[torqueusers] beginning with torqueu

Standa Kunc standa.kunc at gmail.com
Thu Oct 19 16:32:32 MDT 2006


Thank you. Now I am one step further but now I have different problem.
I cannot submit job via qsub. This is the error message:

root at pc-xubuntu:~/torque-2.1.3# echo "sleep 30" | qsub
qsub: Bad UID for job execution

My current configuration is:

xubuntu distribution, user root is enabled


I modified torque.setup to this:
--

root at pc-xubuntu:~/torque-2.1.3# cat torque.setup
#!/bin/sh
# torque.setup

# USAGE:  torque.setup <USERNAME>

if [ "$1" = "" ] ; then
  echo "USAGE:  torque.setup <USERNAME>"
  exit 1
  fi

# create default queue
# enable operator privileges

#USER=$1@`hostname`
USER=$1 at localhost

echo "initializing TORQUE (admin: $USER)"

pbs_server -t create

qmgr -c "set server scheduling=true"

echo set server operators += $USER | qmgr
echo set server operators += sk at localhost | qmgr
echo set server managers += $USER | qmgr
echo set server managers += sk at localhost | qmgr
echo set server submit_hosts = localhost | qmgr

qmgr -c 'create queue batch'
qmgr -c 'set queue batch queue_type = execution'
qmgr -c 'set queue batch started = true'
qmgr -c 'set queue batch enabled = true'
qmgr -c 'set queue batch resources_default.walltime = 1:00:00'
qmgr -c 'set queue batch resources_default.nodes = 1'

qmgr -c 'set server default_queue = batch'


This is result of running torque.setup:
--

root at pc-xubuntu:~/torque-2.1.3# ./torque.setup root
initializing TORQUE (admin: root at localhost)
PBS_Server pc-xubuntu.local: Create mode and server database exists,
do you wish to continue y/(n)?y
Max open servers: 4
Max open servers: 4
Max open servers: 4
Max open servers: 4
Max open servers: 4


Server settings:
--
root at pc-xubuntu:~/torque-2.1.3# 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 enabled = True
set queue batch started = True
#
# Set server attributes.
#
set server scheduling = True
set server managers = root at localhost
set server managers += sk at localhost
set server operators = root at localhost
set server operators += sk at localhost
set server default_queue = batch
set server log_events = 511
set server mail_from = adm
set server scheduler_iteration = 600
set server node_check_rate = 150
set server tcp_timeout = 6
set server pbs_version = 2.1.3
set server submit_hosts = localhost



Queue is set up properly:
--

root at pc-xubuntu:~/torque-2.1.3# qstat -q

server: pc-xubuntu

Queue            Memory CPU Time Walltime Node  Run Que Lm  State
---------------- ------ -------- -------- ----  --- --- --  -----
batch              --      --       --      --    0   0 --   E R
                                               ----- -----
                                                   0     0

Compute node is replying properly:
--

root at pc-xubuntu:~/torque-2.1.3# pbsnodes -a
localhost
     state = free
     np = 1
     ntype = cluster
     status = opsys=linux,uname=Linux pc-xubuntu 2.6.16-xen #1 SMP Thu
Apr 13 18:46:07 BST 2006 i686,sessions=2949 3005 3007 3298 3339 3351
3356 3380 3464,nsessions=9,nusers=3,idletime=0,totmem=1641312kb,availmem=1552080kb,physmem=131244kb,ncpus=1,loadave=0.03,netload=0,state=free,jobs=?
0,rectime=1161291975

There is trust between server and compute node:
--
root at pc-xubuntu:~/torque-2.1.3# cat /var/spool/torque/mom_priv/config
$pbsserver localhost

root at pc-xubuntu:~/torque-2.1.3# cat /var/spool/torque/localhost
localhost


Node is defined:
--
root at pc-xubuntu:~/torque-2.1.3# cat /var/spool/torque/server_priv/nodes
localhost


But submiting job via qsub is not running:
--
root at pc-xubuntu:~/torque-2.1.3# echo "sleep 30" | qsub
qsub: Bad UID for job execution


User sk at localhost is also server operator and manager but this user
does not help:
--
pbs_iff: file not setuid root, likely misconfigured
pbs_iff: cannot connect to pc-xubuntu:15001 - fatal error, errno=13
(Permission denied)
No Permission.
qsub: cannot connect to server pc-xubuntu (errno=15007)


Do I need installed sshd?
Do I need configured .rhost file?
Is it possible to submit job via qsub as root?

You have said that you use one pc configuration too. Would you be so
kind and provide me some short version of your configuration process?
I mean some checklist which I can follow. Probably settings for just
one computer can be easier than setting for network.

And yes, I have tried to search resolution in previous mails. I have
tried google too.

Thank you very much
S. Kunc










On 19/10/06, Glen Beane <glen.beane+torque at gmail.com> wrote:
> On 10/18/06, Standa Kunc <standa.kunc at gmail.com> wrote:
> > Hello,
> >
> > I am beginning with Torqueu and I would like to test it. Just simple
> > test on one PC (compute node will be same computer as server).
> >
> > I have installed Torque 2.1.3 successfully but next phase called L.2
> > Initialize/Configure TORQUE on the Server (pbs_server) in QuickStart
> > Manual is not working for me.
> >
> > Problem is that running command ./torqueu.setup root results in error messages.
> >
> > root at pc-xubuntu:~/torque-2.1.3# ./torque.setup root
> > initializing TORQUE (admin: root at pc-xubuntu)
> > Max open servers: 4
> > qmgr obj= svr=default: Bad ACL entry in host list MSG=First bad host: pc-xubuntu
> > Max open servers: 4
> > qmgr obj= svr=default: Bad ACL entry in host list MSG=First bad host: pc-xubuntu
> >
> > Some errors in /var/log/daemon.log
> >
> > Oct 19 00:25:03 localhost PBS_Server: Connection refused (111) in
> > contact_sched, Could not contact Scheduler - port 15004
> > Oct 19 00:25:03 localhost PBS_Server: Bad ACL entry in host list
> > (15073) in manager_oper_chk, bad entry in acl: root at pc-xubuntu
> > Oct 19 00:25:03 localhost PBS_Server: Bad ACL entry in host list
> > (15073) in manager_oper_chk, bad entry in acl: root at pc-xubuntu
> >
> > It seems to me that I have problem with hostname resolution but I am
> > linux beginner and I can not fix it myself. Probably it is problem
> > with forward/reverse name resolution.
> >
> > I hope I do not need DNS server for this. Can you advice me or give me
> > some simple material to set up name resolution properly? Have you any
> > other idea?
>
> use localhost for your server name.  I often test torque on my workstation
>


More information about the torqueusers mailing list