[torqueusers] jobs stuck

Bas van der Vlies basv at sara.nl
Tue Sep 19 00:30:23 MDT 2006


Garrick Staples wrote:
> On Mon, Sep 18, 2006 at 10:48:20PM +0200, Bas van der Vlies alleged:
>> On Sep 18, 2006, at 7:43 PM, Garrick Staples wrote:
>>
>>> On Mon, Sep 18, 2006 at 10:46:56AM +0200, bill alleged:
>>>> Hello
>>>>
>>>> Come back to work on monday and I saw every jobs stucks.
>>>> CPUs are up to 0% working.
>>>> show_pbs_res.py shows me:
>>>> Total nodes : 2
>>>> 	Nodes with 4 CPU
>>>> 		1 node with -8 CPU free
>>>> 		1 node with 0 CPU free
>>> Where does show_pbs_res.py pull this information coming from?  From  
>>> the
>>> name, I can't tell if it talks to TORQUE or maui/moab.
>>>
>> It was posted as example to show how many resources there are  
>> left ;-) It get its info  from the pbs_server.
> 
> So how does it get a negative number from pbs_server?  number of cpus
> minus the number of cpus assigned to jobs from a pbs_nodestat()?
> 

Is does a pbs_statnode() and parses te node attributes:
      np = 2
      jobs = 0/227572.batch-ng.irc.sara.nl

resources left: np - len(jobs. split on ,)

Show the negative number is that er more jobs running then cpu's available.

Regards


-- 
--
********************************************************************
*                                                                  *
*  Bas van der Vlies                     e-mail: basv at sara.nl      *
*  SARA - Academic Computing Services    phone:  +31 20 592 8012   *
*  Kruislaan 415                         fax:    +31 20 6683167    *
*  1098 SJ Amsterdam                                               *
*                                                                  *
********************************************************************


More information about the torqueusers mailing list