Dear Ross and Jason,
Thanks for your suggestions. I went through your suggestions and
successfully installed the parallel version of Amber 12. But this time I am
facing some new type of error during make test_paralllel and mpi run. The
screenshots of test are as ---
./Run.1NLN_MD_ntb1_vsolv: Program error
make[2]: [test.sander.QMMM] Error 1 (ignored)
cd qmmm2/PM3-MMX && ./Run.crambin_md_pm3mmx
/home/kshatresh/amber12//bin/sander.MPI: symbol lookup error:
/home/kshatresh/amber12//bin/sander.MPI: undefined symbol: __svml_acos2
/home/kshatresh/amber12//bin/sander.MPI: symbol lookup error:
/home/kshatresh/amber12//bin/sander.MPI: undefined symbol: __svml_acos2
--------------------------------------------------------------------------
mpirun has exited due to process rank 1 with PID 14443 on
node ensemble.bsbe.iitk.ac.in exiting without calling "finalize". This may
have caused other processes in the application to be
terminated by signals sent by mpirun (as reported here).
--------------------------------------------------------------------------
./Run.crambin_md_pm3mmx: Program error
make[2]: [test.sander.QMMM] Error 1 (ignored)
cd qmmm2/PM3-MMX2 && ./Run.crambin_md_pm3mmx2
/home/kshatresh/amber12//bin/sander.MPI: symbol lookup error:
/home/kshatresh/amber12//bin/sander.MPI: undefined symbol: __svml_acos2
/home/kshatresh/amber12//bin/sander.MPI: symbol lookup error:
/home/kshatresh/amber12//bin/sander.MPI: undefined symbol: __svml_acos2
--------------------------------------------------------------------------
mpirun has exited due to process rank 1 with PID 14451 on
node ensemble.bsbe.iitk.ac.in exiting without calling "finalize". This may
have caused other processes in the application to be
terminated by signals sent by mpirun (as reported here).
--------------------------------------------------------------------------
./Run.crambin_md_pm3mmx2: Program error
make[2]: [test.sander.QMMM] Error 1 (ignored)
cd qmmm2/crambin_PM6 && ./Run.crambin
/home/kshatresh/amber12//bin/sander.MPI: symbol lookup error:
/home/kshatresh/amber12//bin/sander.MPI: undefined symbol: __svml_acos2
/home/kshatresh/amber12//bin/sander.MPI: symbol lookup error:
/home/kshatresh/amber12//bin/sander.MPI: undefined symbol: __svml_acos2
and I found similar errors during mpirun sander.MPI.
I am thankful to your valuable suggestions.
Regards
Kshatresh
On Wed, Mar 13, 2013 at 12:47 AM, Jason Swails <jason.swails.gmail.com>wrote:
> As a note, you can also get this error by compiling with two different
> versions of the Intel compiler. It is worth comparing the output of "which
> ifort" with "mpif90 -show" to make sure that the same version of ifort is
> being used in each case.
>
> HTH,
> Jason
>
> On Tue, Mar 12, 2013 at 12:55 PM, Ross Walker <ross.rosswalker.co.uk>
> wrote:
>
> > Hi Kshatresh,
> >
> > Did you build the serial version with the Intel compilers as well? My
> > guess is that somehow your netcdf in AMBER was previously built with gnu
> > compilers.
> >
> > Your quickest option is probably to start with a completely clean tree,
> > from untarring the distribution files, run ./configure intel a few times
> > until it no longer asks if you want to patch things. Then run make, then
> > ./configure -mpi intel; make - and see if it then works.
> >
> > All the best
> > Ross
> >
> >
> > On 3/12/13 8:59 AM, "Kshatresh Dutta Dubey" <kshatresh.gmail.com> wrote:
> >
> > >Dear Ross,
> > >
> > >Thanks for your quick reply. I checked and followings are the out of
> > >mpif90
> > >-show' and 'mpicc -show' respectively-
> > >
> > >/opt/intel/Compiler/11.1/046/bin/intel64/ifort
> > >-I/opt/mpi/openmpi/1.4.5/intel/include -fexceptions
> > >-I/opt/mpi/openmpi/1.4.5/intel/lib -L/opt/mpi/openmpi/1.4.5/intel/lib
> > >-lmpi_f90 -lmpi_f77 -lmpi -lopen-rte -lopen-pal -ldl
> -Wl,--export-dynamic
> > >-lnsl -lutil
> > >
> > >and
> > >
> > >/opt/intel/Compiler/11.1/046/bin/intel64/icc
> > >-I/opt/mpi/openmpi/1.4.5/intel/include -fexceptions -pthread
> > >-L/opt/mpi/openmpi/1.4.5/intel/lib -lmpi -lopen-rte -lopen-pal -ldl
> > >-Wl,--export-dynamic -lnsl -lutil
> > >
> > >Regards
> > >Kshatresh
> > >
> > >
> > >
> > >On Tue, Mar 12, 2013 at 9:23 PM, Ross Walker <ross.rosswalker.co.uk>
> > >wrote:
> > >
> > >> Hi Kshatresh,
> > >>
> > >>
> > >> It looks to me like you MPI installation is incorrectly configured.
> Note
> > >> the error:
> > >>
> > >> ncsu-umbrella.F90(9): error #7013: This module file was not generated
> by
> > >> any release of this compiler. [NETCDF]
> > >> use netcdf, only : nf90_double, nf90_float
> > >> ----^
> > >>
> > >>
> > >> This suggests that there is a compiler mismatch. You should check what
> > >> 'mpif90 -show' and 'mpicc -show' returns.
> > >>
> > >> All the best
> > >> Ross
> > >>
> > >> On 3/12/13 8:41 AM, "Kshatresh Dutta Dubey" <kshatresh.gmail.com>
> > wrote:
> > >>
> > >> >Dear Users
> > >> >
> > >> >I am installing Amber 12 on IBM cluster Rocks 6, Centos. I have
> > >> >successfully installed the serial version. However, when I am
> > >>installing
> > >> >parallel version using ifort compiler, it is giving several errors.
> The
> > >> >log
> > >> >file during make install and configure is attached.
> > >> >I'll deeply appreciate your suggestions.
> > >> >
> > >> >Regards
> > >> >_______________________________________________
> > >> >AMBER mailing list
> > >> >AMBER.ambermd.org
> > >> >http://lists.ambermd.org/mailman/listinfo/amber
> > >>
> > >>
> > >>
> > >> _______________________________________________
> > >> AMBER mailing list
> > >> AMBER.ambermd.org
> > >> http://lists.ambermd.org/mailman/listinfo/amber
> > >>
> > >_______________________________________________
> > >AMBER mailing list
> > >AMBER.ambermd.org
> > >http://lists.ambermd.org/mailman/listinfo/amber
> >
> >
> >
> > _______________________________________________
> > AMBER mailing list
> > AMBER.ambermd.org
> > http://lists.ambermd.org/mailman/listinfo/amber
> >
>
>
>
> --
> Jason M. Swails
> Quantum Theory Project,
> University of Florida
> Ph.D. Candidate
> 352-392-4032
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Mar 13 2013 - 04:30:03 PDT
