PBS

Joey Raheb joeyraheb at usa.net
Fri Jun 8 10:49:41 PDT 2001


Hello,

I was hoping that maybe someone on this mailing list would know how to do
this.  I am trying to create a queue system based on the following:  1 route
queue and 2 execute queues.  I plan to have everyone submit jobs to the route
queue (ROCIT) and have the route queue submit the jobs to the execute queues
(switch1,switch2).  I have done the following in PBS:

#
# Create queues and set their attributes.
#
#
# Create and define queue ROCIT
#
create queue ROCIT
set queue ROCIT queue_type = Route
set queue ROCIT acl_host_enable = False
set queue ROCIT acl_hosts = *.rocit.cluster
set queue ROCIT route_destinations = switch1
set queue ROCIT route_destinations += switch2
set queue ROCIT enabled = True
set queue ROCIT started = True
#
# Create and define queue switch1
#
create queue switch1
set queue switch1 queue_type = Execution
set queue switch1 acl_host_enable = True
set queue switch1 acl_hosts = n3
set queue switch1 acl_hosts += n2
set queue switch1 enabled = True
set queue switch1 started = True
#
# Create and define queue switch2
#
create queue switch2
set queue switch2 queue_type = Execution
set queue switch2 acl_host_enable = True
set queue switch2 acl_hosts = n5
set queue switch2 acl_hosts += n22
set queue switch2 enabled = True
set queue switch2 started = True
#
# Set server attributes.
#
set server scheduling = True
set server acl_host_enable = True
set server acl_hosts = *.rocit.cluster
set server acl_roots = root
set server managers = jraheb@*.rocit.cluster
set server default_queue = ROCIT
set server log_events = 511
set server mail_from = adm
set server query_other_jobs = True
set server resources_default.nodect = 1
set server resources_default.nodes = 1
set server scheduler_iteration = 600
set server default_node = n2
set server node_pack = False

Everything above looks right to me, however, when I try to submit a job to the
route queue it tells me that my job was rejected by all possible destinations.
 This leads me to believe that the settings in the nodes that are given to
each execute queue are not correct.  I have tried giving the execution queues
the nicknames (n1,n2,etc..) and the full hostname (n1.rocit.cluster,...) but
neither way works.  Anyone know whats going on?? Please let me know...

Joey

____________________________________________________________________
Get free email and a permanent address at http://www.amexmail.com/?A=1




More information about the Beowulf mailing list