LAMMPS WWW Site - LAMMPS Documentation - LAMMPS Mailing List Archives
Re: [lammps-users] combining read_data and fix property/atom
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [lammps-users] combining read_data and fix property/atom


From: Andrew Jewett <jewett.aij@...24...>
Date: Wed, 9 May 2018 20:08:43 -0700

Thanks for reporting the solution, and glad you fixed it.
Nevertheless, if you can create a small test case which reproduces the problem (input script, small data file, short simulation duration), hopefully one of us will eventually figure out why it's crashing.  LAMMPS should at least print an error message.

Cheers
Andrew

On Tue, May 8, 2018 at 10:16 PM, Quang Ha <quang.t.ha.20@...12...24...> wrote:
Hi Axel,

I fixed it! Or should I say I realised what was wrong... 

Here for others to avoid this problem. Somehow I used a command create_atoms before the read_data. Removing the create_atoms get rid of the problem. 

I guess it has something to do with duplicating IDs or some sort...

Thanks,
Quang

On 8 May 2018 at 19:26, Axel Kohlmeyer <akohlmey@...24...> wrote:
On Tue, May 8, 2018 at 7:19 PM, Quang Ha <quang.t.ha.20@...24...> wrote:
> Hi all,
>
> I am trying to use write_data and read_data to restart simulation. At the
> end of the first stage, I have:
>
> [...some simulation...]
> write_data      dump.data nofix
>
> Then, at the beginning of the second stage, I follow the instruction
> provided at http://lammps.sandia.gov/doc/fix_property_atom.html:
>
> # Adding a new property - need to be added before simulation box
> fix             prop all property/atom d_foo ghost yes
> read_data       dump.data
> [... more simulation...]
>
> The simulation seems to be working until the very end:
>
> MPI task timing breakdown:
> Section |  min time  |  avg time  |  max time  |%varavg| %total
> ---------------------------------------------------------------
> Pair    | 1.7229     | 1.81       | 1.9589     |   5.4 | 70.16
> Neigh   | 0.04775    | 0.04942    | 0.051717   |   0.7 |  1.92
> Comm    | 0.010183   | 0.16158    | 0.25001    |  18.2 |  6.26
> Output  | 0.51714    | 0.52038    | 0.52466    |   0.3 | 20.17
> Modify  | 0.029224   | 0.030705   | 0.033574   |   0.8 |  1.19
> Other   |            | 0.007594   |            |       |  0.29
>
> Nlocal:    7875.12 ave 7938 max 7772 min
> Histogram: 1 0 0 0 1 0 4 0 0 2
> Nghost:    1187.38 ave 1261 max 1158 min
> Histogram: 3 2 1 0 0 0 1 0 0 1
> Neighs:    102144 ave 104750 max 100667 min
> Histogram: 3 0 1 1 0 0 2 0 0 1
> FullNghs:  204288 ave 209003 max 201481 min
> Histogram: 3 0 0 1 1 2 0 0 0 1
>
> Total # of neighbors = 1634300
> Ave neighs/atom = 25.9409
> Neighbor list builds = 10
> Dangerous builds = 10
> [hyperion:20184] *** Process received signal ***
> [hyperion:20184] Signal: Segmentation fault (11)
> [hyperion:20184] Signal code: Address not mapped (1)
> [hyperion:20184] Failing at address: 0xa01bd1e78
> *** Error in `/home/qth20/Documents/lammps-sph/lammps/src/lmp_mpi':
> corrupted size vs. prev_size: 0x0000000002af2fd0 ***
> *** Error in `*** Error in `*** Error in
> `/home/qth20/Documents/lammps-sph/lammps/src/lmp_mpi[hyperion:20185] ***
> Process received signal ***
> *** Error in
> `/home/qth20/Documents/lammps-sph/lammps/src/lmp_mpi/home/qth20/Documents/lammps-sph/lammps/src/lmp_mpi':
> [hyperion:20185] Signal: Aborted (6)
> [hyperion:20185] Signal code:  (-6)
> corrupted size vs. prev_size[hyperion:20182] *** Process received signal ***
> /home/qth20/Documents/lammps-sph/lammps/src/lmp_mpi': munmap_chunk():
> invalid pointer: 0x0000000001dde9c0 ***
> [...]
>
> which is such a weird stage to fail. I tried to debug with valgrind, but the
> tool complained that I have way too many errors to report (!).

then you should fix those issues *first*.

the way to identify the location of the issue is to compile with the
-g flag included and to enable core dumps.
but then again, the error message indicates memory corruption, which
means, that likely the issue can be anywhere.

> Any suggestions on where I could look into? If further information is
> required please let me know.

you forgot the most basic info: the LAMMPS version you are using.
and it would be helpful to have a small demo input to reproduce it.

axel.


> Thanks,
> Quang
>
>
> ------------------------------------------------------------------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> _______________________________________________
> lammps-users mailing list
> lammps-users@...655....net
> https://lists.sourceforge.net/lists/listinfo/lammps-users
>



--
Dr. Axel Kohlmeyer  akohlmey@...24...  http://goo.gl/1wk0
College of Science & Technology, Temple University, Philadelphia PA, USA
International Centre for Theoretical Physics, Trieste. Italy.


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
lammps-users mailing list
lammps-users@...6297....sourceforge.net
https://lists.sourceforge.net/lists/listinfo/lammps-users