Dear Everyone,
When I run the MD simulation using VASP, there always interrupts at hundred MD steps. You can see the error as follows:
-------------------------------------------------------
Primary job terminated normally, but 1 process returned
a non-zero exit code. Per user-direction, the job has been aborted.
-------------------------------------------------------
Could anybody know any possible reasons to induce this issue, such as the incorrect installation of vasp or some problems in the supercomputer?
Thanks ALL
VASP interrupt when run MD simulation
Moderators: Global Moderator, Moderator
-
- Administrator
- Posts: 282
- Joined: Mon Sep 24, 2018 9:39 am
Re: VASP interrupt when run MD simulation
Please upload a full error report and read the forum posting guide lines
-
- Newbie
- Posts: 11
- Joined: Sun Nov 17, 2019 4:55 am
Re: VASP interrupt when run MD simulation
Dear Sir,
Due to the OUTCAR file is too large, I deleted the middle part of the simulation. Please kindly find the error report files in attached.
Thanks so much for your help.
Due to the OUTCAR file is too large, I deleted the middle part of the simulation. Please kindly find the error report files in attached.
Thanks so much for your help.
You do not have the required permissions to view the files attached to this post.
-
- Global Moderator
- Posts: 505
- Joined: Mon Nov 04, 2019 12:41 pm
- Contact:
Re: VASP interrupt when run MD simulation
Hi, to reduce the size of your OUTCAR you can use the NWRITE variable:
wiki/index.php/NWRITE
For relatively long MD runs you might want to use NWRITE=0.
With the information you provide its hard to exactly pin-point where the issue might come from.
Here are a couple of hints:
1. the job might be killed by due to walltime limit. How long does the job run for?
2. the job is hitting some kind of memory limit.
Recently we've found a memory leak that occurs in MD runs when NWRITE>=1. This issue has been fixed in VASP6
Let me know if any of these points helps you.
wiki/index.php/NWRITE
For relatively long MD runs you might want to use NWRITE=0.
With the information you provide its hard to exactly pin-point where the issue might come from.
Here are a couple of hints:
1. the job might be killed by due to walltime limit. How long does the job run for?
2. the job is hitting some kind of memory limit.
Recently we've found a memory leak that occurs in MD runs when NWRITE>=1. This issue has been fixed in VASP6
Let me know if any of these points helps you.