I forgot to put it in the instruction didn’t I :-)
Since it is already added to my .profile for the current version as of amber18 (root folder being the same) I keep forgetting to add it to the instructions. I suppose that since I am manually exporting $AMBERHOME and adding it to my path during the install, it should not make any difference anyway.
Meanwhile, I finished the GCC compilation and in defence of my slightly obsessive need to build using multiple available options to compare the results:
Installation of AmberTools19 serial is complete at Sat Apr 27 15:20:30 CEST 2019.
Finished test suite for AmberTools at Sat Apr 27 16:17:37 CEST 2019.
2432 file comparisons passed
2 file comparisons failed ( 2 ignored)
0 tests experienced errors
Test log file saved as /amber18/logs/test_at_serial/2019-04-27_15-21-34.log
Test diffs file saved as /amber18/logs/test_at_serial/2019-04-27_15-21-34.diff
Installation of AmberTools19 MPI is complete at Sat Apr 27 18:56:43 CEST 2019.
Installation of AmberTools19 OpenMP is complete at Sat Apr 27 19:12:53 CEST 20
So, using the gnu gcc compilers and building openMPI from source rather then using the repository/package manager version it is possible to build AmberTools19 with fewer errors and including the CPPTRAJ openMPI parallell NetCDF trajectory features.
I’ll post the rundown momentarily
All the best and have a fantastic rest of you evening!
// Gustaf
> On 27 Apr 2019, at 18:36, Hai Nguyen <nhai.qn.gmail.com> wrote:
>
> hi, you forget to "source amber.sh"
>
> Hai
>
> On Sat, Apr 27, 2019 at 8:58 AM Gustaf Olsson <gustaf.olsson.lnu.se> wrote:
>
>> So, this time around the serial configuration and compilation proceeded
>> with some warnings though without errors. The tests did, however, report
>> possible errors/failures that in part seems to stem from some python (or
>> just pytraj) issue.
>>
>> The parallell installation using external and global openMPI version
>> 4.0.1. also proceeded with some warnings though without errors. Once more
>> the -openmp configuration for cpptraj parallell NetCDF support failed with
>> an incompatibility issue.
>>
>> Attached are the registered potential failures/errors for the clang serial
>> installation. Running the parallell tests is not possible in the current
>> manner as every test generates a “Allow incoming network connection” popup
>> and turning the firewall off just produces a direct failure for all tests.
>>
>> ### INSTALLATION ###
>> # As I have already done this, I cannot comment on the step being needed
>> or not
>> $ open
>> /Library/Developer/CommandLineTools/Packages/macOS_SDK_headers_for_macOS_10.14.pkg
>>
>> $ export AMBERHOME=/Users/guolaa/SOFTWARE/amber18
>> $ export PATH=${AMBERHOME}:$PATH
>> $ export MACOSX_DEPLOYMENT_TARGET='10.14'
>> $ ./configure -macAccelerate clang
>> $ make install
>>
>> Installation complete
>>
>> $ ./configure -macAccelerate -mpi clang
>> $ make install
>>
>> Installation complete
>>
>> $ export DO_PARALLEL="mpirun -np 4"
>> make test
>>
>> —> First firewall issues, turning off the firewall produces the following:
>>
>> Finished test suite for AmberTools at Sat Apr 27 09:22:45 CEST 2019.
>>
>> 3 file comparisons passed
>> 693 file comparisons failed ( 0 ignored)
>> 689 tests experienced errors
>> Test log file saved as
>> /amber18/logs/test_at_parallel/2019-04-27_09-21-56.log
>> Test diffs file saved as
>> /amber18/logs/test_at_parallel/2019-04-27_09-21-56.diff
>> make[1]: *** [test.parallel] Error 1
>> make: [test.parallel] Error 2 (ignored)
>>
>> ./configure -macAccelerate -openmp clang
>> Using the AmberTools miniconda installation in
>> /Users/guolaa/SOFTWARE/amber18/miniconda
>> version 2.7.15
>> OpenMP and clang are not currently compatible
>> Configure failed due to the errors above!
>>
>> #### POTENTIAL SERIAL TEST ISSUES ###
>>
>>
>> ********************************************************************************
>> TEST: /amber18/AmberTools/src/cpptraj/test/Test_XtalSymm
>> ==============================================================
>> cd pytraj && /Applications/Xcode.app/Contents/Developer/usr/bin/make
>> Testing serial pytraj
>> /amber18/bin/amber.python test.py
>> Traceback (most recent call last):
>> File "test.py", line 2, in <module>
>> import pytraj as pt
>> ImportError: No module named pytraj
>> test.pytraj: Program error
>> ============================= test session starts
>> ==============================
>> platform darwin -- Python 2.7.15, pytest-4.4.1, py-1.8.0, pluggy-0.9.0 --
>> /amber18/bin/amber.python
>> cachedir: .pytest_cache
>> rootdir: /amber18/AmberTools/src/pytraj
>> collecting ... collected 0 items / 1 errors
>>
>> ==================================== ERRORS
>> ====================================
>> __________ ERROR collecting tests/test_energy/test_sander_energies.py
>> __________
>> ImportError while importing test module
>> '/amber18/AmberTools/src/pytraj/tests/test_energy/test_sander_energies.py'.
>> Hint: make sure your test modules/packages have valid Python names.
