[torqueusers] Max nodes seems to be 9 despite config

mhuhtala at abo.fi mhuhtala at abo.fi
Tue Dec 28 01:47:35 MST 2004


I have Torque managing 14 dual-processor nodes. The maximum number of
nodes for parallel jobs is set to 28 (the number of processors), both in
the queue config and server attributes. However, jobs requesting
anything more then 9 nodes are rejected with the error 'qsub: Job
exceeds queue resource limits'. Requesting either nodes=9:ppn=2 or just
nodes=9 works, but nodes=10 (regardless of ppn) or more does not. The
pbs_server log shows

12/28/2004 10:46:50;0080;PBS_Server;Req;req_reject;Reject reply
code=15036(Job exceeds queue resource limits), aux=0, type=1, from
mhuhtala at volvox.abo.fi

I am mystified. As far as I can see there should be nothing that limits
nodes to just 9. Is there something about queue config that I have not
understood? Config the queue in question and server attributes are below.

Mikko


-- qmgr print server -----------------------------------------------

#
# Create and define queue parallel
#
create queue parallel
set queue parallel queue_type = Execution
set queue parallel Priority = 100
set queue parallel max_running = 14
set queue parallel resources_max.cput = 25920:00:00
set queue parallel resources_max.nodes = 28
set queue parallel resources_min.cput = 00:30:00
set queue parallel resources_min.nodes = 2
set queue parallel resources_default.cput = 25920:00:00
set queue parallel resources_default.nodes = 2
set queue parallel enabled = True
set queue parallel started = True
#
# Set server attributes.
#
set server scheduling = True
set server max_user_run = 28
set server acl_host_enable = True
set server acl_hosts = volvox.abo.fi
set server acl_hosts += *.molmol.fi
set server default_queue = serial
set server log_events = 63
set server mail_from = mhuhtala
set server query_other_jobs = True
set server resources_default.cput = 4320:00:00
set server resources_default.neednodes = 1
set server resources_default.nodect = 1
set server resources_default.nodes = 1
set server resources_max.nodes = 28
set server scheduler_iteration = 60
set server node_ping_rate = 300
set server node_check_rate = 600
set server tcp_timeout = 6
set server default_node = 1


More information about the torqueusers mailing list