[Mauiusers] maui segfaults trying to schedule a job

DuChene, StevenX A stevenx.a.duchene at intel.com
Tue Nov 29 12:13:51 MST 2011


Did you also increase MMAX_REQ_PER_JOB as well?
--
Steven DuChene

-----Original Message-----
From: Michel Béland [mailto:michel.beland at calculquebec.ca] 
Sent: Tuesday, November 29, 2011 11:08 AM
To: DuChene, StevenX A
Cc: mauiusers at supercluster.org
Subject: Re: [Mauiusers] maui segfaults trying to schedule a job

Steven DuChene wrote:
> No, the systems in this cluster only have 4 cores per node.
>
> BTW, in case this has some bearing on the problem here are the things I have set in my maui.cfg file:
>
>
> The job the user is submitting requests the following:
>
> #PBS -l nodes=39:SeaMicro:diskgrp1+39:SeaMicro:diskgrp2+39:SeaMicro:diskgrp3+39:SeaMicro:diskgrp4+16:SeaMicro:diskgrp6
>   

That's it, you have five node requests with properties and the maximum 
is four. In your later message you wrote that you changed 
MAX_MREQ_PER_JOB from 4 to 8. This is what solved the problem. On our 
cluster, I bumped MAX_MREQ_PER_JOB to 64, but still I think that anybody 
with malicious intentions could write a dummy job that would kill Maui 
all the time.

-- 
Michel Béland, analyste en calcul scientifique
michel.beland at calculquebec.ca
bureau S-250, pavillon Roger-Gaudry (principal), Université de Montréal
téléphone : 514 343-6111 poste 3892     télécopieur : 514 343-2155
Calcul Québec (www.calculquebec.ca)
Calcul Canada (calculcanada.org)



More information about the mauiusers mailing list