Error in running cg-dynamics by atdyn?

[wuchaohao-wl]

Hi there, i built a system include 250 protein particles and 50 RNA particles, generated by cg_tool and performed single-chain simulation for each.
initially, it can run properly but trigger this bug after 41000 steps:
[intel016:1666758:0:1666758] Caught signal 11 (Segmentation fault: address not mapped to object at address 0x11ccfbeb4)
[intel016:1666777:0:1666777] Caught signal 11 (Segmentation fault: address not mapped to object at address 0xebf60984)
==== backtrace (tid:1666758) ====
0 0x0000000000012dd0 .annobin_sigaction.c() sigaction.c:0
1 0x00000000004a1be4 __at_pairlist_mod_MOD_update_pairlist_pbc_cg_exv._omp_fn.8() at_pairlist.f90:0
2 0x00000000000116a6 GOMP_parallel() ???:0
3 0x00000000004ab246 __at_pairlist_mod_MOD_update_pairlist() ???:0
4 0x000000000061e5f3 __at_md_vverlet_cg_mod_MOD_vverlet_dynamics_cg() ???:0
5 0x0000000000684488 at_dynamics_mod_MOD_run_md() ???:0
6 0x00000000006a5fa8 MAIN() atdyn.f90:0
7 0x0000000000402e3d main() ???:0
8 0x00000000000236a3 __libc_start_main() ???:0
9 0x0000000000402e6e _start() ???:0

Program received signal SIGSEGV: Segmentation fault - invalid memory reference.

Backtrace for this error:
==== backtrace (tid:1666777) ====
0 0x0000000000012dd0 .annobin_sigaction.c() sigaction.c:0
1 0x00000000004a1be4 __at_pairlist_mod_MOD_update_pairlist_pbc_cg_exv._omp_fn.8() at_pairlist.f90:0
2 0x00000000000116a6 GOMP_parallel() ???:0
3 0x00000000004ab246 __at_pairlist_mod_MOD_update_pairlist() ???:0
4 0x000000000061e5f3 __at_md_vverlet_cg_mod_MOD_vverlet_dynamics_cg() ???:0
5 0x0000000000684488 at_dynamics_mod_MOD_run_md() ???:0
6 0x00000000006a5fa8 MAIN() atdyn.f90:0
7 0x0000000000402e3d main() ???:0
8 0x00000000000236a3 __libc_start_main() ???:0
9 0x0000000000402e6e _start() ???:0

Program received signal SIGSEGV: Segmentation fault - invalid memory reference.
I cannot find this error in guide, what is it meaning?

[wuchaohao-wl]

I checked the .dcd file, it was not a normal output, i cannot find trajectory.
I think it is because I directly mix the rna.gro and protein.gro by gmx insert-molecule command and rewrite the .top in a wrong way.

[jaewoonjung]

To see trajectory, please control crdout_period in the control input file.

[wuchaohao-wl]

[DYNAMICS]
integrator = VVER_CG
nsteps = 50000000
timestep = 0.010
rstout_period = 1000000
crdout_period = 10000
eneout_period = 10000
nbupdate_period = 20
I think it is not problem on crdout_period

[jaewoonjung]

If you have an error after 41,000 steps and if you wrote crdout_period=10000, there should be 4 snapshots in the dcd file. I asked you because you said that there is no trajectory. I think it's a problem of initial coordinate. Did you run minimization?

[wuchaohao-wl]

there is a .dcd file output, but I cannot find any atom in trajectory. I didnot run minimization, I just follw the tutorial for FUS protein. I run atdyn for the single chain of protein and rna, and then, duplicated them by gmx insert-molecules. finally run atdyn for the whole system. I will try minimization