[gmx-users] Segmentation fault after mdrun for MD simulation
Mark Abraham
Mark.Abraham at anu.edu.au
Thu Aug 18 07:03:21 CEST 2011
On 18/08/2011 2:41 PM, rainy908 wrote:
> Hi Justin,
>
> THanks for the input. So I traced back to my energy minimization steps, and am getting the error message after I execute the following line:
>
> $mdrun -s 1JFF_em.tpr -o 1JFF_em.trr -c 1JFF_b4pr.gro -e em.edr
>
> output:
> Back Off! I just backed up md.log to ./#md.log.2#
> Reading file 1JFF_em.tpr, VERSION 4.5.3 (single precision)
> Starting 24 threads
>
> Will use 15 particle-particle and 9 PME only nodes
> This is a guess, check the performance at the end of the log file
>
> -------------------------------------------------------
> Program mdrun, VERSION 4.5.3
> Source code file: domdec.c, line: 6428
>
> Fatal error:
> There is no domain decomposition for 15 nodes that is compatible with the given box and a minimum cell size of 2.92429 nm
> Change the number of nodes or mdrun option -rdd
> Look in the log file for details on the domain decomposition
> For more information and tips for troubleshooting, please check the GROMACS
> website at http://www.gromacs.org/Documentation/Errors
>
> -------------------------------------------------------
>
> I figure the problem must lie within my em.mdp file:
It could, but if you follow the above advice you will learn about some
other considerations.
Mark
>
> title = 1JFF
> cpp = /lib/cpp ; location of cpp on SGI
> define = -DFLEX_SPC ; Use Ferguson’s Flexible water model [4]
> constraints = none
> integrator = steep
> dt = 0.001 ; ps !
> nsteps = 10000
> nstlist = 10
> ns_type = grid
> rlist = 0.9
> coulombtype = PME ; Use particle-mesh ewald
> rcoulomb = 0.9
> rvdw = 1.0
> fourierspacing = 0.12
> fourier_nx = 0
> fourier_ny = 0
> fourier_nz = 0
> pme_order = 4
> ewald_rtol = 1e-5
> optimize_fft = yes
> ;
> ; Energy minimizing stuff
> ;
> emtol = 1000.0
> emstep = 0.01
> ~
>
> I figure this is an issue related to with PME and the Fourier spacing?
>
> Thanks,
>
> rainy908
>
>
>
> On 17 August 2011 17:55, Justin A. Lemkul<jalemkul at vt.edu> wrote:
>
>
>
> rainy908 wrote:
>
> Dear gmx-users:
>
> Thanks Justin for your help. But now I am experiencing a Segmentation fault error when executing mdrun. I've perused the archives but found none of the threads on segmentation faults similar to my case here. I believe the segmentation fault is caused by the awkward positioning of atoms 8443 and 8446 with respect to one another, but am not 100%. Any advice would be especially welcome.
>
> My files are as follows:
>
> md.mdp
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> title = 1JFF MD
> cpp = /lib/cpp ; location of cpp on SGI
> constraints = all-bonds
> integrator = md
> dt = 0.0001 ; ps
> nsteps = 25000 ;
> nstcomm = 1
> nstxout = 500 ; output coordinates every 1.0 ps
> nstvout = 0
> nstfout = 0
> 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
> tc-grps = system
> ref_t = 310
> ; 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 310 K.
> gen_vel = yes
> gen_temp = 310.0
> gen_seed = 173529
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
>
>
> error output file:
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> ..
> ..
> Back Off! I just backed up md.log to ./#md.log.1#
> Getting Loaded...
> Reading file 1JFF_md.tpr, VERSION 4.5.3 (single precision)
> Starting 8 threads
> Loaded with Money
> Making 3D domain decomposition 2 x 2 x 2
> Back Off! I just backed up 1JFF_md.trr to ./#1JFF_md.trr.1#
>
> Back Off! I just backed up 1JFF_md.edr to ./#1JFF_md.edr.1#
>
> Step 0, time 0 (ps) LINCS WARNING
> relative constraint deviation after LINCS:
> rms 0.046849, max 1.014038 (between atoms 8541 and 8539)
>
> Step 0, time 0 (ps) LINCS WARNING
> relative constraint deviation after LINCS:
> rms 0.001453, max 0.034820 (between atoms 315 and 317)
> bonds that rotated more than 30 degrees:
> atom 1 atom 2 angle previous, current, constraint length
> bonds that rotated more than 30 degrees:
> atom 1 atom 2 angle previous, current, constraint length
>
> If mdrun is failing at step 0, it indicates that your system is physically unreasonable. Either the starting configuration has atomic clashes that have not been resolved (and thus you need better EM and/or equilibration) or that the parameters assigned to the molecules in your system are unreasonable.
