Hi,
I have very recently installed Amber on a new supercomputer with intel toochain. (intel/2019)
Installation works fine, however I get 4 Errors on the test suite for serial AmberTools (no errors on Amber).
As I intend to do QMMM calculations I am worried that these Errors are important to fix, however I
Am having trouble in finding solutions on the net. I understand that SIGSEGC errors do not say much, but maybe some of you
might have a suggestion here.
Also for the last error I am wondering if it has to do with he python miniconda installed through amber.
I have now tried on several machines and I consistently get this error when running ./configure the first time and installing miniconda;
============ Error from ./configure when installing miniconda
Done. Miniconda installed in /home/francesb/60-programs/amber18_mpi/amber18/miniconda
Traceback (most recent call last):
File "/home/francesb/60-programs/amber18_mpi/amber18/miniconda/bin/conda", line 7, in <module>
from conda.cli import main
ModuleNotFoundError: No module named 'conda'
Problem installing Miniconda... see error above.
Configure failed due to the errors above!
==
Rerunning ./configure results in no problems, but is there a way to check that the miniconda installation is not faulty?
Below are the Errors from the logfile (test_at_serial)
==========ERROR 1========
Testing the QM/MM non-periodic interface
forrtl: severe (174): SIGSEGV, segmentation fault occurred
Image PC Routine Line Source
libifcoremt.so.5 00002BA42474B522 for__signal_handl Unknown Unknown
libpthread-2.17.s 00002BA426F4D5D0 Unknown Unknown Unknown
libsander.so 00002BA40CB4F264 get_qm2_forces_re Unknown Unknown
libsander.so 00002BA40CB3D1EA sander_api_mp_san Unknown Unknown
testf 0000000000403AEC Unknown Unknown Unknown
testf 0000000000402322 Unknown Unknown Unknown
libc-2.17.so 00002BA42717C3D5 __libc_start_main Unknown Unknown
testf 0000000000402239 Unknown Unknown Unknown
Program error
Testing the Fortran file-less API
ifort -fPIC -I/cluster/home/francesb/src/amber18_gpu/amber18/include -I/cluster/h
ome/francesb/src/amber18_gpu/amber18/include -I/cluster/home/francesb/src/amber18_
gpu/amber18/include -L/cluster/home/francesb/src/amber18_gpu/amber18/lib -shared-int
el test2.F90 -o testf2 -lsander
Running testf2
Testing GB sander interface
PASSED
============ERRORs 2 and 3========
Testing the QM/MM periodic interface (PM3-PDDG)
| QMMM: ERROR dspevd failed to converge and returned 55
| QMMM: Falling back on internal diagonalizer for the remainder of this run.
| QMMM: ERROR dsyevd failed to converge and returned 55
| QMMM: Falling back on internal diagonalizer for the remainder of this run.
forrtl: severe (174): SIGSEGV, segmentation fault occurred
Image PC Routine Line Source
libifcoremt.so.5 00002ADC53FE2522 for__signal_handl Unknown Unknown
libpthread-2.17.s 00002ADC567E45D0 Unknown Unknown Unknown
libsander.so 00002ADC3C3E6264 get_qm2_forces_re Unknown Unknown
libsander.so 00002ADC3C3D41EA sander_api_mp_san Unknown Unknown
testf2 0000000000403CE7 Unknown Unknown Unknown
testf2 00000000004022A2 Unknown Unknown Unknown
libc-2.17.so 00002ADC56A133D5 __libc_start_main Unknown Unknown
testf2 00000000004021B9 Unknown Unknown Unknown
Program error
Testing the C API
icc -fPIC -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -DBINTRAJ -DHASGZ -DHASBZ2 -D__PLUMED_HAS_DLOPEN -I/cluster/home/francesb/src/amber18_gpu/amber18/include -L/cluster/home/francesb/src/amber18_gpu/amber18/lib -shared-intel test.c -o testc -lsander
Running testc
ERROR: Flag "ATOM_NAME" not found in PARM file
unreasonably small cut for non-periodic run: 5.00
/bin/sh: line 1: 3434 Segmentation fault (core dumped) ./testc
Program error
Testing the C++ API
icpc -fPIC -I/cluster/home/francesb/src/amber18_gpu/amber18/include -L/cluster/home/francesb/src/amber18_gpu/amber18/lib -shared-intel test.cpp -o testcpp -lsander
Running testcpp
Testing GB sander interface
PASSED
===========ERROR 4 ========
/bin/sh: line 1: 3447 Segmentation fault (core dumped) ./testcpp
Program error
Testing the Python API
/cluster/home/francesb/src/amber18_gpu/amber18/bin/amber.python test.py
| QMMM: ERROR dspevd failed to converge and returned 55
| QMMM: Falling back on internal diagonalizer for the remainder of this run.
| QMMM: ERROR dsyevd failed to converge and returned 55
| QMMM: Falling back on internal diagonalizer for the remainder of this run.
Testing GB sander interface
PASSED
Do you have any suggestions for how to solve these problems? The parallel and cuda (not parallel) installations also give no errors during installation, but I have not run the tests yet.
Best regards
Francesca
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Francesca L. Bleken, Ph.D.
Research Scientist
Process Chemistry and Functional Materials, SINTEF Industry
Norway
Mobile: (+47) 95 20 79 71
www.sintef.no<
http://www.sintef.no/>
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Jul 03 2019 - 01:30:02 PDT
