Opened 10 years ago
Closed 10 years ago
#1168 closed Bug (invalid)
BDY MPP issues
| Reported by: | detraceyb | Owned by: | nemo |
|---|---|---|---|
| Priority: | low | Milestone: | |
| Component: | OCE | Version: | v3.6 |
| Severity: | Keywords: | ||
| Cc: |
Description
Merge of trunk rev 4056 and time splitting updates in NEMOGCM/branches/2013/dev_r3867_MERCATOR1_DYN : 3970|3 months|cbricaud: Time splitting update, see ticket #1079
My config:
bld::tool::fppkeys key_mpp_mpi key_vectopt_loop key_ec01 key_bdy key_dynspg_ts key_ldfslp key_traldf_c3d key_dynldf_c3d key_traldf_smag key_dynldf_smag key_zdftke key_lim2
Where key_ec01 is my domain definition.
BDY has MPP issues which have two manifestations.
1) Model crashes/hangs when using a single(non-tiled) restart file.
2) Model crashes/hangs when using land suppression
Whether the model crashes or hangs depends on whether I include LIM2
With LIM2 both 1) and 2) result in a crash. In this case the ocean.output ends with:
...
lim_sbc_init_2 : LIM-2 sea-ice - surface boundary condition
~~~~~~~~~~~~~~~
EVP rheology - C-grid case
albedo : set albedo parameters
~~~~~~~
Namelist namsbc_alb : albedo
correction for snow and ice albedo rn_cloud =
6.000000000000000E-002
albedo of melting ice in the arctic and antarctic rn_albice =
0.530000000000000
coefficients for linear rn_alphd =
0.800000000000000
interpolation used to compute albedo rn_alphdi =
0.720000000000000
between two extremes values (Pyane, 1972) rn_alphc =
0.650000000000000
and the OPA.ERROR file starts with a strange namelist error:
forrtl: severe (17): syntax error in NAMELIST input, unit 18, file /export/scratch3/brendan/EC01_EXP00/namelist_ice, line 54, position 23 Image PC Routine Line Source opa 0000000000BF5F0D Unknown Unknown Unknown opa 0000000000BF4A15 Unknown Unknown Unknown opa 0000000000B8B299 Unknown Unknown Unknown opa 0000000000B2669F Unknown Unknown Unknown opa 0000000000B25ED2 Unknown Unknown Unknown opa 0000000000B546A4 Unknown Unknown Unknown opa 000000000098C11D Unknown Unknown Unknown opa 000000000094964A Unknown Unknown Unknown opa 000000000041DD7F Unknown Unknown Unknown opa 000000000043E84C Unknown Unknown Unknown opa 0000000000404A4E Unknown Unknown Unknown opa 0000000000404960 Unknown Unknown Unknown opa 000000000040490C Unknown Unknown Unknown libc.so.6 000000330941D994 Unknown Unknown Unknown opa 0000000000404819 Unknown Unknown Unknown MPI Application rank 0 exited before MPI_Finalize() with status 17 forrtl: error (78): process killed (SIGTERM) Image PC Routine Line Source libmpi.so.1 00002B61BA376AE0 Unknown Unknown Unknown libmpi.so.1 00002B61BA375E95 Unknown Unknown Unknown libmpi.so.1 00002B61BA36F659 Unknown Unknown Unknown opa 000000000085CFA0 Unknown Unknown Unknown opa 0000000000863CEA Unknown Unknown Unknown opa 00000000009A02BB Unknown Unknown Unknown opa 000000000098928D Unknown Unknown Unknown opa 000000000094964A Unknown Unknown Unknown opa 000000000041DD7F Unknown Unknown Unknown opa 000000000043E84C Unknown Unknown Unknown opa 0000000000404A4E Unknown Unknown Unknown opa 0000000000404960 Unknown Unknown Unknown opa 000000000040490C Unknown Unknown Unknown libc.so.6 000000330941D994 Unknown Unknown Unknown opa 0000000000404819 Unknown Unknown Unknown forrtl: error (78): process killed (SIGTERM) Image PC Routine Line Source libmpi.so.1 00002AAFDC5F6AEE Unknown Unknown Unknown ...
Without LIM2 case 1) and 2) cause the model to hang. The model appears to hang during bdy_init. The last line in ocean.ouput is:
bdy_init : initialization of open boundaries ~~~~~~~~
When compiled without BDY the model runs for both 1) and 2).
Commit History (0)
(No commits)
Change History (2)
comment:1 Changed 10 years ago by detraceyb
- Priority changed from major to minor
comment:2 Changed 10 years ago by detraceyb
- Resolution set to invalid
- Status changed from new to closed
Bug was reported for "non-standard" code version.