the detail of error is
Calculating the average structure ...
Last frame 1000 time 10000.000
Back Off! I just backed up average.pdb to ./#average.pdb.2#
Constructing covariance matrix (4266x4266) ...
Last frame 1000 time 10000.000
Read 1001 frames
Trace of the covariance matrix: 26.3911 (nm^2)
Diagonalizing ...
Sum of the eigenvalues: 26.3911 (nm^2)
Segmentation fault (core dumped)
You're probably running out of memory. Is this the latest version of GROMACS (5.0.4)? Do you get the same problem with fewer atoms or when analyzing fewer frames? The last two points will tell you whether it's a memory problem.
Dear Justin,
I have enough memory in my system. Yes it's in gromacs 5.0.4 version. It's reading all the frames (10ns) and while diagonalizing the matrix its giving the error..
The reason is unlikely due to memory issues because it is the matrix diagonalisation which requires a lot of memory. In the above example, however, the diagonalisation is finished already, the sum of eigenvalues is computed and agrees with the trace of the covariance matrix.
So the diagonalisation went through alright. Which is good.
What normally happens next is normally writing of output files. Did you verify that you have sufficient writing privileges in the working directory?
Hi Ulf,
Yes while generating output file only it's terminating.. Now I got the correct results..
Thanks to Justin & Ulf for your suggestions.
Christian Bharathi Ulf Hensen I am also facing the same problem regarding the 10NS run. So, How can you solve that issue?
Hi Sourav,
My error was with the altered trajectory (that was contained less atoms than the selected atoms) for generating covariance matrix of the protein. If you have the trajectory with complete atom of your protein then you will success, I hope.
- Badges
- Science topic