VMD-L Mailing List
From: Mayne, Christopher G (cmayne2_at_illinois.edu)
Date: Tue Feb 02 2016 - 09:23:36 CST
OK, so we’ve figured out the cause of the error, so that’s a start. The way that the parsers are currently setup, there’s no built-in way to deal with restarts based on wall clock timeouts. I don’t know what the header of the restart LOGs look like, so it’s hard for me to design a work around without that information. Could you send me those the LOGs, PSF, and PDB files directly (no need to cc VMD-L on that email). I’m wondering if you can just concatenate them.
On Feb 1, 2016, at 8:07 PM, Soumya Lipsa Rath <soumyalipsabt_at_gmail.com<mailto:soumyalipsabt_at_gmail.com>> wrote:
Dear Dr. Mayne,
Thanks! that worked. But again what do I do about the other log files. Say if I scan the dihedral for 90 degrees (18 steps with 5 degrees each), in the first time it runs 10 steps and terminates due to wall time error, then I restart from the checkpoint file and get another log file. Should I ignore the rest of the log files?
On Tue, Feb 2, 2016 at 10:58 AM, Mayne, Christopher G <cmayne2_at_illinois.edu<mailto:cmayne2_at_illinois.edu>> wrote:
Without seeing the log files, I would assume that #1 is the problem. The ffTK parser reads some information from the beginning of the LOG files to define the dihedral that is being scanned. My guess is that this data isn’t replicated in the LOG files that result from a restart. Try loading in the first LOG file.
On Feb 1, 2016, at 6:19 PM, Soumya Lipsa Rath <soumyalipsabt_at_gmail.com<mailto:soumyalipsabt_at_gmail.com>> wrote:
Dear Dr. Mayne,
Thanks for your reply.
1) I am passing on the log files from gaussian. Although, during the calculation, I had to restart gaussian 2-3 times, and I am loading the last log file. Is that the problem? If so, could you please tell me how to rectify the same.
2) I am using vmd version 1.9.2
3) Yes, the gaussian terminated normally. They also ran few steps before terminating.
On Tue, Feb 2, 2016 at 3:34 AM, Mayne, Christopher G <cmayne2_at_illinois.edu<mailto:cmayne2_at_illinois.edu>> wrote:
Something is amiss with the parser. There’s not much information go on here, so I have a few questions:
1) Make sure that you’re passing it the LOG files output from the Gaussian calcaulation.
2) What version of VMD are you using?
3) Did you check the Gaussian LOG files to ensure that they ran? (they don’t necessarily need to complete without error, but they need to make it a few steps)
On Feb 1, 2016, at 12:27 AM, Soumya Lipsa Rath <soumyalipsabt_at_gmail.com<mailto:soumyalipsabt_at_gmail.com>> wrote:
I am using the FFTK toolkit of vmd for developing CHARMM compatible parameters. During the "Dihedrals I: Generating Dihedral Scan Target Data" step, when I am trying to load the log files, I am getting the following error.
can't read "scanDihDef": no such variable. I am using Gaussian 09 for running the scans. Please help me figure out the problem.