mdrun runs infinitely when checkpoint file is beyond the designated end point - Redmine #2717
Archive from user: zhiyi wu
Say I have a tpr file which I say in the mdp file for it to runs for 50
ns and I stopped the simulation at 31 ns.
If I then use convert-tpr to change the end time to 30 ns and use mdrun
to continues the simulation,
The following message will be generated and the simulation runs forever.
WARNING: This run will generate roughly 4827577325564461056 Mb of data
starting mdrun ‘GkApcT YneM POPE POPG in water’
15000000 steps, infinite ps (continuing from step 15083300, 30166.6 ps).
It will be nice if the mdrun checks if the current time is longer than the end time and either write gro file and quit or quit with a non-zero exit state.
(from redmine: issue id 2717, created on 2018-10-29 by gmxdefault, closed on 2018-11-15)
- Relations:
- relates #1781 (closed)
- relates #2569 (closed)
- relates #2757 (closed)
- Changesets:
- Revision 4dcb2a1a by Paul Bauer on 2018-11-08T21:25:17Z:
Issue fatal error if checkpoint does not suit the .tpr
If the step number in the checkpoint is out of range for that
described in the .tpr, then the user has provided mismatching
inputs. We do not intend then to be able to address this with the
mdrun -nsteps option.
Fixes #2717
Change-Id: I827bdc1b92ee69bf6287e2fd552ace7583b62028
- Revision 61285613 by Mark Abraham on 2018-11-15T16:37:21Z:
Fix checkpoint restart of tpr with infinite step count
The recent fix of #2717 did not account for the way a user's .tpr file
can require an infinite number of steps by using the special value
-1. Such special values are difficult to remember when maintaining the
code, so we should tend to avoid introducing them.
Fixes #2757
Change-Id: I6570c4f4e7d63b2375dbb595a514c9e709f18856