>
> -Justin
>
>
> Step 0, time 0 (ps) LINCS WARNING
> relative constraint deviation after LINCS:
> rms 0.048739, max 1.100685 (between atoms 8422 and 8421)
> bonds that rotated more than 30 degrees:
> atom 1 atom 2 angle previous, current, constraint length
> ..
> ..
> <snip>
> ..
> ..
>
> starting mdrun 'TUBULIN ALPHA CHAIN'
> 25000 steps, 50.0 ps.
> Warning: 1-4 interaction between 8443 and 8446 at distance 2.853 which is larger than the 1-4 table size 2.000 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
> or with user tables increase the table size ..
> ..
> <snip>
> ..
> ..
> step 0: Water molecule starting at atom 23781 can not be settled.
> Check for bad contacts and/or reduce the timestep if appropriate.
>
> Back Off! I just backed up step0b_n0.pdb to ./#step0b_n0.pdb.1#
>
> Back Off! I just backed up step0b_n1.pdb to ./#step0b_n1.pdb.1#
>
> Back Off! I just backed up step0b_n5.pdb to ./#step0b_n5.pdb.2#
>
> Back Off! I just backed up step0b_n3.pdb to ./#step0b_n3.pdb.2#
>
> Back Off! I just backed up step0c_n0.pdb to ./#step0c_n0.pdb.1#
>
> Back Off! I just backed up step0c_n1.pdb to ./#step0c_n1.pdb.1#
>
> Back Off! I just backed up step0c_n5.pdb to ./#step0c_n5.pdb.2#
>
> Back Off! I just backed up step0c_n3.pdb to ./#step0c_n3.pdb.2#
> Wrote pdb files with previous and current coordinates
> Wrote pdb files with previous and current coordinates
> Wrote pdb files with previous and current coordinates
> Wrote pdb files with previous and current coordinates
> Wrote pdb files with previous and current coordinates
> ^Mstep 0/opt/sge/jacobson/spool/node-2-05/job_scripts/1097116: line 21: 1473 Segmentation fault (core dumped) $MDRUN -machinefile $TMPDIR/machines -np $NSLOTS $MDRUN -v -nice 0 -np $NSLOTS -s 1JFF_md.tpr -o 1JFF_md.trr -c 1JFF_pmd.gro -x 1JFF_md.xtc -e 1JFF_md.edr
>
>
>
> On 16 August 2011 10:58, Justin A. Lemkul<jalemkul at vt.edu> wrote:
>
>
>
> rainy908 wrote:
>
> Hi,
>
> I get the error "Atomtype CR1 not found" when I execute grompp. After perusing the gmx archives, I understand this error has to do with the lack of "CR1" being specified in the force field. However, I did include the appropriate .itp files in my .top file (shown below). As you can see, obviously CR1 is specified in taxol.itp and gtp.itp. Therefore, I'm not sure what exactly is the problem here.
>
>
> You're mixing and matching force fields. PRODRG produces Gromos-compatible parameters (and does a poor job of that - I can tell you now that the charges assigned to the aromatic C and H of taxol are total junk). You're then trying to combine these parameters with OPLS. Even if you could somehow hack the topology together, the results would be useless.
>
> For more about the hazards of PRODRG, see http://pubs.acs.org/doi/abs/10.1021/ci100335w.
>
> The solution is to pick a force field and apply it uniformly to the system you're working with.
>
> -Justin
>
> Any input on this issue would be greatly appreciated.
>
> 1JFF.top
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> ; File '1JFF.top' was generated
> ; By user: onbekend (0)
> ; On host: onbekend
> ; At date: Mon Aug 15 17:22:01 2011
> ;
> ; This is a standalone topology file
> ;
> ; It was generated using program:
> ; pdb2gmx - VERSION 4.5.3
> ;
> ; Command line was:
> ; /software/gromacs/gromacs-4.5.3-x86_64/bin/pdb2gmx -f 1JFF_new.pdb -o 1JFF.gro -p 1JFF.top
> ;
> ; Force field was read from the standard Gromacs share directory.
