Lost atoms while atoms are still in the box

Dear LAMMPS users,

I created 2 atoms in simulation box, and moved them to new positions using “set” command. Even though I placed atoms to new positions inside the box, LAMMPS warned me of lost atoms. And this warning was, to my surprise, dependent on where I created my atoms in the beginning. Here is my input script:

#------------------------------------------INPUT---------------------------------------------

units metal
atom_style charge

region box block 0 120 0 20 0 20 units box
create_box 2 box

mass 1 65
mass 2 16

create_atoms 1 single 5 5 5
create_atoms 2 single 4 10 10

If in “create_atoms 2”, I change “4 10 10” to “40 10 10” then there is no atom lost

pair_style lj/cut 2.5
pair_coeff * * 1 1

neighbor 1.0 bin

thermo 1
thermo_style custom step atoms
#thermo_modify lost warn

run 0
set atom 1 x 10 y 10 z 10
run 0
set atom 2 x 40 y 10 z 10
run 10

#---------------------------------------------LOG---------------------------------------------
(1) If I use "create_atoms 2 single 4 10 10 " then:

set atom 2 x 40 y 10 z 10
1 settings made for x
1 settings made for y
1 settings made for z
run 10
WARNING: No fixes defined, atoms won’t move (…/verlet.cpp:55)
Per MPI rank memory allocation (min/avg/max) = 0.442 | 0.7623 | 3.192 Mbytes
Step Atoms
ERROR: Lost atoms: original 2 current 1 (…/thermo.cpp:433)
Last command: run 10

(2) If I use “create_atoms 2 single 40 10 10” then:

set atom 2 x 40 y 10 z 10
1 settings made for x
1 settings made for y
1 settings made for z
run 10
WARNING: No fixes defined, atoms won’t move (…/verlet.cpp:55)
Per MPI rank memory allocation (min/avg/max) = 0.442 | 0.8873 | 2.817 Mbytes
Step Atoms
0 2
1 2
2 2
3 2
4 2
5 2
6 2
7 2
8 2
9 2
10 2
Loop time of 0.000372484 on 16 procs for 10 steps with 2 atoms

Could anyone tell where the problem lies, please?

Best regards,
Trung Quoc Phung

Dear LAMMPS users,

I created 2 atoms in simulation box, and moved them to new positions using
"set" command. Even though I placed atoms to new positions inside the box,
LAMMPS warned me of lost atoms. And this warning was, to my surprise,
dependent on where I created my atoms in the beginning. Here is my input
script:

​i cannot reproduce this with the current LAMMPS version 24 Jul 2017.

axel.​

Dear Professor Kohlmeyer,

Sorry that I waste your time. Today I run the script again and it works!
It should be a problem with our system, so I will ask my lab technician to look at this.

Best regards,
Trung Quoc Phung