[gmx-users] g_cluster segmentation fault

[Mark Abraham]

On 26/03/11, Craig Kitchen  <ck363 at cam.ac.uk> wrote:
> Dear All,
> 
> When running g_cluster on a file generated with g_mdmat I receive a segmentation fault. I have tried different versions of Gromacs (4.0.5 and 4.5.3) on different workstations with both large and small matrices - all without success.
> 

Good trouble-shooting and report. You'd likely have been asked to check a recent version and with a small matrix :-)

This was a two-part bug - a clean-up function was called when the corresponding initialization hadn't been called, and that clean-up function assumed that initialization had happened. I fixed both in git.

You can work around this by hacking in src/tools/gmx_cluster.c. Around line 1260 there will be code like

if (bPBC) {
  gmx_rmpbc_done(gpbc);
}

which you should move one line up (inside a set of braces that mean the clean-up is called only after initialization occurs). Don't worry about the indenting.

Mark


> I am calling g_cluster with:
> 
> g_cluster -dm dmat_1.xpm -cutoff 3.4 -g cluster_1.log
> 
> Whereupon I receive the error:
> 
> Using linkage method for clustering
> Reading rms distance matrix 100%
> Segmentation fault
> 
> The distance matrix (dmat_1.xpm) was generated with:
> 
> g_mdmat -f onlyba.gro -s onlyba.gro -t 4.0 -mean dmat_1.xpm
> 
> The configuration (onlyba.gro) and matrix file (dmat_1.xpm) are attached (and since they are short also pasted below).
> 
> Any help on this would be greatly appreciated!
> 
> Regards,
> 
> Craig
> 
> Craig Kitchen
> Department of Chemistry
> University of Cambridge
> Lensfield Road
> Cambridge
> CB2 1EW
> UK
> 
> 
> 
> 
> Generated by trjconv : Coarse-Grained Model of EAS Barrels and Loop Residues (including bonds) t= 0.00000
>  10
>   1BA      BA    1  44.780  21.592  37.304  0.0067  0.0035  0.0073
>  29BA      BA    2  12.191  38.323   0.704 -0.0084  0.0108 -0.0014
>  57BA      BA    3   2.092   9.298  47.957  0.0053 -0.0144  0.0053
>  85BA      BA    4  15.664  33.182  25.406  0.0291 -0.0067 -0.0069
> 113BA      BA    5  47.194  32.613  24.031 -0.0146 -0.0132  0.0184
> 141BA      BA    6  11.731  25.247  11.921 -0.0100 -0.0049  0.0025
> 169BA      BA    7  33.751  20.349  25.599 -0.0103 -0.0131  0.0078
> 197BA      BA    8   2.467  11.277  47.988 -0.0063  0.0188 -0.0001
> 225BA      BA    9  25.948  26.283  35.007  0.0191  0.0058  0.0047
> 253BA      BA   10  26.103  27.621  36.697 -0.0057  0.0281 -0.0110
> 48.00000  48.00000  48.00000
> 
> 
> /* XPM */
> /* Generated by g_mdmat */
> /* This file can be converted to EPS by the GROMACS program xpm2ps */
> /* title:   "Mean smallest distance" */
> /* legend:  "Distance (nm)" */
> /* x-label: "Residue Index" */
> /* y-label: "Residue Index" */
> /* type:    "Continuous" */
> static char *gromacs_xpm[] = {
> "10 10   40 1",
> "A  c #FFFFFF " /* "0" */,
> "B  c #F8F8F8 " /* "0.103" */,
> "C  c #F2F2F2 " /* "0.205" */,
> "D  c #EBEBEB " /* "0.308" */,
> "E  c #E5E5E5 " /* "0.41" */,
> "F  c #DEDEDE " /* "0.513" */,
> "G  c #D8D8D8 " /* "0.615" */,
> "H  c #D1D1D1 " /* "0.718" */,
> "I  c #CBCBCB " /* "0.821" */,
> "J  c #C4C4C4 " /* "0.923" */,
> "K  c #BEBEBE " /* "1.03" */,
> "L  c #B7B7B7 " /* "1.13" */,
> "M  c #B1B1B1 " /* "1.23" */,
> "N  c #AAAAAA " /* "1.33" */,
> "O  c #A3A3A3 " /* "1.44" */,
> "P  c #9D9D9D " /* "1.54" */,
> "Q  c #969696 " /* "1.64" */,
> "R  c #909090 " /* "1.74" */,
> "S  c #898989 " /* "1.85" */,
> "T  c #838383 " /* "1.95" */,
> "U  c #7C7C7C " /* "2.05" */,
> "V  c #767676 " /* "2.15" */,
> "W  c #6F6F6F " /* "2.26" */,
> "X  c #696969 " /* "2.36" */,
> "Y  c #626262 " /* "2.46" */,
> "Z  c #5C5C5C " /* "2.56" */,
> "a  c #555555 " /* "2.67" */,
> "b  c #4E4E4E " /* "2.77" */,
> "c  c #484848 " /* "2.87" */,
> "d  c #414141 " /* "2.97" */,
> "e  c #3B3B3B " /* "3.08" */,
> "f  c #343434 " /* "3.18" */,
> "g  c #2E2E2E " /* "3.28" */,
> "h  c #272727 " /* "3.38" */,
> "i  c #212121 " /* "3.49" */,
> "j  c #1A1A1A " /* "3.59" */,
> "k  c #141414 " /* "3.69" */,
> "l  c #0D0D0D " /* "3.79" */,
> "m  c #070707 " /* "3.9" */,
> "n  c #000000 " /* "4" */,
> /* x-axis:  1 2 3 4 5 6 7 8 9 10 */
> /* y-axis:  1 2 3 4 5 6 7 8 9 10 */
> "nnnnnnnnVA",
> "nnnnnnnnAV",
> "nnUnnnnAnn",
> "nnnnnnAnnn",
> "nnnnnAnnnn",
> "nnnnAnnnnn",
> "nnnAnnnnnn",
> "nnAnnnnUnn",
> "nAnnnnnnnn",
> "Annnnnnnnn"
> 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20110326/02efb618/attachment.html>