>> Traceback:
>> ../../../miniconda/lib/python2.7/site-packages/six.py:709: in exec_
>> exec("""exec _code_ in _globs_, _locs_""")
>> ../../src/pytraj/tests/test_energy/test_sander_energies.py:4: in <module>
>> import pytraj as pt
>> E ImportError: No module named pytraj
>> !!!!!!!!!!!!!!!!!!! Interrupted: 1 errors during collection
>> !!!!!!!!!!!!!!!!!!!!
>> =========================== 1 error in 0.20 seconds
>> ============================
>>
>>
>> --> So it seems that the pytraj module was not correctly installed.
>>
>>
>> ==============================================================
>> cd ../src/FEW/examples/test && AMBER_SOURCE=/amber18 ./Run.few.test
>> #########################################################################
>> PERFORMING FEW TESTS
>>
>> If all of the tests below fail then check the system specific path to
>> the locally installed Perl libraries.
>> For details on test issues and on how to fix them read:
>> /amber18/AmberTools/src/FEW/examples/test/few.log
>>
>> CLEANING
>> PERFORMING TEST 01: Step1 of RESP charge calculation
>> CHECKING IN TEST 01: Step1 of RESP charge calculation
>> possible FAILURE: file
>> /amber18/AmberTools/src/FEW/examples/test/gauss/L51a/L51a.gcrt does not
>> exist.
>> ==============================================================
>> mv: rename /amber18/AmberTools/src/FEW/examples/test/gauss to
>> /amber18/AmberTools/src/FEW/examples/test/gauss_setup: No such file or
>> directory
>> PERFORMING TEST 02: Step2 of RESP charge calculation and setup of MD
>> simulations
>> CHECKING IN TEST 02: Step2 of RESP charge calculation and setup of MD
>> simulations
>> possible FAILURE: file
>> /amber18/AmberTools/src/FEW/examples/test/gauss/L51a/L51a.gout.xyz does
>> not exist.
>> ==============================================================
>> possible FAILURE: file
>> /amber18/AmberTools/src/FEW/examples/test/leap/L51a/L51a_resp.lib does not
>> exist.
>> ==============================================================
>> possible FAILURE: file
>> /amber18/AmberTools/src/FEW/examples/test/leap/L51a/L51a.frcmod does not
>> exist.
>> ==============================================================
>> possible FAILURE: file
>> /amber18/AmberTools/src/FEW/examples/test/leap/L51a/L51a_com.pdb does not
>> exist.
>> ==============================================================
>> possible FAILURE: file
>> /amber18/AmberTools/src/FEW/examples/test/leap/L51a/L51a_rec.pdb does not
>> exist.
>> ==============================================================
>> possible FAILURE: file
>> /amber18/AmberTools/src/FEW/examples/test/leap/L51a/L51a_lig.pdb does not
>> exist.
>> ==============================================================
>> possible FAILURE: file
>> /amber18/AmberTools/src/FEW/examples/test/MD_resp/L51a/cryst/L51a_solv_com.crd
>> does not exist.
>> ==============================================================
>> possible FAILURE: file
>> /amber18/AmberTools/src/FEW/examples/test/MD_resp/L51a/cryst/L51a_solv_rec.crd
>> does not exist.
>> ==============================================================
>> possible FAILURE: file
>> /amber18/AmberTools/src/FEW/examples/test/MD_resp/L51a/cryst/L51a_solv_lig.crd
>> does not exist.
>> ==============================================================
>> possible FAILURE: file
>> /amber18/AmberTools/src/FEW/examples/test/MD_resp/L51a/lig/equi/
>> md_nvt_red_06.in does not exist.
>> ==============================================================
>> possible FAILURE: file
>> /amber18/AmberTools/src/FEW/examples/test/MD_resp/rec/equi/
>> md_nvt_red_06.in does not exist.
>> ==============================================================
>> possible FAILURE: file
>> /amber18/AmberTools/src/FEW/examples/test/MD_resp/L51a/lig/prod/
>> md_prod_001.in does not exist.
>> ==============================================================
>> possible FAILURE: file
>> /amber18/AmberTools/src/FEW/examples/test/MD_resp/rec/prod/md_prod_001.in
>> does not exist.
>> ==============================================================
>>
>>
>> --> Don't exactly know what happened though something is apparently
>> missing.
>>
>>
>> ==============================================================
>> Testing the Python LES API
>> /amber18/bin/amber.python test.py
>> Could not import sanderles. Skipping test
>>
>> --> Seems like there is an issue with python
>>
>> Finished test suite for AmberTools at Sat Apr 27 08:27:19 CEST 2019.
>>
>> 2077 file comparisons passed
>> 16 file comparisons failed ( 1 ignored)
>> 20 tests experienced errors
>> Test log file saved as /amber18/logs/test_at_serial/2019-04-27_07-40-53.log
>> Test diffs file saved as
>> /amber18/logs/test_at_serial/2019-04-27_07-40-53.diff
>>
>> --- DIFFS ---
>>
>> possible FAILURE: check addles.out.dif
>> /amber18/test/PIMD/part_pimd_water
>>
>> --> This file is empty...
>>
>>
>> possible FAILURE: (ignored) check min.out.dif
>> /amber18/test/sebomd/AM1-d-CB1/dimethylether
>> 99c99
>> < 4 -4.5795E+1 4.0472 7.7317 H13 5
>>> 4 -4.5795E+1 4.0472 7.7317 H12 4
>>
>> --> Don't really know what has happened here though I assume the
>> difference is the hydrogen atom name?
>>
>> Next up is the GCC installation
>>
>> Have a fantastic weekend and best regards
>> Best regards
>> // Gustaf
>> _______________________________________________
>> 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
Received on Sat Apr 27 2019 - 11:30:02 PDT
