[torqueusers] Free placement.

Roy Dragseth Roy.Dragseth at cc.uit.no
Tue Feb 8 13:07:47 MST 2005


On Monday 07 February 2005 19:35, David Jackson wrote:
> Roy,
> 
>   You should be able to set the queue resources_available.nodect to a
> desired value.  See 
> 
> http://clusterresources.com/torquedocs/1.2queueconfig.shtml
> 

Nope, I've tried both queue and server limits with no success.  I have a 5 
node  cluster where I try to up the limit to 8 nodes and submit a 7 node job:

> qmgr -c 'print server'
#
# Create queues and set their attributes.
#
#
# Create and define queue default
#
create queue default
set queue default queue_type = Execution
set queue default resources_max.nodect = 8
set queue default resources_max.nodes = 8
set queue default resources_available.nodect = 8
set queue default resources_available.nodes = 8
set queue default enabled = True
set queue default started = True
#
# Set server attributes.
#
set server scheduling = True
set server acl_host_enable = False
set server managers = maui at snykov.uit.no
set server managers += root at snykov.uit.no
set server default_queue = default
set server log_events = 511
set server mail_from = adm
set server query_other_jobs = True
set server resources_available.nodect = 8
set server resources_available.nodes = 8
set server resources_max.nodect = 8
set server resources_max.nodes = 8
set server scheduler_iteration = 600
set server node_ping_rate = 300
set server node_check_rate = 600
set server tcp_timeout = 6

> qsub -lnodes=7 run.sh
qsub: Job exceeds queue resource limits

This is for torque 1.2.0p0.

r.
-- 

  The Computer Center, University of Tromsø, N-9037 TROMSØ, Norway.
	      phone:+47 77 64 41 07, fax:+47 77 64 41 00
     Roy Dragseth, High Performance Computing System Administrator
	 Direct call: +47 77 64 62 56. email: royd at cc.uit.no



More information about the torqueusers mailing list