Segmentation fault(rrtmg_lw) and upperbound error (Thompson); WRFV4.6.0

[kturb]

Hello, I've used debugged WRFV4.6.0 version with initial and boundary conditions created using ERA5 data. I encountered several issues.

Firstly, I got the error on RRTMG 'taugb3' as below,
1) Program terminated with signal SIGSEGV, Segmentation fault.
#0 0x00000000028e5477 in rrtmg_lw_taumoltaumol_mp_taugb3_ () at module_ra_rrtmg_lw.f90:5458
5458 fracs(lay,ngs2+ig) = fracrefa(ig,jpl) + fpl * &
(gdb) where
#0 0x00000000028e5477 in rrtmg_lw_taumoltaumol_mp_taugb3_ () at module_ra_rrtmg_lw.f90:5458
Backtrace stopped: Cannot access memory at address 0x7ffdeb2d24d8

So, I recompiled the wrf with '-check bounds', and now I got different errors as below..

2)
forrtl: severe (408): fort: (2): Subscript #3 of the array TMR_RACG has value 5 which is greater than the upper bound of 1
Image PC Routine Line Source
wrf.exe 000000000C4E49CF Unknown Unknown Unknown
wrf.exe 00000000047582F1 module_mp_thompso 2474 module_mp_thompson.f90
wrf.exe 00000000046ED2BE module_mp_thompso 1257 module_mp_thompson.f90
wrf.exe 000000000377EE75 module_microphysi 1054 module_microphysics_driver.f90
wrf.exe 0000000002E13B99 solve_em_ 4478 solve_em.f90
wrf.exe 0000000002708E50 solve_interface_ 123 solve_interface.f90
wrf.exe 00000000005B5432 module_integrate_ 329 module_integrate.f90
wrf.exe 0000000000415378 module_wrf_top_mp 326 module_wrf_top.f90
wrf.exe 0000000000414B35 MAIN__ 29 wrf.f90
wrf.exe 0000000000414AE2 Unknown Unknown Unknown
libc-2.28.so 000014FD23CFE7E5 __libc_start_main Unknown Unknown
wrf.exe 00000000004149EE Unknown Unknown Unknown

line #2464 prg_rcg(k) = tmr_racg(idx_g1,idx_g,idx_bg(k),idx_r1,idx_r) &
Subscript #3 is idx_bg(k) and this apparently has a value of 5, but the upper bound for this 3rd subscript is 1.

Could you please help me to solve this problem..? I attached my namelist.input and rsl.error file as below.

Thank you very much for your time ..!

 

[kwerner]

Hi, Apologies for the delay. Take a look at this previous post that addresses this issue and let me know if you're able to correct the problem.

 

[kturb]

Hello, thank you for your reply. I am still having trouble in running wrf.exe.
I decreased the domain size and adopted default vertical spacing setting (instead of eta_levels) for creating input and boundary data. But still I am not able to solve the problem. I failed to succesfully run wrf.exe using either wrfinput created using either GFS or ERA5 (2-3 hours running and stopped..).
I also received another error message as below..
>>
forrtl: severe (408): fort: (3): Subscript #1 of the array KA_MN2O has value -2147483647 which is less than the lower bound of 1
Image PC Routine Line Source
wrf.exe 000000000C4E4A1F Unknown Unknown Unknown
wrf.exe 000000000690CCED rrtmg_lw_taumolta 5401 module_ra_rrtmg_lw.f90
wrf.exe 00000000068F5A10 rrtmg_lw_taumol_m 5047 module_ra_rrtmg_lw.f90
wrf.exe 0000000006A11922 rrtmg_lw_rad_mp_r 11095 module_ra_rrtmg_lw.f90
wrf.exe 0000000006A6D5B7 module_ra_rrtmg_l 12636 module_ra_rrtmg_lw.f90
wrf.exe 0000000003912759 module_radiation_ 1928 module_radiation_driver.f90
wrf.exe 0000000003F53725 module_first_rk_s 264 module_first_rk_step_part1.f90
wrf.exe 0000000002D78145 solve_em_ 943 solve_em.f90
wrf.exe 0000000002708E50 solve_interface_ 123 solve_interface.f90
wrf.exe 00000000005B5432 module_integrate_ 329 module_integrate.f90
wrf.exe 0000000000415378 module_wrf_top_mp 326 module_wrf_top.f90
wrf.exe 0000000000414B35 MAIN__ 29 wrf.f90
wrf.exe 0000000000414AE2 Unknown Unknown Unknown
libc-2.28.so 00001464617D27E5 __libc_start_main Unknown Unknown
wrf.exe 00000000004149EE Unknown Unknown Unknown
>>

Could you please help me to figure it out this issue?
Thank you very much for your time.


** Solved **
I don't know exact reason but it's solved after recompiling, redownloading GEOG data and reanalysis grib data.

 

[kwerner]

I'm so glad to hear that you were able to solve the issue, even if you don't know why. Thank you for the update.

