When using DFT, an error occur while trying to read or write to disk space


Clicked A Few Times
Hi, I am new to Nwchem. This error occurs and I can't overcome it. Please help!

My input while is as follows,

"BASIS cartesian
  • library 6-31G
END

dft
mult 2
iterations 2000
XC b3lyp
direct
end

set task:theory dft "

And the output is like this,


     Screening Tolerance Information
-------------------------------
Density screening/tol_rho: 1.00D-10
AO Gaussian exp screening on grid/accAOfunc: 14
CD Gaussian exp screening on grid/accCDfunc: 20
XC Gaussian exp screening on grid/accXCfunc: 20
Schwarz screening/accCoul: 1.00D-08


Loading old vectors from job with title :



  Time after variat. SCF:   8734.9
Time prior to 1st pass: 8734.9
------------------------------------------------------------------------
utilscratchavail: eaf_open -10002
------------------------------------------------------------------------
------------------------------------------------------------------------
current input line :
0:
------------------------------------------------------------------------
------------------------------------------------------------------------
An error occured while trying to read or write to disk space
------------------------------------------------------------------------
For more information see the NWChem manual at
http://nwchemgit.github.io/index.php/NWChem_Documentation

Forum Vet
Could you provide more detail about the NWChem binary you are using?
For example: version, how did you compile it, etc ...

Clicked A Few Times
Quote:Edoapra Mar 29th 9:56 am
Could you provide more detail about the NWChem binary you are using?
For example: version, how did you compile it, etc ...

Thanks for help, Edoapara.

The version for NWChem is 6.1.1.

The script for compile is like this,

  1. !/bin/bash
echo " "
export LD_LIBRARY_PATH=/usr/lib64:$LD_LIBRARY_PATH
export mpidir=/opt/software/mpich2-intel
cd /opt/software/NWC/nwchem-6.1.1-src
export NWCHEM_TOP=`pwd`
length=`pwd | wc -m`
if [ $length -gt 64 ]
then
 echo "The directory name chosen for NWCHEM_TOP is longer"
echo "than the maximum allowed value of 65 characters"
echo "Please shorten the path"
echo "Exiting"
exit 1
fi
cd src
export TCGRSH=/usr/bin/ssh
export NWCHEM_TARGET=LINUX64
export USE_MPI=y
export USE_MPIF=y
export USE_MPIF4=y
export MPI_LOC=$mpidir
export MPI_LIB=$MPI_LOC/lib
export LIBMPI="-L$MPI_LIB -L/usr/lib64 -lmpichf90 -lmpich -lopa -lmpl -lrt -lpthread"
export MPI_INCLUDE=$MPI_LOC/include
export LARGE_FILES="true"
export NWCHEM_MODULES="all"
export USE_NOFSCHECK="true"
export USE_NOIO="true"
export F77=ifort
export F90=ifort
export FC=ifort
export FL=ifort
export CXX=icpc
export CC=icc
export ARMCI_NETWORK = MPI-MT
export MSG_COMMS = MPI
export USE_INTERNALBLAS=y
  1. make realclean
find . -name include_stamp -exec rm {} \; -print
  1. make clean
make nwchem_config FC=ifort CC=icc > make.log
make CC=icc FC=ifort CXX=icpc >> make.log
echo "Done Making Nwchem"

Forum Vet
Since NWChem 6.6 as been our for a while, could you please try to install it this newer version, instead?

Clicked A Few Times
I tried the newest version, but the same problem occurred.

export TCGRSH=/usr/bin/ssh
export NWCHEM_TARGET=LINUX64
export ARMCI_NETWORK=MPI_TS
export USE_INTERNALBLAS=y
export USE_MPI=y
export USE_MPIF=y
export USE_MPIF4=y
export MPI_LOC=/opt/software/openmpi-2.1.0
export MPI_LIB=$MPI_LOC/lib
export LIBMPI="-lmpi_usempif08 -lmpi_usempi_ignore_tkr -lmpi_mpifh -lmpi"
export MPI_INCLUDE=$MPI_LOC/include
export LARGE_FILES="true"
export NWCHEM_MODULES="all"
  1. export USE_NOFSCHECK="true"
  2. export USE_NOIO="true"
export F77=ifort
export F90=ifort
export FC=ifort
export FL=ifort
export CXX=icpc
export CC=icc
make nwchem_config FC=ifort> make.log
make >> make.log
echo "Done Making Nwchem"
export NWCHEM_MODULES="qm geninterface"
make nwchem_config NWCHEM_MODULES=venus
make nwchem_config
make FC=ifort CC=icc CXX=icpc >> make.log
make stubs.o
cd geninterface
make FC=ifort CC=icc
cd ../../../
cd venus-nwchem66
make clean > make-venus.log
make FC=ifort >> make-venus.log

Quote:Edoapra Mar 30th 4:15 pm
Since NWChem 6.6 as been our for a while, could you please try to install it this newer version, instead?

