[Mauiusers] Questions regarding "tasks" and the cfg file.

Michael Homa mhoma at uic.edu
Wed Sep 17 13:03:06 MDT 2008


> out of curiosity, if you remove the lines you singled out in your first
email from your maui.cfg (nodecfg[argo17-1] and classcfg[dedicated]) which
you think may be causing the problem, does the system then work?

Same result. I submitted a new job, 318:

318  mhoma    dedicate hello_worl    --      1   6    --  00:30 Q   -- --

job cannot run in partition DEFAULT (idle procs do not meet requirements :
0 of 6 procs found)
idle procs:  28  feasible procs:   0

Rejection Reasons: [CPU          :    1][HostList     :    6]

Detailed Node Availability Information:

argo17-1                 rejected : CPU
argo17-2                 rejected : HostList
argo17-3                 rejected : HostList
argo17-4                 rejected : HostList
argo18-1                 rejected : HostList
argo18-2                 rejected : HostList
argo18-3                 rejected : HostList

>
> --Joe
>
> ________________________________
>
> From: Michael Homa [mailto:mhoma at uic.edu]
> Sent: Wed 9/17/2008 1:46 PM
> To: Greenseid, Joseph M.
> Cc: mauiusers at supercluster.org
> Subject: RE: [Mauiusers] Questions regarding "tasks" and the cfg file.
>
>
>
> On Wed, 17 Sep 2008, Greenseid, Joseph M. wrote:
>
> > what batch queue are you using?
>
> I have six queues defined in torque: one route queue called batch and five
> execution queues (two of the queues are staff/faculty and the other three
> are only for students). The job is being submitted directly to "dedicated"
> which is one of the two staff/faculty queues. By the way, my id is in the
> acl_user entry for the queue dedicated. Here's the script
>
> #PBS -N hello_world
> #PBS -q dedicated   <------------
> #PBS -l nodes=argo17-1
> /home/homes51/mhoma/a.out
>
> > are the nodes showing up as available to your queue?
>
> Queue            Memory CPU Time Walltime Node  Run Que Lm  State
> ---------------- ------ -------- -------- ----  --- --- --  -----
> batch              --      --       --      --    0   0 --   E R
> staff              --      --    720:00:0    12   0   1 --   E R
> student_long       --      --    240:00:0     4   0   0 16   E R
> student_short      --      --    04:00:00     4   0   0 10   E R
> dedicated          --      --    00:30:00     3   0   3  1   E R  <-----
> student_medium     --      --    24:00:00     6   0   0 10   E R
>
> > for example, if you're using torque, what is the output of the command
> > `pbsnodes -a`?
>
> argo17-1
>      state = free
>      np = 4
>      properties = Linux2.i86pc,dualcore,amd,smp,dedicated <------
> argo18-2                                        ---------
>      state = free
>      np = 4
>      properties = Linux2.i86pc,dualcore,amd,smp,dedicated <------
> argo18-3                                        ---------
>      state = free
>      np = 4
>      properties = Linux2.i86pc,dualcore,amd,smp,dedicated <------
>                                                 ---------
> list queue dedicated
> Queue dedicated
>         queue_type = Execution
>         max_queuable = 48
>         max_user_queuable = 24
>         total_jobs = 3
>         state_count = Transit:0 Queued:3 Held:0 Waiting:0 Running:0 Exiting:0
>         max_running = 1
>         resources_max.ncpus = 6
>         resources_max.nodect = 3
>         resources_max.walltime = 00:30:00
>         resources_default.neednodes = dedicated   <-------------
>
> As you can see, the property dedicated associated with the three nodes is
> associated with the queue dedicated.
>
> Michael
>
>
>
>
>


More information about the mauiusers mailing list