> ;
>
> ; Include forcefield parameters
> #include "oplsaa.ff/forcefield.itp"
> #include "taxol.itp"
> #include "gdp.itp"
> #include "gtp.itp"
>
> ; Include chain topologies
> #include "1JFF_Protein_chain_A.itp"
> #include "1JFF_Protein_chain_B.itp"
> #include "1JFF_Ion_chain_A2.itp"
>
> ; Include water topology
> #include "oplsaa.ff/spc.itp"
>
> #ifdef POSRES_WATER
> ; Position restraint for each water oxygen
> [ position_restraints ]
> ; i funct fcx fcy fcz
> 1 1 1000 1000 1000
> #endif
>
> ; Include topology for ions
> #include "oplsaa.ff/ions.itp"
>
> [ system ]
> ; Name
> TUBULIN ALPHA CHAIN; TUBULIN BETA CHAIN in water
>
> [ molecules ]
> ; Compound #mols
> Protein_chain_A 1
> Protein_chain_B 1
> Ion_chain_A2 1
> Taxol 1
> GDP 1
> GTP 1
> SOL 21016
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
>
>
> taxol.itp
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> ; This file was generated by PRODRG version AA081006.0504
> ; PRODRG written/copyrighted by Daan van Aalten
> ; and Alexander Schuettelkopf
> ;
> ; Questions/comments to dava at davapc1.bioch.dundee.ac.uk
> ;
> ; When using this software in a publication, cite:
> ; A. W. Schuettelkopf and D. M. F. van Aalten (2004).
> ; PRODRG - a tool for high-throughput crystallography
> ; of protein-ligand complexes.
> ; Acta Crystallogr. D60, 1355--1363.
> ;
> ;
>
> [ moleculetype ]
> ; Name nrexcl
> TA1 3
>
> [ atoms ]
> ; nr type resnr resid atom cgnr charge mass
> 1 CR1 1 TA1 C05 1 -0.015 12.0110
> 2 HC 1 TA1 H05 1 0.015 1.0080
> 3 CR1 1 TA1 C06 2 -0.012 12.0110
> 4 HC 1 TA1 H06 2 0.019 1.0080
> 5 CR1 1 TA1 C07 2 -0.012 12.0110
> 6 HC 1 TA1 H07 2 0.019 1.0080
> 7 CR1 1 TA1 C08 2 -0.012 12.0110
> 8 HC 1 TA1 H08 2 0.019 1.0080
> 9 CR1 1 TA1 C09 2 -0.012 12.0110
> 10 HC 1 TA1 H09 2 0.019 1.0080
> 11 C 1 TA1 C04 2 -0.028 12.0110
> 12 C 1 TA1 C03 3 0.357 12.0110
> 13 O 1 TA1 O03 3 -0.710 15.9994
> 14 OA 1 TA1 O02 3 -0.186 15.9994
> 15 CH1 1 TA1 C02 3 0.235 13.0190
> 16 CH1 1 TA1 C10 3 0.148 13.0190
> 17 CH1 1 TA1 C11 3 0.156 12.0110
> 18 OA 1 TA1 O04 4 -0.175 15.9994
> 19 C 1 TA1 C12 4 0.378 12.0110
> 20 O 1 TA1 O05 4 -0.670 15.9994
> ...
> ...
> <snip>
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
>
>
> gtp.itp
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> [ moleculetype ]
> ; Name nrexcl
> GTP 3
>
> [ atoms ]
> ; nr type resnr resid atom cgnr charge mass
> 1 O 1 GTP O6 1 -0.266 15.9994
> 2 C 1 GTP C6 1 0.216 12.0110
> 3 NR 1 GTP N1 1 -0.008 14.0067
> 4 H 1 GTP H1 1 -0.020 1.0080
> 5 C 1 GTP C2 1 0.322 12.0110
> 6 NT 1 GTP N2 1 -0.001 14.0067
> 7 H 1 GTP H22 1 -0.020 1.0080
> 8 H 1 GTP H21 1 -0.020 1.0080
> 9 NR 1 GTP N3 1 -0.203 14.0067
> 10 C 1 GTP C4 2 0.133 12.0110
> 11 C 1 GTP C5 2 0.048 12.0110
> 12 NR 1 GTP N7 2 -0.321 14.0067
> 13 CR1 1 GTP C8 2 -0.037 12.0110
> 14 HC 1 GTP H8 2 -0.019 1.0080
> 15 NR 1 GTP N9 2 0.112 14.0067
> 16 CH1 1 GTP CAU 2 0.084 13.0190
> ..
> ..
> <snip>
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
>
>
> -- ========================================
>
> Justin A. Lemkul
> Ph.D. Candidate
> ICTAS Doctoral Scholar
> MILES-IGERT Trainee
> Department of Biochemistry
> Virginia Tech
> Blacksburg, VA
> jalemkul[at]vt.edu | (540) 231-9080
> http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin
>
> ========================================
>
>
> --
> ========================================
>
> Justin A. Lemkul
> Ph.D. Candidate
> ICTAS Doctoral Scholar
> MILES-IGERT Trainee
> Department of Biochemistry
> Virginia Tech
> Blacksburg, VA
> jalemkul[at]vt.edu | (540) 231-9080
> http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin
>
> ========================================
More information about the gromacs.org_gmx-users
mailing list