Re: [AMBER] AMBER error slurmstepd

From: Adrian Roitberg via AMBER <amber.ambermd.org>
Date: Mon, 18 Sep 2023 09:12:12 -0400

There has to be something else, hidden somewhere. Basically, it seems
that you tried the time limit, and you report back that the error is
still there, within maybe one hour.

I suggest the following.

Resend your job with a VERY small nstlim, maybe 100. If that does not
run, then you need to talk to the system administrators, because it
should run very quickly.

If it runs, increase nstlim and see where it "breaks". That will give
you some idea of real runtime versus what the queue allows you to do.

Adrian


On 9/18/23 8:45 AM, Carlos Simmerling via AMBER wrote:
> [External Email]
>
> It's possible that the queue itself has a limit that you can't change. That
> isn't an amber issue, it's something to discuss with the cluster
> administrator. If it is indeed a time limit, you will need to either adjust
> your simulation length, or restart from where it ended.
>
> On Mon, Sep 18, 2023, 8:09 AM SATYAJIT KHATUA via AMBER <amber.ambermd.org>
> wrote:
>
>> Your log file says "slurmstepd: error: *** JOB 1871970 ON cn6015 CANCELLED
>> AT 2023-09-15T23:08:05 DUE TO TIME LIMIT ***". As it is showing time limit,
>> then this should be the solution in my opinion.
>>
>> Satyajit
>> ________________________________
>> From: VERONICA MARTIN HERNANDEZ <veronica.martin.hernandez.uva.es>
>> Sent: 18 September 2023 15:04
>> To: SATYAJIT KHATUA <satyajitkhatua09.gmail.com>; AMBER Mailing List <
>> amber.ambermd.org>
>> Subject: RE: AMBER error slurmstepd
>>
>> Thanks
>>
>> I've tried your solution but the error remains. Is there another
>> posibility?
>>
>> Thanks
>> ________________________________
>> De: SATYAJIT KHATUA <satyajitkhatua09.gmail.com>
>> Enviado: lunes, 18 de septiembre de 2023 10:29
>> Para: VERONICA MARTIN HERNANDEZ <veronica.martin.hernandez.uva.es>; AMBER
>> Mailing List <amber.ambermd.org>
>> Asunto: Re: AMBER error slurmstepd
>>
>> Hi,
>>
>> So, you are using SLURM script. In you script try to increase the time
>> limit (given in -t option). It is because of the time constraints.
>> If -t option is not there try to add something like
>>
>> #SBATCH -t 72:00:00 (if max limit is 72 hrs)
>>
>> Best,
>> Satyajit
>> ________________________________
>> From: VERONICA MARTIN HERNANDEZ via AMBER <amber.ambermd.org>
>> Sent: 18 September 2023 13:15
>> To: AMBER Mailing List <amber.ambermd.org>
>> Subject: [AMBER] AMBER error slurmstepd
>>
>> I have recently installed AMBER in a new cluster and I'm trying this
>> tutorial: https://ambermd.org/tutorials/advanced/tutorial3/section1.php<
>> https://urldefense.com/v3/__https://ambermd.org/tutorials/advanced/tutorial3/section1.php__;!!D9dNQwwGXtA!X_KYwoFq-3u7IeFSln2zCeBDQGhLlF9i0IKESFHQ_-I3jPqM7BoC_vDtPQmshVXLh52dSEXNpVoGh7G3GFRXq_BM6htG8bXR0ug3uA$>
>> and in the equilibration step I use this command: srun pmemd.cuda -O -i
>> equil.in -o ras-raf-equil.out -p ras-raf-solv.prmtop -c
>> ras-raf-density.rst -r ras-raf-equil.rst -x ras-raf-equil.mdcrd -ref
>> ras-raf-density.rst
>> The program does not finish because it doesn't print the and I get the
>> following error:
>> slurmstepd: error: *** JOB 1871970 ON cn6015 CANCELLED AT
>> 2023-09-15T23:08:05 DUE TO TIME LIMIT ***
>> In other clusters I don't have this error, It's the first time that I see
>> it
>> How can I fix it?
>>
>> _______________________________________________
>> AMBER mailing list
>> AMBER.ambermd.org
>> http://lists.ambermd.org/mailman/listinfo/amber<
>> https://urldefense.com/v3/__http://lists.ambermd.org/mailman/listinfo/amber__;!!D9dNQwwGXtA!X_KYwoFq-3u7IeFSln2zCeBDQGhLlF9i0IKESFHQ_-I3jPqM7BoC_vDtPQmshVXLh52dSEXNpVoGh7G3GFRXq_BM6htG8bV3M7Y0ug$
>> _______________________________________________
>> AMBER mailing list
>> AMBER.ambermd.org
>> http://lists.ambermd.org/mailman/listinfo/amber
>>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber

-- 
Dr. Adrian E. Roitberg
V.T. and Louise Jackson Professor in Chemistry
Department of Chemistry
University of Florida
roitberg.ufl.edu
352-392-6972
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Sep 18 2023 - 06:30:02 PDT
Custom Search