# Grompp fails by md or sd as integrator

**URL:** <https://gromacs.bioexcel.eu/t/grompp-fails-by-md-or-sd-as-integrator/167>\
**Category:** User discussions\
**Created:** [May 21, 2020, 8:36pm UTC](https://gromacs.bioexcel.eu/t/grompp-fails-by-md-or-sd-as-integrator/167 "2020-05-21T20:36:55Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![Alex](https://avatars.discourse-cdn.com/v4/letter/a/db5fbb/32.png) [@Alex](https://gromacs.bioexcel.eu/u/Alex)\
**Post date:** [May 21, 2020, 8:36pm UTC](https://gromacs.bioexcel.eu/t/grompp-fails-by-md-or-sd-as-integrator/167/1 "2020-05-21T20:36:55Z")

</div>

GROMACS version: 2020.1  
GROMACS modification: No

Grompp works fine when the steep or cg is used as integrator in mdp file (even the mdrun also works afterward), but it would fail by switching only the integrator to md or sd, giving the below segmentation fault:

\_pmiu\_daemon(SIGCHLD): [NID 05465] [c0-2c1s6n1] [Thu May 21 15:03:23 2020] PE RANK 0 exit signal Segmentation fault  
Application 17692065 exit codes: 139

The GROMACS has been compiled in a cray machine using intel.

Any helpful comment is greatly appreciated.

Regards,  
Alex

---

<div class="post-metadata">

**Author:** ![cblau](https://dub1.discourse-cdn.com/flex017/user_avatar/gromacs.bioexcel.eu/cblau/32/5_2.png) [@cblau](https://gromacs.bioexcel.eu/u/cblau)\
**Post date:** [May 25, 2020, 8:19am UTC](https://gromacs.bioexcel.eu/t/grompp-fails-by-md-or-sd-as-integrator/167/2 "2020-05-25T08:19:13Z")

</div>

Hi Alex,

As I understand you only changed the integrator field in the .mdp file? This should work smoothly and I would suspect that something else must have changed here.

Do you get any more output from your system to the command line?  
Did you maybe try to run grompp in an mpi process?

---

<div class="post-metadata">

**Author:** ![Alex](https://avatars.discourse-cdn.com/v4/letter/a/db5fbb/32.png) [@Alex](https://gromacs.bioexcel.eu/u/Alex)\
**Post date:** [May 25, 2020, 10:38am UTC](https://gromacs.bioexcel.eu/t/grompp-fails-by-md-or-sd-as-integrator/167/3 "2020-05-25T10:38:37Z")

</div>

Thanks for the response.  
Yes, the only change was the integrator for sure.

And that was the whole output printed out in terminal after invoking the aprun -n 1 gmx\_mpi grompp … .  
Again, the "aprun -n 1 gmx\_mpi grompp … " works with integrator = steep or cg. But not with the integrator = md or sd.

Regards,

---

<div class="post-metadata">

**Author:** ![cblau](https://dub1.discourse-cdn.com/flex017/user_avatar/gromacs.bioexcel.eu/cblau/32/5_2.png) [@cblau](https://gromacs.bioexcel.eu/u/cblau)\
**Post date:** [May 25, 2020, 12:36pm UTC](https://gromacs.bioexcel.eu/t/grompp-fails-by-md-or-sd-as-integrator/167/4 "2020-05-25T12:36:36Z")

</div>

Hi Alex,

running grompp should just work fine on your local machine, as long as you use the same gromacs version.

Still, this error should not happen - can you share the input files that you used for grompp?