 

[struggling_phd]

Hi, I have the exact same error. Are there any updates or leads on how to handle this?

I tried using the precompiles WPS and the geog directly on Derecho, yet I am still facing this.


At line 5399 of file module_ra_rrtmg_lw.f90
Fortran runtime error: Index '-2147483647' of dimension 1 of array 'ka_mn2o' below lower bound of 1
Error termination. Backtrace:
#0 0x33b69e5 in taugb3
at /glade/work/lmatak/wrfv4.5.2/phys/module_ra_rrtmg_lw.f90:5399
#1 0x337e873 in __rrtmg_lw_taumol_MOD_taumol
at /glade/work/lmatak/wrfv4.5.2/phys/module_ra_rrtmg_lw.f90:5045
#2 0x33f12bf in __rrtmg_lw_rad_MOD_rrtmg_lw
at /glade/work/lmatak/wrfv4.5.2/phys/module_ra_rrtmg_lw.f90:11102
#3 0x341a324 in __module_ra_rrtmg_lw_MOD_rrtmg_lwrad
at /glade/work/lmatak/wrfv4.5.2/phys/module_ra_rrtmg_lw.f90:12643
#4 0x22bfde1 in __module_radiation_driver_MOD_radiation_driver
at /glade/work/lmatak/wrfv4.5.2/phys/module_radiation_driver.f90:1965
#5 0x251947b in __module_first_rk_step_part1_MOD_first_rk_step_part1
at /glade/work/lmatak/wrfv4.5.2/dyn_em/module_first_rk_step_part1.f90:323
#6 0x1b0876e in solve_em_
at /glade/work/lmatak/wrfv4.5.2/dyn_em/solve_em.f90:973
#7 0x18469a7 in solve_interface_
at /glade/work/lmatak/wrfv4.5.2/share/solve_interface.f90:141
#8 0x47d1c7 in __module_integrate_MOD_integrate
at /glade/work/lmatak/wrfv4.5.2/frame/module_integrate.f90:325
#9 0x406884 in __module_wrf_top_MOD_wrf_run
at ../main/module_wrf_top.f90:326
#10 0x405b8d in wrf
at /glade/work/lmatak/wrfv4.5.2/main/wrf.f90:29
#11 0x405be2 in main
at /glade/work/lmatak/wrfv4.5.2/main/wrf.f90:6

 

[kturb]

Hi, I have the exact same error. Are there any updates or leads on how to handle this?

I tried using the precompiles WPS and the geog directly on Derecho, yet I am still facing this.


At line 5399 of file module_ra_rrtmg_lw.f90
Fortran runtime error: Index '-2147483647' of dimension 1 of array 'ka_mn2o' below lower bound of 1
Error termination. Backtrace:
#0 0x33b69e5 in taugb3
at /glade/work/lmatak/wrfv4.5.2/phys/module_ra_rrtmg_lw.f90:5399
#1 0x337e873 in __rrtmg_lw_taumol_MOD_taumol
at /glade/work/lmatak/wrfv4.5.2/phys/module_ra_rrtmg_lw.f90:5045
#2 0x33f12bf in __rrtmg_lw_rad_MOD_rrtmg_lw
at /glade/work/lmatak/wrfv4.5.2/phys/module_ra_rrtmg_lw.f90:11102
#3 0x341a324 in __module_ra_rrtmg_lw_MOD_rrtmg_lwrad
at /glade/work/lmatak/wrfv4.5.2/phys/module_ra_rrtmg_lw.f90:12643
#4 0x22bfde1 in __module_radiation_driver_MOD_radiation_driver
at /glade/work/lmatak/wrfv4.5.2/phys/module_radiation_driver.f90:1965
#5 0x251947b in __module_first_rk_step_part1_MOD_first_rk_step_part1
at /glade/work/lmatak/wrfv4.5.2/dyn_em/module_first_rk_step_part1.f90:323
#6 0x1b0876e in solve_em_
at /glade/work/lmatak/wrfv4.5.2/dyn_em/solve_em.f90:973
#7 0x18469a7 in solve_interface_
at /glade/work/lmatak/wrfv4.5.2/share/solve_interface.f90:141
#8 0x47d1c7 in __module_integrate_MOD_integrate
at /glade/work/lmatak/wrfv4.5.2/frame/module_integrate.f90:325
#9 0x406884 in __module_wrf_top_MOD_wrf_run
at ../main/module_wrf_top.f90:326
#10 0x405b8d in wrf
at /glade/work/lmatak/wrfv4.5.2/main/wrf.f90:29
#11 0x405be2 in main
at /glade/work/lmatak/wrfv4.5.2/main/wrf.f90:6

That's the only thing I did is 1) redownload geog data and ERA5 data (ERA5 are downloaded in NCAR RDA), 2) just setting the vertical layers instead of defining eta levels, 3) decrease the time step, 4) recompile WRF and WPS. Hope it helps to solve your problem.