[torqueusers] Allocating resources by CPUs

Martins, Flavio fmartins at fttinc.com
Mon Mar 12 06:08:58 MDT 2007


I am requesting nodes on the command line as part of the qsub command.
Qsub -l nodes=8 test.csh

I get the message that "not enough of the right type of nodes are
available"
Here is the complete output from qstat -f:

[/u1/cfd/user/torque]% qstat -f 2
Job Id: 2.hostname
    Job_Name = test.csh
    Job_Owner = user at hostname
    job_state = Q
    queue = batch
    server = hostname
    Checkpoint = u
    ctime = Fri Mar  9 00:34:24 2007
    Error_Path = hostname:/u1/cfd/user/torque/test.csh.e2
    Hold_Types = n
    Join_Path = n
    Keep_Files = n
    Mail_Points = a
    mtime = Fri Mar  9 00:34:24 2007
    Output_Path = hostname:/u1/cfd/user/torque/test.csh.o2
    Priority = 0
    qtime = Fri Mar  9 00:34:24 2007
    Rerunable = True
    Resource_List.ncpus = 20
    Resource_List.neednodes = 8
    Resource_List.nodect = 8
    Resource_List.nodes = 8
    Resource_List.walltime = 72:00:00
    substate = 10
    Variable_List = PBS_O_HOME=/u1/cfd/user,PBS_O_LANG=en_US.UTF-8,
        PBS_O_LOGNAME=user,
 
PBS_O_PATH=/sfw/PROD/compilers/pgi6/linux86-64/6.0/bin:/opt/torque/bin:.
/:/usr/local/bin:/bin:/usr/bin:/sfw/PROD/launch:/usr/share/pvm3/lib:/usr
/X11R6/bin:/export/home/user/bin,
        PBS_O_MAIL=/var/spool/mail/user,PBS_O_SHELL=/bin/csh,
        PBS_O_HOST=hostname,
        PBS_O_WORKDIR=/u1/cfd/user/torque,PBS_O_QUEUE=batch
    euser = user
    egroup = users
    queue_rank = 2
    queue_type = E
    comment = Not Running: Not enough of the right type of nodes are
available

    etime = Fri Mar  9 00:34:24 2007
    submit_args = -l nodes=8

Again, thanks for any help. Everything I have read so far indicates that
this should work. Maybe it's a scheduler problem? I'm running the
default PBS_sched.

Flavio Martins


>On Sat, 10 Mar 2007, Martins, Flavio wrote:

> I tried setting the resources_available_nodect parameter to 16 as
mentioned
> in previous suggestions to this problem. This allows me to submit the
job,
> but then it just sits in q status waiting for resources to become
> available.

What do you get from a qstat -f of the job ID ?

I'm assuming you are requesting nodes as:

#PBS -l nodes=7

for example ?

-- 
 Christopher Samuel - (03)9925 4751 - VPAC Deputy Systems Manager
 Victorian Partnership for Advanced Computing http://www.vpac.org/
 Bldg 91, 110 Victoria Street, Carlton South, VIC 3053, Australia

_______________________________________________
torqueusers mailing list
torqueusers at supercluster.org
http://www.supercluster.org/mailman/listinfo/torqueusers


More information about the torqueusers mailing list