Error When Running DNTMC module


Just Got Here
Hi Everyone,
    I was trying to run the example file (given in the manual) for Dynamic Nucleation Theory Monte Carlo in NWCHEM6.0. However I got errors constantly. The errors are like following:

everyone should have an open file at this point
********** Setting Up DNTMC Info **********
*******************************************
Made it through the initial config setup
NSPC = 2
NMOLS = 10
TEMP = 243.00 K
MCSTPS= 1.00E+06
Trans Max Disp = 4.00E-02 Ang
Rot Max Disp = 6.00E-02 rad
Restart files written
Start Run
RMIN = 3.25
RMAX = 12.25
R-Config Methodology
NGROUPS = 1
Made it through initialization
********** Starting MC Loop ***************
*******************************************
0:Segmentation Violation error, status=: 11
(rank:0 hostname:login002 pid:13464):ARMCI DASSERT fail. signaltrap.c:SigSegvHandler():301 cond:0
Last System Error Message from Task 0:: No such file or directory
 0: ARMCI aborting 0 (0).
0: ARMCI aborting 0 (0).
system error message: Illegal seek

I checked the manual everywhere and still can figure out what happened here. Does anyone have some idea about it and help me out. I really appreciate it. Thank you very much.

Sincerely,
Wen Xu
Chemistry Department
Texas A&M University, College Station

Forum Vet
Can you post the full input deck so we can test.

Bert



Quote:Wxu1024 Feb 20th 10:39 pm
Hi Everyone,
    I was trying to run the example file (given in the manual) for Dynamic Nucleation Theory Monte Carlo in NWCHEM6.0. However I got errors constantly. The errors are like following:

everyone should have an open file at this point
********** Setting Up DNTMC Info **********
*******************************************
Made it through the initial config setup
NSPC = 2
NMOLS = 10
TEMP = 243.00 K
MCSTPS= 1.00E+06
Trans Max Disp = 4.00E-02 Ang
Rot Max Disp = 6.00E-02 rad
Restart files written
Start Run
RMIN = 3.25
RMAX = 12.25
R-Config Methodology
NGROUPS = 1
Made it through initialization
********** Starting MC Loop ***************
*******************************************
0:Segmentation Violation error, status=: 11
(rank:0 hostname:login002 pid:13464):ARMCI DASSERT fail. signaltrap.c:SigSegvHandler():301 cond:0
Last System Error Message from Task 0:: No such file or directory
 0: ARMCI aborting 0 (0).
0: ARMCI aborting 0 (0).
system error message: Illegal seek

I checked the manual everywhere and still can figure out what happened here. Does anyone have some idea about it and help me out. I really appreciate it. Thank you very much.

Sincerely,
Wen Xu
Chemistry Department
Texas A&M University, College Station

Just Got Here
Sure. Thank you very much Bert. My input file is just the example file provided with NWCHEM 6.0. When I run the calculation, first there is error saying missing nob value. Then I just added the nob=10 in the file. I didn't do any further modifications on the example input file.
The input is in the following


  start
# start or restart directive if a restart run
MEMORY 1000 mb

PERMANENT_DIR /scratch/wenxu/NWCHEM/
# Globally accessible directory which the
# rtdb (*.db) file will/does reside.

basis "ao basis" spherical noprint
* library 6-31++G**
end
# basis set directive for scf energies

scf
singlet
rhf
tol2e 1.0e-12
vectors input atomic
thresh 1.0e-06
maxiter 200
print none
end
# scf directive for scf energies

geometry geom1 units angstroms noautosym nocenter noprint
O 0.393676503613369 -1.743794626956820 -0.762291912129271
H -0.427227157125777 -1.279138812526320 -0.924898279781319
H 1.075463952717060 -1.095883929075060 -0.940073459864222
symmetry c1
end
# geometry of a monomer with title "geom1"

