On Tue, 2014-08-19 at 20:15 +0800, Huang wrote:
> I encounter the same situation when `make test` Amber14, for a serial
> version. Here is the setting of my platform:
> Intel(R) Xeon(R) CPU E7310 . 1.60GHz
> Debian jessie/sid
> gcc (Debian 4.9.1-4) 4.9.1
> mpirun (Open MPI) 1.6.5
>
> I also do `make test` for a parallel version with DO_PARALLEL='mpirun
> -np2', and more module reports similar failure.
> Thus bin/pmemd.MPI is also broken. Failed modules are:
>
> export TESTsander='../../../bin/pmemd.MPI'; cd pmemdTI/campTI && ./Run.campTI
> diffing campTI.out.save with campTI.out
> possible FAILURE: check campTI.out.dif
> ==============================================================
> export TESTsander='../../../bin/pmemd.MPI'; cd pmemdTI/pheMTI && ./Run.0
> diffing out.0.save with out.0
> PASSED
> ==============================================================
> export TESTsander='../../../bin/pmemd.MPI'; cd pmemdTI/pheMTI && ./Run.1
> diffing out.1.save with out.1
> PASSED
> ==============================================================
> export TESTsander='../../../bin/pmemd.MPI'; cd pmemdTI/pheMTI && ./Run.lambda0
> diffing pheMTI0.out.save with pheMTI0.out
> possible FAILURE: check pheMTI0.out.dif
> ==============================================================
> export TESTsander='../../../bin/pmemd.MPI'; cd pmemdTI/pheMTI && ./Run.lambda1
> diffing pheMTI1.out.save with pheMTI1.out
> possible FAILURE: check pheMTI1.out.dif
> ==============================================================
> export TESTsander='../../../bin/pmemd.MPI'; cd pmemdTI/sodium && ./Run.sodium
> diffing md1.o.save with md1.o
> possible FAILURE: check md1.o.dif
> ==============================================================
> export TESTsander='../../../bin/pmemd.MPI'; cd pmemdTI/ti_ggcc && ./Run.test1
> diffing out_icfe1_klambda_1_clambda_0.5.p1.save with
> out_icfe1_klambda_1_clambda_0.5.p1
> possible FAILURE: check out_icfe1_klambda_1_clambda_0.5.p1.dif
> ==============================================================
> export TESTsander='../../../bin/pmemd.MPI'; cd pmemdTI/ti_ggcc &&
> ./Run.test2
> diffing out_icfe1_klambda_6_clambda_0.5.p1.save with
> out_icfe1_klambda_6_clambda_0.5.p1
> possible FAILURE: check out_icfe1_klambda_6_clambda_0.5.p1.dif
> ==============================================================
>
> ......
>
>
> Protein-Ligand complex test
> diffing out.save with out
> possible FAILURE: check out.dif
> ==============================================================
> diffing restrt.save with restrt
> PASSED
> ==============================================================
>
> ==============================================================
> Solvation free energy test
> diffing out.0.save with out.0
> possible FAILURE: check out.0.dif
> ==============================================================
> diffing restrt.0.save with restrt.0
> PASSED
> ==============================================================
> diffing out2.0.save with out2.0
> possible FAILURE: check out2.0.dif
> ==============================================================
> diffing restrt2.0.save with restrt2.0
> PASSED
> ==============================================================
>
> ==============================================================
> Dynamic lambda test
> diffing out.0.save with out.0
> possible FAILURE: check out.0.dif
> ==============================================================
> diffing restrt.0.save with restrt.0
> PASSED
> ==============================================================
>
> ==============================================================
> Restrained complex test
> diffing out.0.save with out.0
> possible FAILURE: check out.0.dif
> ==============================================================
> diffing restrt.0.save with restrt.0
> PASSED
> ==============================================================
>
> ==============================================================
> Using softcore electrostatics
> diffing out.0.save with out.0
> possible FAILURE: check out.0.dif
> ==============================================================
> diffing restrt.0.save with restrt.0
> PASSED
> ==============================================================
> .....
>
> ==============================================================
> export TESTsander='../../../bin/pmemd.MPI'; cd emil/emil_pmemd_tip3p
> && ./Run.emil EMIL
> DO_PARALLEL is set: mpirun -np 4
> diffing mdout.emil.0.1.save with mdout.emil.0.1
> possible FAILURE: check mdout.emil.0.1.dif
> ==============================================================
> diffing emil.log.0.1.save with emil.log.0.1
> PASSED
> ==============================================================
> diffing mdout.emil.0.5.save with mdout.emil.0.5
> possible FAILURE: check mdout.emil.0.5.dif
> ==============================================================
> diffing emil.log.0.5.save with emil.log.0.5
> PASSED
> ==============================================================
> diffing mdout.emil.0.9.save with mdout.emil.0.9
> possible FAILURE: check mdout.emil.0.9.dif
> ==============================================================
> diffing emil.log.0.9.save with emil.log.0.9
> PASSED
> ==============================================================
>
>
> I guess Dr Case is right, since all differences are around the section
> "DV/DL, AVERAGES".
>
> What can I do to fix the error, or to obtain more specific information?
Make sure you've updated Amber 14 to the latest version. You can do
this by recompiling. If you are asked to update your installation
during the configure step, you need to recompile Amber in serial and
parallel. If not, it's fine.
Also, you should look at the dif files that the output suggests you look
at. For instance, see where it says "check mdout.emil.0.9.dif" -- look
at that file. If the differences appear very small, then it's not
really a problem, it's just round-off error.
Also, what compilers are you using?
HTH,
Jason
--
Jason M. Swails
BioMaPS,
Rutgers University
Postdoctoral Researcher
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Aug 19 2014 - 06:00:02 PDT
