[torqueusers] np = 2?

Leroy van Logchem leroy.vanlogchem at wldelft.nl
Fri Aug 3 02:37:10 MDT 2007


Hi All,

After upgrading to dualcore cpu's we changed np=1 to np=2.
Now we're in trouble since Torque sometimes starts more then two jobs on 
one node.
The state stays 'free' while it's working on 5 jobs already..

Thanks!
-- 
Leroy

For example pbsnodes -a shows host x342:
     state = free
     np = 2
     properties = q-normal
     ntype = cluster
     jobs = 0/25843.h3, 0/25842.h3, 0/25841.h3, 0/25838.h3, 0/25778.h3
     status = opsys=linux,uname=Linux x342 2.6.9-42.0.3.ELsmp #1 SMP Fri 
Oct 6 06:21:39 CDT 2006 i686,sessions=2700 27705 27778 27792 31023 31114 
31222,nsessions=7,nusers=4,idletime=68543,totmem=7830136kb,availmem=6943448kb,physmem=4016048kb,ncpus=2,loadave=3.93,netload=1698392591,state=free,jobs=25778.h3 
25838.h3 25841.h3 25842.h3 25843.h3,rectime=1186129220

--- Qmgr: print server
#
# Create queues and set their attributes.
# Create and define queue normal
#
create queue normal
set queue normal queue_type = Execution
set queue normal resources_max.nodes = 1
set queue normal resources_min.nodes = 1
set queue normal resources_default.neednodes = 1
set queue normal resources_default.nodes = 1
set queue normal enabled = True
set queue normal started = True
#
# Set server attributes.
#
set server scheduling = True
set server managers += logch_l@*
set server default_queue = normal
set server log_events = 511
set server mail_from = torque
set server query_other_jobs = True
set server scheduler_iteration = 60
set server node_ping_rate = 30
set server node_check_rate = 60
set server tcp_timeout = 6
set server node_pack = False
set server job_stat_rate = 30
set server pbs_version = 2.1.0p0

And for completeness:

--- Qmgr: print node x342
# Create and define node x342
set node x342 state = free
set node x342 properties = q-normal
set node x342 ntype = cluster
set node x342 status = opsys=linux
set node x342 status += uname=Linux x342 2.6.9-42.0.3.ELsmp #1 SMP Fri 
Oct 6 06:21:39 CDT 2006 i686
set node x342 status += sessions=2700 27705 27778 27792 31023 31114 31222
set node x342 status += nsessions=7
set node x342 status += nusers=4
set node x342 status += idletime=69417
set node x342 status += totmem=7830136kb
set node x342 status += availmem=6942504kb
set node x342 status += physmem=4016048kb
set node x342 status += ncpus=2
set node x342 status += loadave=3.77
set node x342 status += netload=4237953047
set node x342 status += state=free
set node x342 status += jobs=25778.h3 25838.h3 25841.h3 25842.h3 25843.h3
set node x342 status += rectime=1186130094


More information about the torqueusers mailing list