ABINIT Discussion Forums

Dear colleagues,

I'm having problems with calculating the phonon dispersions of TiSe2 in T1 phase. Otherwise, I have no issues when calculating these in other materials on the Linux cluster using abinit 5.8.4.
After converging the 1st dataset, it just stops in the beginning of the 2nd dataset when the perturbation part at the gamma point begins. Here's the message from the log file:

8 - MPI_COMM_FREE : Null communicator
3 - MPI_COMM_FREE : Null communicator
5 - MPI_COMM_FREE : Null communicator
6 - MPI_COMM_FREE : Null communicator
7 - MPI_COMM_FREE : Null communicator
[5] [] Aborting Program!
2 - MPI_COMM_FREE : Null communicator
[7] [] Aborting Program!
9 - MPI_COMM_FREE : Null communicator
[2] [] Aborting Program!
[8] [] Aborting Program!
[9] [] Aborting Program!
1 - MPI_COMM_FREE : Null communicator
[3] [] Aborting Program!
[1] [] Aborting Program!
4 - MPI_COMM_FREE : Null communicator
[6] [] Aborting Program!
[4] [] Aborting Program!
15 - MPI_COMM_FREE : Null communicator
Exit code -3 signaled from node003.srce.cro-ngi.hr
Killing remote processes...[15] [] Aborting Program!
10 - MPI_COMM_FREE : Null communicator
[10] [] Aborting Program!
11 - MPI_COMM_FREE : Null communicator
[11] [] Aborting Program!
Abort signaled by rank 5: Aborting program !
12 - MPI_COMM_FREE : Null communicator
[12] [] Aborting Program!
MPI process terminated unexpectedly
DONE



Has enyone even encountered such a problem before? I would appreciate any help with this. My input file is given bellow.

Thnak you in advance!

Yours,
Igor Lukacevic




# Crystalline TiSe2 - T1 phase
# computation of the response to homogeneous
# electric field and atomic displacements, at q=0
# and phonon dispersions

ndtset 9

#Set 1 : ground state self-consistency

getwfk1 0 # Cancel default
kptopt1 1 # Automatic generation of k points, taking
# into account the symmetry
nqpt1 0 # Cancel default
tolvrs1 1.0d-18 # SCF stopping criterion (modify default)
rfphon1 0 # Cancel default

#Q vectors for all datasets

#Complete set of symmetry-inequivalent qpt chosen to be commensurate
# with kpt mesh so that only one set of GS wave functions is needed.
#Generated automatically by running GS calculation with kptopt=1,
# nshift=0, shiftk=0 0 0 (to include gamma) and taking output kpt set
# file as qpt set. Set nstep=1 so only one iteration runs.

nqpt 1 # One qpt for each dataset (only 0 or 1 allowed)
# This is the default for all datasets and must
# be explicitly turned off for dataset 1.

qpt2 0.00000000E+00 0.00000000E+00 0.00000000E+00
qpt3 2.50000000E-01 0.00000000E+00 0.00000000E+00
qpt4 5.00000000E-01 0.00000000E+00 0.00000000E+00
qpt5 2.50000000E-01 2.50000000E-01 0.00000000E+00
qpt6 0.00000000E+00 0.00000000E+00 5.00000000E-01
qpt7 2.50000000E-01 0.00000000E+00 5.00000000E-01
qpt8 5.00000000E-01 0.00000000E+00 5.00000000E-01
qpt9 2.50000000E-01 2.50000000E-01 5.00000000E-01

#Sets 2-9 : Finite-wave-vector phonon calculations (defaults for all datasets)

getwfk 1 # Use GS wave functions from dataset1
kptopt 3 # Need full k-point set for finite-Q response
rfphon 1 # Do phonon response
rfatpol 1 3 # Treat displacements of all atoms
rfdir 1 1 1 # Do all directions (symmetry will be used)
tolvrs 1.0d-8 # This default is active for sets 2-9

#######################################################################
#Common input variables

#Definition of the unit cell - p = 0 GPa
acell 2*6.6388970478E+00 1.0952628882E+01
angdeg 90 90 120
spgroup 164
brvltt -1

#Definition of the atom types
ntypat 2
znucl 22 34 34

#Definition of the atoms
natom 3
typat 1 2 2
xred 0.0 0.0 0.00
1/3 2/3 0.27086866503
2/3 1/3 -0.27086866503

#Gives the number of band, explicitely (do not take the default)
nband 10
occopt 4
tsmear 0.04

#Exchange-correlation functional
ixc 1

#Definition of the planewave basis set
ecut 50
ecutsm 0.5
dilatmx 1.2

#Definition of the k-point grid
ngkpt 4 4 2
nshiftk 1
shiftk 0.5 0.5 0.5

#Definition of the SCF procedure
iscf 7
npulayit 16 # Number of Pulay iterations
nnsclo 12 # Number of non-self consistent loops
nline 14
nstep 100

Hi Igor,

it's practically impossible to tell this way what is going wrong. Try version 6.6 for kicks, and note that dilatmx 1.2 is quite high, so your effective ecut is very high, and you may just be running out of memory.

Otherwise try running without re-directing output to a file, or running under a debugger: you can go onto a run node, and launch
gdb PATHTOABINIT/abinit <pid>

with the pid of the running abinit. Then type
gdb> cont

and it will continue the calculation, and gdb should catch the failure when it happens.

Matthieu

Hi Matthieu,

thank you for your care. Actually, I found it to be an issue with the cluster I use. Although the cluster admin stubbornly claimed that it was not. But your answer could be useful in the future when I need to figure out smth more specific concerning the cluster running.

Cheers!

Igor