocean.output

Recommendation

Change:

WRITE(numout,*) '      "COARE 3.5" algorithm   (Edson et al. 2013)         ln_COARE_3p5 = ', ln_COARE_3p0

to

WRITE(numout,*) '      "COARE 3.5" algorithm   (Edson et al. 2013)         ln_COARE_3p5 = ', ln_COARE_3p5
#2502 fixed DOMcfg: improve reproducibility of the domcfg.nc mathiot mathiot
Description

Context

When you found a domaincfg.nc, it is difficult to know how it has been built and how to reproduce it. MEOM/DRAKKAR build a script to add a specific the namelist_cfg in it as netcdf variable and input file name as attribute. A script to generate the namelist from the domcfg.nc is also available.

Analysis

To partially fix this, namelist_cfg should be introduced in the domcfg.nc output file AND bathymetry, coordinate (…) files name used to generate the input file should be added to the namelist_cfg (default name being the one specified in namelist_ref of the domain_cfg).

Steps

  • Introduced in the namelist the ability to use non generic name for coordinates, bathy, isf draft:
   cn_fcoord   =  'coordinates.nc'             ! external coordinates file (jphgr_msh = 0)
   cn_topo     =  'bathy_meter.nc           '  ! external topo file (nn_bathy =1/2)
   cn_topolvl  =  'bathy_level.nc           '  ! external topo file (nn_bathy =1) 
   cn_fisfd    =  'isf_draft_meter.nc'         ! external isf draft (nn_bathy =1 and ln_isfcav = .true.)
   cn_bath     =  'Bathymetry'                 ! topo name in file  (nn_bathy =1/2)
   cn_bathlvl  =  'Bathy_level'                ! lvl name in file   (nn_bathy =1) 
   cn_visfd    =  'isf_draft'                  ! isf draft variable (nn_bathy =1 and ln_isfcav = .true.)
  • Add the sript to include the namelist_cfg.nc into the domcfg.nc netcdf:
    ./dom_doc.exe -h
      usage : dom_doc -n NAMELIST-file 
                            -d DOMAIN_CFG-file
           
          PURPOSE :
             Add information in the domain_cfg.nc file after its creation for
             NEMO4. The additional information consists in a new netcdf variable
             called namelist_cfg, holding the content of the used namelist_cfg.
           
          ARGUMENTS :
             -n NAMELIST-file : name of the namelist_cfg. file required
             -d DOMAIN_CFG-file : name of the domain_cfg file to document. file requ
     ired
           
          OUTPUT : 
              input DOMAIN_CFG-file is modified on output.
    
  • Add the reverse script (ie to generate the namelist from the namelist_cfg netcdf variable):
    ./xtrac_namelist.bash domain_cfg.nc namelist_output
    (base) mathiotp@f-dahu:.../test_trunk$ cat namelist_output 
    !!>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
    !! NEMO/OCE :   Reference namelist_ref                                !!
    !!>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
    !! NEMO/OCE  :  1 - Domain & run manager (namrun, namcfg, namdom, namzgr, namzgr_sco )
    !!              8 - diagnostics      (namnc4)
    !!             10 - miscellaneous    (nammpp, namctl)
    !!>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
    !-----------------------------------------------------------------------
    &namrun        !   parameters of the run
    !-----------------------------------------------------------------------
    ...
    /
    !-----------------------------------------------------------------------
    &namdom        !   space and time domain (bathymetry, mesh, timestep)
    !-----------------------------------------------------------------------
       ln_read_cfg = .false.   !  Read from a domain_cfg file
       nn_bathy    =    1      !  compute analyticaly (=0) or read (=1) the bathymetry file
                               !  or compute (2) from external bathymetry
       nn_interp   =    1                          ! type of interpolation (nn_bathy =2)
       cn_domcfg   = ' '       ! Name of the domain_cfg input file
       cn_fcoord   =  'coordinates.nc'             ! external coordinates file (nn_bathy =2)
       cn_topo     =  'bathy_meter.nc'             ! external topo file (nn_bathy =2)
       cn_topolvl  =  'bathy_level.nc'             ! external topo file (nn_bathy =2)
       cn_fisfd    =  'isf_draft_meter.nc'         ! external isf draft (nn_bathy =2 and ln_isfcav = .true.)
       cn_bath     =  'Bathymetry_isf'             ! topo name in file  (nn_bathy =2)
       cn_visfd    =  'isf_draft'                  ! isf draft variable (nn_bathy =2 and ln_isfcav = .true.)
       cn_lon      =  'nav_lon'                    ! lon  name in file  (nn_bathy =2)
       cn_lat      =  'nav_lat'                    ! lat  name in file  (nn_bathy =2)
       rn_bathy    =    0.     !  value of the bathymetry. if (=0) bottom flat at jpkm1
       nn_msh      =    1      !  create (=1) a mesh file or not (=0)
       rn_hmin     =   10.     !  min depth of the ocean (>0) or min number of ocean level (<0) (out of ice shelf cavity)
       rn_e3zps_min=   20.0     !  partial step thickness is set larger than the minimum of
       rn_e3zps_rat=    0.1    !  rn_e3zps_min and rn_e3zps_rat*e3t, with 0<rn_e3zps_rat<1
                               !
       rn_rdt      =  720.0    !  time step for the dynamics (and tracer if nn_acc=0)
       rn_atfp     =    0.1    !  asselin time filter parameter
       ln_crs      = .false.      !  Logical switch for coarsening module
       jphgr_msh   =       2               !  type of horizontal mesh
      ...
    
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Note: See TracQuery for help on using queries.