[torqueusers] Problem to allocate more than one node
Albino Aveleda
bino at coc.ufrj.br
Mon Dec 17 09:13:46 MST 2007
Hi,
I have a cluster with 4 nodes with 8 cores each node.
I installed torque 2.2.1, but I can use more than one
node in same job. Please, look the smaple bellow that
I asked 2 nodes (16 cores), but torque allocate only
one node as showed in exec_host line.
What's wrong in my configuration?
[bino at adm test]$ cat test.job #PBS -l nodes=2:ppn=8
#PBS -l walltime=01:00:00
#PBS -j oe
#PBS -N test
# change directory
cd ${PBS_O_WORKDIR}
# get the number of processors and nodes
NUM_PROC=`cat ${PBS_NODEFILE} | wc -l`
# run
echo "Num Proc: $NUM_PROC"
cat ${PBS_NODEFILE}
[bino at adm test]$ qsub test.job; qstat -f 8
8.adm
Job Id: 8.adm
Job_Name = test
Job_Owner = bino at adm
job_state = R
queue = batch
server = adm
Checkpoint = u
ctime = Mon Dec 17 12:40:57 2007
Error_Path = adm:/home/users/bino/test/test.e8
exec_host = node01/7+node01/6+node01/5+node01/4+node01/3+node01/2+node01/1
+node01/0
Hold_Types = n
Join_Path = oe
Keep_Files = n
Mail_Points = a
mtime = Mon Dec 17 12:40:57 2007
Output_Path = adm:/home/users/bino/test/test.o8
Priority = 0
qtime = Mon Dec 17 12:40:57 2007
Rerunable = True
Resource_List.nodect = 2
Resource_List.nodes = 2:ppn=8
Resource_List.walltime = 01:00:00
Variable_List = PBS_O_HOME=/home/users/bino,PBS_O_LANG=en_US.UTF-8,
PBS_O_LOGNAME=bino,
PBS_O_PATH=/usr/kerberos/bin:/home/users/utils/bin:/usr/local/mpich2
/bin:/usr/local/torque/bin:/usr/local/bin:/bin:/usr/bin:/home/users/bi
no/bin,PBS_O_MAIL=/var/spool/mail/bino,PBS_O_SHELL=/bin/bash,
PBS_SERVER=adm,PBS_O_HOST=adm,PBS_O_WORKDIR=/home/users/bino/test,
PBS_O_QUEUE=batch
comment = Job started on Mon Dec 17 at 12:40
etime = Mon Dec 17 12:40:57 2007
submit_args = test.job
[bino at adm test]$ qmgr -c 'p s'
#
# Create queues and set their attributes.
#
#
# Create and define queue batch
#
create queue batch
set queue batch queue_type = Execution
set queue batch resources_default.nodes = 1
set queue batch resources_default.walltime = 01:00:00
set queue batch resources_available.nodect = 999999
set queue batch enabled = True
set queue batch started = True
#
# Set server attributes.
#
set server scheduling = True
set server managers = root at adm
set server operators = root at adm
set server default_queue = batch
set server log_events = 511
set server mail_from = pbs_adm
set server resources_available.nodect = 999999
set server scheduler_iteration = 600
set server node_check_rate = 150
set server tcp_timeout = 6
set server mom_job_sync = True
set server pbs_version = 2.2.1
set server keep_completed = 5
Best regards,
Bibo
More information about the torqueusers
mailing list