RE : RE : [gmx-users] segmentation fault in mdrun when using PME

Wed May 3 17:45:54 CEST 2006

It seems that in my case this is a bug (see Bugzilla, bug # 74) related to using the Intel  Math Kernel Library (MKL) v. 8.0.1 for Fourier transforms. The team managing the Altix are trying different FFT libraries. Eric Lindahl says that using a FFT library that is not optimized for Itanium 2 shouldn't hamper the perfomance very much since FT doesn't represent a very big part of the computation.

I think the error I get with the version compiled without Fortran is boggus, because this tutorial (John Kerrigan's) was tried sucessfully by many people, so there shouldn't be any mistakes in the .mdp file. Also, the runs I do with that version have strange output and anyway, I don't get that error with the Fortran-enabled version. Segmentation fault has been documented in the position restrained dynamics stage of this tutorial with gromacs 3.3.0 though, and was solved when upgrading to 3.3.1. 

An exploding system is often caused by extreme forces due to bad contacts, which can be relieved by a minimization step. Have you used a steepest descents minimization step on your system before doing the pr ?

________________________________

De: gmx-users-bounces at gromacs.org de la part de Arneh Babakhani
Date: mar. 2006-05-02 23:21
À: Discussion list for GROMACS users
Objet : Re: RE : [gmx-users] segmentation fault in mdrun when using PME

Hello, I'm experiencing the exact same problem, when trying to do some restrained molecular dynamics of a small peptide in a water box. Have you had any luck in trouble-shooting this? (I've pasted my mdp file below, for your reference).  Also running Gromacs 3.3.1

Arneh

title = ResMD
warnings = 10
cpp = /usr/bin/cpp ; location of cpp on SGI
define = -DPOSRES
constraints = all-bonds
integrator = md
dt = 0.002 ; ps !
nsteps = 25000 ; total 50.0 ps.
nstcomm = 1
nstxout = 500 ; output coordinates every 1.0 ps
nstvout = 1000 ; output velocities every 2.0 ps
nstfout = 0
nstlog = 10
nstenergy = 10
nstlist = 10
ns_type = grid
rlist = 0.9
coulombtype = PME
rcoulomb = 0.9
rvdw = 1.0
fourierspacing = 0.12
fourier_nx = 0
fourier_ny = 0
fourier_nz = 0
pme_order = 6
ewald_rtol = 1e-5
optimize_fft = yes
; Berendsen temperature coupling is on in four groups
Tcoupl = berendsen
tau_t = 0.1 0.1
tc_grps = protein sol
ref_t = 300 300
; Pressure coupling is on
Pcoupl = berendsen
pcoupltype = isotropic
tau_p = 0.5
compressibility = 4.5e-5
ref_p = 1.0
; Generate velocites is on at 300 K.
gen_vel = yes
gen_temp = 300.0
gen_seed = 173529

Diane Fournier wrote: 

	________________________________

	De: gmx-users-bounces at gromacs.org de la part de David van der Spoel
	Date: lun. 2006-05-01 13:33
	À: Discussion list for GROMACS users
	Objet : Re: [gmx-users] segmentation fault in mdrun when using PME

	Have you enabled fortran at the compilation stage? In that case try it
	without, otherwise please file a bugzilla, such that we can document
	this problem (and try to fix it of course).

		Still doesn't work. logfile ends in the usual way. Except this time, I get this output:

	Reading file trp_em.tpr, VERSION 3.3.1 (single precision)

	Steepest Descents:

	Tolerance (Fmax) = 1.00000e+03

	Number of steps = 500

	Warning: 1-4 interaction between 1 and 7 at distance 39513.957 which is larger than the 1-4 table size 1.000 nm

	These are ignored for the rest of the simulation

	This usually means your system is exploding,

	if not, you should increase table-extension in your mdp file

	Wrote pdb files with previous and current coordinates

	Back Off! I just backed up step0.pdb to ./#step0.pdb.1#

	Wrote pdb files with previous and current coordinates

	and then these files get written:

	step0.pdb

	#step0.pdb.1# 

	step-1.pdb 

	step1.pdb 

	Will file a bugzilla.

			------------------------------------------------------------------------

			_______________________________________________
			gmx-users mailing list    gmx-users at gromacs.org
			http://www.gromacs.org/mailman/listinfo/gmx-users
			Please don't post (un)subscribe requests to the list. Use the
			www interface or send it to gmx-users-request at gromacs.org.
			Can't post? Read http://www.gromacs.org/mailing_lists/users.php

		------------------------------------------------------------------------

		_______________________________________________
		gmx-users mailing list    gmx-users at gromacs.org
		http://www.gromacs.org/mailman/listinfo/gmx-users
		Please don't post (un)subscribe requests to the list. Use the
		www interface or send it to gmx-users-request at gromacs.org.
		Can't post? Read http://www.gromacs.org/mailing_lists/users.php

	--
	David.
	________________________________________________________________________
	David van der Spoel, PhD, Assoc. Prof., Molecular Biophysics group,
	Dept. of Cell and Molecular Biology, Uppsala University.
	Husargatan 3, Box 596,          75124 Uppsala, Sweden
	phone:  46 18 471 4205          fax: 46 18 511 755
	spoel at xray.bmc.uu.se    spoel at gromacs.org   http://folding.bmc.uu.se
	++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
	_______________________________________________
	gmx-users mailing list    gmx-users at gromacs.org
	http://www.gromacs.org/mailman/listinfo/gmx-users
	Please don't post (un)subscribe requests to the list. Use the
	www interface or send it to gmx-users-request at gromacs.org.
	Can't post? Read http://www.gromacs.org/mailing_lists/users.php

________________________________

	_______________________________________________
	gmx-users mailing list    gmx-users at gromacs.org
	http://www.gromacs.org/mailman/listinfo/gmx-users
	Please don't post (un)subscribe requests to the list. Use the 
	www interface or send it to gmx-users-request at gromacs.org.
	Can't post? Read http://www.gromacs.org/mailing_lists/users.php

-------------- next part --------------
A non-text attachment was scrubbed...
Name: winmail.dat
Type: application/ms-tnef
Size: 8925 bytes
Desc: not available
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20060503/f1ab4de7/attachment.bin>