LAMMPS WWW Site - LAMMPS Documentation - LAMMPS Mailing List Archives
Re: [lammps-users] input cause MPI error
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [lammps-users] input cause MPI error

From: liyi bai <liyibai2011@...24...>
Date: Thu, 16 Nov 2017 23:34:22 +0900

Dear User,

Thank you very much for your suggestions.
The problem I was facing has been fixed after installing newest lammps version.Newest lammps will not face this problem.

Thank you

All Best

Liyi Bai

On 11/16/2017 22:14, Axel Kohlmeyer wrote:

On Thu, Nov 16, 2017 at 12:46 AM, liyi bai <liyibai2011@...24...> wrote:
Dear Users,

I can run some stuff on MPI  and is succeed.However,It occurs some MPI errors when I using the input below.Could you please let me know,which command might be cause this MPI problem?For my previous succeed stuff.I didn't use the commands as I marked in red color with below input.

​there is not enough information here to give any recommendations. while this may be triggered by the commands you highlight, this kind of MPI error usually originates from some *other* command causing stray MPI messages.

you need to report which LAMMPS version you are using and on what platform you are running and provide a *complete* input deck and the exact command line, that you are using, so that somebody else can investigate and try to reproduce the failure you are seeing.



Thank you very much for your time in advance.

[marge11:40473] *** An error occurred in MPI_Allreduce
[marge11:40473] *** on communicator MPI_COMM_WORLD
[marge11:40473] *** MPI_ERR_TRUNCATE: message truncated
[marge11:40473] *** MPI_ERRORS_ARE_FATAL: your MPI job will now abort
mpirun has exited due to process rank 3 with PID 40473 on
node marge11 exiting improperly. There are two reasons this could occur:

1. this process did not call "init" before exiting, but others in
the job did. This can cause a job to hang indefinitely while it waits
for all processes to call "init". By rule, if one process calls "init",
then ALL processes must call "init" prior to termination.

2. this process called "init", but exited without calling "finalize".
By rule, all processes that call "init" MUST call "finalize" prior to
exiting or it will be considered an "abnormal termination"

This may have caused other processes in the application to be
terminated by signals sent by mpirun (as reported here).
[marge11:40469] 1 more process has sent help message help-mpi-errors.txt / mpi_errors_are_fatal
[marge11:40469] Set MCA parameter "orte_base_help_aggregate" to 0 to see all help / error messages

LAMMPS input:

echo both

# Define temperature parameters
variable        temperature equal 300.0 # Simulation temperature in Kelvin
variable        tempDamp equal 100.0 # Relaxation time of themostat - 0.1 ps

# Define pressure parameters
variable        pressure equal 1.0 # Pressure in bars
variable        pressureDamp equal 500.0 # Relaxation time of the barostat 0.5 ps

# Define random seed
variable        seed equal 74582

# Simulation setup
units           real
atom_style      full
variable        out_freq equal 100 # Thermo output frequency
variable        out_freq2 equal 100 # Dump output frequency
neigh_modify    delay 10 every 1 # Neighbor list setup
include         in.nacl # Load initial configuration
timestep        2.0 # Timestep 2 fs
thermo          ${out_freq}
thermo_style    custom step temp pe ke etotal press lx ly lz xy xz yz
restart         ${out_freq} restart restart2

# Perform minimization
minimize 1.0e-2 1.0e-3 100 1000

# Write trajectory to file
dump            myDump all dcd ${out_freq2} out.dcd
dump_modify     myDump unwrap yes

# Start simulation

reset_timestep  0

fix             1 all temp/csvr ${temperature} ${temperature} ${tempDamp} ${seed} # Bussi's thermostat
fix             2 all shake 0.0001 20 100000 t 1 2 # Shake algorithm
fix             3 all nph iso ${pressure} ${pressure} ${pressureDamp} # Isotropic barostat
fix             4 all momentum 10000 linear 1 1 1 # Set total linear momentum to zero every 10000 steps
#fix             5 all plumed plumedfile plumed.dat outfile plumed.out # Use plumed

velocity        all create ${temperature} ${seed} dist gaussian

run             500000 # 1 ns

unfix           1
unfix           2
unfix           3
unfix           4
#unfix           5

# Write final configuration

include file:

### TIP4P/Ew and NaCl Potential Parameters ###
pair_style      hybrid lj/cut/coul/long 5.0 buck/coul/long 5.0 # This cutoff is too short for a real calculation
# Load configuration
read_data       data.nacl

# create groups ###

group hy type 1
group ox type 2
group na type 3
group cl type 4

## set charges - beside manually ###
set group ox charge -0.830
set group hy charge 0.415
set group na charge 1.0
set group cl charge -1.0

# Water TIP3P
pair_coeff      1 1 lj/cut/coul/long 0.0 0.0
pair_coeff      2 2 lj/cut/coul/long 0.102 3.188
pair_coeff      1 2 lj/cut/coul/long 0.0 0.0
# Mixed from Pettitt 86
pair_coeff      2 3 lj/cut/coul/long 0.1338549 2.72
pair_coeff      1 3 lj/cut/coul/long 0.1338549 1.31
pair_coeff      2 4 lj/cut/coul/long 0.359825 3.55
pair_coeff      1 4 lj/cut/coul/long 0.359825 2.14
# NaCl from Pettitt 86
pair_coeff      3 4 buck/coul/long 28929.93 0.317 161.2016
pair_coeff      3 3 buck/coul/long 9772.847 0.317 24.18024 # This is not used
pair_coeff      4 4 buck/coul/long 80312.94 0.317 1683.981 # This is not used
bond_style      harmonic
bond_coeff      1 450 0.9572
angle_style     harmonic
angle_coeff     1 55 104.52
kspace_style    pppm 1e-2 # This value is too large for a real calculation

Best Regards

Liyi Bai

Check out the vibrant tech community on one of the world's most
engaging tech sites,!
lammps-users mailing list

Dr. Axel Kohlmeyer  akohlmey@...12...24...
College of Science & Technology, Temple University, Philadelphia PA, USA
International Centre for Theoretical Physics, Trieste. Italy.