geometry geom2 units angstroms noautosym nocenter noprint
N 6.36299e-08 0.00000 -0.670378
H 0.916275 0.00000 -0.159874
H -0.458137 0.793517 -0.159874
H -0.458137 -0.793517 -0.159874
symmetry c1
end
# geometry of another monomer with title "geom2"
# other monomers may be included with different titles

set subgroups_number 8
# set directive which gives the number of subgroups
# each group runs a separate MC simulation

dntmc
# DNTMC input block
nspecies 2
# The number of unique species (number of titled geometries
# above)
species geom1 geom2
# An array of geometry titles (one for each
# nspecies/geometry)
nmol 5 5
# An array stating the number of each
# monomer/nspecies/geometry in simulation.
temp 243.0
mcsteps 1000000
rmin 3.25
rmax 12.25
nob 10
mprnt 10
tdisp 0.04
rdisp 0.06
#print none fdist mcdata
print debug fdist mcdata
# this print line first sets the print-level to none
# then it states that the *.fdist and *.mcdata.(#num)
# files are to be written
rconfig
#dntmc_dir /home/bill/largefile
# An accessible directory which to place the *.fdist,
# *.mcdata.(#num), and *.restart.(#num) files.
convergence 1.0D+00
end

task scf dntmc
# task directive stating that energies are to be done at the scf
#level of theory.


Quote:Bert Feb 21st 11:23 pm
Can you post the full input deck so we can test.

Bert



Quote:Wxu1024 Feb 20th 10:39 pm
Hi Everyone,
    I was trying to run the example file (given in the manual) for Dynamic Nucleation Theory Monte Carlo in NWCHEM6.0. However I got errors constantly. The errors are like following:

everyone should have an open file at this point
********** Setting Up DNTMC Info **********
*******************************************
Made it through the initial config setup
NSPC = 2
NMOLS = 10
TEMP = 243.00 K
MCSTPS= 1.00E+06
Trans Max Disp = 4.00E-02 Ang
Rot Max Disp = 6.00E-02 rad
Restart files written
Start Run
RMIN = 3.25
RMAX = 12.25
R-Config Methodology
NGROUPS = 1
Made it through initialization
********** Starting MC Loop ***************
*******************************************
0:Segmentation Violation error, status=: 11
(rank:0 hostname:login002 pid:13464):ARMCI DASSERT fail. signaltrap.c:SigSegvHandler():301 cond:0
Last System Error Message from Task 0:: No such file or directory
 0: ARMCI aborting 0 (0).
0: ARMCI aborting 0 (0).
system error message: Illegal seek

I checked the manual everywhere and still can figure out what happened here. Does anyone have some idea about it and help me out. I really appreciate it. Thank you very much.

Sincerely,
Wen Xu
Chemistry Department
Texas A&M University, College Station

Forum Vet
please try this patch
Please apply the following patch and recompile.
It should fix the segmentation violation error

http://nwchemgit.github.io/images/Gibbs.patch.gz

Edo

Just Got Here
Thank you Edo,

   I downloaded the newest source file from newchem website. The version is 6.1. I compiled it and it works this time. I think before either because there is some problem in the source file or because my supercomputer administrator didn't compiled it correctly. Thank you for your help.

Wen


Quote:Edoapra Feb 22nd 12:15 am
Please apply the following patch and recompile.
It should fix the segmentation violation error

http://nwchemgit.github.io/images/Gibbs.patch.gz

Edo

Forum Vet
OK.
I saw that when Intel compilers were used, the segmentation violation error was happening. GNU gfortran was exempted from that.
My fix cured the Intel problem.

Edo

Quote:Wxu1024 Feb 22nd 11:30 am
Thank you Edo,
I downloaded the newest source file from newchem website. The version is 6.1. I compiled it and it works this time. I think before either because there is some problem in the source file or because my supercomputer administrator didn't compiled it correctly. Thank you for your help.

Wen


Quote:Edoapra Feb 22nd 12:15 am
Please apply the following patch and recompile.
It should fix the segmentation violation error

http://nwchemgit.github.io/images/Gibbs.patch.gz

Edo


Forum >> NWChem's corner >> Running NWChem