Dalton 2016.0 does not terminate properly with
Posted: 02 Aug 2018, 13:33
Dear Dalton Users,
I have compiled Dalton2016.0. The details are as follows:
Compilation information
-----------------------
Who compiled | addiw17
Host | ui.plgrid.wcss.wroc.pl
System | Linux-3.10.0-693.17.1.el7.x86_64
CMake generator | Unix Makefiles
Processor | x86_64
64-bit integers | OFF
MPI | ON
Fortran compiler | /usr/local/openmpi/intel-15.0/1.8.4/bin/mpif90
Fortran compiler version | ifort (IFORT) 15.0.1 20141023
C compiler | /usr/local/openmpi/intel-15.0/1.8.4/bin/mpicc
C compiler version | icc (ICC) 15.0.1 20141023
C++ compiler | /usr/local/openmpi/intel-15.0/1.8.4/bin/mpicxx
C++ compiler version | unknown
BLAS | /usr/local/intel/15.0.1.133/composer_xe_2015.1.133
| /mkl/lib/intel64/libmkl_intel_lp64.so;/usr/local/i
| ntel/15.0.1.133/composer_xe_2015.1.133/mkl/lib/int
| el64/libmkl_sequential.so;/usr/local/intel/15.0.1.
| 133/composer_xe_2015.1.133/mkl/lib/intel64/libmkl_
| core.so;/usr/lib64/libpthread.so;/usr/lib64/libm.s
| o
LAPACK | /usr/local/intel/15.0.1.133/composer_xe_2015.1.133
| /mkl/lib/intel64/libmkl_lapack95_lp64.a;/usr/local
| /intel/15.0.1.133/composer_xe_2015.1.133/mkl/lib/i
| ntel64/libmkl_intel_lp64.so
Static linking | OFF
Last Git revision | 693735158eee69ff73df6846b9911e94803b8d7a
Configuration time | 2018-08-01 19:39:52.587738
I intend to use my installation in portable batch queuing system and here an issue appears that when I run calculations they finish in a sense
that proper output with results is generated (I can see it in scratch directory) however Dalton seems not to communicate with PBS to finish the
process. Hence the finished calculations do not "call" PBS to finish my task and output files are not copied to the directory from which I start
the job.
I have tried also intel16.0 and intel13.1 compilers but it does not help. What can be the source of this problem? In the same computer centre,
there is Dalton2015.0 installation but I need my own with some parameters changed. Anyway, this Dalton2015.0 does not show similar issues.
Best wishes,
Dawid
I have compiled Dalton2016.0. The details are as follows:
Compilation information
-----------------------
Who compiled | addiw17
Host | ui.plgrid.wcss.wroc.pl
System | Linux-3.10.0-693.17.1.el7.x86_64
CMake generator | Unix Makefiles
Processor | x86_64
64-bit integers | OFF
MPI | ON
Fortran compiler | /usr/local/openmpi/intel-15.0/1.8.4/bin/mpif90
Fortran compiler version | ifort (IFORT) 15.0.1 20141023
C compiler | /usr/local/openmpi/intel-15.0/1.8.4/bin/mpicc
C compiler version | icc (ICC) 15.0.1 20141023
C++ compiler | /usr/local/openmpi/intel-15.0/1.8.4/bin/mpicxx
C++ compiler version | unknown
BLAS | /usr/local/intel/15.0.1.133/composer_xe_2015.1.133
| /mkl/lib/intel64/libmkl_intel_lp64.so;/usr/local/i
| ntel/15.0.1.133/composer_xe_2015.1.133/mkl/lib/int
| el64/libmkl_sequential.so;/usr/local/intel/15.0.1.
| 133/composer_xe_2015.1.133/mkl/lib/intel64/libmkl_
| core.so;/usr/lib64/libpthread.so;/usr/lib64/libm.s
| o
LAPACK | /usr/local/intel/15.0.1.133/composer_xe_2015.1.133
| /mkl/lib/intel64/libmkl_lapack95_lp64.a;/usr/local
| /intel/15.0.1.133/composer_xe_2015.1.133/mkl/lib/i
| ntel64/libmkl_intel_lp64.so
Static linking | OFF
Last Git revision | 693735158eee69ff73df6846b9911e94803b8d7a
Configuration time | 2018-08-01 19:39:52.587738
I intend to use my installation in portable batch queuing system and here an issue appears that when I run calculations they finish in a sense
that proper output with results is generated (I can see it in scratch directory) however Dalton seems not to communicate with PBS to finish the
process. Hence the finished calculations do not "call" PBS to finish my task and output files are not copied to the directory from which I start
the job.
I have tried also intel16.0 and intel13.1 compilers but it does not help. What can be the source of this problem? In the same computer centre,
there is Dalton2015.0 installation but I need my own with some parameters changed. Anyway, this Dalton2015.0 does not show similar issues.
Best wishes,
Dawid