Re: namd-l digest V1 #2289

From: Abhishek TYAGI (atyagiaa_at_connect.ust.hk)
Date: Wed Oct 25 2017 - 05:42:24 CDT

please attach your log file

thanks



Abhishek Tyagi,PhD

Chemical and Biological Engineering

Hong Kong University of Science and Technology

Clear Water Bay, Hong Kong


________________________________
From: Zahra Mohammadi <mohammadizahra94_at_gmail.com>
Sent: Wednesday, October 25, 2017 3:58:36 PM
To: namd-l_at_ks.uiuc.edu; owner-namd-l_at_ks.uiuc.edu
Cc: namd-l-digest_at_ks.uiuc.edu
Subject: Re: namd-l digest V1 #2289

Hi everyone,

And thank you

Abhi for replying me. I did the "

+idlepoll" in my command line. The command line is:

/home/.../NAMD_2.11_Linux-x86_64-multicore-CUDA/charmrun ++local +p24 /home/.../NAMD_2.11_Linux-x86_64-multicore-CUDA/namd2 +idlepoll File.conf > File.log

1- Another question is the syntax of above command, is it correct? Because you said add the "

+idlepoll" next to charmrun, while in mine it is written next to namd2. Would it make any difference?

2- Also the problem still exist, so any advice, suggestion or solution by anyone is deeply appreciated.

Thanks,
Zahra





On Wed, Oct 25, 2017 at 9:37 AM, namd-l digest <owner-namd-l-digest_at_ks.uiuc.edu<mailto:owner-namd-l-digest_at_ks.uiuc.edu>> wrote:

namd-l digest Wednesday, October 25 2017 Volume 01 : Number 2289



In this issue:

    Re: namd-l: FATAL ERROR: cuda_check_progress polled 1000000 times over ...

----------------------------------------------------------------------

Date: Tue, 24 Oct 2017 11:42:32 +0000
From: Abhishek TYAGI <atyagiaa_at_connect.ust.hk<mailto:atyagiaa_at_connect.ust.hk>>
Subject: Re: namd-l: FATAL ERROR: cuda_check_progress polled 1000000 times over ...

Hi,

It seems to me that on hpc you need to add +idlepoll next to charmrun

Best
Abhi

Get Outlook for Android<https://aka.ms/ghei36>





________________________________
From: owner-namd-l_at_ks.uiuc.edu<mailto:owner-namd-l_at_ks.uiuc.edu> <owner-namd-l_at_ks.uiuc.edu<mailto:owner-namd-l_at_ks.uiuc.edu>> on behalf of Zahra
Mohammadi <mohammadizahra94_at_gmail.com<mailto:mohammadizahra94_at_gmail.com>>
Sent: Tuesday, October 24, 2017 7:12:55 PM
To: namd-l_at_ks.uiuc.edu<mailto:namd-l_at_ks.uiuc.edu>
Subject: namd-l: FATAL ERROR: cuda_check_progress polled 1000000 times over
...

Hi eveyone,
I was wondering if anybody who knows about my problem or has any related
experiences, help me with my problem.

I am runnig a common MD simulation of a protein in a water box. My simulation
is running on High Performance Computer Cluster (HPCC). After the minimization
process (3000 steps) when it ends my problem occurs which is at the beginning
of MD simulation process. I get this FATAL error which is below:

Charm++ fatal error:
FATAL ERROR: cuda_check_progress polled 1000000 times over 100.825342 s on
step 1181

I searched and found nothing about it to solve my problem. In advance, I
appreciate all of the help.

Thanks
Zahra

------------------------------

End of namd-l digest V1 #2289
*****************************


This archive was generated by hypermail 2.1.6 : Sun Dec 31 2017 - 23:21:44 CST