Forum Vet
The error message you reported on March 30 2017
utilscratchavail: eaf_open -10002
can occur when more than 1024 files are open at the same time ... is this your case?

Clicked A Few Times
Quote:Edoapra Apr 20th 4:09 pm
The error message you reported on March 30 2017
utilscratchavail: eaf_open -10002
can occur when more than 1024 files are open at the same time ... is this your case?


Thanks, I will check. If it was the case, how to avoid it?

Forum Vet
Quote:Gougourun Apr 20th 7:49 pm
Quote:Edoapra Apr 20th 4:09 pm
The error message you reported on March 30 2017
utilscratchavail: eaf_open -10002
can occur when more than 1024 files are open at the same time ... is this your case?


Thanks, I will check. If it was the case, how to avoid it?


Not sure what happens in your case. It might be that you keep opening files without closing them (just a guess ...).
I could not remember seeing this failure in DFT runs ... are you using just DFT or is hooked up to the VENUS interface?
If you re-compile NWChem with the env. variable USE_NOIO=y, the problem might disappers

Forum Vet
Quote:Gougourun Apr 20th 7:49 pm
Quote:Edoapra Apr 20th 4:09 pm
The error message you reported on March 30 2017
utilscratchavail: eaf_open -10002
can occur when more than 1024 files are open at the same time ... is this your case?


Thanks, I will check. If it was the case, how to avoid it?


Not sure what happens in your case. It might be that you keep opening files without closing them (just a guess ...).
I could not remember seeing this failure in DFT runs ... are you using just DFT or is hooked up to the VENUS interface?
If you re-compile NWChem with the env. variable USE_NOIO=y, the problem might disappear

Clicked A Few Times
Quote:Edoapra Apr 21st 9:51 am
Quote:Gougourun Apr 20th 7:49 pm
Quote:Edoapra Apr 20th 4:09 pm
The error message you reported on March 30 2017
utilscratchavail: eaf_open -10002
can occur when more than 1024 files are open at the same time ... is this your case?


Thanks, I will check. If it was the case, how to avoid it?


Not sure what happens in your case. It might be that you keep opening files without closing them (just a guess ...).
I could not remember seeing this failure in DFT runs ... are you using just DFT or is hooked up to the VENUS interface?
If you re-compile NWChem with the env. variable USE_NOIO=y, the problem might disappear


Great! When using just DFT, I did not meet with this failure. Indeed Nwchem is hooked up to the VENUS interface, I will re-complile to have a try. Thanks again!

Clicked A Few Times
Quote:Edoapra Apr 21st 9:51 am
Quote:Gougourun Apr 20th 7:49 pm
Quote:Edoapra Apr 20th 4:09 pm
The error message you reported on March 30 2017
utilscratchavail: eaf_open -10002
can occur when more than 1024 files are open at the same time ... is this your case?


Thanks, I will check. If it was the case, how to avoid it?


Not sure what happens in your case. It might be that you keep opening files without closing them (just a guess ...).
I could not remember seeing this failure in DFT runs ... are you using just DFT or is hooked up to the VENUS interface?
If you re-compile NWChem with the env. variable USE_NOIO=y, the problem might disappear


Great! When using just DFT, I did not meet with this failure. Indeed Nwchem is hooked up to the VENUS interface, I will re-complile to have a try. Thanks again!

Clicked A Few Times
It helps after re-compiling with the env. variable USE_NOIO=y. Thanks again!

Quote:Edoapra Apr 21st 9:51 am
Quote:Gougourun Apr 20th 7:49 pm
Quote:Edoapra Apr 20th 4:09 pm
The error message you reported on March 30 2017
utilscratchavail: eaf_open -10002
can occur when more than 1024 files are open at the same time ... is this your case?


Thanks, I will check. If it was the case, how to avoid it?


Not sure what happens in your case. It might be that you keep opening files without closing them (just a guess ...).
I could not remember seeing this failure in DFT runs ... are you using just DFT or is hooked up to the VENUS interface?
If you re-compile NWChem with the env. variable USE_NOIO=y, the problem might disappear


Forum >> NWChem's corner >> Running NWChem