Re: [AMBER-Developers] parallel sander QMMM tests broken?

From: Mengjuei Hsieh <mengjueh.uci.edu>
Date: Wed, 17 Feb 2010 11:46:39 -0800

On Tue, Feb 16, 2010 at 08:35:11AM -0500, case wrote:
> For me, with the current (Feb. 16) CVS, the test.parallel.QMMM tests fail.
> Jobs get an MPI_Abort after printing "TOTAL SIZE OF NONBOND LIST".
> Mail from Kim suggests that these were working on Feb. 13 or so.
> My tests were on MacOSX, gfortran 4.4.0, openmpi. It would be of interest
> to know if others do *not* see this problem. And, of course, debug volunteers
> are most welcome.
Hi,

I didn't see the problem, my setup is osx 10.6.2, gfortran 4.4.2, openmpi
compiled from amber11 source. (Had to put $AMBERHOME/bin in the front of
$PATH to prevent mis-usage of system mpif77/mpirun.)

Check the test log in the end of mail for references.

Best,

-- 
Mengjuei Hsieh, Luo Research Group, Molecular Biology & Biochemistry,
University of California Irvine.
[test/] mjhsieh% env DO_PARALLEL="mpirun -np 2" AMBERHOME=/opt/amber11 make test.parallel.QMMM | grep -v -e diffing -e 'cd ' -e '===' -e PASSED
export TESTsander=/opt/amber11/exe/sander.MPI; make test.sander.QMMM
possible FAILURE:  check crambin_md_qmgb2_saltcon.out.dif
possible FAILURE:  check mdout.1NLN_dspevd.dif
possible FAILURE:  check mdout.1NLN_MD_ntb1_aq1.dif
possible FAILURE:  check mdout.notimaged_md_pme_qmewald.dif
possible FAILURE:  check tip3p_cap_md.out.dif
export TESTsander=/opt/amber11/exe/sander.MPI; make test.sander.DFTB
possible FAILURE:  check crambin_md.out.dif
possible FAILURE:  check crambin_md_qmgb2_aq2.out.dif
possible FAILURE:  check mdout.aladip_ewald_ntb1.dif
possible FAILURE:  check mdout.aladip_ewald_ntb1_link_atoms.dif
possible FAILURE:  check MG_QM_water_MM_MD.out.dif
possible FAILURE:  check tip3p_cap.out.dif
possible FAILURE:  check tip3p_cap_md_shake.out.dif
[test/] mjhsieh% cat */*/crambin_md_qmgb2_saltcon.out.dif */*/mdout.1NLN_dspevd.dif */*/mdout.1NLN_MD_ntb1_aq1.dif */*/mdout.notimaged_md_pme_qmewald.dif */*/tip3p_cap_md.out.dif */*/crambin_md.out.dif */*/crambin_md_qmgb2_aq2.out.dif */*/mdout.aladip_ewald_ntb1.dif */*/mdout.aladip_ewald_ntb1_link_atoms.dif */*/MG_QM_water_MM_MD.out.dif */*/tip3p_cap.out.dif */*/tip3p_cap_md_shake.out.dif
288a289
> 1:07:44.80  on 02/17/2010
254a255
> 8:15.69  on 02/17/2010
256a257
> 1:09:46.78  on 02/17/2010
264a265
> 1:31.24  on 02/17/2010
227a228
> 7.49  on 02/17/2010
265a266
> 1.84  on 02/17/2010
300a301
> 3.47  on 02/17/2010
308a309
>  done at 11:14:06.83  on 02/17/2010
511a512
>   Job began  at 11:14:19.64  on 02/17/2010
266a267
> 5:56.80  on 02/17/2010
168a169
> 7:07.22  on 02/17/2010
243a244
> 7:09.189  on 02/17/2010
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers

Received on Wed Feb 17 2010 - 12:00:02 PST