[GE users] Queue Resource Allocation

Daniel Templeton Dan.Templeton at Sun.COM
Tue Jul 24 15:52:47 BST 2007


    [ The following text is in the "iso-8859-1" character set. ]
    [ Your display is set for the "ISO-8859-10" character set.  ]
    [ Some special characters may be displayed incorrectly. ]

Sathish,

Yes, the details of why your jobs were left pending, even though there 
was a free job slot, are now lost.  You will have to wait until it 
happens again before you can diagnose it.

My comment about changing the default was intended for the rest of the 
engineering team.  It means that I think the current default setting for 
job_load_adjustments in the scheduler configuration does more harm than 
good.  It should be changed to NONE so that people don't have mysterious 
problems with free-but-unschedulable job slots.

Daniel

Chandran, Sathish wrote:
> Daniel,
>
> Cool. I agree with you. The results for the scheduler overloaded may not
> appear all at once.
>
> 1. Do you confirming that the details for the jobs waited although the
> resources are free will not be available any more.
>
> Why I'm throwing this is bcoz if I can get the details for the delay in
> the scheduler so that I can make some configurations and make my grid
> perfectly running all time.
>
> 2. Daniel, in your comment please elaborate the " We really 
> should change the default to NONE...". What it means???
>
> Really I'm hanging with the above said problems. Finding someone for
> their valuable considerations in this.
>
>
>
> Thanks,
> Sathish.C
>
>
> -----Original Message-----
> From: Dan.Templeton at Sun.COM [mailto:Dan.Templeton at Sun.COM] 
> Sent: Tuesday, July 24, 2007 8:01 PM
> To: users at gridengine.sunsource.net
> Subject: Re: [GE users] Queue Resource Allocation
>
> Sathish,
>
> You can only get scheduling information for jobs that are still 
> running.  Once a job has ended, you can get information about it with 
> qacct -j <job_id>, but it does not include any information about why it 
> was or wasn't scheduled immediately.
>
> Could the problem be the job_load_adjustments?  That can sometimes make 
> a host appear overloaded to the scheduler when it isn't.  We really 
> should change the default to NONE...
>
> Daniel
>
> Chandran, Sathish wrote:
>   
>> Rob,
>>
>>  
>>
>> Really that sounds great!!!
>>
>>  
>>
>> Here the job-2738 by now is completed its execution. But still I need 
>> to know why the job was in queue.
>>
>>  
>>
>> Any commands or files/folders to explore the completed job that was in
>>     
>
>   
>> queue. So that I can get a clear idea for what resource the job was 
>> waiting.
>>
>>  
>>
>> Awaiting for your warm support.
>>
>>  
>>
>> **Thanks,***
>> **Sathish.C***
>>
>>
>>     
> ------------------------------------------------------------------------
>   
>> *From:* Robert Griffiths [mailto:Robert.Griffiths at int.sc.mufg.jp]
>> *Sent:* Tuesday, July 24, 2007 4:50 PM
>> *To:* users at gridengine.sunsource.net
>> *Subject:* RE: [GE users] Queue Resource Allocation
>>
>>  
>>
>> Try qstat -j <jobid> which, in your printout below, appears to be 2738
>>
>>  
>>
>> This should spit out a whole heap of information for you about job 
>> 2378 and you should be able to tell (possibly from the hard 
>> resource_list or other) what the problem is.
>>
>>  
>>
>> Hope that helps.
>>
>>  
>>
>> Rob
>>
>>  
>>
>> **Dr. Robert K. Griffiths**
>>
>> **Associate Director**
>>
>> **Exotic Derivatives Group**
>>
>> **Robert.Griffiths at int.sc.mufg.jp 
>> <mailto:Robert.Griffiths at int.sc.mufg.jp>**
>>
>>     -----Original Message-----
>>     *From:* Chandran, Sathish [mailto:Sathish.Chandran at Honeywell.com]
>>     *Sent:* 24 July 2007 11:05
>>     *To:* list
>>     *Subject:* [GE users] Queue Resource Allocation
>>
>>     Hi,
>>
>>      
>>
>>     My Grid have a problem about pending queues, please see below, 2
>>     nodes have resource, yet.
>>
>>      
>>
>>     But the job in pending queue can not be assigned to them.
>>
>>      
>>
>>     Could you please solve this problem?
>>
>>      
>>
>>      
>>
>>      
>>
>>     And may I know the cause of the problem in detail.
>>
>>      
>>
>>     Your reply appreciated.
>>
>>      
>>
>>     **Thanks,***
>>     **Sathish.C***
>>
>>      
>>
>>
>>
>> **********************************************************
>> Mitsubishi UFJ Securities International plc ("MUSI") is registered in 
>> England, company number 1698498 and registered office at 6 Broadgate, 
>> London EC2M 2AA. MUSI is part of the Mitsubishi UFJ Financial Group 
>> and is authorised and regulated by The Financial Services Authority. 
>> This message is intended solely for the individual addressee named 
>> above. The information contained in this e-mail is confidential and 
>> may be legally privileged. If you are not the intended recipient 
>> please delete in its entirety. Messages sent via this medium may be 
>> subject to delays, non-delivery and unauthorised alteration. The 
>> information contained herein or attached hereto has been obtained from
>>     
>
>   
>> sources we believe to be reliable but we do not represent that it is 
>> accurate or complete. Any reference to past performance should not be 
>> taken as an indication of future performance.
>>
>> The information contained herein or attached hereto is not to be 
>> construed as an offer or solicitation to buy or sell any security, 
>> instrument or investment. MUSI or any affiliated company, may have an 
>> interest, position, or effect transactions, in any investment 
>> mentioned herein. Any opinions or recommendations expressed herein are
>>     
>
>   
>> solely those of the author or analyst and are subject to change 
>> without notice.
>>
>>     
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe at gridengine.sunsource.net
> For additional commands, e-mail: users-help at gridengine.sunsource.net
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe at gridengine.sunsource.net
> For additional commands, e-mail: users-help at gridengine.sunsource.net
>
>   

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe at gridengine.sunsource.net
For additional commands, e-mail: users-help at gridengine.sunsource.net




More information about the gridengine-users mailing list