[mesa-users] solar_calibration in the test suite

Warrick Ball wball at astro.physik.uni-goettingen.de
Tue Nov 12 10:07:28 EST 2013


João,

> If I set search_type to 'simplex' or 'bobyqa', MESA throws a core dumped error when going to the second model*:
> 
> *note: this is after reading the inlist, setting the composition and right after printing
> 'OMP_NUM_THREADS           8'
>       Program received signal SIGABRT: Process abort signal.
>
>       Backtrace for this error:
>       #0  0x7F7DEFC054C7
>       #1  0x7F7DEFC05ACE
>       #2  0x3F7243291F
>       #3  0x3F724328A5
>       #4  0x3F72434084
>       #5  0x3F724707B6
>       #6  0x3F724760E5
>       #7  0x53285D in __mod_pgstar_support_MOD_pgstar_clear
>       #8  0x485AA9 in __run_star_support_MOD_run1_star
>       #9  0x46BE9F in __run_star_extras_astero_MOD_eval1
>       #10  0x4678B9 in __run_star_extras_astero_MOD_simplex_f
>       #11  0x8C68B8 in get_val.1971.constprop.0 at mod_simplex.f:0
>       #12  0x8C7F5F in __mod_simplex_MOD_do_simplex
>       #13  0x886DB9 in __num_lib_MOD_nm_simplex
>       #14  0x465EF9 in __run_star_extras_astero_MOD_do_simplex
>       #15  0x46CC40 in __run_star_extras_astero_MOD_do_run_star_astero
>       #16  0x46D07E in MAIN__ at run.f:0
>       ./re: line 13: 21152 Aborted                 (core dumped) ./star_calibrate
> 
> 
> From the backtrace it seems to have something to do with pgstar but it happens even when pgstar_flag = .false.
> Any idea why?

Although it isn't really helpful, I haven't have any problem with either 
simplex or bobyqa.  Both run absolutely fine.  I'm using MESA 5527 on 
Scientific Linux 6.4, MESA SDK April 2013 edition, and 4 threads.  I 
generally leave PGSTAR turned off but I guess it's unrelated if you have 
the same problem when pgstar_flag = .false..

> Also, when I try to use GYRE instead of ADIPLS (after copying the files gyre.l0.in, etc from the astero/work
> folder) I get this:
>       failed to get l=0 frequencies

I currently do the post-processing myself but I'll try using the built-in 
version and see if I can recover this error.  (I'm waiting for the current 
run to finish...)

> Is there anything else missing for GYRE to work? And in a related question, how do I go about calculating
> non-adiabatic frequencies?

I'd like to know too!  I get errors when I try with FGONG output, so I 
guess that for manual post-processing, I need MESA models?

Cheers,
Warrick

------------
Warrick Ball
Postdoc, Institut für Astrophysik Göttingen
wball at astro.physik.uni-goettingen.de
+49 (0) 551 39 5069


More information about the Mesa-users mailing list